Planning, design, construction and rehabilitation

ОРГАНИЗАТОР/ORGANIZER: Град Бања Лука, Република Српска, Босна и ХерцеговинаCity of Banja Luka, Republic of Srpska, Bosnia and Herzegovina

СУОРГАНИЗАТОРИ/CO-ORGANIZERS:

Завод за изградњу а.д. Бања Лука - ЗИБЛ, Република Српска, Босна и ХерцеговинаInstitute of Construction Banja Luka - ZIBL, Republic of Srpska, Bosnia and Herzegovina

Институт за земљотресно инжењерство и инжењерску сеизмологију - ИЗИИС Скопље - МакедонијаInstitute of Earthquake Engineering and Engineering Seismology - IZIIS Skopje, Macedonia

PUBLISHER: Institute for Construction Banja Luka - ZIBLFOR THE PUBLISHER: Director Čedo Savić, B.LL.

EDITORS: Prof. Mirko Aćić, Ph.D.Prof. Drago Trkulja, Ph.D.

TECHNICAL EDITORS: Aleksandar Cvijanović, Technical Director ZIBLNovak Pupavac, M.Sc.Čedomir Radulović, B.Sc.EE

PREPARATION FOR PRINTING: Marija ĐurićSnježana Lepir, B.Sc.EE Čedomir Radulović, B.Sc.EE

COVER DESIGN: Ljubiša Gornik

PRINTED BY: N.I.G.D. Nezavisne novine d.o.o., Banja Luka

CIRCULATION: 500

Banja Luka, October 2009

Bawa Luka, Republika Srpska, 26-28. Oktobar 2009.

ПЛАНИРАЊЕ, ПРОЈЕКТОВАЊЕ, ИЗГРАДЊА И РЕХАБИЛИТАЦИЈАЗГРАДА И ДРУГИХ ИНЖЕЊЕРСКИХОБЈЕКАТА У СЕИЗМИЧКИ АКТИВНИМ ПОДРУЧЈИМА

PLANNING, DESIGN, CONSTRUCTIONAND REHABILITAION OF BUILDINGSAND OTHER ENGINEERING FACILITIES IN SEISMICALLY ACTIVE AREAS

Zemqotresi 26. i 27. oktobra 1969.

godine na povr{ini od 9.000km2, ostvar-

ili su seizmi~ki intenzitet 70, 8

0i 9

0

skale MCS. U Bawoj Luci i 15 Kraji -

{kih op{tina poginulo je 15, a teè i

lak{e povrije|eno 1117 qudi. Poru{eno

je ili jako o{te}eno 86.000 stanova, 266

{kola i 592 kulturna, zdravstvena, so-

cijalna i privredna objekta. Zemqotres

od 26. oktobra shva}en je i kao mogu}a

najava glavnog, znatno jaèg udara, koji

se narednog dana i dogodio, ali je ve}

ve}ina ìteqa bila pod vedrim nebom, u

parkovima, poqanama, ... To je bila sre}a

u nesre}i, pa je broj poginulih i povrije -

|enih relativno mali u pore |ewu sa

ru{ila;kom snagom katastrofalnog

zemqotresa od 27. oktobra.

U pro{losti je zabiqeèno vi{e

jakih zemqotresa, koji su u Bawalu~kom

podru~ju, izazivali pravu pusto{, ali je

Bawa Luka, ponovo, iz ru{evina i pe-

pela, izrastala u jo{ ve}i i qep{i grad.

Danas Bawa Luka ima preko 250.000

stanovnika, {to je skoro ~etiri puta

vi{e nego u vrijeme zemqotresa od prije

40 godina, a urbani dio grada pro{irio

se za pet puta. Bawa Luka je sjedi{te

Republike Srpske, entiteta Bosne i

Hercegovine.

Uvjereni smo da }e ova Konferen-

cija, koja se odràva povodom 40 godina

od zemqotresa koji je pogodio Bawa

Luku, biti pravo mjesto za sumirawe

znawa i iskustva iz zemqotresnog

inèwerstva i da }e, u tom pogledu,

dati svoj doprinos razvoju ne samo u re-

gionu i podru~ju Balkana, ve} i u Evropi

pa i u svijetu. Imaju}i u vidu zna~ajan

broj prispjelih nau~no-stru~nih radova

me|u ~ijim autorima se nalazi i ve}i

broj, danas u svijetu, veoma poznatih

imena iz zemqotresnog inèwerstva,

Konferencija }e biti vrlo aktuelna za

sve struke u graditeqstvu, a posebno za

inèwere koji se bave istraìvawem,

planirawem, urbanizmom, projekto-

vawem, izvo|ewem, nadzorom i odrà v-

awem gra|evinskih objekata i sistema,

ali }e biti vrlo zna~ajna i za organe

vlasti - donosioce odluka, zasnivane na

smawewu seizmi~kog rizika.

Bawa Luka je poznata po svom gosto -

primstvu i otvorenosti i oduvijek je

bila doma}in mnogim uglednim li~nos-

tima i delegacijama. Na{ grad je pretr-

pio mnoge po{asti i promijenio mnogo

svojih lica, ali bogatstvo koje se mjeri

spomenicima, reprezentativnim arhite k-

tonskim nasqe|em, prirodnim qepotama

i bogatim iskustvom wegovih gra|ana

ostaje da plijeni i do~ekuje goste i

danas.

Na kraju, èlim da Vam svima izra -

zim zahvalnost u ime Grada i li~no,

{to ste, Va{im odzivom, omogu}ili

odràvawe ove Konferencije, a posebno

autorima saop{tewa koja su publiko-

vana u Zborniku radova. Tako|e, izraà -

vam zahvalnost ~lanovima Nau~nog i

Organizacionog komiteta koji su svo-

jim zalagawem u~inili da se ovaj skup

odrì na vrlo zavidnom nivou.

Sa posebnim zadovoqstvom, èlim

Vam dobrodo{licu i prijatan boravak u

Bawoj Luci.

Dragoqub Davidovi}

Gradona~elnik Bawa Luke

Predgovor

The earthquakes that struck Banja Lukaon October 26 and 27 1969, affected thearea of 9 000 km² with seismic intensities of7°, 8° and 9° on the MCS scale, and left 15killed and 1117 severely and slightly injuredin Banja Luka and in fifteen other municipal-ities of the Krajina Region. Eighty-six thou-sand apartments, 266 schools and 592cultural, health, social and public facilitieswere completely destroyed or severely dam-aged by this disaster. The first earthquake ofOctober 26 was interpreted to be a foreshockof the main, considerably stronger event,which in fact occured the following day. Atthe time, most of the inhabitants were al-ready out in the open, which turned out to bemost fortunate as the number of the killedand injured did not increase in proportion tothe destructive force of the second devastat-ing earthquake on October 27th.

Several strong earthquakes were regis-tered in the past and some of them destroyedthe area of Banja Luka. In spite of this, thecity grew from the ruins and ashes to becomeeven bigger and more beatiful. There aremore than 250 000 inhabitants in BanjaLuka today, which is almost four times morethan forty years ago. The urban area of thecity is now five times larger than what it usedto be. Today Banja Luka is the administrativecenter of the Republic of Srpska, which isone of the two entities in Bosnia and Herze-govina.

We beleive that this Conference, which istaking place on the occasion of the fortyethanniversary of the earthquake, is the idealplace for reviewing the knowledge and expe-rience in the field of earthquake engineering,and that this conference will contribute tofurther developments not only in our region

and the Balkans, but also in Europe and inthe world. Judging from the significant num-ber of the scientific and expert papers we re-ceived, from distinguished authors in the fieldof earthquake engineering worldwide, theConference will address the state of the artand will be most informative for all civil en-gineering prifessionals, and in particular forthe engineers dealing with research, urbanplaning, design, supervision and managingthe construction sites and systems. Further-more, the Conference will prove very usefulfor the local city and government officials.

Having welcomed and hosted many dis-tinguished visitors and delegations so far,Banja Luka has always been known for itshospitality and openness. Our city sufferedmany calamities and changed its appearencemany times, but its treasures, which are re-flected in its monuments, architectural her-itage, beautiful natural envoronment, andrich experience of its citizens, will continueto attract and to host its dear guests.

In conclusion, allow me to express mygratitute to you personally and on behalf ofthe City of Banja Luka for your willingnessto participate, and which made this Confer-ence possible. I would also like to thank themembers of the Scientific and OrganizationalCommittees, who helped to make it possiblefor this event to take place and at suh an ad-vanced level.

It is with great pleasure that I extend mycordial welcome to all of you, and wish youthe most enjoyablet stay in Banja Luka.

Dragoljub DavidovićMayor of Banja Luka

Forward

Међународни научни комитет

•Проф. др Мирко АЋИЋ, Београд, Србија •Проф. др Петар АНАГНОСТИ, Београд, Србија •Проф. др Ставрос АНАГНОСТОПУЛОС, Атина, Грчка •Проф. др Атила АНСАЛ , Генерални Секретар ЕАЕЕ, Турска •Проф. др Феликс АПТИКАЈЕВ , Москва, Русија •Проф др Џек Г. БАУКАМП,Дамштат, Њемачка •Проф. др Хонг ЧЕН, Харбин, Кина •Проф. др Радомир ФОЛИЋ, Нови Сад, Србија •Проф. др Михаил ГАРЕВСКИ, Скопље, Македонија •Проф. др Предраг ГАВРИЛОВИЋ, Скопље, Македонија •Др Бранислав ГЛАВАТОВИЋ, Подгорица, Црна Гора •Др Хазим ХОРВАТОВИЋ, Сарајево, БиХ •Проф. др Марин КОСТОВ, Софија, Бугарска •Мр Огњен КУРАЈИЦА, Пасадена, САД •Проф. др Сугито МАСАТА, Јапан •проф. др Зоран МИЛУТИНОВИЋ, Скопље, Македонија •Проф. др Маринко ОЛУЈИЋ, Загреб, Хрватска •Проф. др Божидар С. ПАВИЋЕВИЋ, Подгорица, Црна Гора •Акад. Бошко ПЕТРОВИЋ, Београд, Србија •Проф. др Мирослав СТОЈКОВИЋ, Скопље, Македонија •Мр Славица РАДОВАНОВИЋ, Београд, Србија •Проф. др Душко СУНАРИЋ, Београд, Србија •Проф. др Миха ТОМАЖЕВИЧ, Љубљана, Словенија •Проф. др Михаило ТРИФУНАЦ, Лос Анђелос, САД •Проф. др Драго ТРКУЉА, Бања Лука, БиХ •Проф. др Ђорђе ВУКСАНОВИЋ, Београд, Србија

Организациони комитет

•Драгољуб ДАВИДОВИЋ, Градоначелник Бањалуке - Предсједник •Чедо САВИЋ, директор Завод за изградњу а.д. Бања Лука •Рајко ПУЦАР, Завод за изградњу а.д. Бања Лука •Проф. др Мирко АЋИЋ, Грађевински факултет Београд •Проф. др Михаил ГАРЕВСКИ, директор ИЗИИС-а, Скопље •Проф. др Предраг ГАВРИЛОВИЋ, ИЗИИС, Скопље •Верица КУНИЋ, начелник Одјељења за просторно уређење Града •Будимир БАЛАБАН, начелник Одјељења за комуналне и стамбене послове

и послове саобраћаја Града •Љиљана РАДОВАНОВИЋ, начелник Одјељења за друштвене дјелатности Града •Драшко ИЛИЋ, одборник Скупштине Града

•Слободан БУЋМА, одборник Скупштине Града •Проф.др. Миленко СТАНКОВИЋ, декан Архитектонско-грађвинског факултета

Бања Лука •мр Радован БЕЛЕСЛИН, Архитектонско-грађвински факултет Бања Лука •Проф. др Драго ТРКУЉА, Завод за изградњу а.д. Бања Лука •Проф. др Владимир ЛУКИЋ, ИГ, Бања Лука •мр Борко ЂУРИЋ, предсједник Привредне коморе Републике Српске •мр Новак ПУПАВАЦ, Завод за изградњу а.д. Бања Лука

Почасни комитет

•Акад. Рајко КУЗМАНОВИЋ, предсједник Републике Српске •Игор РАДОЈЧИЋ, предсједник Народне Скупштине Републике Српске •Милорад ДОДИК, предсједник Владе Републике Српске •Фатима ФЕТИБЕГОВИЋ, Министар за просторно уређење

грађевинарство и екологију РС •Проф. др Станко СТАНИЋ, Ректор Универзитета, Бања Лука

International Scientific Comitee

•Prof. dr Mirko AĆIĆ, Belgrade, Serbia •Prof. dr Petar ANAGNOSTI, Belgrade, Serbia •Prof. dr Stavros ANAGNOSTOPULOS, Atena, Greece •Prof. dr Atila ANSAL , Secretary General of EAEE, Turkey •Prof. dr Felix APTIKAEV , Moscow, Russia •Prof dr Jack G. BOUWKAMP, Darmshtad, Germany•Prof. dr Hong CHEN, Harbin, China •Prof. dr Radomir FOLIĆ, Novi Sad, Serbia •Prof. dr Mihail GAREVSKI, Skopje, Macedonia •Prof. dr Predrag GAVRILOVIĆ, Skopje, Macedonia •Dr Branislav GLAVATOVIĆ, Podgorica, Montenegro •Dr Hazim HORVATOVIĆ, Sarajevo, Bosnia and Herzegovina •Prof. dr Marin KOSTOV, Sofia, Bulgaria •Mr Ognjen KURAJICA, Pasadena, USA •Prof. dr Sugito MASATA, Japan •Prof. dr Zoran MILUTINOVIĆ, Skopje, Macedonia •Prof. dr Marinko OLUJIĆ, Zagreb, Croatia •Prof. dr Božidar S. PAVIĆEVIĆ, Podgorica, Montenegro •Akad. Boško PETROVIĆ, Belgrade, Serbia •Prof. dr Miroslav STOJKOVIĆ, Skopje, Macedonia •Mr Slavica RADOVANOVIĆ, Belgrade, Serbia •Prof. dr Duško SUNARIĆ, Belgrade, Serbia •Prof. dr Miha TOMAŽEVIČ, Ljubljana, Slovenia •Prof. dr Mihailo TRIFUNAC, Los Angeles, USA •Prof. dr Drago TRKULJA, Banja Luka, Bosnia and Herzegovina •Prof. dr Đorđe VUKSANOVIĆ, Belgrade, Serbia

Organizing Committee

•Dragoljub DAVIDOVIĆ, Mayor of Banja Luka - Chairman •Čedo SAVIĆ, director, Zavod za izgradnju Banja Luka •Rajko PUCAR, Zavod za izgradnju Banja Luka •Prof. dr Mirko AĆIĆ, Faculty of civil engineering – Belgrade •Prof. dr Mihail GAREVSKI, director, IZIIS, Skopje •Prof. dr Predrag GAVRILOVIĆ, IZIIS, Skopje •Verica KUNIĆ, Section head for spatialy organization of City of Banja Luka•Budimir BALABAN, Section head for municipal and residental dealing of City of Banja Luka•Ljiljana RADOVANOVIĆ, Section head for public activity of City of Banja Luka•Draško ILIĆ, parliamentarian

•Slobodan BUĆMA, parliamentarian •Prof.dr. Milenko STANKOVIĆ, dean of Faculty of civil eng., Banja Luka •mr Radovan BELESLIN, Faculty of civil eng, Banja Luka •Prof. dr Drago TRKULJA, Zavod za izgradnju Banja Luka •Prof. dr Vladimir LUKIĆ, Banja Luka •mr Borko ĐURIĆ, entity president of Chamber of economy •mr Novak PUPAVAC, Zavod za izgradnju Banja Luka

Honorary Committee

•Acad. Rajko KUZMANOVIĆ, President of Republic of Srpska •Igor RADOJČIĆ, President of House of commons RS •Milorad DODIK, Premier of Republic of Srpska •Fatima FETIBEGOVIĆ, Minister of arch. arrangement RS •Prof. dr Stanko STANIĆ, Rector of University, Banja Luka

SADRŽAJ

Prof.dr Miha TomaževičASEIZMIČKO PROJEKTOVANJE ZIDANIH ZGRADA: 40 GODINA POSLE ZEMLJOTRESA U BANJA LUCIEARTHQUAKE RESISTANT DESIGN OF MASONRY BUILDINGS: 40 YEARS AFTER THE BANJA LUKA EARTHQUAKE ................................................21

Prof.dr Paata RekvavaUTICAJ INTERAKCIJE TLO-INTERFEJS-KONSTRUKCIJA NA SEIZMIČKI ODGOVOR PANEL ZGRADASOIL – INTERFACE – STRUCTURE INTERACTION EFFECT ON PANEL BUILDING SEISMIC RESPONS.....................................................................................35

Prof.dr Zoran MilutinovićABU DABI, UAE, SISTEM ZA MONITORING I MENADŽMENT SEIZMIČKOG RIZIKAEMIRATE OF ABU DHABI, UAE, SYSTEM FOR SEISMIC RISK MONITORING AND MANAGEMENT ..........................................................................45

Prof.dr Zaven KhlghatyanZAŠTITA OBJEKATA OD SEIZMIČKIH EFEKATA RAZVIJANJEM “SPREGNUTIH SISTEMA”, ISTRAŽIVANJA I REALIZACIJASEISMIC PROTECTION OF STRUCTURES BY “COUPLED SYSTEMS”- DEVELOPMENT, RESEARCH AND REALIZATION .................................................................................63Prof.dr A.H. BarbatOCENA SEIZMIČKE SIGURNOSTI ARMIRANOBETONSKIH ZGRADA PROJEKTOVANIH NA OSNOVU EVROKODOVA 2 I 8EVALUATION OF THE SEISMIC SAFETY OF RC BUILDINGS DESIGNED BY USING EUROCODES 2 AND 8 ................................................................................75

Prof. dr Mirko Aćić Prof. dr Goran ĆirovićULOGA I ZNAČAJ GRADITELJSTVA U SMANJENJU SEIZMIČKOG RIZIKATHE ROLE AND THE SIGNIFICANCE OF CONSTRUCTION IN REDUCTION OF SEISMIC RISK ..............................................................................................................89

Akademik Mandžić Enver Mr sc. Kulukčija Salko Dr sc. Mandžić Kenan Humo Mustafa OCJENA STANJA POSTOJEĆIH OBJEKATA KULTURNOG I ISTORIJSKOG NASLIJEĐA PRIMJENOM REFRAKCIONE SEIZMIKE .......................................105

Prof. dr Radomir Folićmr Lidija BabićPARAMETARSKA SEIZMIČKA ANALIZA AB DIMNJAKA

PARAMETRIC SEISMIC ANALYSIS OF RC CHIMNEYS ......................................115

Prof. dr Đorđe Lađinović Prof. dr Radomir Folić mr Mladen ĆosićUPOREDNA ANALIZA SEIZMIČKIH ZAHTEVA REGULARNIH

BETONSKIH VIŠESPRATNIH OKVIRA

COMPARATIVE ANALYSIS OF SEISMIC DEMANDS OF REGULAR MULTI–STORY CONCRETE FRAMES .................................................129

mr Goran Simonović Prof.dr Branislav VerbičPRAĆENJE STANJA KONSTRUKCIJE I ISTRAŽIVANJE SEIZMIČKE OTPORNOSTI ZIDANIH ZGRADA RESEARCH IN SEISMIC RESIDENCE OF

THE MASONRY BUILDING ...........................................................................................145

Prof. dr Miodrag ManićINTERAKCIJA TLA I ZGRADE BK-2 U NASELJU ''BORIK'' U BANJA LUCISOIL STRUCTURE INTERACTION OF BK-2 BUILDING IN BANJA LUKA...........155

Prof. dr D.Šumarac Z.PetraškovićKONTROLA OŠTEĆENJA I POPRAVKA RADI BEZBEDNOSTI ZGRADA

DAMAGE CONTROL AND REPAIR FOR SECURITY OF BUILDINGS...............165

Zoran PetraškovićHISTEREZISNO PONAŠANJE KONSTRUKCIJE ČELIČNIH DAMPERA U POLJU ZAMORA ZEMLJOTRESNIM OPTEREĆENJEM VRLO MALOG BROJA CIKLUSA HYSTERESIS BEHAVIOR OF

STEEL DAMPERS CONSTRUCTION IN FATIGUE ZONE UNDER

EARTHQUAKE LOADING

OF EXTREMLY LOW NUMBER OF CYCLES ..............................................................183

Prof. dr Milenko Stanković Srđan StankovićŽIVOT BEZ STRAHA OD ZEMLJOTRESA-POTREBA-VIZIJA-IMPERATIVLIFE WITHOUT FEAR OF EARTHQUAKE-NEEDS-A VISION-AN IMPERATIVE...............................................................................................197

Prof. dr Lidija Krstevska mr Ljubiša Živković ISPITIVANJE OBJEKTA "NOVA BANKA" U BANJA LUCI METODOM AMBIJENTALNIH VIBRACIJAIN SITU TESTING OF "NOVA BANKA" IN BANJA LUKA BY AMBIENT VIBRATION MEASUREMENTS .................................................................207

Prof. dr Dragan MilašinovićAleksandar Borković STOHASTIČKE VIBRACIJE SAVIJAJUĆIH PLOČA PRIMJENOMMETODA KONAČNIH TRAKASTOHASTIC VIBRATIONS OF PLATE IN BENDING USING THE FINITE STRIP METHOD.........................................................................................215

Prof. dr Mihail Garevski Prof. dr Veronika Šendova mr Blagojče Stojanoski REKONSTRUKCIJA PRAVOSLAVNOG SABORNOG HRAMA "SV. BOGORODICA" U SKOPLJURECONSTRUCTION OF THE ORTHODOX CATHEDRAL CHURCH OF THE VIRGIN MARY IN SKOPJE........................................................225

Prof. dr Veronika Šendova Prof. dr Predrag Gavrilović mr Blagojče Stojanoski Goran Jekić INTEGRIRANI PRISTUP SANACIJE I SEIZMIČKOG OJAČANJA MUSTAFA PAŠINE DŽAMIJE U SKOPLJUINTEGRATED APPROACH TO REPAIR AND SEISMIC STRENGTHENING OF MUSTAFAPASHA MOSQUE IN SKOPJE .....................................................................................233

Prof. dr Veronika Šendova dr Zoran Rakićević Prof. dr Predrag Gavrilović Prof. dr Dimitar Jurukovski SEIZMIČKA ZAŠTITA VIZANTIJSKIH CRKVI PRIMENOM SISTEMA ZA PASIVNU KONTROLURETROFITTING OF BYZANTINE CHURCH USING PASSIVE BASE CONTROL SYSTEM .........................................................................245

Prof. dr Roberta Apostolska Prof. dr Golubka Necevska-Cvetanovska dr Zdravko Bonevdr Elena Vasseva dr Dylian Blagov Julijana CvetanovskaMETODA SPEKTRA KAPACITETA ZA OCENU SEIZMIČKOG PONAŠANJA AB ZGRADA NA DEFORMABILNOM TLUCAPACITY SPECTRUM METHOD FOR SEISMIC PERFORMANCE OF RC BUILDINGS INCLUDING SOIL FLEXIBILITY ..........................................257

Prof. dr Golubka Necevska-Cvetanovska Prof. dr Roberta Apostolska mr Natasa Mirčić Julijana CvetanovskaKONSTRUKTIVNE MERE ZA POBOLJŠANJE SEIZMIČKOG PONAŠANJA BEZGREDNIH KONSTRUKTIVNIH SISTEMASTRUCTURAL MEASURES FOR IMPROVING SEISMIC PERFORMANCE OF FLAT-SLAB BUILDING STRUCTURAL SYSTEMS...............................................265

dr Milan Trifković Žarko Nestorović MOGUĆNOSTI KORIŠĆENJA 2D GEODETSKIH MREŽA ZA UTVRĐIVANJE SEIZMIČKIH POJAVAPOSSIBILITIES FOR 2D GEODETIC NETWORKS UTILIZATION FOR SEISMIC EVENTS DETERMINATION..................................................................273

Prof. dr Božidar Pavićević PROSTORNO-URBANISTIČKO PLANIRANJE KAO KLJUČNI ASPEKT INTEGRALNOG UPRAVLJANJA SEIZMIČKIM RIZIKOMLAND USE PLANNING AS THE KEY ASPECT OF INTEGRATED SEISMIC RISK MANAGAMENT....................................................................................281

Prof. dr Božidar Pavićević Jadranka Mihaljević UPRAVLJANJA ZEMLJOTRESNIM RIZIKOM U SAVREMENIM USLOVIMAGENERAL CONCEPT OF INTEGRATED SEISMIC RISK MANAGAMENT .............295

Prof. dr Violeta Mirčevska Prof. dr Vladimir BičkovskiBEM – REŠENJE HIDRODINAMIČKOG PRITISKAUSE OF BEM IN SOLVING FLUID – STRUCTURE INTERACTION....................311

Prof. dr Zoran Rakičević Aleksandra Bogdanović Prof. dr Dimitar JurukovskiASEIZMIČKO PROJEKTOVANJE ČELIČNIH RAMOVSKIH KONSTRUKCIJA SA DODATNIM PRIGUŠENJEMASEISMIC DESIGN OF STEEL FRAME STRUCTURES WITH ADDED DAMPING .........................................................................................................317

Prof. dr Viktor Hristovski mr Marta Stojmanovska Prof. dr Mihail Garevski 1 DEO: PREDLOŽENE KONSTITUTIVNE RELACIJE ZA VEZE KOD LAMELIRANIH DRVENIH PANELA (XLAM)PART 1: PROPOSED CONSTITUTIVE RELATIONSHIPS FOR CONNECTIONS OF XLAM PANELS ................................................................329

Prof. drViktor Hristovski mr Marta Stojmanovska Prof. dr Mihail Garevski2 DEO: PREDLOŽENE KONSTITUTIVNE RELACIJE ZA VEZE KOD LAMELIRANIH DRVENIH PANELA-MKE PART 2: PROPOSED CONSTITUTIVE RELATIONSHIPS FOR CONNECTIONS OF XLAM PANELS-FE IMPLEMENTATION ............................................................339

mr Radmila Šalić Prof. dr Mihail Garevski Prof. dr Zoran MilutinovićODGOVOR KONSTRUKCIJA IZOLIRANIH OLOVNO-GUMENIM LEŽIŠTIMARESPONSE OF LEAD-RUBBER BEARING ISOLATED STRUCTURE ................349

Prof.dr Ljubomir Taškov Prof. dr Lidija KrstevskaSEIZMIČKA BAZNA IZOLACIJA REZERVOARA I ZGRADA SA PRIMENOM SISTEMA ALSCSEISMIC BASE ISOLATION ON RESERVOIRS AND BUILDINGS BY APPLICATION OF THE ALSC SYSTEM ......................................357

mr Novak Pupavac KALIBRACIJA FUNKCIJE POVREDLJIVOSTI UZ POMOĆ STANDARDACODE BASED CALIBRATION OF VULNERABILITY MODELS...............................365

Prof. dr Violeta Mirčevska Prof. dr Vladimir Bičkovski Prof. dr Mihail GarevskiBENCHMARK TEST softvera PROC3DN-IZIISA BENCHMARK TEST OF THE SOFTWARE PROC3DN-IZIIS ............................379

Prof. dr Violeta Mirčevska Prof. dr Vladimir Bičkovski Prof. dr Mihail GarevskiBENCHMARK TEST softwera FILT3D-IZIISA BENCHMARK TEST OF THE SOFTWARE FILT3D-IZIIS..................................389

Prof.dr Dragan Lukićmr Elefterija ZlatanovićМЕРОДАВНИ ЧИНИОЦИ СЕИЗМИЧКЕ АНАЛИЗЕ ДВА ПАРАЛЕЛНА БЛИСКА ТУНЕЛСКА ОБЈЕКТАRULING PARAMETERS FOR SEISMIC ANALYSIS OF TWIN-TUNNELS.......................................................................................................395

Prof. dr Ratko Salatić DISIPACIJA ENERGIJE U POLUKRUTIM VEZAMAENERGY DISSIPATION IN SEMI-RIGID CONNECTIONS ....................................407

dr Renato Vidrih Matjaž Godec Peter Sinčič OSMATRANJE SEIZMIČNOSTI NA PODRUČJU VELIKIH BRANA U SLOVENIJISEISMOLOGICAL MONITORING OF LARGE DAMS IN SLOVENIA ......................417

Prof. dr Radenko Pejović mr Radivoje Mrdak mr Jelena Pejović mr Nina SerdarSEIZMIČKI ODGOVOR VISOKE LUČNE BRANE “MRATINJE” SEISMIC ANALYSIS OF HIGH ARC DAM “MRATINJE”............................................427

21

Miha Tomaževi 1

ASEIZMI KO PROJEKTOVANJE ZIDANIH ZGRADA: 40 GODINA POSLE ZEMLJOTRESA U BANJA LUCI

Rezime:

Razmatrani su neki aspekti aseizmi kog projektovanja zidanih zgrada, koji su posledica novih tehnologija gra enja i traženja savremenih propisa. Tako su sa rezultatima eksperimentalnih ispitivanja upore eni rezultati ra una otpornosti zidova na smicanje, dobijeni na osnovu razli itih modela rušnih mehanizama. Analiza pokazuje, da za sada još nema jedinstvenog ra unskog modela, koji bi dao prihvatljive rezultate za razli ite uslove optere enja. Na osnovu ispitivanja na seizmi koj platformi i uzimaju i u obzir kapacitet duktiliteta i kriterijum za ograni enje ošte enja, izra ena je ocena faktora ponašanja konstrukcije.

Klju ne re i: zidarija, otpornost na smicanje, faktor ponašanja

EARTHQUAKE RESISTANT DESIGN OF MASONRY BUILDINGS: 40 YEARS AFTER THE BANJA LUKA EARTHQUAKE Summary:

Some issues of earthquake resistant design of masonry buildings, resulting from recent technologies of masonry construction and contemporary code requirements, are discussed. In particular, the results of calculations based on different shear failure mechanism models are compared with experimental results. The analysis has indicated that numerical models, which yield reliable results in different loading conditions, need yet to be developed. On the basis of the shaking table test results and taking into consideration damage limitation requirements and ductility capacity, the values of elastic force reduction factors have been assessed.

Key words: masonry, shear resistance, behavior factor

1 Professor, Ph.D., Civil Engineer, Slovenian National Building and Civil Engineering Institute, Ljubljana, Slovenia

22

1 INTRODUCTIONIn 1969, when the earthquake struck the city of Banja Luka, the first Yugoslav

seismic code of 1964 has been enforced for only a few years. Consequently, not many buildings subjected to earthquake, especially masonry buildings, have been designed by the code. Nevertheless, many buildings survived the earthquake with repairable damage or even undamaged, hence giving us opportunity to learn important lessons regarding the seismic resistant design. For example, it is not easy to find a better example of influence of quality of construction on the seismic resistance of masonry buildings than two houses, located in the vicinity of Banja Luka and shown in Figure 1. The house, built with hollow clay units in good quality lime-cement mortar, survived the earthquake without damage (Fig. 1a), whereas the neighboring house of the same structural configuration, built with solid clay bricks but in poor quality mortar without cement, was severely damaged (Fig. 1b). Since the houses have been built at about only 5 m distance between them, the difference in the observed behavior cannot be attributed to site effects of the earthquake.

a.) b.)

Figure 1 Banja Luka, 1969: the house on the right, built with clay bricks in poor quality mortar, was severely damaged, whereas the house on the left, built with hollow clay units in good quality lime-cement mortar, survived the earthquake without damage (photo by

S.Ter elj)

Many additional lessons of similar character have been learned since the earthquake of Banja Luka in 1969. In the last few decades, considerable research in the behavior of masonry walls and buildings subjected to seismic actions has been carried out in many countries. The behavior of masonry buildings during earthquakes has been analyzed, and experiments to determine the basic parameters of the seismic resistance of masonry walls and buildings have been carried out.

Based on experimental research, new data on the strength and stiffness degradation and deterioration, ductility and energy dissipation capacity of different types of masonry have been obtained. The results of investigations made possible many improvements in masonry construction, and provided basis for the development of analytical models and mathematical tools for earthquake resistance verification and design of masonry structures for seismic loads.

The investigations have also resulted into the development of new codes for masonry construction and design. In Eurocode 6, European standard for the design of masonry structures [CEN, 2005], a clear concept for limit states verification has been

23

introduced also for masonry structures. In Eurocode 8, standard for the design of structures for earthquake resistance [CEN, 2004], additional requirements are specified to be taken into consideration in seismic regions. However, the validity of some important design equations, required by the code, needs yet to be verified, and the values of some important design parameters, proposed in the code, need to be confirmed.

In this contribution, some issues of earthquake resistant design of masonry buildings, resulting from contemporary code requirements, will be discussed. The results of experimental and analytical research, recently carried out at Slovenian National Building and Civil Engineering Institute in Ljubljana, will support the discussion.

2 SHEAR RESISTANCE OF MASONRY WALLS: MODELS AND CALCULATION

Shear failure, characterized by the occurrence of diagonally oriented cracks, is a typical failure mode of unreinforced and confined masonry walls subjected to in-plane seismic loads. Although other mechanisms are also possible, seismic resistance of a regular masonry structure depends predominantly on the shear resistance of structural walls. Therefore, the parameters which define the behavior of masonry walls subjected to shear and equations for the calculation of the shear resistance of the walls are of relevant importance for the seismic resistance verification of masonry buildings in seismic-prone areas.

Figure 2 Shear failure mechanisms: a) shear sliding on the bed-joints, b) shear failure characterized by the formation of diagonal cracks [adapted from Tomaževi , 1999]

If the vertical compressive stresses in the wall are low and the quality of mortar is poor, seismic forces may cause sliding of a part of the wall along one of the bed-joints (Fig. 2a). Sliding shear failure of unreinforced walls usually takes place in the upper parts of masonry buildings below rigid roof structures, where the compressive stresses are low and the response accelerations are high. However, this phenomenon is seldom observed in the buildings’ bottom parts where, typically, diagonally oriented cracks develop in the walls when subjected to seismic loads (Fig.2b). Because of the orientation of cracks, the failure of the wall in such a case is also called diagonal tension shear failure. Depending on the quality of masonry units and mortar, diagonally oriented cracks may either follow the bed- and head-joints or pass through the units or partly follow the joints and partly pass through the units.

24

Various methods and equations have been already proposed for the assessment of the shear resistance of unreinforced masonry walls, characterized by diagonal cracking. In all cases, the sectional stresses and forces are used and the gross dimensions of masonry walls are taken into consideration in order to simplify the analysis. Turnšek and a ovi[1970] introduced the hypothesis that the tensile strength of masonry, conventionally defined as the principal tensile stress developed at the attained maximum resistance of a masonry wall, assuming that the wall is elastic, homogeneous and isotropic panel, determines the shear resistance of the wall. Following this idea, the situations where the diagonal cracks pass either the mortar joints or masonry units, or both, are covered by the same equation. According to Turnšek and a ovi [1970], shear resistance of an unreinforced masonry wall, Rw,ft, is calculated by:

1t

otwftw, fb

fAR , 1)

where Aw = the area of the horizontal cross-section of the wall, o = the average compressive stress due to vertical load, and ft = the tensile strength of masonry, conventionally defined as the principal tensile stress developed at the attained maximum resistance of a masonry wall, assuming that the wall is elastic, homogeneous and isotropic panel:

2)(

2o2

max

2o

tt bf . (2)

where max = the average shear stress in the horizontal section of the wall at the attained maximum resistance. This approach has been implemented into the Yugoslav seismic code of 1981. The method to determine the tensile strength, ft, as defined by Turnšek and

a ovi , is not standardized. Three different testing methods are used to determine the parameter: cyclic lateral resistance tests of symmetrically fixed or cantilever walls at constant vertical load, simple racking tests or diagonal compression test of walls. It has been already shown that comparable results can be obtained by cyclic lateral resistance tests, simple racking tests as well as diagonal compression tests of masonry walls [Bernardini et al., 1981]. The specimens with geometry aspect ratio (height/length) h/l = 1.5 are usually tested.

According to Mann and Müller [1982], however, the shear resistance is calculated depending on the path of the diagonally oriented cracks. In the case where the cracks pass through the vertical and bed joints (step-formed cracks - friction failure of the bed joints), the resistance is defined by the friction law introducing cohesion and friction coefficients as the critical parameters.

In the case of the friction failure of the bed joints, the shear resistance of the wall is calculated by:

Rw,fr = Aw = (k’ + ’ o) Aw, (3) where Aw = tl = the area of the horizontal cross section of the wall and = k’ + ’ o = the shear strength of masonry. Reduced cohesion is given in the form of:

yx

k'k21

1 , (4a)

25

and the reduced coefficient of friction in the form of:

yx' 21

1 , (4b)

where x = lb and y = hb, i.e. the length and height of the unit, respectively. In the case where the cracks pass through the units, however, the tensile strength

of the unit is critical and the equation for the calculation of the shear resistance of the wall is similar to the one proposed by Turnšek and a ovi :

zst

ozstww,cr 1

32.AR , (5)

where zst = the tensile strength of masonry units, determined by the diagonal compression test of units.

Although substantial amount of experimental and analytical research to study the behavior of masonry walls subjected to shear has been carried out, the recent European standard for the design of masonry structures, Eurocode 6 [CEN, 2005], requires that only sliding shear failure mechanism (similar as the friction failure of the bed joints according to Mann and Müller), with the initial shear strength instead of cohesion and prescribed value of the friction coefficient as the governing parameters, be used for the assessment of the shear resistance of unreinforced and confined masonry walls.

According to Eurocode 6, the design shear resistance of a masonry wall, Rdw,EC6, is calculated by assuming that pure sliding shear mechanism determines the shear resistance of a wall:

cM

vkEC6dw, t lfR , (6)

where: fvk = fvko + 0.4 d, (7)

ell2

3c , (8)

and fvk = the characteristic shear strength of masonry, M = partial safety factor for masonry, t = the thickness of the wall, lc = the length of the compressed part of the wall, d = the average vertical stress over the compressed part of the wall that is providing shear resistance in design situation, fvko = the characteristic initial shear strength of masonry at zero compression, e = Hh/V is the eccentricity of the vertical load, h = the height of the wall. The expression for lc should be considered in the case where the eccentricity of axial load, e, exceeds 1/6 of the wall’s length.

The results of two series of tests of clay hollow unit masonry walls, tested as vertical cantilevers by subjecting them to constant vertical load and imposed cyclic lateral in-plane displacements, have been used to analyze the validity of the before mentioned commonly used calculation methods [Tomaževi and Gams, 2009]. Within the first series of tests (walls type B), 20 walls with height/length ratio h/l = 1.5 have been tested. The specimens have been constructed with 5 different types of clay hollow units and tested at vertical precompression equal to 20 %, 30 %, and 40 % of the compressive strength of

26

masonry. In the second series (walls type A), however, 9 equal walls with height/length ratio h/l = 0.7, have been tested at vertical precompression equal to 5 %, 10 %, and 15 % of compressive strength of masonry.

In both series of tests, the final mode of failure was of typical shear type with diagonally oriented cracks formed in the walls. In the case of walls type B with geometry aspect ratio h/l = 1.5, which have been tested at higher levels of precompression (20 40 % of the compressive strength of masonry, f), cracking and crushing of units because of insufficient robustness has been also observed. In the case of walls type B, the brittleness of clay hollow masonry units determined the behaviour and failure mechanism. In the case of the long walls type A with geometry aspect ratio h/l = 0.7, the influence of precompression on failure mechanism can be clearly seen. Whereas diagonal orientation of cracks can be clearly identified in all cases, the mechanism depended on the level of precompression. At lowest level of precompression, step-formed cracks developed which passed through the bed- and head-joints, whereas diagonally oriented cracks passed through the units at high precompression. Also, at lowest precompression level, the units rotated at increased amplitudes of imposed lateral displacements, so that distinct gaps formed at the interface between the mortar and units in the joints. At higher precompression, however, the rotation was prevented and the units started cracking and crushing. The difference in the behavior can be clearly seen in Figures 3 and 4.

a) b)

Figure 3 Typical damage pattern observed during testing walls type B at o = 0.37f (a) and o = 0.2f (b)

27

a) b) Figure 4 Typical damage patterns observed during testing walls type A at

o = 0.05f (a) and o = 0.15f (b)

The experiments have been used to analyze the reliability of models and respective equations for the calculation of the shear resistance of walls. The calculated and experimentally obtained results are compared in Tables 1 and 2. It should be mentioned that the mid-height section of the walls has been considered as the resisting section of the walls in the case where the Eurocode 6 methodology has been used for the evaluation of the shear resistance (Rw,EC6).

Table 1 Comparison of experimentally obtained and calculated values of the shear resistance of the tested walls: shear friction failure

Wall Hmax,exp(kN) o/f

Rw,EC6(kN)

expmax,

EC6w,

HR Rw,fr

(kN)expmax,

frw,

HR

B1/1 141 0.40 302 2.14 184 1.30 B1/2 92 0.20 155 1.68 106 1.15 B2/1 134 0.35 256 1.91 172 1.29 B2/2 91 0.20 155 1.70 109 1.20 B2/3 118 0.28 209 1.78 143 1.21 B3/1 129 0.37 256 1.98 170 1.32 B3/2 84 0.20 146 1.73 99 1.18 B4/1 142 0.34 248 1.75 223 1.57 B4/2 94 0.21 155 1.65 147 1.56 B6/1 131 0.36 293 2.24 267 2.04 B6/2 92 0.18 175 1.90 166 1.81 A/1 303 0.15 406 1.34 254 0.84 A/2 221 0.10 310 1.40 190 0.86 A/3 130 0.06 220 1.69 131 1.01

28

Table 2 Comparison of experimentally obtained and calculated values of the shear resistance of the tested walls: diagonal tension failure

Wall Hmax,exp(kN) o/f

Rw,cr(kN)

expmax,

crw,

HR Rw,ft

(kN)expmax,

ftw,

HR

B1/1 141 0.40 148 1.05 137 0.97 B1/2 92 0.20 116 1.26 102 1.11 B2/1 134 0.35 187 1.40 133 0.99 B2/2 91 0.20 157 1.72 104 1.14 B2/3 118 0.28 174 1.47 121 1.02 B3/1 129 0.37 155 1.21 131 1.02 B3/2 84 0.20 124 1.48 99 1.18 B4/1 142 0.34 134 0.95 139 0.98 B4/2 94 0.21 110 1.17 110 1.17 B6/1 131 0.36 265 2.02 130 0.99 B6/2 92 0.18 226 2.46 99 1.07 A/1 303 0.15 304 1.00 303 1.00 A/2 221 0.10 272 1.23 259 1.17 A/3 130 0.06 238 1.83 209 1.60

On the basis of the observed behavior and development of cracks in the tested walls, the shear friction failure mechanism models have not been expected to yield good results. However, great difference between the measured and calculated values, as indicated in Table 1, is not acceptable. In the particular case studied, Mann-Müller proposal yielded better results than Eurocode 6 proposed method. In the case of the Eurocode 6 calculations, the resistance values significantly overestimated the measured ones, although the initial shear strength values of masonry have been obtained by standardized testing and the compressed part of the walls’ length at the mid-height level has been taken into consideration when assessing the resistance. In the case of the Mann-Müller proposal, where the whole length of the walls has been considered as resisting to shear and the experimentally obtained initial shear strength values have been reduced as recommended by the authors of the proposal (reduced cohesion), the correlation is improved, especially in the case of the long walls. Although the shear friction mechanism has not been observed during the tests, good correlation between the measured and calculated shear resistance values has been obtained in the case of the long walls with geometry aspect ratio h/l = 0.7, subjected to low level of precompression.

It has been found that the existing methods of calculation of the shear resistance of masonry walls do not have general validity. The idea proposed by Turnšek and a ovi ,that the diagonal tension failure of masonry, which is the result of the principal tensile stresses occurring in the walls under the combination of vertical and lateral loads, determines the shear resistance capacity, yielded good correlation between the experimentally obtained and calculated resistance values in the case of the walls where the working compressive stresses exceeded 10 % of masonry’s compressive strength. In the case of the brick cracking failure mechanism model, proposed by Mann and Müller, the correlation with the experimental results was not consistent, although the tensile strength of units has been determined by testing.

29

3 DISPLACEMENT CAPACITY AND BEHAVIOUR FACTOR qAccording to European standard for earthquake resistant design of structures,

Eurocode 8 [CEN, 2004], the structure should be designed to withstand the earthquake with return period 475 years and 10 % probability of exceedance in 50 years, “without local or global collapse, thus retaining its structural integrity and a residual load bearing capacity after the seismic events” (no collapse requirement). However, the structure shall be also designed to withstand an earthquake having a larger probability of occurrence than the design earthquake, i.e. earthquake with return period 95 years with 10 % probability of exceedance in 10 years, “without the occurrence of damage and limitation of use, the costs of which would be disproportionately high in comparison with the costs of the structure itself” (damage limitation requirement). When verifying the seismic resistance, it has to be verified that for all structural members as well as for the structure as a whole, the design resistance capacity Rd, calculated by taking into account the characteristic strength values and partial safety factors M of members' materials, is greater than the design value of combined action effect Ed, which includes seismic actions.

The form of seismic action to be used in seismic resistance verification depends on the importance and complexity of the structure under consideration. In the case of structures with regular structural configuration, such as masonry structures, the calculations are simplified by taking into account only one horizontal component of the seismic ground motion and analyzing the structure in each orthogonal direction separately. Non-linear dynamic response analysis is replaced by equivalent elastic static analysis, where the design seismic loads are evaluated on the basis of the design response spectra, considering the structure as an equivalent single-degree-of-freedom system. To obtain the design spectra, the ordinates of the elastic response spectra are reduced by a factor, which takes into account the displacement and energy dissipation capacity of the structure under consideration. This factor is generally called “force reduction factor” or, by terminology used in European standard for earthquake resistant design, “structural behavior factor q”.

According to Eurocode 8, “the behavior factor q is an approximation of the ratio of the seismic forces that the structure would experience if its response was completely elastic with 5 % viscous damping to the minimum seismic forces that may be used in the design - with a conventional elastic analysis model - still ensuring a satisfactory response of the structure”. The definition can be expressed by:

q = Se/Su,d, (9) where Se = seismic force developed in a completely elastic structure and Su,d = the

design seismic load. The structural behavior factor q can be also expressed in terms of the global ductility factor of the structure under consideration, u = du/de, where de = the displacement of the structure at the idealized elastic limit and du = the displacement at ultimate limit, as follows:

q = (2 u 1)1/2. (10) Whereas the basic definition expresses the behavior factor q in terms of forces

(Eq. 9), Eq. 10 determines the minimum ductility and energy dissipation requirements, i.e. displacement capacity requirement, which should be fulfilled if the behavior factor q is used for seismic resistance verification:

q(Eq. 9) q(Eq.10). (11)

30

This makes possible the use of the idea of reduction of seismic forces also in the case where the resistance curve of the structure, calculated by pushover methods, is used for seismic resistance verification. However, in such a case not only the resistance, but also the displacement capacity of the structure should be verified. In other words, if the seismic resistance of a masonry structure is verified for the design seismic loads, calculated by taking into account structural behavior factor q, its global ductility should not be less than:

u,min = (q2+1)/2. (12) The following ranges of values of structural behavior factor q are proposed in Eurocode

8 for different masonry construction systems: For unreinforced masonry: q = 1.5 2.5,For confined masonry: q = 2.0 3.0,For reinforced masonry: q = 2.5 3.0.

To enhance the existing information regarding the possible ranges of values of structural behavior factor q, the seismic behavior of typical Central European masonry buildings with different structural configurations and qualities of masonry materials has been investigated also at Slovenian National Building and Civil Engineering Institute [Tomaževi and Weiss, 2009]. Six models, built at 1:5 scale and representing buildings of two different structural configurations and constructed with two different types of masonry materials have been tested on a simple uni-directional seismic simulator. Models type M1 (Figure 5a) represented a two-story terraced house with the main structural walls orthogonal to seismic motion, whereas models type M2 (Figure 5b) represented a three-story apartment house with uniformly distributed structural walls in both directions. Four models of the first and two models of the second type have been tested. In the case of the terraced house, two models have been built as either partly or completely confined masonry structures.

(a) (b) Figure 5 Typical confined terraced house model type M1 (a) and (b)

apartment house model type M2 at ultimate state before collapse

Story mechanism governed the behavior of the tested models. Therefore, the resistance curve of the model under consideration has been evaluated as a relationship between the maximum resisted base shear and corresponding first story drift (relative story displacement), developed in the structure at the same instant of time. In Figure 6, the obtained relationships are expressed in a non-dimensional form of the base shear

31

coefficient, BSC, and the first story drift angle, , respectively. Base shear coefficient is the ratio between the base shear BS and the weight of the model above the foundation W: BSC= BS/W, whereas the first story drift angle, , is the ratio between the first story drift (relative story displacement), d, and story height, h: (in %) = (d/h) 100.

As can be seen in Figure 6, the values of story drift where the first cracks occurred in the structural walls and the stiffness of models significantly changed (crack limit), and the values of story drift at the attained maximum resistance of the models do not differ much from case to case. An attempt has been therefore made to correlate also the observed damage, displacement capacity and limit states. By analyzing these and other experimental results, trends can be seen and ranges of possible story drift values at the attainment of these two characteristic limit states can be evaluated. The following ranges of values of story rotation can be attributed to characteristic limit states [Tomaževi , 2007]:

Crack limit: cr = 0.2 0.4 %; Maximum resistance: Rmax = 0.3 0.6 %, and Limit state of collapse: u = 2.0 4.0 %.

Story drift at the point where the resistance of the structure degrades to 80 % of the maximum, is usually defined as the ultimate. In other words, story drift at 20 % of strength degradation is considered as the maximum value which can be taken into consideration for the evaluation of the idealized design ultimate global ductility factor of the structure: u,d = 0,8Rmax/ e,id. It is assumed that a ductile structure, although severely damaged, will resist such a displacement without risking collapse (no collapse requirement).

Figure 6 Seismic resistance curves in the form of the base shear coefficient-first story drift angle relationships for terraced-house models M1 (a) and (b) apartment house models M2

However, as the analysis of the observed behavior of tested models indicated, the damage to structural walls exceeds the acceptable limit at this point. The analysis further indicated that such damage generally occurs at story drifts, equal to approximately 3-times story drift at the occurrence of the first cracks in the walls, or 3-times of story drift at the idealized elastic limit. Taking this into consideration, the design ultimate state may be defined by either the story drift where the resistance degrades to 80 % of the maximum, or the story drift equal 3-times the value of story drift at the crack limit, whichever is less:

d,u = min { 0,8Rmax; 3 cr}. (13)

32

The values of behavior factor q, evaluated on the basis of the global ductility of the tested models (Eq. 10), are given in Table 3.

Table 3 Values of structural behavior factor q(Eq. 4), evaluated on the basis of available ductility ( u = 0.8Rmax/ e,id) and damage limitation requirements ( u = 3 cr/ e,id)

q(Eq. 10) = (2 u 1)1/2

u = 0.8Rmax/ ei,d u = 3 cr / ei,dModel ei,d(in %)

0.8Rmax(in %)

cr(in %)

3 cr(in %)

u q u q M1-1 0.24 0.89 0.26 0.78 3.70 2.53 3.25 2.35 M1-2 0.05 1.20 0.60 0.16 24.00 6.85 3.20 2.32 M1-1c 0.17 2.60 0.28 0.84 15.29 5.44 4.94 2.98 M1-1d 0.17 1.81 0.27 0.81 10.65 4.50 4.76 2.92 M2-1 0.07 0.42 0.20 0.60 6.00 3.32 8.57 4.02 M2-2 0.16 1.65 0.33 0.99 10.31 4.43 6.18 3.37

The analysis of experimental results has shown, that the values at the upper limit of the Eurocode 8 proposed range of values of structural behaviour factor q for unreinforced and confined masonry construction systems, i.e. q = 2.5 in the case of the regular unreinforced, and q = 3.0 in the case of the regular confined masonry structures are adequate, if pushover methods are used for seismic resistance verification. However, in this case, the calculated displacement capacity of the structure should be verified and compared with displacement demand.

In the case where elastic analysis methods are used and significant overstrength is expected, the proposed values are conservative. However, additional research and parametric studies are needed to further support and propose the modifications.

4 CONCLUSIONS Some issues of earthquake resistant design of masonry buildings, resulting from

recent technologies of masonry construction and contemporary code requirements, are discussed. In particular, the results of calculations based on different shear failure mechanism models have been compared with experimental results. It has been found that the existing methods of calculation of the shear resistance of masonry walls do not have general validity. The analysis has indicated that numerical models, which yield reliable results in different loading conditions, need to be developed.

On the basis of the shaking table test results and taking into consideration damage limitation requirements and ductility capacity, the values of elastic force reduction factors have been assessed. It has been found that the Eurocode 8 range of values of structural behaviour factor q are adequate, if pushover methods are used for seismic resistance verification, and the calculated displacement capacity of the structure is verified and compared with displacement demand. They are conservative where elastic analysis methods are used. However, additional research is needed to evaluate the overstrength and propose the modifications.

33

REFERENCES

[1] Eurocode 8: Design of structures for earthquake resistance, Part 1: General rules, seismic actions and rules for buildings/EN 1998-1:2004, CEN, 2004, Brussels.

[2] Eurocode 6: Design of masonry structures - Part 1-1: Common rules for reinforced and unreinforced masonry structures/EN 1996-1-1:2005, CEN, 2005, Brussels.

[3] Failure of shear-stressed masonry - an enlarged theory, tests and application to shear walls/W.Mann, H.Müller/Proceedings of the British Ceramic Society, No.30, Shelton House, 1982, Stoke-on-Trent: 223 235.

[4] Damage as a measure for earthquake-resistant design of masonry structures: Slovenian experience/M.Tomaževi /Canadian Journal of Civil Engineering, 2007, 34 (11): 1403 1412.

[5] Displacement capacity of masonry buildings as a basis for the assessment of behavior factor: an experimental study/M.Tomaževi , P.Weiss/Submitted to Bulletin of Earthquake Engineering, 2009.

[6] Shear resistance of unreinforced masonry walls/M.Tomaževi ,M.Gams/Ingegneria Sismica in Italia, in print, 2009.

[7] Some experimental results on the strength of brick masonry walls/V.Turnšek, F. a ovi /Proceedings 2nd International Brick-Masonry Conference, British Ceramic Society, 1970, Stoke-on-Trent: 149 156.

35

Paata Rekvava1

UTICAJ INTERAKCIJE TLO-INTERFEJS-KONSTRUKCIJA NA SEIZMI KI ODGOVOR PANEL ZGRADA Rezime:

U ovom radu predstavljen je metod za ocenu seizmi kog odgovora sistema tlo-interfejs-panel zgrada. Ocena ponašanja konstrukcije izvršena je pomo umetodologije za projektovanje zasnovano na o ekivanom ponašanju konstrukcija izloženih dejstvu seizmi kih sila (Performance-Based Seismic Design -PBSD). U ovoj studiji razmatrana je stokasti ka priroda pomeranja tla u regionu Tbilisija. Metod je predstavljen primenom panel zgrada podvrgnutih seizmi kom pobu ivanju za odre eni hazard na terenu. Razvijeni metod i dobijeni rezultati mogu se korisiti prilikom projektovanja novih zgrada ispitanog tipa, kao i za postoje e armiranobetonske panel zgrade starije generacije za o ekivane seizmi keaktivnosti u budu nosti.

Klju ne re i: seizmi an, interakcija, analiza, PBSD, panel, zgrada, pouzdanost.

SOIL – INTERFACE – STRUCTURE INTERACTION EFFECT ON PANEL BUILDING SEISMIC RESPONSE

Summary:

A method is presented for the evaluation of the seismic response of soil-interface- panel building system. The Structural behavior is evaluated by means of the methodology of Performance-Based Seismic Design (PBSD). This study has taken into account the stochastic nature of the ground motion in Tbilisi region. The method is demonstrated with an application to panel building subjected to seismic excitation for the specified hazard at the site. The developed method and results can be used in seismic risk study for new buildings of examined type under design, as well as for existing RC panel buildings of old generation for future seismic activity.

Key words: seismic, interaction, analysis, PBSD, panel, building, reliability.

_______________________________________________________ 1 Prof. Dr., Director, Kiriak Zavriev Institute of Structural Mechanics and Earthquake Engineering (ISMEE), 8, M. Aleksidze st. Tbilisi 0193, Georgia

36

1. INTRODUCTIONPanel buildings with many stories and spacing normally are analyzed with

simplified nonlinear models which neglect effect of spatial structural performance Caccese and Harris [1]. Some of the analytical models Rekvava [2], Astarlioglu et al. [3] are suitable for modeling 3D performance of panel buildings under seismic loading. The nonlinear seismic analysis method based on the Finite Element Method (FEM) and substructures procedures for panel building considers more scalar parameters as measures of the damage sustained Rekvava and Mdivani [4].

These parameters are also known as Engineering Demand Parameters (EDPs). The most common EDPs are the maximum story drift ratios, the maximum roof drift ratios or the maximum floor accelerations, that will be used with fragility relations to determine performance of building system and components. This study focuses on the analytical model of soil-interface-structure to assess the panel building seismic response using some phase of Performance-Based Seismic Design (PBSD).

2. OUTLINE OF NUMERICAL ANALYSIS The practical approach to PBSD considers a ground motion Intensity Measure

(IM), structural response to calculate EDP, resulting damage analysis, which relates the EDP to Damage Measure (DM) and calculation of Decision Variable (DV), in terms that are useful to decision makers such as direct losses, downtime (or restoration time), and life safety risks Moehle and Deierlein [5].

Direct assessment of seismic response EDP and DM of building begins with a careful assessment of the various modes of deterioration in the structural components that make up a building. In reinforced concrete panel building the primary structural components of the seismic force resisting system are the wall panels, horizontal panel slabs and key joints. Deterioration of key joints is associated with axial tension/compression, shear or a combination of these.

Based on the actual performance of panel buildings under strong earthquakes the idealized mechanical model, shown in fig.1, and plastic hinge joints method is developed by the author for simulating the inelastic response of the soil-interface-panel building system Rekvava [6]. The panel building is represented by a spatial system of elastic substructures in plane stress - wall panels and panel slabs connected in points, corresponding to location of key joints, by the nonelastic hinge links (lumped plasticity model). The soil is simulated by the ensemble of 3D elastic finite elements in the form of elastic nonhomogeneous isotropic half-space.

The conditions of interconnection as separation and sliding (constructive nonlinearities) on the interface between the building and surrounding soil are modeled by contact elements, not passing the tension strength to surfaces belonging to the building and soil. The contact element is assumed to have zero thickness and can be conceptually thought of as consisting of springs and Goodman joint element Heuze and Barbour [7]. The Mohr-Coulomb yield criterion is used to simulate interface behavior.

The moment of entering into the phase of cracking, yielding and failure of reinforced concrete key joints is specified by the von Mises plasticity formulation with the Mroz hardening theory Rekvava [6].

The model of the building for dynamic analysis is idealized as a multi degree of freedom

37

Fig. 1- Diagram of the soil-interface-panel building model

(dof) system consisting of masses. Thus, each mass is lumped at the level of the floor at nodes of structures interaction and at this stage may possess only three translation dof per node.

Equations of dynamic motion for assumed model of the soil-interface-panel building subjected to earthquake ground motion at the time t can be written as follows Rekvava [6]

P(t)F(t)(t)UC(t)UM (1) where:

(t),U-MBP(t) g

M, C are mass and damping matrices, F(t) is a vector of restoring (stiffness) forces, U(t) is the nodal displacement vector, B is matrix of coefficient of quasi-static effects of seismic influence, Üg(t) is a vector of the input ground acceleration time history, whose elements are given by the x-, y- and z-components of ground acceleration. The equations of motion (1) at time =t+ t can be written as

)(UMBFUCUM g

(2) Define the increments in acceleration, velocity, displacement and force occurring in the time increment t by

UUUUUUUUU ttt ttt

UKUUF

FFF Tt

tt ttt

(3)

Substituting these expressions in eq.(2), the incremental form of the equations of motion is obtained as follows:

]FUCUM[)(UMBUKUCUM tttgtTtt (4)

38

where: KT is the tangent stiffness matrix of the model at time t, which is a function of the nodal displacements at time t; Solution of eq. (4) involves time integration to generate the response at discrete times at intervals t, 2 t, etc.Initial conditions are required at time 0, usually the building carrying gravity loads in the at-rest state. The numerical integration of the nonlinear eqs. (4) is performed employing the Newmark constant average acceleration method ( =1/4 and =1/2) with Newton-Raphson type iterative technique to achieve equilibrium at the end of each time step. Constant average acceleration, an implicit time integration scheme, uses the following time-stepping relations

(t)U-(t)Ut

4-U(t))-t)(U(tt)(4t)U(t 2

(5a)

t)) t(tU(t)U(21(t)Ut)(tU

(5b) To define the kth iteration in the step from t to t+ t, replace U(t+ t) in eq.(5a) by (Uk(t+ t)+ Uk), linearize the stiffness forces as

kkk t) U(tKt)(tFt)F(t T (6) and substitute eqs. (5) and (6) into eq. (1) written at time (t+ t) to obtain

kkT2 Ut)+(tK+C

t2

+M)t(

4t)+(tUC

t2

+M)t(

4-t)+(tF-t)+P(t= k

2k

(t)UM+(t)UC+Mt

4+U(t)C

t2

+M)t(

4+ 2

(7) To carry out iteration k, eq.(7) is solved for Uk.. The new displacement approximation is found as Uk+1(t+ t)= Uk(t+ t) + Uk (8) and the updated stiffness forces Fk+1(t+ t) are computed from Fk(t+ t) by following the actual nonlinear behavior through the increment. After convergence, using the last approximation to U(t+ t), and the next time step commences. Thus, calculated values of displacements are used as structural response model for EDPs.

3. GROUND MOTION MODEL The recorded accelerograms may be used to represent earthquakes at a site. But

there is a scarcity of strong motion records for Tbilisi region. Because of the lack of records, in this study, synthetic earthquake time histories are generated to reflect the region (100 km environment) site conditions.

The model of seismic ground motion used in this paper is a set of discrete nonstationary Gaussian process that differ from one another by dominant frequencies, duration and other parameters.

39

Each j element of this set or the ground acceleration Üg (t, j) is found as the tjetejjjg ))x(t,(t,)(t,U (9)

where: j. is dominant j-th process frequency, its boundary values min and max are assumed on

the basis of empirical data, ( j) is root mean square value of acceleration, determines the effective duration and process nonstationarity,

x(t, j) is normalized random function that is characterized by function of correlation as )||sin/(cose)K(

||

jjjjj

(10) where:

is correlation coefficient, characterizing width of the spectrum. The computation of the parameters of the predicted earthquakes was carried out

at the eight seismogenic zones of Tbilisi region (100 km environment), that can reveal maximum seismic effect on the territory of the city Rekvava [8].

Calculated parameters considering earthquake magnitude and hypocentral distance for the generation of synthetic accelerograms for Tbilisi territory are given in Tab. 1.

4. RELIABILITY ASSESSMENT

After sample response histories of sufficient size are generated statistics are taken on the significant response quantities to determine their probabilistic parameters, which are in turn used for the reliability analysis of the building by means of Monte Carlo techniques by the computer code BUILDING-NL Rekvava and Mdivani [4].

The building failure criterion is considered the moment when the roof relative deflection value exceeds its permissible one |Ur/H| > [Ur/H]

(11) Table 1 - Parameters of design accelerograms

No zone (sec-1) (sec-1) (sec-1) cm/sec2

1st group with M=6 1217167

34.88 34.88 33.05 27.30

17.44 17.44 16.52 13.65

0.56 0.56 0.53 0.44

91 87 77 51

2nd group with M=6.5 20 19.03 9.51 0.3 33 3rd group with M=7 11430

19.62 17.44 14.60

9.81 8.72 7.30

0.31 0.28 0.23

120 74 32

where: Ur is the general roof horizontal deflection of the building, H is the building height,

40

[Ur/H] = 1/1200 is allowable value of the given parameter for design Poliakov [9]. For an alternative approach the reliability of structure Rs can be also evaluated on the basis of statistical method by Rs=1–No/Nt (12) where: No is number of failure event which is connected with the fulfillment of the condition

Ur >[Ur] under seismic influence, Nt is total number of roof deflections during seismic influence considered as a realization of random function. Under the equal probability condition the following value of the reliability is defined

n

1isis R

n1R

(13) where: n is number of seismic influence.

The seismic resistance criterion of the panel building generally is written Rs> Rul (14) where: Rul is the admissible reliability value and is adopted to be equal to 0.9-0.99.

5. SEISMIC ANALYSIS AND DISCUSSIONS The structure used in the analysis is 5-story and 15 m high panel building with

wide 7.2-8.4 m spacing. Structural members are prefabricated from the lightweight concrete. The story weight is 2952 kN for span of wall 7.2 m and 3240 kN for span of wall 8.4 m. The building is situated in Tbilisi area at the class II soil type (medium, with soil shear wave velocity 300-800 m/s) according to soil classification DC 01.01.09 [10].

The building is founded on the ground presented in appearance of a rectangular prism with sizes in plan 280 x 170 m. The ground segment from a surface to basic bedrock consists of two layers ( H1=10 m loam, E=58 MPa and H2=50 m clay, E=33 MPa) and 1426 elastic three dimensional finite elements with three translation degrees of freedom at each node. The 36 contact elements are arranged along the interaction surface between building and the soil. The maximum values of stiffness for contact elements in shear and compression are Kx=25.6x104 and Kz=36.5x104 kN/m, respectively.

The design gravity loads adopted are: 1) the structure self-weight; 2) the design live load.

The building was subjected to the three components of the Tbilisi earthquake of April 25, 2002 with scaled PGA of 0.2g and three-component synthetic accelerograms generated using the data of Tab. 1 for Tbilisi region of seismogenic zones 12, 16, 7 and 11 were used.

The calculation was carried out on the basis of design model considering the nonlinear ductility of connections of structural elements, the contact surface between the building and ground and the initial conditions (strained state from static load) at mixed system of bearing walls spacing in the first version – 4.2 and 7.2 m, and in the second version – 4.2 and 8.4 m.

In fig. 2 the first seven periods of natural vibration of the system are shown. It can be seen that increasing of the wall spacing up to 8.4 m conditions the increasing of initial

41

fundamental period per 5%, whereas the difference in the values of higher tones of period composes 30-60 %.

Fig.2- Values of the periods

Figure 3 shows the sample values of the maximum roof deflection for different dominant frequencies of seismic action.

Fig. 3- Roof deflection 1- for 7.2 m; 2 - for 8.4 m.

Figures 4 and 5 summarize the maximum storey responses to five ground motions.

Fig. 4- Story horizontal deflections for spacing 7.2 m

42

Fig. 5 - Story horizontal deflections for spacing 8.4 m

The calculated extreme story drift distribution is illustrated in fig. 6.

a b

c d Fig. 6 - Story drift of building for longitudinal (a,c) and transversal (b,d) directions: spacing 7.2 (a,b); spacing 8.4 (c,d). 1 –due to “Tbilisi” earthquake ; 2 – due to accelerogram from seismogenic zone 11

The analysis shows that during the elastic-plastic vibration of building with spacing of 7.2 and 8.4 m maximum roof horizontal relative deflection and stories drift ratios are less than 1/1200 and 1/200, respectively that indicates the great rigidity and the ability of a panel system with wide spacing under examination to resist to earthquakes of

43

various spectral content. This is also confirmed by the displacement ductility demand ( B)of the building that is ratio of the maximum displacement to the yield displacement and as numerical results showed its value equals 2.

The value of slippage at elastic-plastic vibration on the interface of building-ground reaches its maximum at the spacing of 8.4 m at generated earthquake with the prevailing frequency 19.62 sec-1 (zone 11) and composes 0.0006 m that is 2.8 times greater then the effect of the 2002 Tbilisi earthquake.

The building maximum subsidence is 0.013 m that is less than the maximum allowable one for panel buildings.

In consideration of real and generated accelerograms there is no disturbance of the contact along the vertical axis of the building.

The increase of spacing up to 8.4 m does not cause damage of the panel slabs with exhausting of carrying capacity of the compressed cross-section.

The concentration of main tensile stress zones is observed in the joints of connection and exceeds the concrete design resistance in tension that conditions the local damages in these places.

The maximum compression stresses in the panels of external and internal walls at increasing of spacing up to 8.4 m are raised per 19-30%, but they remain less then design compression resistance of the concrete of respective class. In the most strained panels of the first floor the cracks appear under action of accelerograms of zone 11, whereas the other earthquakes under consideration do not affect significantly the structure operation.

The deformation of key joints of structural elements has a complex character. The number of elastic-plastic cycles of deformation depending on the duration and spectral content of real earthquake and generated accelerograms reaches 10-30.

Cracks in the key joints and local damages are developed, but permanent displacements do not exceed the permissible ones in the horizontal (0.03 m) and vertical (0.01 m) directions. Here the normal (compression, tension) and shear forces are increased per 1.2-1.3 times in comparison with spacing of 7.2 m, and compose respectively 56% and 41 % of ultimate strength of indicated joints.

CONCLUSIONS 1. The method has been presented that is capable of accurate reproduction of the

complete three-dimensional nonlinear behavior of the soil-interface-panel building system under strong ground motion in Tbilisi region and used to study the reliability of a residential 5-story panel building of new generation.

2. The bearing capacity of the panel building with the super-wide spacing 8.4 m is not exceeded. It resists the effect of an earthquake of high intensity and retains the ability of further deformation.

3. The code reduction coefficient K1=0.25 (it is equal to 1/q, where q is the behavior factor used for design in Eurocode 8) considering the panel building capacity to develop the inelastic deformations, in this study composes 0.5, that indicates the low degree of nonlinear deformability of the 5-story panel building.

4. The building reliability for both versions are equal to 0.95 that is greater than ultimate admissible one (0.9) for the dwelling houses that guarantees the structure safety from collapse and allows to recommend the expediency of experimental design and the construction of proposed building with super-wide wall spacing.

44

5. Degree of total damage of the building is less than 3 determined by MSK-64, that is connected with DM and does not cause the stopping of a building function and is not required a lot of financial expenses for repairing. Obtained results can be used for many decision-makers.

7. Further research needs tools for improved nonlinear analyses of the soil-interface-panel building system that includes a successful methodology for evaluation of monetary loss or DV as the last stage of PBSD of the examined type panel building for future seismic activity.

REFERENCES

[1] Seismic resistance of precast concrete shear walls-correlation of experimental and analytical results /Caccese V., and Harris H.G. //J. Earthquake Engineering and Structural Dynamics, 1987, 15, pp. 661-677.

[2] Nonelastic ductility influence of nodes on seismic response of large panel building s/ Rekvava P.A.// Proc. 9th European Conference on Earthquake Engineering, 1990, Moscow, vol.7-A, pp.118-126.

[3] Modeling strategies for three dimensional analysis of precast panel buildings under seismic load/ Astarioglu S., Memariz Ali M .and A.Scanlon A. //Proc. 12-th World Conference on Earthquake Engineering, 2000, London, Paper Reference 1356.

[4] Investigation on seismic performances of precast R/C panel buildings /Rekvava P.A and Mdivani K.I. // Proc. International Turkey Symposium on Advances in Earthquake and Structural Engineering, 2007, Isparta, pp. 17-27.

[5] A framework methodology for performance-based earthquake engineering/ Moehle J. and Deierlein G.G. // Proc.13-th World Conference on Earthquake Engineering, 2004, Vancover, Paper Reference 679.

[6] Method of plastic hinge joints in design panel building under seismic influenc /Rekvava P.A. // Proc.14-th World Conference on Earthquake Engineering, 2008, Beijing, Paper Reference 14-014.

[7] New models for rock joints and interfaces/ Heuze F.E. and Barbour T.C. // 1982, J. Geotechnical Engineering Devision, 108, pp.757-776.

[8] Use of the regional models of seismic effect in building design/ Rekvava P.A. // Proc.10-th European Conference on Earthquake Engineering, 1994, Vienna, vol.1, pp.253-256.

[9] Seismic resistant constructions of Buildings/ Poliakov S.V. // Building Publishing House, 1983, Moscow, pp. 303.

[10] Design code for earthquake engineering of Georgia /DC 01.01.09, 2009, Tbilisi, pp.103.

[11] Guidelines of the panel dwelling-houses design/ BCR 2.08.01-85//Building Publishing House, 1989, Moscow, pp. 245.

45

Zoran V. MILUTINOVI and Mihail A. GAREVSKI

ABU DABI, UAE, SISTEM ZA MONITORING I MENADŽMENT SEIZMI KOG RIZIKA Rezime:

Da bi se obezbedio i zaštitio održljiv razvoj Abu Dabi emirata, UAE, a posebno Opštine i grada Abu Dabija i zaštitila zivotna sredina njegovih stanovnika, Sektor za Planiranje Opštine Abu Dabi inicirao je projekt “Ocena seizmi kog hazarda i rizika Abu Dabija”. Tokom konceptualne pripreme projekta i izrade tender dokumenta, primarni koncept je modificiran i teritorijalni obuhvat projekta je proširen, rezultirajuci u “Abu Dabi sistem za monitoring i menadzment seizmi kog rizika”. Rad diskutira i u neophodnim detaljima prezentira strukturu projekta, njegove ciljeve i domet, kao i druge krucijalne aspekte.

Klju ne re i: Sistem, seizmi ki rizik, monitoring, menadzment.

EMIRATE OF ABU DHABI, UAE, SYSTEM FOR SEISMIC RISK MONITORING AND MANAGEMENT Summary:

To assure and maintain sustainable development of the Emirate of Abu Dhabi, UAE, in general, of Abu Dhabi Municipality and Abu Dhabi City in particular, and disaster free living environment for its citizens, the Abu Dhabi Municipality Town Planning Department initiated project "Assessment of Seismic Hazard and Risk in Abu Dhabi". During the conceptual design of the system and development of the Tender document, the primary concept and territorial coverage of the project have been enhanced ending up into “Emirate of Abu Dhabi System for Seismic Risk Monitoring and Management”. This paper discusses and in necessary details presents the project structure, its objectives and scope and some other the crucial aspects.

Key words: System, Seismic Risk, Monitoring, Management.

Professor, Head RDM-IZIIS, Institute of Earthquake Engineering and Engineering Seismology, Skopje, Macedonia; ADM, UAE, consultant. Professor, Director, Institute of Earthquake Engineering and Engineering Seismology, Skopje, Macedonia;

ADM, UAE, consultant.

46

1 OBJECTIVES

To assure and maintain sustainable development of the Emirate of Abu Dhabi in general, of Abu Dhabi Municipality (ADM) and Abu Dhabi City in particular, and disaster free living environment for its citizens, the Abu Dhabi Municipality initiated the Project "Assessment of Seismic Hazard and Risk in Abu Dhabi". During the latest stage of ToR development (December 2008), Al Ain (AAM) and Western Region (WRM) Municipalities joined the ADM initiative. The project has been enhanced to the national, the Emirate of Abu Dhabi, scale and the title was reformulated in "Assessment of Seismic Hazard and Risk in Emirate of Abu Dhabi". The Project is aimed in setting up a system that in the most effective, but technically and scientifically consistent, manner will provide data, technology and state-of-the-art know-how for developing sound strategy and acceptable policy environment for protection of current and planned development against adversity of potential environmental impacts. The Project, endeavored in creating Abu Dhabi System for Seismic Risk Monitoring and Management, is primarily focused on seismicity sector. The historic record on local seismicity of engineering significance is practically nonexistent, however, the unique structural typology emerging rapidly in the region, represented by worldwide unique typology of tall and ultra tall buildings, is calling on particular attention. In particular critical is their behavior when exposed to strong seismic action from neighboring high energy seismic sources, their intrinsic potential of causing significant and psychologically unacceptable discomfort, even panic, of occupants and businesses, including adverse effects’ (physical damage) potential on their seismic stability, structural integrity and seismic safety. The Project shall meet seismic community safety goals by assuring consistent, technically sound and economically justified policies in the domain of:

o Prevention, Regional and Development Planning (Emirate scale) o Urban Planning and Land Development (Municipal, City of Abu Dhabi, Al

Ain and designated urban areas in WRM scale) o Mitigation, Emergency Preparedness and Response (Municipal, City of Abu

Dhabi, Al Ain and designated urban areas in WRM scale) o Engineering and Development (City of Abu Dhabi, Al Ain and designated

urban areas in WRM scale) o Promotion of Risk Prevention Culture and Public Safety (Emirate scale)

Such strategic orientations are made possible by recent worldwide advances in seismic monitoring instrumentation, real-time computational and data transmission and communication technology, as well as the data accumulated, databases developed and established GIS systems for monitoring and management of urbanization processes in major urban areas of Abu Dhabi Emirate.

2 SCOPE The Project pillars are the following broadly defined activities:

o Qualitative and quantitative assessment of seismic hazard, Seismic Zoning of the Abu Dhabi Emirate, Microzoning of areas defined by the "Plan Abu Dhabi 2030: Urban Structure Framework Plan", "Al Ain Urban and Regional Structure Plan 2015" and designated urban areas in WRM

47

o Qualitative and quantitative assessment of seismic risk of buildings, selected nine (9) essential structures, selected Critical Importance lifeline (water supply, electric power, gas and oil transportation) systems, and of urban development’s of Abu Dhabi, Al Ain and Western Region Municipalities)

o Seismic monitoring for emergency management and disaster risk reduction by deployment of network of 50 accelerograph seismic monitoring stations

o Seismic monitoring for development and maintenance of 3D Seismic Simulation Model for simulating long-period seismic waves by deployment of a network of four (4) stations consisting of three-component broad-band sensors and triaxial accelerograph unit

o Installation of Structural Health Monitoring Systems in seven (7) selected unique structures

o Development, including Web display, of Ground Shaking Map System for Abu Dhabi Emirate (EAD-GSM System), for automatic real time collection of ground shaking parameters for emergency impact phase decision making process

o Creation, upload and maintenance of Web for public informing and in a support to architectural and structural design practice

o Creation of reliable seismic data base as an umbrella and the major instrument for seismic risk monitoring and management

o Establishment of disaggregated EAD Seismic Risk Monitoring and Management Center (EAD-SRMMC) consisting of two (2) Data and five (5) Display Centers.

3 STRATEGIC GOALS The strategic goals are:

o Study "Assessment of Seismic Hazard and Risk in Emirate of Abu Dhabi" as a base for Building Seismic Design Code Development, engineering prevention and regional and urban planning

o Creation of an Unit to operate the system and databases – A Centre for Seismic Risk Monitoring and Management (EAD-SRMMC)

o Capability development o Public Awareness

4 FORESEEN STRATEGIC ACHIVEMENTS 4.1 Infrastructural Achievements The major tangible infrastructural achievement is establishment of disaggregated EAD Seismic Risk Monitoring and Management Center (EAD-SRMMC). The EAD-SRMMC Center consists of two (2) Data and five (5) Display Centers. Data Centers are conceptually designed as complex, powerful and independent IT and communication facilities for:

o acquisition and hosting of all metadata accumulated during the project performance:

48

seismic zonation and microzonations 3D seismic simulation model and other geo-spatial data urban (buildings) and infrastructure (4 Critical Importance lifelines) spatial

data/results simulation models and expertise for various project needs and subsequent

simulations host seismic design parameters web specialized client’s software and seismic databases

o acquisition, hosting, real-time processing and dissemination of data acquired from:

seismic monitoring network that presently consisting of 5 stations of National seismic monitoring network, 4 stations to be set-up under this project as well as additional 4 stations to be set-up in coming future as a second phase of programmed development of the National seismic monitoring network

permanent accelerograph (strong motion) network consisting of 50 accelerograph stations

seven (7, ADM – 3, AAM – 2, and WRM-2) structural health monitoring systems (SHMS)

other data relevant for other project needs as presented in ToR document. o in real-time calculation and release EAD Ground Shake Map o host and in real-time maintain needs and operations of 5 display centers. Two, spatially dislocated from each other Data centers, designed to mirror the

function of the other one, are foreseen for assuring: o operability of at least one Data center for real-time function of five Display

centers during emergency impact period o real-time acquisition and transfer of data and other relevant parameters for

decision making process of stakeholders involved o real-time decision making process, and o other emergency needs Considering the interdisciplinarity and complexity of real-time operations the

data centers are designed for, not only one, but several stakeholders shall and will be mandated for their operation and maintenance. Mandates and coordination will be decided during the project implementation phase.

Five Display centers, in fact decision making posts, are to be established in: o Abu Dhabi Municipality (ADM) o Al Ain Municipality (AAM) o Western Region Municipality (WRM) o Department of Municipal Affairs (DMA) o Civil Protection/Emergency Function (CP) The conceptual design of EAD-SRMMC is presented in Img. 1. The EAD-

SRMMC is a system that: o Catalyses and integrates cooperation in the fields of hazard, risk and

emergency/disaster management among different national institutions

49

(NCMS2, UAEU3, PI4, ADNOC5, ADWEA6, EAD7 / Former ERWDA8,other), the decision making authorities (DMA, Municipalities /ADM, AAM and WRM], Emergency systems), Critical infrastructure operators (power and water supply, gas and oil transportation), and other interested in potential stakeholders

o efficiently bridges the gap between professionals of various proficiencies and decision making authorities

o assures and maintains real-time decision making process among major emergency and crisis management stakeholders (DMA, ADM, AAM, WRM and CP)

In essence, EAD-SRMMC will represent a seed and backbone of an Integrated EAD (Multi) Disaster Management System that, although conceptually designed based on potential seismic threats the Emirate is exposed to, is easy upgradable for any other disaster agent jeopardizing safety, security and quality of life of citizens of the Abu Dhabi Emirate.

4.2 Other Tangible Strategic Achievements

Bellow is summarized only the most tangible achievements: o Integration of worldwide state-of-the-art in the fields of:

Geosciences engineering and earthquake engineering IT & communication emergency management, etc.

o Improved seismic monitoring network o Established seismic monitoring of engineering and decision making

parameters by permanent accelerograph network and structural health monitoring systems

o Seismic zoning of the Emirate and seismic microzonations of areas defined by the "Plan Abu Dhabi 2030: Urban Structure Framework Plan", "Al Ain Urban and Regional Structure Plan 2015" and designated urban areas in WRM

o Assurance of data, parameters and criteria for EAD Seismic and Wind Design Code development including provisions for design of tall and ultra tall buildings

o Technological and IT linkage and integration of major Decision making Authorities

o In a case of a seismic impact and in a support of rapid and coordinated situation management, provision of near-real time (not more that 25-30

2 NCMS: National Centre of Meteorology and Seismology, U.A.E. Ministry of Presidential Affairs 3 UAEU: United Arab Emirates University, Al Ain, Abu Dhabi Emirate 4 PI: Petroleum Geosciences Department, the Petroleum Institute, Abu Dhabi 5 ADNOC: Abu Dhabi National Oil Company 6 ADWEA: Abu Dhabi Water and Electricity Agency 7 EAD: Environment Agency - Abu Dhabi (previously ERWDA7)8 ERWDA: Environmental Research & Wildlife Development Agency of Abu Dhabi Emirate

50

minutes) information on seismic effects (EAD Ground Shaking Map /EAD-GSM/ System):

in major urban areas of the Emirate on major Critical Infrastructure (CI) systems (Water supply, Electric Power,

Oil and Gas transportation) o Build-up of national capacity and capability for which in other part of the

World several decades of progressive research and development were and are usually needed

o Coherent estimates on seismic vulnerability and risk of urban areas defined by the "Plan Abu Dhabi 2030: Urban Structure Framework Plan", "Al Ain Urban and Regional Structure Plan 2015" and designated urban areas in WRM

o Other achievements pertinent to seismic qualification and quantification of the territory of Abu Dhabi Emirate and neighboring regions, process of regional urban planning, development and preventive seismic protection of existing and planned construction.

5 PROJECT STRUCTURE General Project structure is presented in Table 1. More detailed insight on foreseen deliverables and expected completion time frame is provided in Table 2. The Project, as structured and presented in Table 1, consists of 20 tasks and 68 deliverables. Fifteen (15) tasks and fifty (59) deliverables of technical nature (Table 3) are related to qualitative and quantitative understanding of:

o seismic environment the Emirate of Abu Dhabi is exposed to o potential consequences if the Emirate in general, and Abu Dhabi, Al Ain

and/or Western Region Municipalities in particular, are exposed to a major earthquake event from surrounding high energy seismic zones

o performance and seismic risk inherent to current and planned building stock, and selected critical importance lifeline systems; and,

o installation of seismic monitoring systems (accelerograph network, seismic monitoring network and structural health monitoring systems) for research, engineering, planning and emergency management needs

o establishment of disaggregated EAD Seismic Risk Monitoring and Mana-gement Center (EAD-SRMMC) consisting of two (2) Data and five (5) Display Centres.

o definition of parameters for deciding nationally feasible and economically justifiable levels of seismic protection, i.e. – economically acceptable level of seismic risk.

Three (3) tasks and seven (7) deliverables are related to coordinated activities, the Project promotion, validation and creation of public awareness and training, one (1) task and one (1) deliverable to post-implementation phase maintenance and one (1) task and one deliverable (1) are related to BOT services that will enable Client to rapidly start offshored operations, and within a period of 3 (three) years, to take ownership of the offshored developed center (EAD-SRMMC Center) as its own subsidiary.

51

Fourteen (14) tasks and twenty-six (26) deliverables shall be performed at Abu Dhabi Emirate level, while six (6) tasks and forty-two (42) deliverables are municipality specific and shall be conducted either within or for selected urban areas and critical9 and/or unique10 structures in Abu Dhabi, Al Ain and Western Region Municipalities.

6 SPATIAL COVERAGE The spatial coverage of the Project is the entire jurisdiction area of Abu Dhabi Emirate (Img.2), including

o Abu Dhabi Island o East of Abu Dhabi City up to Alkhatem o West of Abu Dhabi City up to Seih Shoaib, and o Al Ain Area

7 OTHER PROJECT ASPECTS Implementation phase of the Project lasts 20 months. Project dominantly relies on use, interpretation and reinterpretation of existing data fund. The additional field (site and building characterization, Table 4) investigations are required to meet international standards for installment of systems as planned, as well as to assure a set of consistently interlinked parameters for calibrating data compiled from literature and from the existing data fund. In summary, additional field investigations comprise drilling of 117 boreholes with total length of 2.75 km, 70 P-S log (suspension) measurements, 190 months of 1 month continuous microtremor measurements, 17.5 km of surface refraction (or MASW) geophysical measurements, sampling of 20 undisturbed samples in cohesive soils and 117 other samples for determination of index properties (water content, sieve analysis, Atteberg's limits), 140 samples and corresponding dynamic laboratory testing for determination of G- / - material curves, 36 samples for liquefaction susceptibility testing, as well as ambient vibration measurements of 16 buildings with estimated total duration of about 6 months.

9 Buildings/structures with high occupancy, critical response services (fire, police, hospitals) and vulnerable populations (schools, nursing homes). Damages to critical structures lead to more life loss, larger economic loss and greater social disruption, and slow community response to earthquakes. Often these buildings are called 'essential'.10 Structures of unique architectural shape and/or size, historic buildings, monuments, structures requiring unique features for their specific purpose and use.

52

8 OVERLAPPING Project has no overlaps. The identified one, overlapping with DMA initiative "Adoption of IBC (2006/2009)" Building Code of International Code Council (US) has been extracted. However, the link between both initiatives has been established through Task 16: "Coordinated Activities" in order to assure mutual transfer of data and outputs and other feed backs. During course of performance, other links will be assured if, and when, other complementary activities are identified.

9 GAPS During the tendering process it was identified that neither DMA initiative "Adoption of IBC (2006/2009)" Building Code of ICC, US nor the ADM project “Assessment of Seismic Hazard and Risk in Emirate of Abu Dhabi” provides elements, guidance and rationales for seismic and wind design of tall and ultra tall buildings favored as current construction typology in the region. To amortize this gap that can seriously affect design practice in the Emirate, in particularly tall and ultra tall buildings and other flexible structures, during the implementation phase of this project adequate ToRs will be developed for:

o Wind Hazard Assessment and Wind Zoning of Emirate of Abu Dhabi; and, o Special Provisions and Guidelines for Wind and Seismic Design of Tall and Ultra Tall Buildings o and both additional deliverables will be accommodated under the Task 16: Coordinated Activities.

10 END USERS Project outputs cover mandates and needs of broad number of stakeholders. The most concerned are listed in the following

o Governmental authorities o Emergency response systems & Emergency Managers o Engineering & scientific community (Seismologists, Architects, Structural and Earthquake engineers, Geologists, etc.) o Regional & Urban planners o Land developers o Operators of critical infrastructure o Owners and/or operators of infrastructure and public utilities o Owners of structures o Businesses o Media o General public; etc.

53

11 TRAINING Training needs of nominated staff, necessary to assure adequate operation of installed systems and the maximum benefit of results achieved are identified, as specified in Table 5. While hands-on trainings are designed to develop national capability to operate and use the system for various technical, planning, research and emergency needs, the purpose oriented trainings are designed for training of decision making authorities and other relevant professionals to familiarize with system outputs and develop skills to efficiently implement them in decision making process.

12 DISSEMINATION/INTRNATIONAL VALIDATION To present Project achievements, position them with respect to worldwide state- of-the-art, receive comments and, if viable, incorporate in the final products and assure national feed-back, three International workshops are planned, as presented in TK- 16:

o Workshop 1: Seismic Hazard Assessment, Seismic Zoning and Microzoning o Workshop 2: Strong Motion Instrumentation and Structural Health Monitoring o Workshop 3: Seismic Risk and Loss Assessment Further project promotion is foreseen to be made by organizing special, the

Project specific, sessions at: o 14th European Conference of Earthquake Engineering

Skopje, Republic of Macedonia (2010) o 15th World Conference of Earthquake Engineering

Lisbon, Portugal (2012) o Other national, regional and international events

13 ADMINISTRATIVE ISSUES The project "Assessment of Seismic Hazard and Risk in Emirate of Abu Dhabi" alias “Emirate of Abu Dhabi System for Seismic Risk Monitoring and Management” by its objectives and scope, as requested by RFP and translated in ToR, in the forthcoming period will worldwide be one of the most outstanding activities in the field of seismology, earthquake engineering, seismic monitoring and protection of population, material property and environment against adverse seismic impacts. The tendering process of the project “Emirate of Abu Dhabi System for Seismic Risk Monitoring and Management” is completed and Contractor is selected. Presently, it is in Contract negotiation and Contract award phase, with expected commencement at November 2009.

14 ACKNOWLEDGEMENT

The Consultant, GECO Engineering, Dubai, represented by Prof. Dr. Zoran Milutinovic, conceptually designed the system and developed Project’s Terms of Reference (ToR) in full cooperation with the Project Technical Committee composed of nominees from:

o U.A.E. Ministry of Energy o Department of Municipal Affairs (DMA)

54

o Al Ain Municipality o Western Region Municipality o National Center for Meteorology and Seismology (NCMS) o United Arab Emirate University (UAEU), Faculty of Science, Geology Department, Al Ain o Petroleum Geosciences Department, the Petroleum Institute, Abu Dhabi, and o Abu Dhabi Municipality, Building Permit Directorate

and with full support of the Abu Dhabi Municipality, Town Planning Sector, Spatial Data Directorate – the initiator of the Project. Consultant expresses his gratitude to Technical Committee for continuous assistance, support and encouragement during the course of conceptual design of the system and its ToR development, as well as established working and cooperation synergy.

Image 1 - Conceptual Design and Functionality of Disaggregated EAD-SRMMC Center

55

Image 2 - Study Area (Emirate of Abu Dhabi)

56

Table 1 - Project Structure Broken Into Tasks (TK) and Deliverables (DL)

57

58

59

60

Table 3 - Brake Down of Task and Deliverables by Nature

Table 4 - Summary on Field Investigations for Site/Building Characterization

61

Table 5 - Nominated Staff Training Needs

63

Zaven Khlghatyan1, Ruben Badalian2

SEISMIC PROTECTION OF STRUCTURES BY “COUPLED SYSTEMS”- DEVELOPMENT, RESEARCH AND REALIZATION

Summary: The Armenia and adjacent territories are situated in the collision zone of the Arabian and Eurasian plates. The Spitak (Armenia), 1988 destructive earthquake has shown that seismic hazard in Armenia was considerably higher than buildings and structures seismic resistance and practically all the territory of Armenia is situated in the area of high seismic risk. It is evident that the increase of existing buildings and structures seismic resistance is essential for Armenia. The results of theoretical and experimental research on application coupled system of strengthening and retrofitting of existing buildings in Armenia are provided. Key words: seismic hazard, seismic protection, strengthening and retrofitting, coupled system

ZAŠTITA OBJEKATA OD SEIZMI KIH EFEKATA RAZVIJANJEM “SPREGNUTIH SISTEMA”, ISTRAŽIVANJA I REALIZACIJA

Rezime:Jermenija i teritorije u njenoj neposrednoj blizini nalaze se u zoni sudaranja arapske i evroazijske plo e. Razorni zemljotres koji se dogodio u Spitaku (Jermenija) 1988. godine pokazao je da je seizmi ki hazard u Jermeniji znatno ve i do seizmi ke otpornosti zgrada i objekata i da se prakti no cela teritorija Jermenije nalazi u zoni visokog seizmi kog rizika. O igledno da je pove anjeseizmi ke otpornosti postoje ih zgrada i objekata od suštinskog zna aja za Jermeniju. Dati su rezultati teorijskih i eksperimentanih istraživanja o primeni spregnutih sistema oja anja i rehabilitacije postoje ih zgrada u Jermeniji. Klju ne re i: seizmi ki hazard, zaštita od seizmi kih efekata, oja anje i rehablitacija, spregnuti sistem

___________________________ 1 Zaven Khlghatyan – Head of Earthquake Engineering Centre, Armenian Western Survey for Seismic Protection SNCO2 Ruben Badalian, RA, J-S CC ARIEE and PC

64

1. OVERVIEW OF SEISMIC PROTECTION SYSTEMS

Seismic protection of damaged buildings during the earthquake and the lack of seismic resistant buildings are traditionally performed by strengthening of their structural systems. Depending on the type of building and strengthening of damaged reinforced concrete, or are implemented by steel jackets, stiffeners, reinforced mortar layers, an additional rigid elements in the form of reinforced concrete diaphragms or steel lateral braces, prefabricated slabs, framing of openings, polymer reinforced bars, injection of cracks with mortar and so on. Seismic Protection Systems be grouped as follows:

- Serving to increase the carrying capacity of individual structural elements and the connections of their joints;

- Intended for the perception of a large part of the inertial forces on the building, increasing its rigidity and stability.

In first group consists primarily strengthen the constructive solution, in second group strengthening actions taken based on the structural analysis and design of the building and its individual elements.

Constructive solutions, methods of analysis and research of strengthening of structures and buildings devoted a lot of valuable publications, including normative documents, being issued and published in Armenia [1-5].

In order to radically improve the reliability and efficiency by significantly reducing operating on these inertial forces, have been used and known in the seismic design methods for active and semi-active seismic protection. There is already developed and to some extent tested a number of such systems. They can be grouped as follows:

- Working on the principle of seismic isolation system, including a flexible lower part of the load-bearing structures, with the rubber supports, with the kinematic supports the principle of rolling, with pendulum bearings, with suspended floors, the same, but with joints of dry friction with sliding supports;

- Adaptive seismic protection systems operating on the principle of change within the prescribed limits of their dynamic characteristics, including: from disconnecting joints; to include the connecting;

- Systems with high damping, acting on the principle of dissipation of energy during the vibrations, including: viscous dampers, with elements of high plastic deformation, including vibration dampers beam, ring type and extrusion, with dry frictional dampers, frictional diaphragm, “sandy-dampers”;

- System with tuned mass absorbers, including: with impact and dynamic, including: spring, pendulum, combined, in the form of a flexible upper floor, etc.

Suffice detailed overview and analysis of such systems given in [6]. New versions of the active seismic protection systems, including the principles of vibration and seismic isolation dampers have been developed and studied in Armenia (J-SCC ARIEE & PC, NSSP of RA). Moreover, they were first implemented not in new construction but in the seismic zone of the Spitak earthquake [7-12].

Along with these systems have been proposed the first varieties of the Coupled Systems. We note constructive solution adopted in developed by us in 1992 to strengthen the draft 3-and 4-storey buildings of the maternity advice bureau, gynecology and obstetrics of the regional maternity hospital in Artik town, built on the basis of the standard frame of the series IIS-04.

65

As provided for them instead of unfinished construction, poor quality and partially damaged large-panel walls, new monolithic reinforced concrete with stone facings. Between them and the frame of the building through joints and reinforcing flexible braces under the angle of 45 were installed, being anchored in walls and pouring in place plates of precast floors. It allowed us to use the walls in its plane as a seismic protection of buildings.

2. THEORETICAL PRINCIPLES OF THE COUPLED SYSTEMS.Determination of the natural frequencies of structures is the most important and difficult

part of the dynamic analysis of their seismic behavior. How, in particular, pointed out in [14], it is connected with the imminent errors occurring from the fact that the design scheme in most cases is rather crude reflection of reality, the parameter value of the designed systems on which they depend for rigidity, can not made precise, etc. By the increase of the same number of degrees of freedom determine the natural frequencies even more complicated.

However, this problem is susceptible to some simplification. This becomes possible if the complex system considered as coupled to each other, two separate, simpler system with a single degree of freedom. Here, “coupled” means that the vibration in separate system, that is in one part of the complex system, influence the vibrations in the other and vice versa.

Designed L.I. Mandelstam, N.D. Papaleksi, A.A. Andronov and their followers Khaikin, and A.A. Vitt, V. Migulin and others [15-16] theory of a bound system is based on the fact that the vibrations of the two systems can be considered as vibrations of a united system with two degrees of freedom. The main question is - how much of the vibrating process of one system affects the presence of the second system and its vibrating process.

L.I. Mandelstam, analyzing this question, introduced the concept of connectedness. The main point of the latter is the following: the character of the systems interaction is revealed not only by the parameters of the joints between them, but the closeness of their partial frequencies. Even a small force in the relationship may have a significant influence on the vibrational process system, if the partial frequencies of each system are close enough to each other. And vice versa - with a large detuning, that is, for a significant difference between the partial frequencies, even at relatively large forces in joints, they can influence the vibrations of each individual system is not so much.

Realized structural system “Building-Multistory Annex” is a kind of two such coupled systems, connected by flexible joints. Developed and other options structural solution of a coupled system.

3. THE STRUCTURAL DECISION OF SEISMIC PROTECTION OF BUILDINGS BY MEAN OF COUPLED SYSTEM.

The coupled system “Building-Multistory Annex” developed and implemented on a typical 5-storey masonry residential building with 90 apartments in the city of Gyumri. Earthquake GPA (Ground Peak Acceleration) was: A= 0,2 g. But now GPA of this territory assigned to: A=0,4 g. Building on the level of damage attributable mainly to the 2-nd and

66

partly to the 3rd degree. To ensure its earthquake GPA, the following activities to provide its seismic resistance:

- the structure of seismic protection system was built as a reinforced concrete multistory annex;

- the structural measures were implemented for transverse walls; - restoration and strengthening measures were implemented for longitudinal bearing

walls with original design bearing capacity can provide the required seismic resistance of building in this direction.

Seismic design of buildings is a prefabricated-monolithic multistory annex, formed of reinforced concrete diaphragm elements and slabs of floors.

By the results of experimental tests of J-SCC ARJEE and PC on performance of compressed reinforcing rods with constrained supports [8] the connected elements were developed and implemented as the joints of reinforcing rods [13]. Their hooked ends are anchored in monolithic areas of the girders of the annex and in special reinforced concrete bindings being installed along the whole building. They are joined with floors to accept the forces not only for the areas of the wall, but also transfer it to the building. The sectional area of the joints was extracted from the forces defined by the design of the system on seismic effects. For the optimization of the system the calculations of the system were executed by 20 different variants of the location of the joints. By the results of the calculation they were fabricated from 12 class A-III rods, with the space 0,85m on the level of the floor of the 5-th storey, and on the level of the room – 1,0m. The free length of the joints was accepted from the necessary values for longitudinal rigidity and forces [13]. The rods of the joints were galvanized to avoid the corrosion.

4. THE CALCULATION OF THE COUPLED SYSTEM “BUILDINGMULTISTORY ANNEX”

During the calculation the following was taken into consideration; its results will allow us to determine the level of reduction of design values for horizontal seismic loads on the building in transverse direction precisely enough and to reveal the possible interaction of the building and the annex in longitudinal direction, that is, to determine and to compare their periods of vibrations separately and together with the system in the same direction. To reveal the efficacy of the system, the building and the annex were calculated separately and together with the system. During the design the building and the annex were modeled by bar finite elements taking into consideration their shear, flexural rigidities, and the joints in the level of the floors – by finite longitudinal rigidity. The masses of the building and the annex were accepted as concentrated in the level of the stories. The calculations (fig. 1) have been made by the computing complex “Mirage”, version 4.2 (NIIASS, Kiev). In fig. 1 the relation of the maximum values of the transverse forces on the building are shown separately and together with the coupled system to reveal the effectiveness of the tested system.

67

a) separately united system building annex building annex

b)

sec007.0sec020.0sec014.0sec054.0

sec020.0sec122.0sec329.0sec043.01075,1011035,304

sec202.0104,939103613

33

22

311

255

12525

TTTT

TTTTkNGAkNGATmkNEImkNEI

c)

d)

e)333,0

1136379

..

..1

sepbQsysbQ

K 175,3222705

..

..1

sepanQsysanQ

K

Fig. 1. The results of the building and the annex separately and the system “Multistory Building-Annex” in transverse direction.

68

a - Displacement (mm) and the values of coefficient of the mode shape ;b - Stiffness characteristics and the vibrations periods (sec); c -The load (ton) on the compartment storey of the building and forces diagrams (in

brackets inertial forces); d - Bending moments diagram (ton.m); e - The ratio of maximum forces on the building and the annex separately for their

values together with the system.

5. EXPERIMENTAL AND THEORETICAL TESTS OF THE REALIZED COUPLED SYSTEM “BUILDING-MULTISTORY ANNEX”

The aim of experimental studies of the system “Building-Multistory Annex” by vibration measurement equipment was to determine the dynamic characteristics- natural periods of the building and the annex separately and together as part of the system, for comparison with the design values.

In order to solve this problem the experimental studies were conducted in the mode of MS and the free vibration of the building and the annex in the horizontal impact of the hanging load and vibrations of the ground during the drop of the load.

Prior to the restoration and strengthening works identified the actual periods of natural vibrations of the building during the drop of the load. From these experimental data, the period of natural vibrations of the building in the transverse direction was 0,37 0,43 sec, that can be conditionally accept an average 0,40 sec.

The experimental values of the natural periods of the building building were slightly higher than the design value (Tdesign. = 0,329 sec). This obviously occurred because of the design load for recording vibrations was absent.

The storey weight of the building constituents: mstgbQfbm /2.08,225 , not taking

into consideration the lack of structures of reinforcement, roof, partitions, snow and other temporary loads.

The ratio of actual mass of building to its designed one is equal to:

777,01dbmf

bm for its initial state.

So, experimental natural period of the building in transverse direction before its

reinforcement is equal to sec454,0777,040,0...

trdesbT by the design load and with

the ratio 1.Such a relatively great value of the natural period of the building is quite obvious in

comparison with that one shown in reference and in Building Codes [14], as all cross walls were performed by lightweight concrete smoke-ventilated blocks with lack of connection with each other and longitudinal walls. Besides, it should be assumed that because of the damages during the earthquake, the wall rigidity of the building was slightly reduced.

69

The reason for such a large period of vibration could be another factor. Analysis of the results shows that the upper layer of the ground was enough soft and it is known, that the more is the foundation deformability, the greater is the natural period of the building.

The foundation base of studied building served to be the loams with the interlayer of the sand loam, while the pedestal piles of the annex were fixed in rocky soils. Such a significant difference of soil foundations of the building and the annex was not assumed by calculations, which evidently caused some divergence of experimental and design values of the natural periods.

It is known [15], that the definition of the natural frequencies of the coupled system includes the conception “mistuning” of partial frequencies (the natural frequencies of

constituent parts of the coupled system). It is equal to: 2)1/2( nn , where jn - is the natural frequency of separate systems with a single degree of freedom ( 2,1j ). Depending on the value of “mistuning” the vibrations occur with the frequency of 1n or 2n , and the transition from the frequency 1n to 2n or vice versa is performed by jumping. In works [15 and 16] the particular case is shown, when the masses of separate systems are equal. And we examined the common case, when the masses of separate systems are differed. The following expression is obtained to determine the natural frequencies of the system 1 and

2 :2/1

24111112,1

2 kkkn ,

(1) where:

212 nn - “mistuning”;

2/1jjj mCn - the natural frequencies of the each system;

13 CCk ,

21 mm - coefficient of the ratio of the system masses. It is obvious, that for k = 0 (that is, for the lack of the braces between the system) we have - 211 nn ; 12 n .In the presented instance, proceeding from the results of calculations listen in fig. 2, we

have the following 132,0233,012,02

212

1n2n TT326,070,289495,9421 mm ;

Where 1sec01,101n , 1sec33,522n , and 1T and 2T , 1m and 2m - are the

natural periods ( sec ), and the revealed masses ( mt /2sec. ) of the annex and the building. It is easy to prove that the curves kj are the parabolic equation. From (1) it is clear, that

2 > 1n > 1 > 2n >0.

70

By rising the coefficient of the brace k , the values of the natural frequencies of the coupled system are removing more from the partial ones ( 1n , 2n ). In addition, the first frequencies of the vibrations asymptotically tend to the certain constant value, and the second one – goes into infinity. And the system itself becomes one-mass with a single degree of freedom and for a given case 1

1 sec75,30 or Hzf 9,41 and sec204,01T , and for a design value of the coefficient 72,6k , that is, for the rigidity of

braces 11 sec08,30 or Hzf 78,41 , sec209,01T , 1

2 sec0,163 or Hzf 94,252 , sec038,01T by design.

Summarizing the above-shown theoretical analysis of the vibrations of the coupled system we have the following its 3 possible states, depending on the value of the rigidity of braces:

1. The brace between the systems is lack or its rigidity is less, that is, 0k . In this case we have 2 separate independent systems with the frequencies:

121 sec01,19n ; Hzf 03,31 ; sec33,01T .

112 sec33,52n ; Hzf 33,82 ; sec12,02T .

2. Separate systems are connected by actual brace, that is, k has two degrees of freedom and correspondingly two natural frequencies, being differed from partial 1n and

2n :1

1 sec08,30 ; Hzf 78,41 ; sec209,01T .1

2 sec0,163 ; Hzf 94,252 ; sec0385,02T .3. Separate systems are jointed by rigid braces, that is k . Then the coupled system

becomes one-mass with a single degree of freedom, the vibration frequency of which is determined by the formula:

2/1

1

2/1

21

211 1

nmmCC

(2)

And it is equal to 11 sec75,30 ( Hzf 90,41 ; sec204,01T ).

From (2) it follows, that the frequency of the vibrations of the coupled system depend both on “mistuning” ( ) and on the ratio of the masses ( ) of separate systems.

Experimental and theoretical studies showed that for design values k and k the values of the vibration period are approximately equal. It can be explained by the fact that the assumed values of the rigidity and the levels of the locations of the joints turned out to be optimum along the whole height of the system.

So, despite the fact, that the results were obtained, proving design reduction of the level of seismic load on the building, there is a chance for optimization of the system and improvement of the structural decision, which can be achieved by the choice of the most effective rigidity of joints, their location and structural decision. The obtained theoretical values of the frequencies of vibrations of the building and the annex in various states of the coupled system have also been registered during the experimental tests. For clear similarity

71

and analysis of experimental and theoretical results of data proceeding of vibrations of the system “Building-Multistory Annex” are listed in transverse direction. When the impact effect the most registration points have the following predominant vibration frequency

Hzyf 1,8;2,48,3 ( sec26,024,0;12,0yT ) on the 1-st and 2-nd phases of experimental study. Based on this, the following values of occurred predominant natural periods can be revealed for the system “Building-Multistory Annex” and its separate parts – the building and the annex:

State 1 – 0,121sec for the annex and 0,342sec for the building; State 2 – 0,19 0,26sec (realized coupled system); State 3 – 0,16 0,18sec (as a system with a single degree of freedom). However, the actual state of the building, the annex and their joints are characterized by

them not providing by the design. Therefore, taking into consideration the fact that the masses of the building and the annex were slightly less during the experimental tests in comparison with the design ones, let’s corrects the above-mentioned average values of the natural periods. Then we shall have the following experimental and theoretical values of the natural periods, complying with the accepted state of the system:

sec124,0022,1122,0sec;359,0052,1342,0 anTbT .

sec197,02/022,1052,119,0sysT .

sec270,02/022,1052,126,0sysT .The verified average experimental and theoretical values of the natural periods can be

used as the criterion for the calculation of the true performance of designs and so, to prove the efficacy of the examining system “Building-Multistory Annex”, accepted by the project.

Let's note, that in the 2-nd state the value k = 0,3 1000 corresponds to the diapason of the vibration period 0,197 0,270sec. And according to the project for realized joints the value k complies with 6,7.

Analyzing the given, we come to the conclusion though the second diapason of the vibration period is close to the actual one, it's rather large.

Let's average the values, and in the result we obtain the following:

sec233,02/207,0197,0exp.sysaverT .

The difference of the estimated average experimental value of the vibration period of the system together with the design one wills constituent the following:

%3,13100233,0/202,0233,0 ,This also says about enough satisfactory convergence of experimental and theoretical

data with the results of the design system. The reduction of the design seismic load more than

0,39,3781136.. systbQcertbQ times is achieved by the evaluation (Fig. 2) instead of twice that one given for designing.

As for the behavior of the system in longitudinal direction, it should be noted, that according to a number of records the predominant vibration period both of the building and of the annex separately and together with the system were in the diapason of 0,28 0,32sec. Or approximately their vibration period was equal to yT =0,3sec., which complies with its design value.

72

It means that in this case the vibration of the building and the annex shouldn't be “antiphasic” during the seismic effects in transverse directions and so the occurrence of any significant extra forces is eliminated in their linking members.

6. CONCLUSION

1. Actually the coupled system “Building - Multistory Annex” fundamentally is a new direction of seismic protection system of buildings, based not on damping or energy absorption, but on redistribution of inertial forces between the building and the new more rigid structure added to it and may be functionally used structure, separated by elastic joints and connected by flexible braces in corresponding levels.

2. As a matter of fact, the system “Building-Multistory Annex” is essentially a kind of the coupled system known in the theory of vibrations. According to obtained experimental eigenvalues of the building and the annex separately and as part of the united system, indicate, that the character of their interaction is defined by the parameters of the braces between them and the proximity of their partial frequencies as it was established by the theory of the Coupled Systems.

3. Analysis of the obtained experimental data for the vibrations of the system “Building-Multistory Annex” it was revealed that there were its 3 possible states depending on rigidity of the braces.

4. According to the results of the calculation, confirmed by experimental and theoretical studies, the estimated value of the transverse force on the building together with the system turned out to be equal to bQ = 379 Ton, being 3 times less its value out of the system, which determines its efficacy and thus the reduction of the seismic risk.

5. As a result, the first experience on using of the structure of seismic protection for concrete residential building showed that strengthening of the building can be performed by industrial method and in a relatively short time. It is characterized by an increasing of the total area of the building by more than 33% while reducing the consumption of the concrete and the steel correspondingly for 25 and 60% per a square meter of the total area, compared with the traditional methods of strengthening.

6. Based on these studies, in comparison with the well-known system of seismic protection, the Coupled System is especially effective for relatively flexible buildings, typical for industrial enterprises, including also especially important buildings and strategic sites.

REFERENCES 1. Building Codes of the Republic of Armenia. Reconstruction, Restoration and Reinforcement of Buildings and Structures. The main items. BCRA 1-4.02-99. Ministry of Urban Development, 2000, Yerevan. /in Armenian/ 2. Recommendations on restoration and reinforcement of bearing structures of residential buildings of the series 1A-450 and 1A-451. ArmEERI, 1991, Yerevan. /in Armenian/

73

3. Recommendations on techniques of injection of polymer composition GIPK-14-32 into the masonry during the restoration and reinforcement of buildings and structures.ARIEE and PC, 1996, Yerevan. /in Russian/ 4. Recommendations on restoration and reinforcement of precast buildings and polymer solutions. M., State Committee of Architecture, TbilZNIIEP, 1990. /in Russian/ 5. Building Codes of the Republic of Armenia. Reconstruction, Restoration and Reinforcement of Buildings and Structures. The main items. BCRA II-6.02-2006. Ministry of Urban Development, 2006, Yerevan. /in Armenian/ 6. Up-to-date methods of seismic protection. Polyakov V.S., Kilimnik L.Sh., Cherkashin A.V. M., Stroyizdat, 1989, 320p. 7. USSR Patent No 1574776 ”Seismic Resistant Multi-storey Building”. Korenev B., Khlghatyan Z., 1990, Moscow. 8. Patent RA No 1313 A2 “Seismic Protection System for Masonry Buildings”. Badalyan R., Khlghatyan Z., 2002, Yerevan. 9. The estimation of buildings vulnerability and experience in application of modern seismic protection systems in Armenia. Internationalni Nauchno-Struchi Skup. Grade-vinarstvo-Nauka i Praksa. Kniga 1, 2006, Zabljak, pp 309-316. 10. On the application of multi mass dynamic vibration dampers for seismic protection of buildings. Korenev B., Khlghatyan Z. M. Proceeding of the 9-th European Conference on Earthquake Engineering. - Moscow, 1990, Moscow, vol.8, pp. 96-103. 11. The methods of seismoprotection of multistoried buildings. Khachiyan E.E., Melkumyan M.G., Khlgatyan Z.M. The International Conference Spitak-88, 22-26 May, 1989. UNESCO and Armenian Academy of Sciences, Publishing House of the Armenian AS, 1989, Yerevan, p.p. 52-53. 12. The System of Seismic Insulation. Melkumyan M.G. Armenian Builders` Bulletin 9,1997, Yerevan, p.p. 6-8. 13. The Behaviour of the compressed reinforcing rods. Badalyan R. Bulletin of the Structural Engineering. 770, October, 1998. M., Publishing House “BSE”, 1998. /in Russian/14. Introduction of the theory of oscillations. Strelkov S.P. GTTL, M. 1951, Leningrad, 341p. 15. The Lectures on the theory of oscillations. Mandelshtamm L.I., M., “Nauka”, 1972, 470p. The theory of oscillation., Andronov A.A., Veett A.A., Khaykin S.E. M., Physmathgiz, 1959,

75

A. H. Barbat111, J. C. Vielma212 and S. Oller3

EVALUATION OF THE SEISMIC SAFETY OF RC BUILDINGS DESIGNED BY USING EUROCODES 2 AND 8

Summary

Modern design codes include prescriptions ensuring a ductile behavior of the elements and of the structure as a whole in order to avoid the sudden collapse of the structures when subjected to strong ground motions. But ductility implies structural damage and it is especially important for the designer to know during the design phase the extent of damage that his structure will undergo under the seismic action specified by the code. This paper evaluates the safety of a set of regular reinforced concrete framed buildings designed according to the EC-2/EC-8 prescriptions using both modal pushover analysis and an incremental dynamic analysis. A seismic global damage index is used and damage thresholds associated with the interstory drift are considered to calculate fragility curves and damage probability matrices. The results show that the response of earthquake resistant buildings designed according to the prescriptions of EC-2/EC-8 ensures that the collapse state is not reached for the specific demand typified by the inelastic design spectra. Key words: Ductility, overstrength, behavior factor, fragility curves, damage index.

OCENA SEIZMI KE SIGURNOSTI ARMIRANOBETONSKIH ZGRADA PROJEKTOVANIH NA OSNOVU EVROKODA 2 I 8

Rezime

Moderni propisi za projektovanje sadrže odredbe kojima se obezbe uje duktilno ponašanje elemenata i cele konstrukcije da bi se izbeglo iznendano rušenje objekata izazvano silnim kretanjima tla usled dejstva zemljotresa. Ali, duktilnost je u vezi sa strukturnim ošte enjima stoga je od posebne važnosti za projektanta da prilikom projektovanja zna koji stepen ošte enja e pretrpeti njegov objekat u uslovima seizmi kih dejstava definisanih propisima. U ovom radu vrši se ocena sigurnosti grupe zgrada armiranobetonske ramovske konstrukcije koje su ....... .....projektovane u skladu sa odredbama EC-2/EC-8 koriš enjem kako modalne pushover analize, tako i inkrementalne dinami ke analize. Koriš en je globalni

1 Technical University of Catalonia, Barcelona, Spain 2 Lisandro Alvarado University, Barquisimeto, Venezuela

76

indeks seizmi kog optere ena i uzete su u obzir granice ošte enja usled me uspratnog drifta za izra unavanje krivih lomljivosti i matrica verovatno eošte enja. Rezultati pokazuju da odgovor zgrada otpornih na zemljotres projektovanih shodno odredbama EC-2/EC-8 obezbe uje da ne do e do stanja u kome dolazi do rušenja za specifi ni zahtev predstavljen nelasti nim projektnim spektrom. Klju ne re i: Duktilnost, pove ana nosivost, faktor ponašanja, krive lomljivosti konstrukcije, indeks ošte enja.

1. INTRODUCTION

Recent advances and developments in the computational tools enabled to develop and to apply more realistic analysis models to evaluate the seismic behavior of new or existent buildings and to take into account main features of the nonlinear seismic behavior of structures, like constitutive laws (plasticity and damage) or large deformations. The non linear analysis has been used in the assessment of buildings designed according to specific design codes [1, 2, 3]. Among the characteristics studied in past works, some examples can be provided: displacement ductility, overstrength and behavior factor. The assessment of these characteristics is possible by applying deterministic procedures in analyzing the nonlinear response of the structures subjected to static or dynamic loads.

The application of the Performance-Based Design concepts required the definition of a set of Limit States, usually starting from engineering demand parameters, such as the interstory drift, the global drift or the global structural damage. These parameters allow defining damage thresholds associated with the Limit States which are applied to calculate the fragility curves and of the damage probability matrices used in the seismic safety assessment of the buildings.

In this paper, the seismic safety of regular reinforced concrete moment-resisting framed buildings (MRB) is studied using static and dynamic nonlinear analysis. Sixteen buildings have been designed according to EC-2 and EC-8 for high ductility class (behavior factor equal to 5.85), with 3, 6, 9 and 12 numbers of stories and 3, 4, 5 and 6 spans, covering a low to medium structural period range and also the relevant range of structural redundancy. The seismic demand is obtained for the B Soil type design spectrum (stiff soil) and for a peak ground acceleration of 0.3g.

The static analysis is performed by means of pushover procedures while the dynamic analysis is performed using the Incremental Dynamic Analysis (IDA). The analysis was performed using the PLCd program [6] which allows incorporating the main characteristics of the rein-forcement and of the confinement provided to the structure members. The results of the non-linear analysis allow calculating the displacement ductility, the overstrength and the behavior factors of the structures. The latter are compared with those prescribed by EC-8. The global performance of the buildings is evaluated using an objective damage index based on the capacity curve. Finally, for the predefined damage thresholds, fragility curves and damage probability matrices are calculated.

77

Normalized roof drift ( /H) %

Bas

e sh

ear

(V/W

)

0 1 2 3 40

0.15

0.3

0.45

y=0.48 u=3.05

Design base shear coefficient, Vd/W=0.15

Yield base shear coefficient, Vd/W=0.34

RR=0.34/0.15=2.27

=3.05/0.48=6.35

1. NON-LINEAR analysis of the buildings

1.1. Characteristics of the computational model The non-linear static analysis with force control was performed using the PLCd

finite element code [7, 8] which allows using two and three-dimensional solid elements as well as prismatic, reduced to one-dimensional, members. This code provides a solution combining both numerical precision and reasonable computational costs [9, 10]. It can deal with kinematics and material nonlinearities. It uses various 3-D constitutive laws to predict the material behavior (elastic, visco-elastic, damage, damage-plasticity, etc. [11]) with different yield surfaces to control its evolution (Von-Mises, Mohr–Coulomb, improved Mohr–Coulomb, Drucker–Prager, etc. [12]). Newmark’s method [13] is used to perform the dynamic analysis. A more detailed description of the code can be found in Mata et al. [9, 10]. The main numerical features included in the code to deal with composite materials are: 1) Classical and serial/parallel mixing theory used to describe the behavior of composite components [14]. 2) Anisotropy Mapped Space Theory enables the code to consider materials with a high level of anisotropy, without the associated numerical problems [15]. 3) Fiber–matrix debonding which reduces the composite strength due to the failure of the reinforced–matrix interface [16].

1.2. Non-linear static analysis To evaluate the inelastic response of the structures, pushover analysis was

performed applying a set of lateral forces representing seismic actions corresponding to the first vibration mode. Before the structure is subjected to the lateral loads simulating seismic action, it is first subjected to the action of gravity loads, in agreement with the combinations applied in the elastic analysis.

Although it is difficult to find a method to obtain the global yield and ultimate displacements [17] in this paper a simplified procedure is applied. The nonlinear static response obtained via finite element techniques is used to generate an idealized bilinear shaped capacity (see Figure 1), which has a secant segment from the origin to a point on the capacity curve that corresponds to a 75% of the maximum base shear [18]. The second segment, which represents the branch of plastic behavior, was obtained by finding the intersection of the aforementioned segment with another, horizontal segment which corresponds to the maximum base shear. The use of the compensation procedure guarantees that the energies dissipated by both nonlinear models are equal.

Figure 1. Idealized capacity curve of the 9 stories MRB There are two variables that characterize the quality of the seismic response of buildings.

78

The first is the displacement ductility , defined as

y

u (1)

where y is the yield drift and u the ultimate drift. The second variable is the overstrength RR of the building, which is defined as the ration of the design base shear Vd to the yielding base shear Vy, both of which are shown in Figure 5

d

yR V

VR (2)

Based on the idealized bilinear curve of this figure, a displacement ductility of 6.35 and an overstrength of 2.27 are obtained. The ductility is higher than that specified in the EC-8 seismic design code, which is 5.85, showing that the MRB designed according to EC-2 and EC-8 have ductile response to seismic actions and adequate overstrength.

1.3. Damage index

A local damage index D is calculated using the finite element program PLCd with a damage and plasticity constitutive model that enables correlation of damage with lateral displacements [19]

0

1in

in

PD

P (3)

where inP

and 0inP

are the norm of current and elastic values of the internal forces vectors, respectively. Initially, the material remains elastic and D=0 but, when the entire

energy of the material is dissipated, 0inP

and 1D . It is useful to know the level which the damage reaches when a structure suffers a certain demand. This is possible if the damage index is normalized respecting the maximum damage which can occur in the

structure. Vielma et al. [20] proposed a capacity curve-based damage index PobjD

which allows assessing the damage level for a specific roof displacement. This objective damage

index 10 P

objDreached by a structure for a given drift corresponding to a point P of

the capacity curve is defined as

1)/1(1 0KK

DD PP

Pobj (4)

For example, P might be the performance point resulting from intersection between the

inelastic demand spectrum and the capacity curve and PK the stiffness corresponding to

this point. Other parameters are the initial stiffness 0K and the displacement ductility ,

calculated with the yield displacement *y which corresponds to the intersection of the

79


Dam

age

inde

x

a)

0 1 2 3 40

0.2

0.4

0.6

0.8

1

Number of spans3_out3_inn4_out4_inn

5_out5_inn6_out6_inn


Dam

age

inde

xb)

0 1 2 3 40

0.2

0.4

0.6

0.8

1




Dam

age

inde

x

c)

0 1 2 3 40

0.2

0.4

0.6

0.8

1




Dam

age

inde

x

d)

0 1 2 3 40

0.2

0.4

0.6

0.8

1



initial stiffness with the maximum shear value (see Figure 2). Figure 3 shows the evolution of the objective damage index with respect to the normalized roof drift, computed for all the frames of the studied buildings.

Figure 2. Parameters used in the calculation of the objective damage index

Figure 3. Evolution of the damage index a) 3 story, b) 6 story, c) 9 story and d) 12 story buildings

80

1.4. Incremental Dynamic Analysis

In order to evaluate the dynamic response of the buildings, the Incremental Dynamic Analysis (IDA) [21] was applied. This procedure consists in performing time-history analyses using real or artificial accelerograms, which are scaled each time in order to induce increasing levels of inelasticity in the structural model. A set of six artificial accelerograms compatible with soil type B of the EC-8 design spectrum were generated. The collapse point is reached when the capacity of the structure drops. A usual criterion is to consider the slope of the curve less than the 20% of the elastic slope [21, 22]. Table 1 summarizes the computed average values of the collapse points for all the studied cases. Dynamic analysis is useful in assessing the collapse point of the buildings. The values of the behavior factors q have been obtained by means of the following equation [2]:

yielddesigng

collapseg

aa

q_

(5)

where collapsega and yielddesignga _ are the collapse and the yield design peak ground

acceleration, respectively. collapsega is obtained from the IDA curves and yielddesignga _ is

calculated from the elastic analysis of the building. Average values of the computed behavior factor q of the studied buildings are show in Table 2 and are compared with the behavior factors prescribed by the design codes.

Table 1. Normalized roof displacement (%) at the collapse of the structures

Static analysis Dynamic analysis (average)

3 2.51 2.51 6 2.63 2.63 9 2.48 2.62 12 2.35 2.39

Table 2. Behavior factors of the buildings qequation(Average) qcode qequation/qcode

3 17.40 5.85 2.97 6 10.79 5.85 1.84 9 15.07 5.85 2.57 12 15.12 5.85 2.58

The computed behavior factors show that the applied code allows designing structures with satisfactory lateral capacity when they are subjected to strong ground motions, regardless of the building height. The relationship between the calculated and the prescribed behavior factors is close to three for the case of low rise buildings.

81

1 SEISMIC SAFETY OF THE BUILDINGS 1.1 Calculation of the performance point

As the main objective of this paper is to study the seismic safety of buildings designed according to the Eurocodes, it is necessary to define a measure of the engineering demand. The global drift of the structure corresponding to the performance point has been selected herein to assess the seismic safety of the buildings. The performance point permits establishing the maximum drift of an equivalent single degree of freedom model produced by the seismic demand. It is determined by using the N2 procedure [23] which requires transforming the capacity curve into a capacity spectrum expressed in terms of the spectral

displacements dS and of the spectral acceleration aS ; dS is obtained as

MPFS c

d

(6)

where c is the roof displacement and MPF the modal participation factor obtained from the response of the first mode of vibration

n

iii

n

iii

m

mMPF

1

2,1

1,1

(7)

In this equation, mi is the mass I and 1,I is the spectral ordinate. The spectral

acceleration aS is given by

WV

Sa

(8)

where V is the base shear, W is the seismic weight and is the coefficient

n

iii

n

iii

m

m

1

2,1

2

1,1

(9)

The values of the spectral displacements corresponding to the performance point are shown in Table 3. An important feature in the non linear response of the buildings is the ratio between the performance point displacement and the ultimate displacement. This ratio indicates whether the behavior of a structure is ductile or fragile. The lower values of this ratio correspond to the 12 story building, which has a weak-beam strong-column failure mechanism.

82

Table 3. Roof drift of performance points for the studied buildings Normalized roof drift (%) Ratio

Storynumber Performance point %) Static analysis

Dynamic analysis(average) Static analysis

Dynamic analysis(average)

3 0.80 3.02 2.51 0.26 0.32 6 0.51 2.48 2.63 0.20 0.19 9 0.39 2.48 2.62 0.16 0.15 12 0.21 2.34 2.39 0.09 0.09

The fragility curves are obtained by using the spectral displacements determined for the damage thresholds and considering a lognormal probability density function for the spectral displacements which define the damage states [27]

2

ds,d

d

dsddsd S

Sln121exp

2S1)S(F (10)

where ds,dS is the mean value of the spectral displacement for which the building reaches

the damage state threshold sd and ds is the standard deviation of the natural logarithm of

the spectral displacement for the damage state sd . The conditional probability )S(P d of

reaching or exceeding a particular damage state sd , given the spectral displacement dS , is defined as

dS

ddd SdSFSP0

)()()( (11)

Figure 4 shows the fragility curves calculated for the four different heights of buildings considered in the analysis. Figure 5 shows the damage probability matrices calculated for the performance points corresponding to all the studied cases. It is worth to observe that the probabilities are not sensitive to the variation of the span number. It is also important to note that for the frames of the same building, the probabilities vary according to the load ratio (seismic load/gravity load).

Another important feature which can be observed in the obtained results is the increase of the probability values that correspond to the higher damage states for low rise buildings, for which the collapse is associated to the soft-storey mechanism as discussed in previous sections. For example, in the case of the inner frames of the 3 levels building, the probability to reach the collapse is four times higher than in the case of the outer frames of the same building. In contrast, the 6, 9 and 12 story buildings show very low probabilities to reach more severe damage states regardless of the load ratio and of the span number. For these buildings, the predominant damage states are the non-damage and the slight damage.

83

Spectral displacement Sd (m)

FD=P

rob.

(ED

> ed

i/Sd=

S di)

0 0.03 0.06 0.09 0.12 0.15 0.18 0.21 0.24 0.27 0.30

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

ND

SL

R

E

ST

C

ND: No damageSL: SlightR: RepairableE: ExtensiveST: StabilityC: Collapse


FD=P

rob.

(ED

> ed

i/Sd=

S di)

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.50

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

ND

S

R

E

ST

C


FD=P

rob.

(ED

> ed

i/Sd=

S di)

0 0.06 0.12 0.18 0.24 0.3 0.36 0.42 0.48 0.54 0.60

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

ND

S

R

EST

C


FD=P

rob.

(ED

> ed

i/Sd=

S di)

0 0.08 0.16 0.24 0.32 0.4 0.48 0.56 0.64 0.72 0.80

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

ND

S

R

EST

C

a)

d)

c)

b)

Figure 4. Fragility curves of the a) 3 story, b) 6 story, c) 9 story and d) 12 story buildings

84

3

456

00.20.40.60.8

1 Spans

Probability

a)

3

456

00.20.40.60.8

1

b)

3

456

00.20.40.60.8

1

c)

3

456

00.20.40.60.8

1

d)

3

456

00.20.40.60.8

1

e)

3

456

00.20.40.60.8

1

f)

3

456

00.20.40.60.8

1

g)

3

456

00.20.40.60.8

1

h)

Figure 5. Damage probability matrices of a) outer and b) inner frames of 3 story buildings, c) outer and d) inner frames of 6story buildings, e) outer and f) inner frames of 9 story buildings and g) outer and h) inner frames of 12 story buildings

85

a) b)

Stories

Dam

age

inde

x

3 6 9 120

0.2

0.4

0.6

0.8

1Number of spans

3456

Stories

Dam

age

inde

x

3 6 9 120

0.2

0.4

0.6

0.8

1Number of spans

3456

Figure 14. Damage index computed for the a) outer frames and b) inner frames

Figure 14 shows the values of the damage index corresponding to the performance point of the different frames. These values were obtained using Eq. 3. First of all, it is possible to observe that low rise buildings (3 levels) reach higher values of the damage index than the other buildings; this is a consequence of the failure mechanism which occurs for this kind of buildings (soft story mechanism). In contrast, the 12 levels buildings exhibits damage index about 0.3 to 0.35 for the outer frames, values that are consistent with the failure mechanism (strong columns-weak beams) and with the probabilities obtained from the fragility curves. Finally it is important to observe that the values of the damage index of the outer frames are lower than the values of the damage index of the inner frames, indicating that the damage index depends on the load ratio.

CONCLUSIONS The local damage distribution of the buildings corresponding to the collapse

threshold shows that low rise buildings have a failure mechanism associated to the formation of the soft storey mechanism. The buildings with 6, 9 and 12 story) exhibit a failure mechanism associated to the weak-beam and strong-column conceptual design objective.

Reinforced concrete framed buildings, designed according to the Eurocodes for a high ductility class show adequate values of overstrength which are greater than the value prescribed in the code (1.5). Behavior factors obtained by means of the dynamic analysis are also adequate and are twice the code values. In the procedure applied to evaluate such factors, no influence of the structural redundancy was detected.

Generally speaking, the studied buildings show adequate ductility behavior, as it is evidenced by the displacement ductility values and by the ratio between the performance point and the ultimate displacement.

The nonlinear response of the buildings depends on the ratio between the seismic and gravity loads. The inner frames which are designed for lower ratios have lower overstrength values. Consequently, the seismic safety of the different frames is influenced by this ratio.

The assessment of the seismic safety of the buildings demonstrates that low rise buildings reach higher damage states than the other studied buildings, when they are subjected to the demand prescribed by the elastic design spectrum. This fact is a

86

consequence of the failure mechanism of the low rise buildings. The probability of damage is not sensitive to the span number of the frames.

ACKNOWLEDGMENTS Most of the developments included in this work are consequences of several

research projects. The institutions and companies responsible of these projects are gratefully acknowledged. These are, CEE–FP6 (LESSLOSS project, ref. FP6-50544(GOCE)); the Spanish Goverment through the Ministerio de Ciencia y Tecnología (RECOMP project, ref. BIA2005-06952, DECOMAR project, ref. MAT2003-08700-C03-02 and DELCOM project, ref. MAT2008-02232/MAT) and the Ministerio de Fomento (project “Retrofitting and reinforcement of reinforced concrete structures with composite materials. Numerical and experimental developments applied to joint of bars and composites anchorage proposal”); AIRBUS through the project FEMCOM and ACCIONA Infraestructuras through the projects SPHERA, CETIC, PROMETEO.

REFERENCES Mwafi, A.M. Elnashai, A. Overstrength and force reduction factors of multistory reinforced-concrete buildings, Structural design of tall buildings, 11, 329-351, 2002. Mwafi AM, Elnashai A. Calibration of force reduction factors of RC buildings, Journal of Earthquake Engineering, 6(2), 239-273, 2002 Sanchez, L. Plumier, A. Parametric study of ductile moment-resisting steel frames: a first step towards Eurocode 8 calibration, Earthquake engineering and structural dynamics, 37,1135-1155, 2008. Comité Européen de Normalisation (CEN). Eurocode 2: Design of concrete structures. BS EN 1992, Brussels, 2001. Comité Européen de Normalisation (CEN). Eurocode 8: Design of Structures for Earthquake Resistance. EN 2004-1-1, Brussels, 2003. PLCd Manual. Non-linear thermo mechanic finite element oriented to PhD student education, code developed at CIMNE, Barcelona, Spain, 2009. Oller, S. and Barbat, A. H. (2006). Moment-curvature damage model for bridges subjected to seismic loads, Computer Methods in Applied Mechanics and Engineering, 195, 4490-4511, 2006. Car, E., Oller, S. and Oñate, E. A large strain plasticity for anisotropic materials: Composite material application, International Journal of Plasticity, 17(11), 1437-1463, 2001. Mata, P., Oller, S. and Barbat, A.H. Static analysis of beam structures under nonlinear geometric and constitutive behaviour, Computer Methods in Applied Mechanics and Engineering, 196, 4458-4478, 2007 Mata, P., Oller, S. and Barbat, A. H. Dynamic analysis of beam structures under nonlinear geometric and constitutive behaviour, Computer Methods in Applied Mechanics and Engineering, 197, 857-878, 2008. Oller S, Onate E, Oliver J, Lubliner J. Finite element non-linear analysis of concrete structures using a plastic-damage model, Engineering Fracture Mechanics, 35(1–3), 219–31, 1990. Lubliner, J. Oliver, J. Oller, S. Oñate, E. A plastic-damage model for concrete, International Journal of Solids & Structures, 25(3), 299–326, 1989

87

Barbat, A.H. Oller, S. Onate, E. Hanganu, A. Viscous damage model for Timoshenko beam structures, International Journal of Solids & Structures, 34(30), 3953–76, 1997. Faleiro, J. Oller, S. Barbat, A.H. Plastic-damage seismic model for reinforced concrete frames, Computer and Structures, 86, 581–97, and 2008. Oller S., Car E., Lubliner J. Definition of a general implicit orthotropic yield criterion, Computer Methods in Applied Mechanics and Engineering, 192(7–8), 895–912, 2003. Martinez, X. Oller, S. Rastellini, F. Barbat, A.H. A numerical procedure simulating RC structures reinforced with FRP using the serial/parallel mixing theory, Computers and Structures, 86, 1604–1618, 2008. Priestley, M.J.N. Calvi, G.M. Kowalsky, M.J. Displacement-based seismic design of structures. IUSS Press. Pavia, Italy, 2007. Park, R. State-of-the-art report: ductility evaluation from laboratory and analytical testing, Proceedings 9th WCEE, IAEE, Tokyo-Kyoto, Japan, 605-616, 1988. Barbat, A. H. Oller, S. Oñate, E. Hanganu, A. Viscous damage model for Timoshenko beam structures, International Journal of Solids and Structures, 34(30), 3953-3976, 1997. Vielma, J. C. Barbat, A. H. Oller, S. Un índice de daño objetivo para la evaluación de los edificios de hormigón armado. Hormigón y acero. 248, 53-64, 2007. Vamvatsikos. D, Cornell, C.A. Incremental dynamic analysis, Earthquake Engineering and Structural Dynamics, 31(3), 491-514, 2002. Han, S.W. Chopra, A. Approximate incremental dynamic analysis using the modal pushover analysis procedure, Earthquake Engineering and Structural Dynamics, 35(3): 1853-1873, 2006. Fajfar, P. A. Nonlinear Analysis Method for Performance Based Seismic Design, Earthquake Spectra, 16(3), 573-591, 2000. SEAOC. Vision 2000 Report on Performance Based Seismic Engineering of Buildings,Structural Engineers Association of California, Volume I, Sacramento, California, 1995. Vielma, J. C., Barbat, A. H. y Oller, S. Umbrales de daño para estados límite de edificios porticados de concreto armado diseñados conforme al ACI-318/IBC-2006, RevistaInternacional de Desastres Naturales, Accidentes e Infraestructura, 8(2), 119-133, 2008. Vielma, J. C. Caracterización de la respuesta sísmica de edificios de hormigón armado mediante la respuesta no lineal, PhD Thesis, Barcelona, ISBN: 978-84-691-3475-7, 2008. Pinto, P.E. Giannini, R. Franchin, P. Seismic reliability analysis of structures. IUSS Press, Pavia, Italy. 2006.

89

113, 214

:

,. ,

, -, -

, , ,, . , ,

, , ,.

, -.

: , ,

THE ROLE AND THE SIGNIFICANCE OF CONSTRUCTION IN REDUCTION OF SEISMIC RISK

Summary:

All construction activities in completing projects of construction structures and infrastructural and other systems in seismic active regions are presented and analyzed in the paper. The significance of constructors are pointed out and also their responsibility in exploring relevant seismic and other parameters in optimal spatial city planning and seismic designing and constructing, as the most significant factors in seismic risk reducing is emphasized, i.e. in protecting people, structures, equipment from earthquake impacts. All these assignments before, during and after an earthquake can be successfully completed only by very developed and well-organized construction industry. Outstanding importance is attached to preventive measures, especially to those that should be implemented by the inhabitants of quake areas. Key words: Seismic Risk, Building Organization, Safeguard

1 , , . . ., , , [email protected] 2 , . . ., , - , ,

[email protected]

90

1

. , ,, , , -

, ,.

, , ,.

, , , -,

., -

, , ,.

. – ,

,, , .

, ,,

, , , ,, ...

. „ “ [9].

, ,-

– .

, , -, .

, , -, , .

,

. , , ,, ,

, . ( : Emergency Management; Disaster Management).

,, ( )

, – ,.

91

2

, , ,, , ,

, ,. , , ,

, , -, ,

( ) , , -. , , (

/ ),,

, ,.

- – ,. -

, , , ( , ...) ,, , , . ,

,. - ( )

. , ,

, , , , ,, , , ....

, , – ,, ,

. , ( ), , , ,

.,

– , , , , ,, , ,

, ,, , -

, , ., . ,. , ,

, ,. ,

- , ,, ,

, - . , , -

, ( - , -),

92

( ) , ,- ,

. , ( ) -, ( ), „ “

( ) , -

( , , , , ...). ,

, , ,. , , ,

, ... , ,, , (

.), , , .

, ,. ,

, , , , .,.

, , ( , ) -

, -, , , ,

[10]. . ( 1:1) ,

, . - – .

– , „ -“, -

. , , -, :

. ..

,, ( ),

, . . 50-60 m.

( ) ( )-

. ( ) [18]. , ,

1969. ,,

. , Ts > Tp, Tp –

0,10 p 0,34 sec. ,.

93

( ) ,. , , , ,

, ., (

), ( , , , …), .

. ,. ,

, -, . „ “ ,

, . ( ).

, ( ) – ( ) -

. , ( -).

,, .

, , ,. ,

, ., -

, ( ) , ( ). , , . -

( ) , -, , , .

, , – , .

,, .

,, , .

, ,, .

( ), ( ),, (

). ( , , , ...) ( ), .

.

3

, ,

.

94

,, -

, , , -, . ,

, ,.

20. : (26. 1963. ), , 1.300 80%

, 10% . (26. 27. 1969. )

5%, (15. 1979. ) 10% [14].

, [11,12,19,20,21,30,31]. ,

.

[22],- , ,

., : ,

- , ,.

3.1

, , -, . ,

., ,

. , , -, ,

, . ,,

, ( , .), ,

( ) ., .

.

: VII, VIII IX , -64, , -

, , ,

95

, ( .),

, ,, .

, ,,

( , ,). „ “,

50 ( , , , . -), ( , )

,, ,

, ., -

, , , ,. -

,.

,: (

), ,, ( )

( .), ., ,

,: , (

,, ,

( ), ,. , ,

( ),.

,: , , , . ,

. [19],

( ),, 8 ( 8),

( )[7,8]. YU , ,

( ) , , 8.

; ,, ,

- . ,: , -

96

/. ( – ) ,

,. „ “ ,

,.

-.

– , , , ,, . -

, -. , - / ,

..

-, ,

, ( ), ,

, , . ,, , . -

- ( -

), -.

, .. , , ,

, , , . ,, ,

, , .„ ” .

, , , , -, , , ,

. , -, , , -

, , . ( -, , .), ,

, ,.

( , , ,...). ,, , , , ...

/ ,.

, (.), ,

,, , ( ) . , -

97

, ( ), ,

( ), . ,, ,

, ,, , , .,

, , . – ,

,, ,

– . 1963. ,

, 10 , ,,

,.

, ,, - , ,

, . - , ,

, , ,. ( , ,

– , ...), , , , .,

. ,, , , ,

, – .

( ) , ( 1964. [18]), ( , ,

, .) , , –

/ , – . ,.

FEMA (Federal Emergency Management Agency) 1997.

NEHRP (Natioanl Earthquake Hazard Reduction Program) [32]. FEMA je

,, , ,

,, , , . je

. FEMA 2.600 , 4.000

.

98

FEMA . je FEMA 273 .je FEMA NEHRP

.FEMA-273 u [32]

/ .,

.FEMA je FEMA 276 - NEHRP -

, (FEMA-273) , :

, je - , , ,

.FEMA je (FEMA 274),

- (FEMA 275). FEMA 356 u -

FEMA 273 274 , „ “ .

: - ,, , , ., a

.ANSS (Advanced National Seismic System) [3,36]

,,

. - National

Earthquake Information Center (NEIC) 15 . -, , , ,

, .SRCE (Seismic Rehabilitation Cost Estimator) je online

, - 2.000 -

FEMA 156 - .

1996. je „ ".

, ( ) -

. je ,. – VULPIP [9],

,. - TOSQA [9], je

„ " . je : ( );

99

( ) ( ). je

. / -

EUROSEISTEST [9], -,

.

3.2

,, , ,

,, . -

,. ,

: , - , , - -, , , , .

, () , .

,,

. 155. YU [19],

( ).

, 10 %, /

.. ,

, ,, ( ),

. ,,

,, .

,. , ,

, , ( ),

, [15].

100

4

, -, , ( )

, ( , ,...) , – , ,

, , ... [4,17,23,24,29,35,37,38]. , ,

,. , ,

, ,:

, , –

. , ,, , , , .

,.

,; .

, , .;

, ,; . ( ,

.), , ,, , – . ,

( , .). ,.

( , , , )..

– ;. , -

., ,

, , ., ,

, ,,

[26]. - . , ,

, , ... , ,

, , ,. .

101

. ..

.

. () ,

[6,33,34]. ,

, [25,28].

, .- ,

[22], ., , - , , - ,

, -, . ,

, .

, , ,.

102

[1] / , . // , III - , , 2007, . 13-42.

[2] Aseizmi ko projektovanje upravljanje seizmi kim rizikom / Pavi evi , S. B. // Univerzitet Crne Gore i Gra evinski fakultet Podgorica, 2000.

[3] Advanced National Seismic System // earthquake.usgs.gov/anss [4] Beehive / http://wa.thebeehive.org/emergencies[5] Building valuation / Preiser W. F. E. // Springer, 2000. [6] Department of Conservation, California / www.conservation.ca.gov/index/Earthquakes[7] EC 8 – Zemljotres potresa Jugoslaviju / Alender, V., A i , M., // asopis „Izgradnja“, Beograd,

1996/6[8] Eurocode 8: Design of Structures for Earthquake Resistances. Part 1: General Rules Seismic

Actions and Rules for Buildings, EN 1998-1 / CEN Brussels, 2004 [9] European Commission – http://ec.europa.eu/research/leaflets/disasters/en/earthqu.html[10] Earthquakes Engineering India’s Standards / www.bis.org.in/other /quake.htm [11] / " . ", .

20/77, 24/85, 27/85, 6/89 52/89 " . ", . 53/93, 67/93, 48/94 101/2005 - .

[12] / " .", . 50/92

[13] Zemljotresno inženjerstvo – visokogradnja / Ani i , M., Fajfar, P. . Petrovi , B., Szavitz – Nossan, A., Tomaševi , M. // Gra evinska knjiga, Beograd, 1990.

[14] Zemljotresi – Seizmi ka opasnost i principi zemljotresnog inženjerstva / Paskalov T. // Univerzitet Union, Beograd, 2008.

[15] Nadogradnja stambenih i javnih zgrada / Grupa autora // Zbornik radova sa savetovanja JUDIMK-e, Beograd, 2000.

[16] Organizovanje gra evinarstva u uslovima katastrofalnih zemljotresa / uranovi , P. // Doktorska disertacija , Gra evinski fakultet, Beograd, 1987.

[17] Optimized resource allocation for emergency response after earthquake disasters / Fiedrich, F., Gehbauer F., and Rickers, U. // Institut für Maschinenwesen im Baubetrieb, Universität Karlsruhe, D-76128 Karlsruhe, Germany, 2001.

[18] / " . ", . 39/64

[19]" / . ", . 31/81, 49/82, 29/83, 21/88 52/92

[20]/ " .

", . 34/78 [21] ,

/ " . ", . 52/85 [22]

, 2009. // www.mup.sr.gov.yu/ /sektorzazastituispasavanje

[23] Post-Occupany Evaluation / Preiser W. F. E., Rabinowitz, H. Z. nd White, T., // Van Nostrand Reinhold, New York, 1988.

[24] Risk factors for injuries due to the 1990 earthquake in Luzon / Roces, M. C., White, M. E., Dayrit, M. M., Durkin M. E. / Philippines, Bulletin of the World Health Organization (WHO), 1992 - bases.bireme.br

103

[25] Roundtable Workshop 7: The National Earthquake Hazards Reduction Program at Twenty-Five Years: Accomplishments and Challenges, http://dels.nas.edu/dr/f7.shtml

[26] / http:// prezentacije.mup.sr.gov.yu /sektorzazastituispasavanje /saveti.html

[27] Seismic Design of Reinforced Concrete and Masonry Buildings / Paulay, T., Prestley, M.N.J. // New York, ..., Singapore, 1992.

[28] Training and Education for Improving Earthquake Disaster Management in Developing Countries / Proceedings of the Sixth International Research and Training Seminar on Regional Development Planning for Disaster Prevention // Nagoya, Japan, 14 December 1992, pp. 170

[29] The Behaviour of Building Occupants in Earthquake / Durkin, M. // Earthquake Spectra, Vol 1, No 2, 1985.

[30] / " . ", . 27/87

[31] o / " . ", . 34/78

[32] FEMA-273 / Guidelines and Commentary for Seismic Rehabilitation of Buildings, ATC, 1996.; http://wwvv.fema.gov/plan/prevent/earthquake/nehrp.shtm

[33] Facilities Development Division / California // www.oshpd.ca.gov/fdd[34] Hospital Facilities Seismic Safety Act (HSSA) / California, USA, 1973. [35] http://www.gadr.giees.uncc.edu[36] California Integrated Seismic Network // www.cisn.orft [37] Centers for Disease Control and Prevention, Atlanta, USA / http://www.bt.cdc.gov /disasters

/earthquakes /prepared.asp [38] www.geo.mtu.edu/UPSeis/bda.html

105

Enver MANDŽI ¹,Salko KULUK IJA², Kenan MANDŽI ³, Mustafa HUMO'

OCJENA STANJA POSTOJE IH OBJEKATA KULTURNOG I ISTORIJSKOG NASLIJE A PRIMJENOM REFRAKCIONE SEIZMIKE

RezimeStanje zidanih objekata kulturnog i istorijskog naslije a naj eš e je vidljivo dostupno samo sa jedne otvorene strane. Na in gradnje i vrsta materijala po dubini debelih zidova kamenom zidanih objekata potpuno je nepoznata. Nepoznati su i uslovi ispunjenosti malterom prostora me u pojedina nim elementima ili slojevima kamenom zidanih konstrukcija. Za ocjenu postoje eg stanja i projektovanje uslova sanacije i rekonstrukcije neophodno je izvesti istraživanja nerazornim metodama. Refrakciona seizmika ili mikroseizmika nudi mogu nosti otkrivanja stanja kamenom zidanih konstrukcija po dubini zidova, na razli itim mjestima koja su od interesa. U radu su pokazani rezultati istraživanja provedeni metodama plitke refrakcione seizmike na objektu Sulejmanpaši a kule kod Bugojna. Dat je i prijedlog odre ivanja koeficijenta kompaktnosti zida.

Summary The determination of the conditions of masonary constructions of cultural and historical heritage is most comonly possible only from one open side. The method of construction and the type of the material inside thik wals is completly unknown. The conditions of mortar fillingof space between individual elements or layers of masonry rock constructions are also unknown. For the assement of present state and recovery and reconstruction conditions planing, it is necessary to conduct research with nondestructable methods. Refractional seismic or microseismic method offers the possibilities for determination of conditions on different depht inside masonry rock wals , in different places of interest. In this work the results of research conducted by the shallow refraction seismic on Sulejmanpaši a tower near Bugojno are presented. The suggestion of koeficient of wall compact is also given in this work.

________________________________¹Akademik Enver MANDŽI , prof., dr sc. Akademija nauka i umjetnosti BiH ²Mr sc. Salko KULUK IJA, dipl.inž., INTERPROJEKT doo Mostar ³Dr sc. Kenan MANDŽI , dipl.inž.geologije, Univerzitet u Tuzli ´Mustafa HUMO, dipl.inž.gra ., INTERPROJEKT doo Mostar

106

1.UVOD

Istraživanja na objektima kulturnog i istorijskog naslije a potrebno je provoditi na na in da se dobije dovoljno potrebnih podataka za izradu projekta restauracije i rekonstrukcije a da se pri tim istraživanjima ne ošte uju bilo koji dijelovi objekta. Jedna od metoda koja pruža mogu nost dobivanja kvalitativnih i kvantitaivnih podataka o zidovima i uslovima kompaktnosti zida zidanog kamenom ili nekim drugim materijalom je i primjena refrakcione seizmike na vanjskim otvorenim i naj eš e jedinim prisutnim dijelovima zida u njegovoj kompletnoj postoje oj strukturi. Refrakciona seizmika omogu ava dobivanje prostiranja seizmi kih talasa koji se izazivaju udarcem eki a u zid na strani gdje se izvodi mjerenje. Stabiliziranjem prijemnika – geofona (za prijem longitudinalnih ili transverzalnih talasa) na površini zida omogu ava registriranje vremena nailaska talasa sa razli itih rastojanja gdje se duž zida udara eki em. Iz poznatog vremena nailaska prvog talasa na prijemnik – geofon i rastojanja od ta ke udara do prijemnika – geofona i rastojanja od ta ke udara do prijemnika – geofona, mogu e je izra unati brzinu talasa. Pove anjem rastojanja ta ke udara od prijemnog ure aja – geofona, mogu e je ispitati stanje materijala ugra enog u zid po cijeloj dubini do druge strane zida. Pomijeranjem prijemnika – geofona duž profila ispitivanja i ponavljanjem procedure mogu e je pratiti promjene stanja zida duž profila na kome se ispitivanja provode. Takav metod je i primijenjen na istraživanju stanja kamenih zidova Sulejmanpaši a kule u mjestu Odžak kod Bugojna.

Primjena seizmi ke metode mogu a je i za uslove postavljanja prijemnika – geofona sa unutrašnje strane zide a mjesto udara da bude sa vanjske strane zida. Mogu e je ponavljati operacije sa geofonom u sredini ili na krajevima profila, kao i druge kombinacije ispitivanja.

12. METODOLOGIJA MJERENJA

Metodologija mjerenja podrazumijeva detaljno prilago avanje uslovima na terenu jer su objekti kulturnog i istorijskog naslije a esto na nepristupa nim ili teško pristupa nim terenima. Kod svakog objekta potrebno je detaljno razmotriti i usvojiti: izbor mogu ih i potrebnih profila za ispitivanje, na in mjerenja duž profila, rastojanje me uta kama udara, na in profiliranja, uslove registracije talasa (frekventno podru je).

2.1. Izbor profila

Profili duž kojih se izvodi ispitivanje mogu biti horizontalni (naj eš e) ili vertikalni (rijetko). Naj eš e je za ispitivanje potrebno postaviti montažnu skelu. Za isptivanje na Sulejmanpaši a kuli korištena je skela a profili su bili horizontalni (vidjeti slike). Obzirom da je kula ravnih zidova sa etiri strane, tako su i profili postavljeni da omogu e utvr ivanje sastava i svojstava kamenog zida u zoni konstrukcije koja se oslanja na teren i slobodnih zidova djelimi no zatrpanih urušenim materijalom sa unutrašnje strane kule.

107

Slika 1. Sulejmanpaši a kula

2.2. Na in mjerenja duž profila

Za istraživanje stanja zidova kule primjenjen je jedinstven na in mjerenja. Prijemnik – gefon je stabiliziran na krajnjem dijelu kamenog zida. Ta ka udara eki em ozna i se na svakih 0,5 m rastojanja od prijemnika – geofona, sa udaljavanjem do kraja zida. Profiliranje se nastavlja tako što se prijemnik – geofon pomjeri na 1/3 dužine zida i postupak udara eki em nastavlja po istoj proceduri kao prethodni. Zatim se prijemnik – geofon pomjera na 2/3 dužine zida i procedura udara eki em ponavlja kao i u prethodnim slu ajevima. Proces pomijeranja geofona za 0,5 m i ponavljanje procedure udara eki em nije primijenjen jer je spor i zahtijeva mnogo sati rada pa u ovom slu aju nije primijenjen, što ne zna i da ne treba biti primijenjen gdje god to uslovi i okolnosti dozvoljavaju.

Kompletna opisana procedura izvedena je na sva etiri zida objekta a dobiveni rezultati zadovoljavaju postavljene uslove o neophodnom nivou informacija koje se o ekuju od ove vrste isptivanja.

108

Slika 2. Lokacije seizmi kih ispitivanja

Seizmi ki profil- južna fasada

Seizmi ki profil- sjeverna fasada

Seizmi ki profil- zapadna fasada

Seizmi ki profil-isto na fasada

109

2.3. Uslovi registracije talasa

Za istraživanje korišten je instrument SEISTRONIC RS 100 kanadske proizvodnje, koji predstavlja najsavremeniji instrument koji se danas koristi i za ovu namjenu. Prije po etka svakog mjerenja namješta se odgovaraju e frekventno podru je koje omogu ava eliminaciju smetnji koje nastaju iz drugih izvora. Instrument registrira vrijeme nailaska prvog talasa i omogu ava sagledavanje kompletnog zapisa talasa za dalju analizu. Mjerenje vremena nailaska prvog talasa za longitudinalne i za transverzalne talase omogu ava crtanje dijagrama vrijeme – rastojanje iz koga se izra unavaju brzine u sredinama kroz koje prolazi talas. Što je brzina manja to zna i da u zidu postoje šupljine i obrnuto. Ako postoji prekid, na primjer zidani vanjski dio zida i kameni ili drugi naba ajizme u vanjskog i unutrašnjeg zidanog dijela zida, brzina e se mijenjati pa možemo registrirati dvoslojnu ili troslojnu strukturu zida.

2.4. Dijagrami brzina

Dijagrami vrijeme – rastojanje omogu avaju izra unavanje brzine talasa koji se prostire kroz ispitanu sredinu.

Za svaki od profila na kome je izvedeno mjerenje nacrtan je dijagram koji je pokazao da je zid kompaktna jednoslojna sredina. Dijagrami jasno pokazuju da je to jednoslojna kamena konstrukcija sa jasno definisanim grani nim površinama sa vanjske i unutrašnje strane. Brzine longitudinalnih talasa su ujedna ene, kao brzine transverzalnih talasa koje su niže po vrijednosti u odnosu na longitudinalne:

Longitudinalni talasi Vl = 1900 - 2700 m/s

110

Slika 3. Fotografije lokacija

SULEJMAN PAŠI A KULA KOD BUGOJNA SJEVERNI ZID

REFRAKCIONI PROFIL RF-1 Tok operacija zapad-istok Longitudinalni talasi VL

Slika 4. Karakteristi an dijagram brzine

V1L =2000 (m/s) V2L=291(m/s)

V1L V2L

111

Refrakcioni profil RF-1/1 (longitudinalni talasi)

02468

101214

0 1 2 3 4 5 6 7 8 9 10Rastojanje d (m)

Vrije

me

t (m

s)

Slika 5. Karakteristi an zapis za jedno mjerenje

3. Koeficijent kompaktnosti zida

Za dalja istraživanja na objektima kulturnog i istorijskog naslije a predlažmo ustanovljavanje koeficijenta kompaktnosti zida (KKZ). Ovaj koeficijent bi omogu iosagledavanje jedinstvenog na ina klasifikacije zidova po kompaktnosti bez obzira da li se radi o jednoslojnim, dvoslojnim ili troslojnim kamenim zidovima ili zidovima od drugog materijala.

Koeficijent kompaktnosti zida (KKZ) izražava stepen ispunjenosti zapremine zida stijenskim materijalom i/ili malterom izme u vanjske i unutrašnje grani ne površine zida. KKZ se izražava u procentima. Lingvisti ki se klasificira u pet klasa: od vrlo malog do veoma velikog. Dobije se istraživanjima provedenim refrakcionom seizmikom na osnovu brzina longitudinalnih talasa utvr enih na zidu i uzorku stijenskog materijala od koga je izgra en zid:

KKZ = (Vlz / Vlu) 100 (%).

gdje je: Vlu – brzina longitudinalnih talasa mjerena u monolitu karakteristi ne stijene koja gradi zid. Etalon uzorak se uzima sa dimenzijama 5 x 5 x 5 cm, bez vidljivo prisutnih pukotina u stijenskom materijalu. Vlz – brzina longitudinalnih talasa mjerena u zidu konstrukcije.

Na osnovu dobivenih podataka o brzinama longitudinalnih talasa može se izra unati KKZ i izvršiti ocjena kvaliteta zida na svi mjestima koja su od interesa za koja su

112

mjerenja izvedena. Klasifikacija zidova zidanih konstrukcija može se izvesti po predloženoj kategorizaciji.

Kategorizacija zidova zidanih konstrukcija 1. kategorija KKZ = 0 – 20%

Veoma mali koeficijent kompaktnosti zida Jednostrano ili dvostrano zidani dvoslojni ili troslojni kameni zid male debljine zidanog materijala. Debljina sloja zidanog vanjskog dijela je oko 1/5 debljine zida. Prazan prostor izme u završnih površina zida ispunjen kamenim naba ajem bez slaganja, bez ili sa malo maltera ili ispunjen zemljanim materijalom. Kod rekonstrukcije potrebno je injektiranje sa ve om potrošnjom injekcione mase, razmak izme u injekcionih bušotina je ve i, materijal za injektiranje vezivo sa punilom od sitnog do prašinastog pijeska ili vezivo sa kamenim brašom.

2. kategorija KKZ = 20 – 40 % Mali koeficijent kompaktnosti zida Zid je jednostrano ili dvostrano zidani dvoslojni ili troslojni kameni zid ve e de-bljine zidanog materijala na vanjskim površinama. Debljina sloja zida vanjskog dijela je ¼ debljine zida. Prazan prostor izme u vanjskih slojeva zida ispunjen je kamenim naba ajem, bez slaganja sa malo ili bez maltera, bez slaganja. Kod rekonstrukcije poželjno injektiranje vezivim materijalom sa malo punila (kameno brašno). Razmak izme u injekcionih bušotina manji.

3. kategorija KKZ = 40 – 60 % Srednji koeficijent kompaktnosti zida Zid je jednostrano ili dvostrano zidani dvoslojni ili troslojni kameni zid sa ve om debljinom zidanih slojeva na vanjskim površinama. Debljina zida vanjskih slojeva je po 1/3 ukupne debljine zida. Prazan prostor izme u završnih površina ispunjen kamenim naba ejem bez slaganja, bez ili sa malterom. Kod rekonstrukcije može se koristiti injektiranje sa vezivim materijalom ve e ili manje migrabilnosti. Razmak izme u injekcionih bušotina manji. Injekcioni ma-terijal bez bilo kakvog punila.

4. kategorija KKZ = 60 – 80 % Veliki koeficijent kompaktnosti zida. Zid je jednoslojno ili dvoslojno ili troslojno zidan kameni zid sa vezivom me uslojem od zidanog materijal sa vezivom kod troslojnog zida. Kod jednoslojnih zidova unutrašnjost je od ne obra enog kamena sa malterom ali sa šuplinama u kojima nema maltera. Kod dvoslojnih zidova samo su vanjeske površine zidane obra enim kamenom a unutrašnjost od ne obra enog kamena sa malterom i sa šupljinama izme u pojedina nih komada kamena. Kod troslojnih zidova vanjski slojevi su zidani obra enim kamenom dok je unutrašnjost zidana neobra enim kamenom sa malterom ali sa šupljinama izme u. U troslojnim zidovima nema kamenog naba aja.Kod rekonstrukcije injektiranje se primjenjuje sa malim razmakom bušotina. Injekciona masa mora biti veoma migrabilnom sa visokim stupnjem pentrabilnosti zbog teškog prodiranja u šupljine koje me usobno ne komuniciraju.

113

5. kategorija KKZ = 80 – 100% Veoma veliki koeficijent kompaktnosti zida. Zid je jednostrano ili dvostrano zidan obra enim kamenim blokovima kao jednoslojna ili dvoslojna konstrukcija sa malterom koji potpuno popunjava prazne prostore izme u pojedina nih blokova. Kod rekonstrukcije injektiranje nepotrebno.

Istraživanja provedena na terenu mjerenjem brzina longitudinalnih talasa na zidovima Sulejmanpaši a kule u Odžaku kod Bugojna i na uzorcima karakteristi nog stijenskog materijala, od koga je zidana kula, svrstavaju 90 cm debele zidove kule u kategoriju sa veoma velikim koeficijentom kompaktnosti, KKZ = 83 – 88 %.

12. Zaklju ak

Primjena nerazornih metoda istraživanja zidanih konstrukcija u svijetu sve više dobiva na zna aju. Me unarodna udruženja koja posve uju punu pažnju rekonstrukciji i restauraciji objekata kulturnog i istorijskog naslije a daju smjernice i preporuke za upotrebu ovih metoda. Obzirom da se u BiH pristupa sve više istraživanju stanja i uslova mogu eg saniranja ovih objekata neophodna su prethodna istraživanja koja e dati uslove kako provesti sanaciju, rekonstrukciju i restauraciju takvih objekata a da se ne doprinese daljem razaranju tih objekata.

Primjena refrakcione seizmike za istraživanje stanja zidova objekata kulturnog i istorijskog naslije a je u BiH vrlo rijetko korištena, pa e po etni, veoma ohrabruju i,rezultati pokazali svu opravdanost korištenja ove metode. Predložena klasifikacija i kategorizacija kamenih zidova po veli ini koeficijenta kompaktnosti zida (KKZ) može zna ajno pomo i kod izbora metoda sanacije, rekonstrukcije i restauracije. Istraživanja provedena za zidove Sulejmanpaši a kule u Odžaku kod Bugojna pokazala su i dokazala mogu nosti primjene metode seizmi kog profiliranja na utvr ivanje stanja kamenih zidova. Na istoj kuli provedena su i geoelektri na istraživanja koja su potvrdila dobivene rezultate.

114

Literatura

1. European Commission, 2006, Onsiteformasonary project, On-site investigation techniques for the structural evaluation of historic masonary buildings, EUR 21696 EN

2. Grupa autora, 2009, Istraživanja stanja zidova Sulejmanpaši a kule kod Bugojna refrakcionom seizmikom, Interprojekt Mostar, ne publikovano

115

Radomir Foli 115,Lidija Babi 216

PARAMETARSKA SEIZMI KA ANALIZA AB DIMNJAKA

Rezime:Zemljotresi su naj eš i i najopasniji uzrok velikog broja ošte enja i rušenja

dimnjaka. Odgovaraju i seizmi ki prora un je neophodan da bi se umanjili mo-gu i rizici. U radu se razmatra ponašanje visokih industrijskih dimnjaka, izloženih seizmi kim dejstvima. Prikazane su metode seizmi ke analize i preporuke date ra-zli itim savremenim propisima. Na osnovu parametarske dinami ke analize izve-denog dimnjaka, upore ene su teorijske vrednosti sa rezultatima eksperimentalnih ispitivanja. U radu se komentariše opravdanost primene pojedinih metoda.

Klju ne re i: seizmi ka dejstva, parametarska analiza, AB dimnjaci, projektovanje

PARAMETRIC SEISMIC ANALYSIS OF RC CHIMNEYS

Summary:

Earthquakes are the most frequent and the most dangerous cause of a large number of damaged and destroyed chimneys. A proper seismic calculation is requisite for risk mitigation. The paper deals with the behaviour of tall industrial chimneys exposed to seismic actions. It displays methods of seismic analysis and recommendations of various provisions of contemporary codes. According to parametric dynamics analysis, theoretical values are compared with experimental results. The paper points out that the application of some methods is justified.

Key words: seismic action, parametric analysis, RC chimneys, design

1 PhD, Professor emeritus, Faculty of technical science, Novi Sad, Serbia 2 MrSci,assistant, Faculty of technical science, Kosovska Mitrovica, Serbia

116

1 INTRODUCTION

Rapid industrial development and building of industrial systems set up new requirements regarding design and construction of chimneys. The application of modern building materials and improvement of construction technology require the choice of adequate mathematical models and more accurate analyses. The design of seismically resistant structures is particularly important for high structures including industrial chimneys. It is the subject of numerous recommendations and technical regulations, among which are European Norms (EN 1998 – Part 6) that are used in designing of such structures in the European Union.

Strong earthquakes cause great damage on structures, (Fig 1) and (Fig 2). Chimneys are particularly sensitive to seismic actions, not only those on residential buildings but industrial chimneys as well.

At the international conference called The International Symposium on Chimney Design held in Edinburgh 1973, the problems in design and construction of industrial chimneys were discussed, together with diverse regulations dealing with the topic [10]. A body was formed, Comité International des Cheminées Industrielles (CICIND), which even now follows and directs the development of regulations in different countries. In 1982 a report named “Proposal for a Model Code for the Design of Chimneys” was published. However, suggested regulations were not accepted by CEB (Comité Euro-International du Béton) for they were not based on the limit state analysis.

2 SURVEYS OF SOME NORMS FOR DYNAMIC CALCULATION OF CHIMNEYS

There are four types of the seismic analysis of structures classified as: linear static, non-linear static, linear dynamic and non-linear dynamic analysis [2].

Fig. 1. Explosive dismantling sequence at the Italsider Bagnoli plant, Naples, Italy, after [6]

117

a.) b.)

Fig. 2. a.) Felling of the 110-F stack at Hanford, USA, using explosives. b.). Time sequence of the stack fall at Marcoule G1, France, after [6]

Linear static analysis is highly applicable in conceptual design [8], and gives acceptable results for low and medium-tall structures in which the first vibration tone is dominant. The linear change of stiffness is adopted and coefficients of ductility and damping that depend on the structural system and applied materials are introduced. Inner forces are computed after the application of pseudo seismic load. In tall chimneys, and generally in high structures, higher modes are dominant and the use of this method does not give reliable results. This procedure is common for simple regular buildings.

Non-linear static analysis (NSA) is more convenient for models where the effects of higher tones are not significant. There are more variants of NSA in use. Good results are obtained with the push-over analysis. Through the application of this method, the structure is exposed to a gradual increase of horizontal seismic load, till the occurrence of a local ortotal failure of the structure. Deformations are also obtained by the analysis, which enables the determination of critical cross-sections of the structure.

Linear dynamic analysis includes methods of spectral response (spectral modal analysis) and the method of time history analysis. It is usually used in systems with more degrees of freedom.

Non-linear dynamic time history analysis gains greater application with the development of computer hardware capacities. It requires complex mathematical operations, and detailed information on the structure and the excitation. Therefore, it is rarely used in designing of new structures, though it is largely applied in scientific investigations. Introducing of the influence of the structure-soil interaction additionally complicates the analysis on the mathematical model.

EN 1998–Part 6 discusses in particular structures of industrial chimneys, and telecommunication and other towers [7]. Requirements that are to be fulfilled by the calculation depending on the category of the structure during an earthquake are:

– protection from jeopardy concerning people, neighbouring structures and nearby goods, and

– preserving of continuous functioning of the facility, industry and communication system.

The former implies the prevention of collapse of the structure, and the latter the limitation of damage of the structure.

Experimental results for TREPCA chimney are compared with conventional European norms. The methods prescribed in the European norm EN 1998-1:2004 for the calculation of structures in seismic areas are given in Table 1:

118

Table 1. Methods prescribed for calculation of structures in European norms

Method STATIC DYNAMICS

Linear Equivalent static analysis Response spectrum

Nonlinear Push-over analysis Time-history analysis

The calculation obtained through the application of equivalent static load gives good results when the first mode dominantly affects the response of the structure, i.e. when the contribution of the higher modes is negligible. For chimneys are recommended the linear analysis with usage of reduced spectrum, or simplified dynamic or multimode analyses. Non-linear methods can be used taking into account seismic actions, corresponding constitutive model, interpretation of results and some specific requirements. Nonlinear dynamics analysis directly calculates seismic actions and displacements by time-history procedure.

When modelling the structure, one should determine the number of degrees of freedom, calculate masses, stiffness, adoption or way of calculation for damping, and the interaction structure-soil should also be introduced [12]. When arranging concentrated masses, equipment, machinery, and possible extensions should be taken into account. In chimneys, the effects of interaction between outer and inner pipes should be considered, as well.

The reinforcing rules are prescribed depending on the outer diameter of the chimney [7]. For outside diameter 4m, minimal vertical reinforcement is more than 0.003. If outside diameter 4m, the ratio of outer layer reinforcement shall not be less than 0.002. The distance between vertical bars should be less than 200mm. The declination of the vertical axis must not exceed 1/1.000 part of the height of shell, or maximum 2 cm. Measured thickness of the wall must not vary more than 1/100 of the specified value.

When designing chimneys on horizontal seismic forces, it is sufficient to, due to symmetry, take into account only one horizontal direction. The vertical component of the earthquake has no significant effect and can be neglected. If the calculation through the elastic spectre is applied, damping must be adopted for possible liners, depending on the confining material. Namely, for steel is assumed the value of 1.5%, for brick 4% and 2% for FRP materials.

Damage limitation requirements are expressed through the maximum displacement of the top of the structure, limited relative displacement of the shell and liner or support platform. Limited displacements contribute to the reduction of second order effects.

The maximum lateral deflection of the top of the structure should be limited to:

max 0.005d v H

(1)

119

where:

maxd … lateral deflection at the top of the chimney, H … height of the structure v … reduction factor to take into account the lower return period of the seismic action.

Simplified dynamic analysis is recommended in structures that can be presented in two plane models, and whose response is not significantly influenced by higher modes. It is allowed for chimneys where the importance factor is 1 1 and height H < 80 m.

Depending on the type and importance of the structure, different importance factors for structures were adopted, by which are multiplied the amplitudes of accelerogram or spectrum.

Chimneys are sensitive to a larger period of seismic excitation. This is particularly expressed in soft soil.

Structures are classified in 4 importance classes (Table 2.). Classification depends on the consequences of collapse for human life, on importance for public safety and civil protection in post-earthquake period, on social and economic consequences [7].

Four importance classes show the recommended I factor, and it is Nationally Determined Parameter (NDP), varied in the National Annex. In EC8, the importance factors are applied to the input motions, as opposed to US practice where importance factors are applied to seismic loads [14]. There is significant difference in nonlinear analysis, because increasing the ground motions may cause less increase in forces, depending of yielding of elements, but more increase in deflections, due to plastic strains and P-delta effects.

Table 2. Importance classes, after [7]

Importance class Buildings I

IBuilding of minor importance for public safety, e.g. agricu-ltural buildings, etc.

0.8

II Ordinary buildings, not belon-ging in the order categories 1.0 (not an NDP)

III

Buildings whose seismic resis-tance is of importance in view of the consequences associa-ted with a collapse, e.g. scho-ols, assembly halls, cultural institutions, etc.

1.2

IV

Buildings whose integrity du-ring earthquakes is of vital importance for civil prote-ction, e.g. hospitals, fire sta-tions, power plants, etc.

1.4

120

The importance factor 1 is given with values from 1 to 1.4, depending on strategic importance, height and environment. If the height of the structure is twice bigger than the maximum dimension in the base, then the influence of soil is great and can significantly increase the effects of second order influence and soil structure interaction.

According to EC8, structures taller than 80 m, in the area of high seismic activity, must be analysed as a special structures. A region where the designed soil acceleration agfor soil type A is smaller than 0.08g is considered to be an area with low seismic activity

ACI 307-98 Building code requirements for structural concrete for analysis, design of the structure and details of RC chimneys in the USA [5] and [1]. Corresponding stiffness is taken into account as RC chimneys are designed to keep elasticity during earthquake. Such a philosophy of designing is in a collision with directions of development of international regulations like the ones given by International Committee on Industrial Chimneys (CICIND 2001) and EN 1998 – Part 6 :2005) which take into account the required ductility of the structure [10]. In ACI 307-98 it is required to apply the modal analysis and elastic spectrum with effective factor of response modification R = 1.33. There is no consideration to involve ductility and capacity design principles. It provides a failure mechanism in foundation area. Associated load factor of 1.43 for seismic forces is also in the function of requirement to keep the behaviour of the structure during earthquake within the elastic area. CICIND 2001 recommends a higher importance factor 1.4, with R = 1 for the elastic response, and R = 2 for ductile behaviour in a highly seismic area, where it is not economic to design a chimney within the elastic domain. According to EN 1998 – Part 6 the reduction of seismic force (in EN 1998 marked with q) is allowed if the ductility is provided with R (q) = 3. Furthermore, the possibility to form one plastic hinge for R = 2 is anticipated. The load level for R = 2 is kept beyond the zone of the plastic hinge [11].

According to CICIND 2001 standard, the response spectrum is given depending on the period T. Seismic actions are obtained from the elastic response increasing the response by importance factor and dividing it by structural response factor. The importance factor depending on the importance class of the chimney has the value IF = 1.0, 1.2 or 1.4 [8].

In November 2008, ACI published Code Requirements for Reinforced Concrete Chimneys, ACI 307-08 [3], and it replaced previous ACI 307-98 [5].Significant revisions are made in the seismic design provisions. One of the key changes from the ACI 307-98 is that selecting the design base earthquake of Maximum Considered Earthquake includes effects of soil amplification in design response spectra, use a response modification factor (R) of 1.5, considering P-delta effects for chimneys design category D., E and F. It defines difference in seismic criterion between the concrete chimney shell and chimney liners.

After effect of 1976 Tangshan Earthquake magnitude 7.8, is damaged 180 m high chimney, at height of 132m, and collapsed top of 48 m during the aftershock, magnitude 7.1, about 15 hours after the earthquake. Influence of higher mode effect is a failure of 75% chimney height. Chimneys shorter than 100m remained with no damage [13].

Earthquake at Izmit, Turkey, 17.08.1999, magnitude 7.4, generated ground shaking of 45 sec. One 115 m tall chimney collapsed at the oil refinery within 20 km of the epicentre, and four chimneys hade been damaged hardly. The collapsed chimney had diameter at the base 10.0m, thickness of 0.45 m, and diameter at the top 5.0 m, and thickness 0.22m. It was based on firm soil. The chimneys were designed to ACI307-69. The chimney had brittle failure of the withshield near the openings. Photographic of failure show little longitudinal reinforcement.

121

During the Loma Pierta earthquake, 154 m tall chimney suffered total cracking between 80%-90% of chimney height. The chimney was designed to the ACI505-1954, which assumed a first mode response. The height mode effects were neglected.

RESULTS OF PARAMETRIC ANALYSIS FOR TREP ACHIMNEY

Fig.3- Chimney “Trepca“, Kosovska Mitrovica

The numerical parametric analysis of dynamics behaviour for chimney Trepca in Kosovska Mitrovica has done. According to parametric analysis, theoretical values are compared with experimental results.

Chimney Trepca is 300 m tall, top thickness t = 0.25m and bottom t = 0.7m. It has five declinations, at 60,100,145 and 190 m (Fig 3).

Base structure is RC round slab, D= 36m, based on 6m depth. Experimental examinations involve environment vibrations, sinusoidally

controllable excitement and free vibration [9]. Environment exciting vibration is addition to random cases. It comprises micro

seismic activities, products of human activities as traffic and machine vibrations, and wind as the prime cause.

Two accelerators FBA-3 are used. Also, 19 spot spans are equidistantly placed with 15 m increment to reach the structure height. For sinusoidal exciting, spot is at the level of 295m.

The chimney is analysed by software SAP2000. The results are obtained for modal analysis, nonlinear static, response spectrum analysis, nonlinear direct-integration time-history analysis (NDIH).

122

Table 3. Experimental and theoretical results for first three modes

f1 [Hz] f2 [Hz] f3 [Hz]

Ambient oscillation (AO) 0.171 0.641 1.498

(AO) compared with theoretical value-shell (+34.6%) (+11.498%) (+5.315%)

Sinus motion (SM) 0.685 1.510

(SM) compared with theoretical value-shell (+19.15%) (+6.159%)

Free oscillation (FO) 0.659 1.52

(FO) compared with theoretical value-shell (+14.629%) (+6.862%)

Theoretical - 20 lumped mass (LM) 0.17 0.66 1.62

(LM) compared with theoretical value-shell (+14.802%) (+13.892%)

Theoretical- shell 0.5m thin 0.127 0.5749 1.4224

123

Fig. 4- Modes 1,3,7 and 12

Fig.5- Stress Smax dijagram for NDIH, response spectrum and nonlinear static analysis

124

Fig.6- Deformed shapes for NDIHA, response spectrum and modal analysis

Fig.7 Response spectrum curves

125

Fig.8 Displacement-time diagram at 300m and 55.08m

Table 4. Displacement of joint 911

Joint Mode 1 Mode 12 Linear response spectrum NDIHA max NDIHA min

911 -3.61E-04 -8.18E-03 4.78E-03 3.631098 -4.27133

Comparison of modal analysis and experimental results is given in the Table 3. Difference between theoretical and experimental results for frequency is the

largest for the first mode, +34.6%. For the second mode, difference is 11.498%-19.15%, and for third, it ranges between 5.315%-6.862%. The Trep a chimney structure is modelled by Finite element method (FEM), as thin shell with 4928 joints.

Stress and displacement values are compared for different analysis types, and some results are shown in Fig. 5 and Fig. 6. Response spectrum curves are shown in Fig. 7 for two joints, and represent significant influence of damping. Displacement-time diagram at two points (top and at 55.08 m) over height of chimneys are shown in Fig. 8 Displacement of one joint for modal analysis, linear response spectrum analysis and nonlinear direct-integration time-history is presented in Table 4, and has appreciably larger value for NDIH analysis.

126

4 CONCLUDING REMARKS

In areas with low seismic activity, a dominant influence on the calculation of the chimney structure is the wind action. In are with high seismic activities, the response of the structure to seismic actions must be examined in detail.

According to [9], steel chimneys are more efficient for heights of 70-90 m, and RC ones for heights above 90 m. For heights exceeding 150 m RC chimneys are built.

Theoretical basis and application in practice of the method for seismic analysis of structures are frequently in disproportion. The analysis of real behaviour of structures in earthquake shows that a well-chosen concept of the structure contributes more to its seismic reliability than the application of complicated numerical models.

ACI 307 norms and earlier recommendations of CICIND (from 1998) give more significance to the elastic behaviour of the structure, neglecting its ductility. The result of this approach is a very expensive structure, which can suffer from a brittle failure in case of strong earthquake. On the contrary, in EN 1998 – Part 6 it is recommended to adopt ductile behaviour by forming a plastic hinge.

A good designing approach is the introduction of limited ductility, when seismic energy is dissipated in plastic hinges. On the bases of the surveyed analyses it can be concluded that the choice of an adequate mathematical model for the seismic analysis of chimneys significantly diminishes the domain of probability regarding the collapse of the structure.

ACKNOWLEDGMENT: This research has been supported by the Ministry of Science of Serbia through the Research Project TR 16017. This support is gratefully acknowledged.

127

5. REFERENCES

[1] Bae, S. et al: Evaluation of the Design Provisions of ACI 307 Standard for Seismic Design of Concrete Chimneys, Proceedings of Sessions of the Structures Congress Long Beach, California, USA, ASCE 2007

[2] Chopra, K.A: Dynamics of Structures (3rd Edition) (Prentice-Hall International Series in Civil Engineering and Engineering Mechanics), 2006

[3] Code Requirements for Reinforced Concrete Chimneys (ACI 307-08)

[4] Customer’s Guide to Specifying Chimneys, CICIND, Switzerland, 1990

[5] Design and Construction of Reinforced Concrete Chimneys (ACI 307-98)

[6] Dismantling Of Contaminated Stacks At Nuclear Facilities, IAA Vienna, 2005

[7] Eurocode 8: Design provisions for earthquake resistance of structures, Part 6: Towers, masts and chimneys, CEN, 2004

[8] Foli , R.: Neke primene dinamike konstrukcija i njihova primena u seizmi koj analizi konstrukcija gra evinskih objekata, Gra evinski kalendar 2007, Vol.39, p. 143-233

[9] Kapsarov, H.,: Doprinos aseizmi kom projektovanju visokih armiranobetonskih postrojenja tipa tornja, Gra evinski fakultet Niš, 1985

[10] Model Code for Concrete Chimneys, with Commentaries, Part A - The Shell,CICIND, Second Edition, Revision 1 - August 2001

[11] Munshi, J., Malushte, S.: Seismic Design of Concrete Chimneys - State of Practice,pp. 1-10, Proceedings of Sessions of the Structures Congress, Long Beach, California, USA, ASCE 2007

[12] Pour, N.S., Chrowdhury, I.: Dynamic Soil-Structure Interaction Analysis of Tall Multi-flue Chimneys under Aerodynamic and Seismic Force, IACMAG, pp. 2696-2703, Goa, India, 2008

[13] Rumman, W.S.: Reinforced concrete chimneys in Handbook of concrete engineering (M. Fintel, e.) 2nd Ed., Van Nostrand Reinhold Co, New York, 1985

[14] Seismic Design of Buildings to Eurocode 8, Edited by Ahmed Y. Elghazouli, Spon Press, New York, 2009

[15] Wilson, L.J., Earthquake Design and Analysis of Tall Reinforced Concrete Chimney, Melbourne ,2000

129

or e La inovi 117, Radomir Foli 218 and Mladen osi 319

UPOREDNA ANALIZA SEIZMI KIH ZAHTEVA REGULARNIH BETONSKIH VIŠESPRATNIH OKVIRA

Rezime:Cilj rada je procena mogu nosti koriš enja uproš enih nelinearnih metoda u analizi regularnih okvira razli itih spratnosti. Procena ciljnog pomeranja je sprovedena primenom poznatih metoda CSM, ELM, CM i DDM. Sve koriš enemetode zasnivaju se na nelinearnoj stati koj analizi i metodi spektra odgovora. Pushover analiza koristi se za konstruisanje pushover krive, koja se idealizuje da bi se odredile karakteristike ekvivalentnog SDOF sistema. Za regularne okvirne konstrukcije uproš ene metode daju zadovoljavaju u ta nost. Sve primenjne raspodele popre nog optere enja daju skoro identi ne nosivosti, krutosti i duktilnosti.Key words: regularni okviri, seizmi ki norme, seizmi ki zahtevi, nelinearna analiza

COMPARATIVE ANALYSIS OF SEISMIC DEMANDS OF REGULAR MULTI–STORY CONCRETE FRAMES Summary:

The aim of the paper is to assess the usability of nonlinear simplified methods for practical application for regular frame structure with different stories. Estimation of target displacement is performed using CSM, ELM, CM and DDM. All applied procedures are based on pushover analysis and response spectrum method. The pushover analysis is used to develop pushover curve, which is idealized to determine the characteristics of SDOF system. For regular frame structure simplified methods yield to results of adequate accuracy. For all the applied distribution of lateral loads were obtained almost identical levels of strength capacity, stiffness and ductility. Key words: regular frames, seismic codes, seismic demands, nonlinear analysis

1 Prof., Dr., University of Novi Sad, Faculty of Technical Sciences, Serbia, E-mail: [email protected] Prof., Dr., University of Novi Sad, Faculty of Technical Sciences, Serbia, E-mail: [email protected] PhD student, University of Novi Sad, Faculty of Technical Sciences, Serbia

130

1 INTRODUCTIONExisting seismic design procedures are predominantly based on elastic structural

models Error! Reference source not found.. The capacity of the structure to dissipate input energy during earthquake with inelastic deformation is taken into account indirectly by using the reduced seismic forces. However, the use of the empirically based reduction factor may fail to predict the actual behaviour of the structure. The need for changes in the existing methodology implemented in existing building seismic codes has been therefore widely recognised. The existing seismic design procedures cannot provide an adequate inspection of damage level of building structures in quantitative terms. These methods are based on the assumption of linear elastic structural behaviour and do not provide information about real strength, ductility and energy dissipation.

In this paper, a simplified nonlinear method for estimation of seismic demands and real response of multi-storey buildings is presented. Two mathematical models are used for the seismic analysis – one mathematical model is a multi degree of freedom (MDOF) system, and the other is a single degree of freedom (SDOF) system. Nonlinear static analysis is used to determine the action effects and the pushover curve of MDOF model, approximated by the bilinear force-displacement relationship to determine the characteristics of the equivalent SDOF system. Developed pushover curve is converted into an acceleration displacement response spectrum (ADRS) format of capacity curve. The ratio of seismic demands and yield strength capacity is determined by comparison of capacity curve and response spectra of excitations. Estimation of target displacement is performed using capacity spectrum method [1], equivalent linearization method [2], coefficient method [3] and displacement modification method [3]. Target displacement determined in this way is again converted into the corresponding displacement of MDOF system. The whole system is afterwards being "pushed" to the target displacement of the multi storey frame, with the determination of the action effects in structure and monitoring of the plastic hinges formation and propagation of nonlinear deformations.

2 NONLINEAR STRUCTURAL MODEL OF MULTISTORY FRAMES

Methods for seismic analysis of structures can be divided into static and dynamic, and structural models into linear and nonlinear. The actual structural behaviour under seismic action can be best simulated using nonlinear time history analysis. However, the nonlinear time history analysis is still too complex for practical usage, which led to a recent development of analysis methods based on a nonlinear static analysis (NSA). Results of these researches are implemented into the latest codes for the design of structures for earthquake resistance: FEMA 356 [2], FEMA 440 [3] and Eurocode EN 1998 [5]. Initial structural model for nonlinear static analysis (NSA) of structures subjected to seismic actions is a multi degree of freedom system, for which is necessary to determine the pushover curve, i.e. the relationship between the base shear force and horizontal displacement of the top of the building. Structural strength capacity, as well as the shape of pushover curve, depends on the applied distribution of seismic forces over the height of the building. Different lateral load distributions can be applied: uniform, triangular, according to the first mode shape (modal distribution), the distribution according to the SRSS combination of modal lateral forces [3] etc.

131

Three dimensional multi storey frame building can be analyzed through the decomposition of structure into certain substructures, which consist of multi storey frames loaded in their own plane. Some codes limit the application of nonlinear static analysis to regular frames in elevation, with the exception to frames with the discontinuity at the ground floor, where the application of nonlinear static analysis is allowed. Plane frames are modelled using beam and column elements of constant cross-sections with two nodes and three degrees of freedom in each node. Structural models with plastic hinges concentrated at the ends of elements (beams and columns) are commonly used for nonlinear analysis of multi-storey frame structures. Recently, structural model of multi storey frames with fibre models of beams and columns are also used, which can include propagation of inelastic deformations along structural elements.

Frame model with plastic hinges is formed using beam finite elements, "placing" plastic hinges at the ends of elements. Nonlinear effects can occur as a result of material and/or geometric nonlinearity. Geometric nonlinear effects are introduced through P–effects and the incremental displacement determination, while the material nonlinearity is introduced using a nonlinear force-deformation relationship in the plastic hinges. The force-deformation relationship in potential plastic hinges must be previously determined (e.g. moment-rotation, moment-curvature relationship, etc.).

The nonlinear static analysis defines the relationship between base shear force and horizontal displacement at the top of the building through the pushover curve. Overall lateral load is divided into increments, and the whole system is observed through different configurations in which the equilibrium equations are solved for the incremental load. Within each increment, it is assumed that the system of equations is linear, so the solution of nonlinear problems is given as the sum of a series of incremental solutions. As a result of linearization, there are unbalanced (residual) forces, which is the reason why iterations are performed within each increment in order to balance residual load. The distribution of seismic loads over the height of the building is taken to be constant during the several increments (conventional analysis) or with the alteration of the load distribution in the incremental situations (adaptive analysis).

Static analysis is first carried out for vertical load in the conventional analysis. Previously should be defined the system geometry, material characteristics, preliminary cross-section dimensions and amount of reinforcement for all elements, the characteristics of plastic hinges (e.g. by [4]) depending on the type of element (beam, column ...). Afterwards, lateral load is progressively applied and the formation of plastic hinges is monitored with a transition of the system to a nonlinear behaviour range. In slender unbraced frames it is possible problem of stability and divergence of solution due to the second-order effects. Lack of conventional methods is that the lateral seismic load does not change with the occurrence of plastic hinges and propagation of inelastic deformation, but the distribution of loads is constant during the entire analysis. Therefore, according to the codes (e.g. [2], [3], [5]), it is required that the analysis must use at least two different load distributions.

Incremental update methods of the lateral load are recently developed, which are using the term adaptive pushover analysis. Update of seismic load vector is implemented using: total updating, incremental updating and hybrid total - incremental updating. In the adaptive analysis is used distribution of lateral loads that are changing during the analysis. The update of the characteristic values and mode shapes is performed due to the nonlinear behaviour of the system. Modification of characteristic vibration analysis is reflected in the

132

fact that after every significant change in stiffness, the model is redefined. Degradation of stiffness is accompanied by extending the period of vibration and changing the characteristic mode shape. Therefore, new periods and mode shapes of vibration and new distribution of horizontal forces for each new refined model is determined through several iterations. In the first phase, as well as in conventional method, the system is analyzed for vertical load. Stiffness matrix K0 at the end of this analysis is used for nonlinear seismic analysis in the next step. The horizontal displacement u is divided into a number of increments within which the incremental-iterative analysis is performed. At the end of each step, stiffness matrix Ki is used for the subsequent modal analysis and the next step of nonlinear static analysis, as long as the condition u = umax is satisfied. Seismic load in i-th step is generated from the characteristic mode shape of vibration determined in the previous step. In this way a series of analysis is performed in which the calculation of structural mode shapes is done alternately and new seismic loads are defined [4].

3 ESTIMATE OF TARGET DISPLACEMENT Analysis of the target displacement is the second phase of nonlinear static analysis.

Estimation of inelastic deformation is based on the analysis of SDOF system, and depending on the procedure applied for the determination of target displacement. In this aim several different procedures are developed. Research in this paper is limited to the analysis verified in practical applications and implemented in technical codes: 1) Capacity Spectrum Method - ATC-40 [1], 2) Coefficient Method - FEMA 356 [2]), 3) Equivalent Linearization Method - FEMA 440 [3]), 4) Displacement Modification Method - FEMA 440 [3]).

3.1 CAPACITY SPECTRUM METHOD

Capacity spectrum method (CSM) is a nonlinear static procedure that determines the nonlinear displacement of system caused by seismic action on the structure. This is an approximate method based on an estimation of the target displacement using the equivalent linear system. Target displacement of the nonlinear system due to the seismic action is determined by dynamic analysis of number of equivalent linear systems with successive update of equivalent vibration period Teq and equivalent damping coefficient eq. The method is based on the application of two mathematical models, one of MDOF and the other of SDOF. The first step in the analysis is the development of MDOF model pushover curve, and then for such developed pushover curve, bilinear force-displacement relationship is determined. Curve obtained in this way is called the capacity curve or capacity spectrum. In determining the bilinear force-displacement relationship of SDOF model, it is necessary to distinguish the initial stiffness of Ki from the effective elastic stiffness Ke. Initial stiffness is the elastic stiffness of MDOF model, while the effective stiffness is determined as an intersection point of the pushover curve for the 60% of the yield strength Vy.

In the previously described procedure participates viscous damping which is introduced through the elastic response spectrum. Idealized, design spectrum or the response spectrum generated from the earthquake records for different levels of viscous damping, ranging from 5% to 40% with incremental growth of 5%, can be used in analysis. It is necessary to take into account the hysteretic energy dissipation which participates in the total damping, for this correction to be included. In this way, the seismic demands

133

expressed by spectral curves are significantly reduced which adjusts the value of the target displacement. The procedure for determining the target displacement is iterative. In the first step the maximum spectral displacement is taken as Sd,i = Sd, and it is determined by the intersection of radial line which corresponds to the initial period of vibration Ti and elastic response spectrum with the viscous damping i = 5%. Afterwards, the calculation of the ductility coefficients is performed according to = Sd,i / Sdy, where Sdy is the yield displacement of the idealized SDOF system. Total damping is calculated as the sum of viscous damping: t = i + 0, where t is the total damping, i is the 5% viscous damping (constant) which is usually used in the analysis for elastic structural behaviour, is the modification factor which simulates possible imperfections in the hysteresis loops, 0 is the hysteretic damping coefficient represented as the equivalent viscous damping coefficient. The simplest procedure for determining the coefficient of equivalent viscous damping is to equalize the energy dissipated in one vibration cycle of inelastic system related to equivalent linear system.

Damping modification factor depends on the structural behaviour, type of structural system and on duration of seismic excitation. The ATC-40 defines three different models of structural behaviour. Type A represents hysteretic behaviour with completely stable hysteresis loops, while type C represents hysteretic behaviour with pinching effects and/or stiffness and strength degradation. Type B refers to inelastic behaviour that is between type A and C.

For the total damping t, the spectral curve and radial lines for the damping are constructed at the same diagram, and then the spectral displacement Sd,j is determined. Verification of the convergence terms is performed according to expression: (Sd,j - Sd,i)/ Sd,j tol, where tol is the adopted value of tolerance. If the previous condition is satisfied, the spectral displacement is Sd =Sd,j, otherwise Sd,i = Sd,j and the iterations continue. In most cases, large number of iterations is not needed for the previous condition to be achieved. Combining the obtained discrete values of Sd,i from the iterations, the demand spectrum with variable damping is formed. The value of the target displacement is determined based on the intersection of the capacity curve and the demand spectrum. This specific value of target displacement of SDOF system is necessary to convert into the MDOF displacement by multiplication of SDOF displacement with the participation factor

1 of the first mode shape.

3.2 COEFFICIENT METHOD

In relation to the capacity spectrum method that is based on the graphic presentation for determining the target displacement, the coefficient method uses the multiplication of elastic displacement with different coefficients. This method also uses idealized bilinear pushover curve, whereby the effective period is determined as Te = Ti (Ki / Ke)0,5. The target displacement is determined by modifying the spectral displacement of equivalent SDOF system as:

2

0 1 2 3 24e

t aT

C C C C S g (1)

Spectral displacement of the equivalent SDOF system is related with the displacement of the control node of MDOF system by the coefficient C0. It is calculated through the participation factors of the first mode shape, but the value of this coefficient must be less than 1,0. Using only the first mode shape of the elastic system, modification

134

coefficient C0 becomes equal to the participation coefficient of the first mode shape. The coefficient C1 associates expected maximum nonlinear displacement with linear elastic response displacement:

1

11 1

e S

S e e S

for T TC

R T T R for T T

(2)

where the characteristic period TS represents the limit between constant acceleration and constant velocity ranges. The coefficient R is the reduction factor, which is calculated by:

/a

my

SR C

V W (3)

where Sa is the spectral acceleration determined from the response spectrum for fundamental period. Coefficient of the effective masses Cm is equal to 0.9 for Ti 1.0 s and 1.0 for Ti 1.0 s.

The coefficient C2 takes into account the pinching effect of the hysteretic loops, stiffness degradation and strength deterioration at maximum response of the system (for details see [2]), while for the nonlinear static analysis it can be taken equal to 1.0. The reduced values of the C2 coefficient are given in [2] for lower levels of damage, such as the performance level of immediate occupancy compared to the performance level of collapse prevention.

The coefficient C3 introduces the increase of displacement due to dynamic P-effects. Positive stiffness in the nonlinear range takes the value of C3 = 1.0, while the negative stiffness for non-linear behaviour is determined as:

3 / 23 1 1 eC R T (4)

where is the ratio of the stiffness in the nonlinear range to the effective elastic stiffness. The increase of displacement caused by dynamic P- effects depends on the coefficient , fundamental period, the hysteretic load-deformation behaviour, frequency characteristics of the earthquake and duration of strong ground motion.

3.3 EQUIVALENT LINEARIZATION METHOD

Conventional method of the capacity spectrum uses the secant period as an effective vibration period in the target displacement determination, as the intersection point of capacity curve and demand curve for effective damping in the ADRS form. The effective vibration period Te of improved procedure is generally less than secant period Tsec, which is defined on the capacity curve for achieved target displacement. Multiplying the ordinate of the ADRS demand curve for the corresponding effective damping e with modification factor M = at / ae gives the modified acceleration displacement response spectrum (MADRS), where at is the acceleration at the target displacement according to MADRS demand, and ae is the acceleration according to ADRS demand (Fig. 1).

Since the values of the acceleration are in direct correlation with the corresponding period and ductility, modification factor can be determined by:

135

2 2 2 1 1e e i

sec i sec

T T TMT T T

(5)

When applying a procedure of equivalent linearization in practice, it is necessary to use the reduction factor for the correction of the initial response spectrum for the appropriate level of effective damping e. This factor is a function of effective damping B( e) and it is used for correction of the spectral acceleration as (Sa) = (Sa)0/B( e), where Bfactor is determined from B( e) = 4/(5,6 – ln e ). This expression is very similar with the expression given in ATC-40.

Figure 1. Modified response spectrum using the secant period Tsec

Figure 2. Bilinear representation of the capacity curve

136

Figure 3. Determination of the target displacement using MADRS

Since the effective period Te and the effective damping e depend on ductility demand, calculation of the target displacement using equivalent linearization is done in iterations, in the following steps:

in the initial response spectrum is constructed with i =5% according to ATC-40, such response spectrum is modified according to FEMA 440 to include the effects

of soil-structure interaction (SSI), modified SSI response spectrum is converted into ADRS format according to

ATC-40 procedure, and becomes the initial response spectrum for iterative analysis, developed pushover curve is converted into capacity curve according to ATC-40, initial target acceleration api and displacement dpi are determined using the

principles of equal of displacement of linear and nonlinear system response, as shown in Fig. 2,

bilinear capacity curve is determined in accordance to ATC-40, then the fundamental period T0, yield displacement dy and yield acceleration ay are determined,

determine the ductility coefficient ( = dpi / dy) and the value of the coefficient ,which represents the ratio of stiffness for non-linear and linear behaviour:

pi y y

pi y y

a a dd d a

(6)

appropriate effective damping e for the bilinear hysteretic model (BLH), for stiffness degrading model (STDG) and for strength degrading model (STRDG) is determined. Effective viscous damping is shown as a percentage of critical damping depending on the ductility coefficient:

2 30

0

2

020

1 4 : 1 14 6,5 : 14 0,32 1

0,64 1 16,5 : 19

0,64 1

e

e

ee

for A Bfor

Tfor

T

(7)

137

The coefficients A, B, C, D, E and F are functions of hysteretic behaviour model and the coefficient . Numerical values of these coefficients can be found in [3].

effective vibration period is determined by (5), where the numerical values of coefficients G, H, I, J, K and L are given in [3].

2 30

0

0

1 4 : 1 1 1

4 6,5 : 1 1

16,5 : 1 1

1 2

e

e

e

for T G H T

for T I J T

for T K TL

(8)

response spectrum in the ADRS format is altered for a previously determined effective damping e, as (Sa) = (Sa)0 / B( e),

values of acceleration response spectrum for e are multiplied by modification factor M, and then the MADRS is generated,

joining the points from intersections of radial lines of secant periods Tsec with MADRS response spectrums, demand spectrum with variable ductility is obtained (Fig. 3),

value of the target displacement is determined from the intersection of the demand spectrum and capacity curve (Fig. 3).

3.4 DISPLACEMENT MODIFICATION METHOD

The displacement modification method (DCM) has certain improvements related to the coefficient method according to FEMA 356. The C1 coefficient is limited to a relatively short vibration period of structures according to FEMA 356, while the FEMA 440 eliminates this restriction and C1 is calculated as:

21 1 1 eC R aT (9)

where a is a constant whose value is 130, 90 and 60 for B, C and D type of soil, respectively. For the vibration period less than 0.2 s, the value of the C1 coefficient is determined by the given expression, while for the vibration period greater than 1.0 s, C1 is determined as C1 = 1.0. Given expression for C1 allows better estimate of the maximum deformation of elasto-plastic SDOF system related to the maximum deformation of the linear SDOF system.

It is well known that the system response is influenced by two types of degradation, stiffness and strength degradation. The C2 coefficient includes only the effect of stiffness degradation according to modification method and may be determined from expression:

2

21 11

800RC

T(10)

For the fundamental period of structure less than 0.2 s, C2 coefficient is taken according to (10), while for the period greater than 0.7 s, it is taken equal to 1.0. Increase of nonlinear deformation due to cyclic degradation depends on the characteristics of hysteretic behaviour. Expression Error! Reference source not found. is determined based on extensive statistical analysis of different types of cyclic degradation of inelastic system.

138

Modified coefficient C3 from the CSM, has been replaced in the DMM with a minimum strength (by using Rmax) necessary to avoid dynamic instability. Reduction factor Rmax can be determined from expression:

max 4

ted

y

R (11)

where t is taken as t = 1 + 0,15 lnT, while d, y and e are determined according to Fig. 4.

Figure 4. Multi–linear approximation of force-displacement relationship

4 NUMERICAL ANALYSIS Four and eight storey frames have been analyzed. To determine the required

reinforcement in beams and columns, preliminary seismic analysis was made, where the seismic effects were determined using the equivalent static method. The RA 400/500 reinforcement and concrete class of MB 30 were used to design. Adopted dimensions of beams and columns, and amount of reinforcement are shown in Fig. 3. Afterwards, force-deformation curves were defined for all plastic hinges according to FEMA 273 and FEMA 356 [2].

139

Figure 5. Input data for considered four and eight storey frame

Figure 6. Pushover curves of four storey frame for different lateral load distribution

140

Table 1. Analysis results of 4 storey frame

Four storey frame – n = 4 Distribution Method: CSM CF ELM DMM

Ft (kN) 578.1 570.1 578.8 570.1 ut (cm) 9.1 16.4 8.7 16.4

Uniform Te (s) 1.534 1.027 1.027 1.027 eff (%) 25.4 – 10.3 –

Ft (kN) 508.5 488.1 509.7 488.1 ut (cm) 10.1 18.4 9.5 18.4

Equivalent Te (s) 1.729 1.521 1.521 1.152 eff (%) 26.2 – 9.6 –

Ft (kN) 505.7 484.9 506.8 484.9 ut (cm) 10.1 18.3 9.5 18.3

Modal Te (s) 1.735 1.146 1.146 1.146 eff (%) 26.2 – 9.9 –

Figure 7. Distribution of interstorey drifts of four storey frame for different load distribution

141

Figure 8. Pushover curves of eight storey frame for different lateral load distribution

Figure 9. Distribution of interstorey drifts of eight storey frame for different load distribution

142

Table 2. Analysis results of 8 storey frame

Eight storey frame – n = 8Distribution Method: CSM CF ELM DMM

Ft (kN) 1215.1 1182.6 1214.7 1182.6 ut (cm) 10.1 18.7 9.8 18.7

Uniform Te (s) 1.679 1.168 1.168 1.168 eff (%) 24.5 – 9.5 –

Ft (kN) 1034.2 1007.2 1034.7 1007.2 ut (cm) 11.9 21.8 11.6 21.8

Equivalent Te (s) 1.973 1.365 1.365 1.365 eff (%) 24.5 – 9.8 –

Ft (kN) 1017.4 990.2 1018.1 990.2 ut (cm) 12 21.9 11.6 21.9

Modal Te (s) 1.999 1.371 1.371 1.371 eff (%) 24.8 – 10.0 –

Based on the previously described methods of analysis and formed numerical models, pushover curves were developed to obtain the control node displacement in seismic design situation. The developed pushover curves for regular structures with plastic hinges at the ends of elements are shown in Fig. 4 for four storey frame and in Fig. 8 for eight storey frame. Results of analysis for various lateral load distributions and by using different procedures for estimation of target displacement are shown in Table 2 (4–storey frame) and Table 2 (8–storey frame). Distribution of inelastic deformation over the height of buildings, i.e. interstorey drifts, for various lateral load distributions that are applied are presented in Fig. 7 for 4–storey frame, and in Fig. 8 for 4–storey frame.

5 DISCUSSIONS AND CONCLUSIONS The paper presents application of various analysis methods, which are used for the

estimation of the structural behaviour under seismic action. They are based on the simplified procedure which combines nonlinear static (pushover) analysis and response spectrum method. Two mathematical models were used for the seismic analysis of multi-storey frames. One mathematical model is a system with multi degrees of freedom, and the other is an equivalent system with one degree of freedom. To calculate action effects of the MDOF model, the nonlinear static analysis is used to develop the pushover curve, which is then idealized to determine the characteristics of the equivalent SDOF system. The target displacements of considered regular frame structures with different stories are performed using capacity spectrum method, equivalent linearization method, coefficient method and displacement modification method.

Most important parameters that can be determined from the developed pushover curves are: stiffness, yield strength and ductility of considered multi storey structure. Uniform distribution of lateral loads leads to higher values of the base shear force in relation to the equivalent and modal distribution. It is obtained as for four stories and also for eight storey frame. However, with uniform distribution less target displacement is

143

obtained than with the equivalent and modal distribution of lateral loads. Also, smaller ductile behaviour is achieved by using uniform distribution in relation to the equivalent and according to the first mode shape distributions. This is particularly expressed in four storey frame. Adaptive analysis also points to smaller ductile behaviour as for four stories and also for eight storey frame.

Comparing methods of analysis for estimation of the target displacements it can be observed two groups of method according to similarity of obtained results. The first group includes capacity spectrum method and equivalent linearization method, and the second group contains coefficient method and displacement modification method. Methods that belong to the first group provide lower values of the target displacements than when using methods that belong to the second group.

The results indicate that the distribution of interstorey drifts over the height of structure significantly depends on the number of stories. Increase of the height of structure leads to highly unequal distribution of inelastic storey deformations along height. Thereby, the distribution of interstorey drifts in tall buildings much more depends on the distribution of applied lateral load distribution than in low-rise buildings.

For all the applied distribution of lateral loads were obtained almost identical levels of strength capacity, stiffness and ductility. This indicates that the regular frames that are sized according to preliminary design using simplified method of analysis develop favourable plastic mechanisms.

ACKNOWLEDGMENTS This paper has been undertaken as part of project No. 16017 funded by the

Ministry of Sciences of Serbia.

REFERENCES [1] ATC-40, Seismic Evaluation and Retrofit of Concrete Buildings, ATC-40 Report, Vol. 1 and 2, Applied Technology Council, Redwood City, California, 1996. [2] FEMA 356, Pre-Standard and Commentary for the Seismic Rehabilitation of Buildings, American Society of Civil Engineers, Federal Emergency Management Agency, Washington D. C., 2000. [3] FEMA 440, Improvement of Nonlinear Static Seismic Analysis Procedures, Applied Technology Council (ATC-55 Project), Federal Emergency Management Agency, Washington D. C., 2005. [4] La inovi ., osi M.: Pushover analiza višespratnih okvira, SGIS, Zemljotresno inženjerstvo i inženjerska seizmologija, Prvo nau no-stru nosavetovanje, Soko Banja, 2008. [5] EN 1998 – Eurocode 8: Design of Structures for Earthquake Resistance, Part 1, General Rules, Seismic Actions and Rules for Buildings, CEN, Brussels, 2004. [6] Earthquake Resistant Regulations – A World List. IAEE: Regulations for Seismic Design, Tokyo, 1992, Supplement – 1996, 2000, 2004 and 2008.

145

Goran Simonovi 120, Branislav Verbi 221,

PRA ENJE STANJA KONSTRUKCIJE I ISTRAŽIVANJE SEIZMI KE OTPORNOSTI ZIDANIH ZGRADA

:

U radu je diskutovan prora un seizmi ke otpornosti zidanih zgrada. Predstavljena su tri softvera, 3MURI koji je razvijen u Italiji, MINEA koja se razvija na Katedri za gra evinsku statiku i dinamiku Univerziteta u Aachenu u Njema koj i pristup razvijen na platformi poznatog programa SAP2000 od autora ovoga rada. Uspore ene su razlike u seizmi kom kapacitetu jedne zidane zgrade dobivene softverima. Pokazano je da uobi ajeni pristup prora unu, uz zanemarivanje preuzimanja seizmi kog optere enja okomito na ravan zidova, podcjenjuje stvarni kapacitet zidane zgrade. Klju ne rije i: Zidane zgrade, zemljotres, elementarni blok, kapacitet,...

RESEARCH IN SEISMIC RESIDENCE OF THE MASONRY BUILDING

Summary:

In the paper are discussed seismic analyses of masonry buildings. Presented are three software, 3MURI developed in Italy, MINEA witch currently being developed at the Technical University in Aachen – Germany and an approach de-veloped using known engineering software SAP2000 developed by authors of this paper. Softwares are compared by seismic analysis of a typical one family house and obtained are small differences in the seismic capacity of the house. It was demonstrated that ordinary design of the construction without limited out-of-plane behavior undervalues the bearing capacity of the masonry buildings Key words: Masonry buildings, earthquake, elementary block, capacity,...

1 Mr.dipl.inž.gra ., viši asistent, Gra evinski fakultet Univerziteta u Sarajevu, Sarajevo 2 Dr.dipl.inž.gra , redovni profesor Gra evinskog fakulteta u Sarajevu u penziji i dopisni lan Akademije nauka i umjetnosti Bosne i Hercegovine, Gra evinski fakultet Univerziteta u Sarajevu, Sarajevo

146

1 UVOD

Za zidane konstrukcije, kao najstariji oblik konstrukcija koje ljudska zajednica koristi, još uvijek ne postoji razvijen numeri ki model kojim bi se uspješno simuliralo stvarno ponašanje zgrada pogotovo pod zemljotresnim optere enjem. Razlozi za nepo-stojanje op e prihva enog modela koji bi obuhvatio sve specifi nosti zidanih konstrukcija su mnogobrojni. Izme u pojedinih zajednica postoje razlike u neimarskoj tradiciji, razlike u mehani kim i geometrijskim osobinama maltera i zidarskih blokova, razlike u konstru-ktivnom sistemu i u me usobnim vezama izme u pojedinih konstruktivnih elemenata itd.

U radu je predstavljen koncept nelinearnog aseizmi kog prora una programom SAP2000 na bazi formiranja makroelemenata koje smo nazvali «Elemetarni blok». Prora unat je kapacitet jedne tipi ne porodi ne zidane zgrade koja je bila subjektom analize opsežnih istraživa kih projekata diljem Europske Unije. Numeri ki model zgrade je prostorni model sa zidovima koji preuzimaju seizmi ko optere enje u sopstvenoj ravni. Modelom je obuhva eno sadejstvo armiranobetonske tavanice sa zidovima kao i torzioni efekti.

U cilju verifikacije predloženog koncepta zgrada je prora unata korištenjem softvera 3MURI zasnovanog na primjeni makro elemenata i softvera MINEA zasnovanog na bibliotekama eksperimentima utvr enih kapaciteta pojedinih zidova na osnovu kojih se formira kapacitet zgrade. Radom su predstavljene specifi nosti ova dva suštinski razli ita suvremena pristupa prora unu zidanih zgrada i izvršena komparacija sa rezultatima dobivenim korištenjem programa SAP2000. Uspore enjem rezultata pokazano je da se korištenjem elementarnih blokova može uspješno numeri ki simulirati ponašanje zidane zgrade uz pretpostavke o nelinearnom ponašanju materijala.

Dalje usavršavanje kapacitetnog prora una primjenom koncepta elementarnih blokova podrazumijeva razvijanje blokova koji prenose optere enje i okomito na ravan zida. Univerzalnost koncepta elementarnih blokova za odre ivanje otpornosti zidane konstrukcije omogu ava njihovu implementaciju u druge softvere kojima se vrši analiza konstrukcija po pretpostavkama o nelinearnom ponašanju materijala uz korištenje osnovnih nelinearnih elemenata. Time model zgrade koji je napravljen korištenjem elementarnih blokova postaje podoban i za druge naprednije oblike analize koje suvremeni softveri pružaju.

2 ELEMENTARNI BLOK

Iz teorije konstrukcija je poznato da se kod okvirnih sisteme najve a naprezanja konstrukcije, za uobi ajena optere enja, javljaju u vezama greda i stubova, odnosno u vezama zidova i plo a, pa je mogu e predvidjeti mjesta u konstrukciji gdje e najprije do ido pojave nelinearnih deformacija. Iz toga je prizišla ideja da se manje napregnuti dijelovi konstrukcije tretiraju kao elasti ni, linearni, homogeni i izotropni, zapreminski ili ravanski elementi, a da se sva nelinearnost dešava tamo gdje su najve a naprezanja, odnosno na potencijalnim mjestima otkazivanja konstrukcije.

Sli an pristup primijenjen je u ovom radu u analizi ponašanja zidane konstrukcije pod dejstvom vanjskog optere enja. Umjesto da se svaka opeka i malterska spojnica modeliraju posebno, cijeli nizovi opeka i maltera u manje napregnutim dijelovima zida zamjenjuju se jednim elementom ve ih dimenzija koji emo nazvati elementarni blok [5,6,7,8].

147

Vezama izme u elementarnih blokova simuliramo otkazivanje malterske spojnice tamo gdje o ekujemo da u stvarnoj konstrukciji ta spojnica i otkaže. Na slijede oj slici prikazani su usporedo stvarni zid od opeke i zid sastavljen od elementarnih blokova unaprijed izabranih dimenzija. Dijagramima su prikazane i karakteristike veznih elemenata. 3d prikaz loma prikazan je za mjerodavnu spojnicu zbirno i komponentalno.

Slika 1. Model zidanog zida napravljen od elementarnih blokova

Sa jedne strane imamo blok unutar koga vrijedi teorija elasti nosti, dok sa druge strane veza elementarnih blokova predstavlja diskontinuitet u elasti nim osobinama zida. Preko veze elementarnih blokova može se unijeti nelinearnost u strukturu zida koji se analizira. U daljim fazama ovoga rada veze izme u elementarnih blokova modelirane su nelinearno pomo u posebnih elemenata kojima se uspostavlja veza izme u dijelova konstrukcije a implementirani su u programu SAP2000. Unutrašnji dio elementarnog bloka deformiše se linearno i uticaje koje prenosi kroz sebe predaje vezi sa susjednim eleme-ntarnim blokom. Veza elementarnih blokova, u slu aju pojave prekora enja mehani kih osobina materijala, simulira nelinearno ponašanja materijala. Sva nelinearnost se dešava isklju ivo na nivou mjerodavne malterske spojnice, odnosno u podru ju n-link elementa. U ovome radu se ograni avamo na materijalnu nelinearnost dok geometrijska nelinearnost (stabilnost zidnih elemenata, uticaj teorije II reda) nije tretirana.

148

Elementarnim blokovima simulira se konstrukcija u prostoru. Ravan pojedinih elementarnih blokova odgovara ravnima zidova u konstrukciji. Debljina i dužina elementarnih blokova odgovara debljini i dužini zida. Visina zida se elementarnim blokovima dijeli na više dijelova tako da pojedini elementarni blokovi imaju kvadratnu formu što ne mora biti pravilo. Na jednoj spojnici moraju se formirati bar etiri n-link elementa. Unutrašnjost elementarnih blokova modelirana je sa kona nim elementima koji imaju naprezanje u svojoj ravni i okomito na svoju ravnu (thin-shell). Podjela elementarnih blokova na mrežu kona nih elemenata treba biti takva da vorovi kona nih elemenata leže u dodirnim ta kama n-link elemenata. Po dužini zida broj kona nih elemenata je za jedan manji od broja n-link elemenata, dok po visini zida kona ni elementi bi trebalo da imaju visinu tako da se podijeljena površina elementarnog broja oblikom približava kvadratnoj formi. Za prostornu analizu u kojoj ponašanje zidova okomito na njihovu ravan ne e biti analizirano mogu se koristiti i membranski elementi (membrane), ali u tom slu aju se mora voditi ra una o stepenima slobode n-link elemenata radi adekvatnog formiranja matrice krutosti.

Zid prikazan na slici 1 modeliran je sa sedam tipova n-link elemenata. Svi n-link elementi su aksijalno nelinearni. Popre nu silu linearno prenose n-link elementi ozna eni sa 3a i 3b, dok n-link element 4 popre nu silu prenosi nelinearno. Ukoliko se analizira i ponašanje zida okomito na njegovu ravan tada n-link elementi ozna eni sa 1a, 2a i 3a se modeliraju linearno a n-link elementi 1b, 2b, 3b i 4 nelinearno. Spojnice «A», «B», «C» i «D» su modelirane aksijalno nelinearne ime se ostvaruje simuliranje loma zida na dejstvo centri ne ili ekscentri ne sile pritiska ili zatezanja. Spojnice «A», «B», i «C» prenose popre nu silu linearno dok spojnica «D» ima implementiranu nelinearnu vezu izme upopre nih sila i popre nih deformacija. Spojnice «A» i «D» su prethodno rotaciono limitirane.

3 DVIJE SUVREMENE METODE PRORA UNA

Prednosti modela kojim bi se obuhvatile realne nelinearne osobine zidanih konstrukcija su o igledne. Pojedina nim lomovima, odnosno prethodno planiranim lomovima, troši se velika koli ina energije. Otkazivanje, odnosno te enje pojedinog zida ne zna i i lom konstrukcije jer po te enju, uslijed preraspodjele, omogu ava se drugim zidovima u konstrukciji da preuzmu optere enje. Pove ava se duktilnost konstrukcije što je naro ito zna ajno kod zemljotresa. Nema potrebe za provjerom napona jer su definirani plasti ni zglobovi. Otvaranjem prslina konstrukcija postaje mekša, period oscilovanja raste, tako da konstrukcija prima manje seizmi ke sile i sli no. Poznavanjem kapaciteta konstrukcije na prijem seizmi kog optere enja mi fakti ki znamo koliku imamo sigurnost što nije slu aj ukoliko dokaz sigurnosti radimo po dopuštenim naponima.

Ukratko e biti predstavljena dva danas korištena koncepta modeliranja kon-strukcija uz pretpostavke o nelinearnom ponašanju materija. Dugo godina je softver 3MU-RI bio vode i europski softver iz oblasti zidanih konstrukcija što je i o ekivano jer Italija kao izrazito trusna zemlja primorana je živjeti sa seizmi kim rizikom. U zadnje vrijeme, po usvajanju novih seizmi kih propisa, može se re i da «Njema ka škola» preuzima primat u analizi zidanih konstrukcija na seizmi ka dejstva. Koncept je implementiran u softver MINEA. Od ostalih softvera, koji svojom u inkovitoš u i rasprostiranjem imaju zna ajno mjesto, izdvojimo i italijanski softver ANDILWALLS.

149

3.1. Prora un zidanih konstrukcija konceptom implementiranim u softveru 3MURI

3MURI je jedan od vode ih europskih softvera za prora un zidanih konstrukcija. Program je godinama evoluirao iz softvera korištenog u DOS-u do suvremene WINDOWS platforme. Grafi ko okruženje softvera je pregledno mada samo korištenje softvera ima svojih specifi nosti pogotovo za korisnike koji su navikli raditi u drugim platformama. Sam tok prora una konstrukcije podijeljen je u tri faze koje se sastoje od unosa podataka, prora una i pregled rezultata. Svi moduli softvera su oboga eni modernim grafi kim su eljem koje omogu ava prostorni pregled konstrukcije u osjen anom obliku, animacije kako ulaznih podataka tako i animacije dostizanja lomova konstrukcije u postprocesiranju, te mogu nosti pravljenja ilustrativnih tekstualnih izvještaja sa grafikom.

Softver je nastao na italijanskoj neimarskoj tradiciji što mu je sa jedne strane velika prednost, ali sa druge strane i mana. Za mnoge zemlje EU i njihovu neimarsku tradiciju njegova upotreba nije adekvatna. Naime, upotreba horizontalnih serklaža i greda u horizontalnoj ravni sa relativno tankim plo ama, ili sitnorebrastim konstrukcijama, je uobi ajena u Italiji. Njema ka tradicija upotrebe zidanih konstrukcija preferira upotrebu debljih tavanica – armiranobetonskih plo a, bez horizontalnih serklaža i gotovo bez ikakvih greda u horizontalnoj ravni. Eventualni serklaži u njema koj tradiciji su samo na vršnim etažama izvedeni kao parapeti. Nadvratnici, nadprozornici i drugi horizontalni elementi se izvode lokalno bez kontinuiziranja u horizontalnoj ravni.

Inspiraciju za postavke softvera autori nalaze u ošte enjima zidanih konstrukcija nakon zemljotresa (slika 2). Karakteristi ne dijagonale i horizontalne prsline nakon zemljotresa dešavaju se oko otvora. Postojanje parapeta i horizontalnih serklaža uokviruje konstrukciju tako da se konstrukcije mogu tretirati kao prostorni okviri.

Slika 2. Prikaz lomova zidova zidanih konstrukcija koje su inspirirale koncept 3MURI [1].

Autori softvera zidani zid zamjenjuju trodijelnim stubom. Gornji i donji dio stuba nalaze se na mjestu su elja sa gredom dok centralni dio stuba zamjenjuje zid na dijelu otvora. Grede su tako er podijeljene na tri dijela pri emu su krajnji dijelovi unutar zida a centralni dijelovi tako podijeljenih greda na dijelu otvora. Prilikom unosa podataka za grede se definiraju geometrijske i mehani ke karakteristika tako da softver može izra unati njihov kapacitet. Prethodno softver napravi raspodjelu gravitacionog optere enja sa plo a na grede i uticaj plo a na daljnje ponašanje modela zanemaruje.

150

Slika 3. Formiranje makro elementa [1].

Slijedi postupno guranje konstrukcije pri emu se u svakom od koraka prati stanje sila u stubovima. Podijeljeni dijelovi stubova rade kao okida i po dostizanju odre enog nivoa optere enja i za dalji prirast horizontalnih pomaka limitiraju razvoj sila u konstrukciji. Princip je jasan kao i njegova implementacija. 3MURI analiza podrazumijeva analizu konstrukcija u prostoru, pri emu se formiraju prostorni okviri, a svi pojedini zidovi te konstrukcije ra unaju isklju ivo na dejstva u svojim ravnima. Zidovi koji su okomiti na pravac dejstva zemljotresa svojim položajem u modelu doprinose seizmi koj otpornosti zidnanih zgrada kroz promjenu normalnih sila u njima pri emu se stvara spreg sila. Softver može imati zna ajnih problema prilikom formiranja mreže makroelemenata. Konstrukcije koje svojom geometrijom, pa i ponašanjem, nisu tipi ne okvirne konstrukcije zahtijevaju veliku domišljatost korisnika kako da te konstrukcije modelira kao okvirne i prilagodi potrebama softvera. To podrazumijeva ubacivanje «lažnih» stubova i «lažnih» greda, odnosno modeliranje konstruktivnih elemenata tamo gdje ih u stvarnosti nema.

Mnogi realizirani objekti, uspješne sanacije objekata ošte enih u zemljotresima, te brojni nau ni radovi koji su napisani zahvaljuju i upotrebi 3MURI-a su zavidne reference softvera bez obzira na njegove mane.

3.2. Prora un zidanih konstrukcija konceptom implementiranim u softveru MINEA

Istraživa ki tim sa katedre za gra evinsku statiku i dinamiku iz Aachena (RWTH-LBB Rheinisch-Westfälische Technische Hochschule - Lehrstuhl für Baustatik und Baudynamik) razvija potpuno novi pristup problemu seizmi ke otpornosti zidanih zgrada. Analiziraju i dosadašnja saznanja iz modeliranja zidanih konstrukcija, ponašanja materijala, nepoznanica u izvo enju, razlika u tradicijama i mnoge druge faktore, zaklju uju da ih razvijanje modela koji bi se oslanjao na postulate teorije konstrukcija ne ezadovoljiti. Okre u se rezultatima eksperimenata kao jedinim pravim pokazateljima stvarnog stanja stvari. Aktivno se povezuju sa drugim Univerzitetima i Institutima u razmjeni podataka i me usobnih iskustava, te po inju sa ogromnim poslom pravljenja biblioteke rezultata eksperimenata koji su ura eni.

Baziraju se na nosivost pojedinog zida u konstrukciji. Razvijaju algoritme za interpolaciju rezultata eksperimenata u cilju dobivanja ponašanja zidova koji svojom geometrijom i nivoom vertikalnog optere enja nisu eksperimentalno obra eni (slika 4). Time se izbjegavaju numeri ke formule za dokaz nosivosti zida na prijem komponentalnih naprezanja. Nosivost zida je jednozna na predstavljena kroz krivu kapaciteta zida, ime se

151

objedinjuju lomovi uslijed otvaranja spojnica na zatezanje, dostizanje gnje enja na pritisnutom dijelu zida pri dejstvu momenta, lomovi uslijed formiranja dijagonala, lomovi uslijed klizanja, te razni drugi kombinirani lomovi koji su opisani u literaturi.

Slika 4. Osnova softvera MINEA i tok prora una [4].

Formiranje krive kapaciteta konstrukcije zapo inje iterativno korak po korak nanošenjem inkrementa pomjeranja u pravcu seizmi kog djelovanja. Rezultanta popre nih sila svih zidova prizemlja objekta nastaje iz kapacitetnih krivih pojedinih zidova. U slu ajuda dispozicija konstrukcije sadrži nesimetri ne glavne pravce nastaje rotacija centra masa. Sistem unosi i inkrement rotacije u proceduru prora una i traže i ravnotežu ostvarenih sila u svim zidovima nastavlja guranje. Na taj na in se formira pojedina to ka u pushover krivoj. Koraci se ponavljaju dok je ravnoteža mogu a. Na slici 4 prikazan je tok prora una.

Proces je implementiran kroz softver MINEA. Iako posve jasan i jednostavan princip njegova implementacija je izuzetno komplicirana. Stoga su autori softvera morali koristiti i odre ene pretpostavke. Pretpostavili su da se lom konstrukcije dešava isklju ivona donjoj etaži konstrukcije. Time se zanemaruju lomovi konstrukcije koji mogu nastati na etažama iznad, a koji su mogu i. Dalje, analizom su obuhva eni samo oni zidovi koji idu od dna do vrha zgrade. Normalne sile u pojedinim zidovima su konstantne veli ineodre ene na osnovu raspodjele gravitacionog optere enja sa tavanica na zidove u zavisnosti od njihovog položaja. Shodno njema koj tradiciji u tavanicama ne postoje grede, a plo e se smatraju dovoljno krutim da zajedno sa zidovima iznad prizemlja osiguraju konturne uslove koje odgovaraju zidovima uklještenim na vrhu i dnu ime se mogu koristiti interpolacione krive dobivene za obostrano uklještene zidove. Zanemareni su me usobni kontakti zidova L ili U oblika i oni se u analizi tretiraju kao razdvojeni zidovi. Autori koriste veliki broj

152

pretpostavki što je i razumljivo obzirom na karakter problema. Koncept u potpunosti odbacuje tradicionalni pristup prora unu i uz naveden pretpostavke daje sliku grani nog stanja konstrukcije kroz mogu i i realni mehanizam loma zgrade.

U trenutku pisanja rada tim intenzivno radi na implementaciji metode kona nih elemenata u softver ime bi se neke od prethodnih pretpostavki eliminisale. Prvenstveno raspodjela optere enja sa tavanica na zidove uzimaju i u obzir realne krutosti tavanica i greda, kao i promjenu normalnih sila u zidovima prilikom guranja konstrukcije.

Rezultati prora una daju pushover krivu koja u se u kombinaciji sa prigušenjima prevodi i prikazuje u obliku CSM (Capacity Spectrum Method) dijagrama koji su prije bili «rezervirani» za armiranobetonske i eli ne konstrukcije. Implementacijom metoda zasnovanih na pomjeranjima modeli zidanih konstrukcija postaju vjerodostojni. Pored naprednog CSM metoda MINEA raspolaže i drugim prora unskim modulima kojim se verificiraju konstruktivna ograni enja shodno DIN-1053 te vrše dokazi naponskih stanja. Ovi moduli su interesantni za inženjersku praksu.

4 PRORA UN KAPACITETA JEDNE PORODI NE ZGRADE

U ovom primjeru izvršena je analiza tipi ne njema ke porodi ne zgrade tzv. «Reihenhaus» koja ima sistem nosivih zidova u dva ortogonalna pravca x i y, pri emu je u kra em pravcu zgrada izrazito seizmi ki neotporna. Arhitektonski, u prostoru, formiranjem nizova ovih jednostavnih zgrada, dobiva se cijelo urbano naselje. Zgrade leže jedna do druge, dilatirane su, tako da konstruktivno predstavljaju neovisne jedinke. Zidovi nazna enikao W1 i W2 u dispoziciji rasprostiru se od podruma do krovišta zgrade i oni su ukru uju ielementi za prijem horizontalnog optere enja u slabijem pravcu. Drugi zidovi u slabijem pravcu nošenja su pregradni i nemaju vertikalni kontinuitet. Shodno njema koj neimarskoj tradiciji zgrade se izvode sa monolitnim tavanicama ra enim na licu mjesta bez ikakvih greda.

Bath room

dining

living

kitchen

y

x

W2

W2

W1

Slika 5. Plan konstrukcije tipi ne njema ke porodi ne zgrade [4].

Navedeni tip zgrade je bio predmet i opsežnih istraživanja u sklopu projekta ESECMaSE iji izvještaj D8.3. opisuje izvršen eksperiment sa stvarnim dimenzijama zgrade (full-scale) pri emu je uslijed simetrije izgra ena i testirana samo jedna polovina koja je bila izložena pseudodinami kom optere enju [2]. Shodno o ekivanim ubrzanjima

153

tla pobuda je preko presa aplicirana na nivoima etaža za ekvivalentna ubrzanja tla intenziteta do 0,22g. Rezultati su javno dostupni na internetu, ali nažalost, svi ulazni podatci tog istraživanja nisu dostupni.

Verifikacija ovog primjera je izvršena po ulaznim podacima dobivenim od strane LBB-RWTH a sa kojima je kontrolni prora un ura en koriste i softvere 3MURI i MINEA. Dvije krive dobivene korištenjem softvera 3MURI su nastale prvenstveno zbog ograni enja softvera u modeliranju zidanih zgrade bez greda pa je dio plo e modeliran kao greda razli itih «sudjeluju ih širina». Usporedno je prikazana i kriva dobivena korištenjem koncepta elementarnih blokova bez doprinosa (kriva SAP b) i sa doprinosom popre nihzidova na ukupnu seizmi ku otpornost zgrade (kriva SAP a).

0

25

50

75

100

125

150

175

200

225

0 1 2 3 4 5 6 7 8 9 10

x (mm)

Fb (kN)

MINEA

SAP a

SAP b

3muri a

3muri b

Slika 6. Pushover krive dobivene korištenjem softvera 3MURI, MINEA i SAP-a.

Razlike u kapacitetu zgrade na prijem horizontalnog optere enja dobivene pomo utri u potpunosti razli ita pristupa postoje. Vidljivo je izuzetno dobro slaganje izme urezultata dobivenih koriste i softver MINEA i SAP2000 te odre eno odstupanje rezultata dolivenih koriste i 3MURI. Razlike u nagibu krivih su izraženije od razlika u kapacitetu što je i o ekivano uslijed razli itih pretpostavki. Naime, nagib krive MINEA odre en je kombiniranjem iteracionih krivih iz biblioteka softvera, nagib krive 3MURI proisti e iz krutosti zidova koje su konstantne vrijednosti do trenutka «okidanja» makroelemenata implementiranih u softver, dok nagib krivih u SAP-u je položeniji uslijed kontinuiranog otvaranja n-link elemenata koji simuliraju otvaranje pojedinih spojnica u zidovima

154

5 ZAKLJU AK

Na ovom jednostavnom primjeru porodi ne ku e vidljivo je odli no slaganje rezultata dobivenih pomo u tri u potpunosti razli ita pristupa odre ivanja kapaciteta zidane zgrade na dejstvo zemljotresa. Implementacijom koncepta elementarnih blokova u inženjerske softvere kao što je SAP2000 otvaraju se mogu nosti modeliranja svih konstruktivnih elemenata unutar konstrukcije ime se u mnogo emu prevazilaze ograni enja korištenjem visokospecijaliziranih softvera (3MURI, MINEA ili sl.) za prora un zidanih konstrukcija. Dodajmo da se unutar modela mogu implementirali i osobine materijala, te koristiti materijale koji su široko u upotrebi bez ograni enja.

Stoga, model zidane konstrukcije, u kom su zidovi modelirani kao elementarni blokovi, u svemu pokazuje svoju univerzalnost i široku mogu nost primjene. Razvijeni elementarni blokovi se mogu implementirati i u druge inženjerske softvere kod kojih je materijalna nelinearnost riješena kroz elemente veze. Prora un može biti u skladu sa svim standardima uklju uju i i EC.

Mana pristupa je nemogu nost ostvarivanja me usobne ovisnosti kapaciteta zidova na prijem popre nih sila u zavisnosti od ostvarenih normalnih napona u pojedinim spojnicama.

LITERATURA

1 3muri, manuale d'uso 3.2.2., S.T.A. DATA srl - C.so Raffaello, Torino 2 ESECMaSE: Enhanced Safety and Efficient Construction of Masonry Structures in

Europe, http://www.esecmase.org3 Eurocode 8, Design of Structures for Earthquake Resistance, Comité Européen de

Normalisation, Brussels, 2004. 4 Capacity design of masonry buildings under cyclic loading /C. Butenweg, C.

Gellert, L. Reindl /, http://www.baustatik.rwth-aachen.de5 Analiza uticaja popre nih zidova na nosivost zidanih zgrada na dejstvo

zemljotresa / G.Simonovi / Magistarski rad, Sarajevo 2008. 6 Doprinos popre nih nosivih zidova zidanih zgrada na nosivost zgrada pri dejstvu

zemljotresa / G. Simonovi , B. Verbi / GNP2008, drugi Internacionalni nau no-stru niskup, Gra evinarstvo – nauka i praksa, zbornik radova – knjiga 1, Strane 463-468, Univerzitet Crne Gore Gra evinski fakultet, Žabljak 2008 7 Doprinos popre nih nosivih zidova seizmi koj otpornosti zidanih zgrada / G.

Simonovi , B. Verbi / Zemljotresno inženjerstvo i inženjerska seizmologija, Zbornik radova, Strane 127-132, Savez gra evinskih inženjera Srbije, Sokobanja 2008 8 Nonlinear model of masonry wall, / G. Simonovi , B. Verbi /, International

Scientfic Symposium «Modeling of structures», Strane 653-664, University of Mostar, Mostar 2008

155

. 122

-2 '' ''

:

-2 '' '' , ,

( L = 2.8 – 4.7) ( L = 5.4). -

, 7. 13. ,

. -2 .

: , , , ,

SOIL STRUCTURE INTERACTION OF BK-2 BUILDING IN BANJA LUKA Summary:

Residential building BK-2 in Banja Luka, Republic of Srpska, is a rare example of an instrumented reinforced concrete building, which was exposed to action of significant number of small earthquake (ML = 2.8 – 4.7) and to one moderate earthquake (ML = 5.4). The recorded accelerograms in the sub-basement, on the 7th

and 13th floors, make it possible to develop and to verify models for the dynamic analysis of the response to earthquake shaking. The results of our investigations of the BK-2 structure are presented in this paper.

Key words: earthquake, soil, structure, interaction, building BK-2, Banja Luka

1 , , , '' '', , 62-64, 11000, , .

156

1,

,. - ,

- , -. -2 ,

- - ,.

1972. , '' '' 5 - 13- - [1-4].

, -2 ( -2, -2), -, 1972. ,

[5]. ,

. 1973. , -2 , 7. 13. .

1974. 1986. ( L = 2.8 – 4.7) ( L = 5.4), 13.

1981. . , , 1982. 1983. , -2

[6]. , 7. 13.

.-2,

(1) ( ) (2) ( ). (2)

. (N-S, E-W V)

, . 7. 13.

. : 1. ,

. 2. ,,

( L = 5.4) [7].

2-2 ( 1)

. , 17,84 17,84 m. 2,47 m, 12 2,80 m,

157

3,40 m .

1 – -2, '' '' ,( .[12])

2 -2 , 7. 13 .

,. -4,24 m,

.

158

2.47

m +

13

x 2.

80 m

= 3

8,87

m

1st

2nd

3rd

4th

5th

6th

7th

8th

9th

10th

11th

12th

13th

Roof

4 bays @ 4.20 m = 16,80 m

6 x

2.80

m =

16,

80 m

2.47

m +

6 x

2.8

0 m

= 1

9,27

m

N

A B C D E

3,40

m

1,35

m

2,47

m

4,24

mSMA-1

SMA-1

SMA-1

13 x

2.8

0 m

=36

,40

m

2 – -2 - (.[12])

-, .

15 cm . , , - (38 38 cm)

159

3 . 22 cm.

.-

. ( ) . -

, ,.

3-2,

( 1972. ), SAP 2000 – 10.0.2.

,. 0.47 m.

, ( ), 12. -2

( )( 1972. ),( 1973. ).

( 1972. )

1972. ,

, . [8]. 1972. 0.59% ,

1972. 0.81% . [8] 40%

N-S 30% E-W .

– ( 1972. ) :MFINAL = 1.33MNOSIV (1) MEKSPL = 1.67MNOSIV (2) KEKSPL-N-S = KFINAL-N-S = 1.67KNOSIV-N-S (3) KEKSPL-E-W = KFINAL-E-W = 1.43KNOSIV-E-W (4)

-2, MODELNOSIV, MODELFINAL MODELEKSPL,

. ,

( 3). ,, -

160

, . , -- -

, ,

-2. Depth [m]

0

1.1

4.3

8.2

Vp = 350 m/s, Vs = 150 m/s, = 1.60 g/cm2

Vp = 690 m/s, Vs = 250 m/s, = 1.80 g/cm2

Vp = 1400 m/s, Vs = 400 m/s, = 1.85 g/cm2

Vp = 2150 m/s, Vs = 650 m/s, = 2.00 g/cm2

Gritty clay

Sandy gravel

Gritty gravel with clay

Marly clay

3 – -2 (.[12])

7. 13. .

,, , , .

.

[7], , 10 ,,

-2 [5,6] -2 [8-11].

4

4.1 ( 1972. )

E-W N-S E-W N-S , -

( 5.1 [7]). -

161

, 1972. (PVJUL72), (MN-E-2).

4% E-W , 8% N-S 3% , 42% E-W , 1% N-S 4%

. E-W

E-W , ,.

1972. . , , MN-E-2 ( 4 5.1 [7]),

MN-K ( 2 5.1 [7]), ,

[5,6,8-11].

4.2 ( 1972. )

.E-W N-S a

. ( 5.2 [7]). ,

, 1972. (PV 72), (MF-E-2).

1% E-W , 5% N-S 2%, 20% E-W , 4% N-S

9% . , ( ), MF-E-2 ( 4 5.2 [7]),

MF-K ( 2 5.2 [7]), ,

[5,6,8-11].

4.3 ( 1973. )

. ( ),

.

ML = 3.3 4.7 ( 5.4 8 9 [7]).

- ML = 5.4

13. 7. ( 4 , )

162

0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00Time [sec]

-6.00

-5.00

-4.00

-3.00

-2.00

-1.00

0.00

1.00

2.00

3.00

4.00

5.00

6.00Po

mer

anje

[cm

]

Registrovano pomeranje na 13 spratu, Komp. N-S

Proracunato pomeranje na 13 spratu, Komp. N-S

ZGRADA BK-2, BANJA LUKAELASTICAN MODEL NA 1B SLOJEmodel = Epocetno, Lamda = 4.5%Reversno zadate NS i EW, K=Kkonst

ZEMLJOTRES DOGODJEN13 08 1981 02:58 (GMT)Ml = 5.4, Edis = 8.4 km

0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00Time [sec]

-6.00

-5.00

-4.00

-3.00

-2.00

-1.00

0.00

1.00

2.00

3.00

4.00

5.00

6.00

Pom

eran

je [

cm]

Registrovano pomeranje na 7 spratu, Komp. N-S

Proracunato pomeranje na 7 spratu, Komp. N-S



4 – ( ) () 13. 7. -2

ML = 5.4 N-S ( [7])

( 5.4 10 [7]) .

163

0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00Time [sec]

-6.00

-5.00

-4.00

-3.00

-2.00

-1.00

0.00

1.00

2.00

3.00

4.00

5.00

6.00

Pom

eran

je [

cm]

Registrovano pomeranje na 13 spratu, Komp. E-W

Proracunato pomeranje na 13 spratu, Komp. E-W



0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00Time [sec]

-6.00

-5.00

-4.00

-3.00

-2.00

-1.00

0.00

1.00

2.00

3.00

4.00

5.00

6.00

Pom

eran

je [

cm]

Registrovano pomeranje na 7 spratu, Komp. E-W

Proracunato pomeranje na 7 spratu, Komp. E-W



4b – ( ) () 13. 7. -2 ML = 5.4 E-W

( [7])

N-S (f = 0.91 Hz) E-W (f = 1.00 Hz) N-S E-W .

164

1 World Housing Encyclopedia Report (country: Yugoslavia) / . // Earthquake Engineering Research Institute and International Association of Earthquake Engineering, 2002, Oakland, California, Report 68, http://www.world-housing.net/.

2 - / . // , 1982,

, IX, 1, 2-7. 3 / . // , 1982, ,

IX, 1, 8-17. 4

/ . // , 1982, , IX, 1, 18-25. 5 Forced-vibration test of 13 story building in Banja Luka, constructed by the system

IMS-Zezelj / J. Petrovski, D. Jurukvski, S. Per inkov // Institute of Earthquake Engineering and Engineering Seismology, University “Kiril i Metodij”, 1975, Skopj , Report DTL 3-75.

6 -2 -4 +12 , / . , . //

, 1983, , 83-86.

7 -2 : / . // , 2009, , LXIII, 5-6, 189-210.

8 Comparison of analytically and experimentally determined dynamic behavior of a multistory RC building / P. Fajfar, M. auševi , Y. Yaing // Proceedings of the EUROBUILD 87, 1987, Dubrovnik, 134-139.

9 Mathematical model formulation of a fourteen story RC building using motion recorded and parameter system identification / D. Jurukvski, Lj. Taškov, V. Trajkovski // Proceedings of the 8th World Conference on Earthquake Engineering, 1984, San Francisco, California, Vol. IV, 615-619.

10 Zemljotresno inženjerstvo – visokogradnja / D. Ani i , P. Fajfar, B. Petrovi , A. Szavis-Nossan // Gra evinska knjiga, 1990, Beograd.

11

/ . // , '' . '', 1992, ,

.12 Impulse response analysis of the Borik-2 13-story residential building in Banja Luka

during 20 earthquake 1974-1986) / M. Trifunac, M. Todorovska, M. Mani , B. Bulaji // University of Southern California, Department of Civil Engineering and University “Ss. Cyril and Methodius”, Institute of Earthquake Engineering and Engineering Seismology (IZIIS), Report CE07-02, 2007, Los Angeles, California.

165

D. Šumarac123, Z. Petraškovi 2 24

KONTROLA OŠTE ENJA I POPRAVKA RADI BEZBEDNOSTI ZGRADA

Rezime:Gra evinski objekati, naro ito zgrade, podložni su ošte enjima i rušenju pod uticajem zemljotresa. Novi DC damper sistem može pove ati otpornost kon-strukcije i oja ati konstrukciju, ime se omogu ava ve a stabilnost objekata pod uticajem seizmi kih dejstava. Iskustva za dalji razvoj ste ena su tokom ekspe-rimentalnih istraživanja, kao i kroz praksu saniranja više od 350 ošte enih obje-kata na etiri kontinenta. Klju ne re i: zavareni spoj, elik, nesavršenosti, mikrostrukturna nehomogenost, servisni uslovi, strukturni integritet

DAMAGE CONTROL AND REPAIR FOR SECURITY OF BUILDINGS

Summary: Civil engineeiring object, esspecially buildings, are prone to damage and failure under earthquake impact. Invented DC Damper System can increase the resistance and strengthen the construction, enabling tougher behaviour under seismic load. The experience for next development has been achieved during experimental research and in practice of repairing more than 350 damaged object on four continents. Keywords: welded joint, steel, imperfections, microstructural inhomogeneity, service condition, structural integrity

INTRODUCTION

Earthquakes are very dangerous impacts on civil engineering structures. It is practically impossible to understand the behaviour of buildings subjected to the seismic

1 Faculty of Civil Engineering Faculty of Civil Engineering University of Belgrade, Kralja Aleksandra 73, 11000 Belgrade, Serbia e-mail: [email protected] 2Faculty of Civil Engineering University of Belgrade, Kralja Aleksandra 73, 11000 Belgrade, Serbia e-mail: [email protected]

166

loads without essential knowledge of behaviour of the construction members from the view point of low-cycle fatigue process as well as the complete stability of the construction during all phases of earthquake activity, because masonry structures are very sensitive to them. It is well known that those structures have large mass, and consequently, because of bad cohesion between bricks (stones) and mortar they crack and suffer damage when exposed to earthquakes, since they can’t avoid non linear post elastic condition. Strong need and desire to find a new effective object protection from seismic loads and realization of the tougher concrete constructions resulted a new, The DC 90 Construction System1 and associate devices, presented in this article.

*To whom correspondence should be addressed: Zoran Petraškovi , Earthquake engineering innovation center System DC 90, Vele Nigrinove 1, 11000 Belgrade, Serbia E-mail: [email protected]

DEVELOPMENT OF THE DC 90 CONSTRUCTION SYSTEM

System DC 90 comprises a number of structural elements which strengthen brittle walls and make them ductile and tough. They make floor slabs and ceilings stiff and capable to transmit the load in their own plane, and connect them by foundation collars. These elements make structure stronger to accept the horizontal and vertical loads. This invention is based on the construction system with damper- absorber which makes the building structures more resistant and lets them withstand the highest values of earth tremors through elastic plastic work and plastic deformation (flow) control. The damper member is tearing in low cyclic fatigue in accordance with engineered values of accumulated dilatations so that can accept more than three or four high quakes. The construction is very effective at the masonry objects of historical value, at the modern nuclear power stations and other objects of any security importance during the life of structure. The constructions capable to achieve higher level of post elastic non linear condition (that means higher ductility) are likely to survive the damage that seismic loading may cause to such structures. Analyzing the behaviour of different type of materials commonly used for construction building (concrete, bricks, stone, wood, plastics) the outstanding ductility of steel elements in building constructions is doubtless. The need to provide deformation control that due to the inadmissibly large deformation scale may cause the destruction of the elements or total structure collapse was particularly considered.

The essence of the new device (damper-absorber) is to provide an accurate and controlled elastic-plastic work. The most important parameters that define the damper construction and its properties are shown in Figs. 1 to 3, where , present stress, strain, and P, – force, displacement, respectively. It can be seen the damper behaviour under cyclic activity through time, which depends on the following factors: accumulated strain, frequency, cycle number and damper properties, defined by model testing in the laboratory for dynamic testing2.

Special contribution represents the inovative design Damper DC 90. It is protected by patent in U.S.A.3 and awarded by Gold Medail in Bruxels. Damper is involved as a part of vertical stiffener elements, and thanks controlled fatigue defines position of plastic hinges, instant of its initiation, intensity of force and deformation, and their control,

167

affecting structure behaviour in seismic condition. The DC 90 Construction System defines the points for DC Damper installation and the parameters of low-cycle fatigue.

The hard work from inventing and innovating to testing and final realization of the technology all over four continents contribute to the understanding of new technology. The permanent innovation of the device and technological process, model testing and in-situ testing as well as technology transfer are of the significant importance.

Pecularity of damper design is the middle part shape (“dog bone”)3, with finished surface to eliminate surface cracks and prevent undercuts. By three limit rings snd one movable ring it controls displacement at predetirmened length of controlled low cycle fatigue due to effect of variable sesmic loading, and by reduction of cross section the load intensity is controlled. Local buckling stability of pressed elements is achieved by special elements. In this way it is possible to locate and control the position of plastic hinge. Changing system stifness and dynamic characteristics of the structure its operation is controlled.

Three diagrams show the ability of the System DC 90 Damper design. The principal feature of the hysteresis loop diagram force vs. displacement (Fig. 1)

is the possibility of deformation control.

-250

-200

-150

-100

-50

0

50

100

150

200

-5,0 -4,0 -3,0 -2,0 -1,0 0,0 1,0 2,0 3,0 4,0 5,0

Displacement [mm]

Forc

e [k

N]

Figure 1. Hysteresis loops diagram force vs. displacement of System DC 90 Damper

Diagram displacement vs. number of cycles (Fig. 2) shows a very good damper performance even at high number of cycles. It is also obvious that accumulated strain increase with decreasing number of cycles to collapse. This feature is used to determine the damper dimensions and application field.

Energy increases with increasing accumulated strain, but decreasses with number of cycles to collapse of damper, as presented in diagram in Fig. 3.4

168

-5

-4

-3

-2

-1

0

1

2

3

4

5

0 200 400 600 800 1000 1200 1400

No of cycles

Dis

plac

emen

t [m

m]

-20000

0

20000

40000

60000

80000

100000

120000

140000

160000

0 200 400 600 800 1000 1200 1400

No of cycles

Ener

gy [k

Nm

m]

Figure 2. Diagram displacement vs. number of cycles

Figure 3. Energy decreasses with number of cycles to collapse of damper

NUMERICAL MODELLING OF BUILDING AND ITS VERIFICATION

Earthquakes are more dangerous impact on civil engineering structures, and masonry structures are sensitive to them. It is known that those structures have large mass, and because of very bad cohesion between bricks (stones) and mortar they suffer damage by cracking under the effect of earthquakes.

Development of numerical model

Addtional contribution to the System DC 90 Damper design is obtained by finite elements (FE) modelling of building. It is necessary for retrofitting by DC 90 technology of damaged building. Testing of specified dampers, including vibro platform, was performed in order to verify obtained models

Typical damage of two store building made from bricks and mortar is presented in Fig. 4, showing that cracks grow in direction of cross diagonals started from openings (windows, doors). The building is modeled by shell FE (Fig. 5)5.

Figure 4. Damage of two store building Figure 5. Building modelled by shell FE

169

The earthquake is represented by El Centro response6. The analysis, performed for two directions of earthquake effect, shows that the largest tensile stresses, responsible for cracking, occur between holes (Fig. 6).

The building design was strengthened, and new model is presented in Fig. 7. Vertical elements - walls are strengthened by vertical stiffeners that connect all

horizontal slabs and the foundation. Vertical stiffeneers – trussess consist of the vertical ties, which are pre-stressed, and the other elements are diagonals with the seismic energy absorber and horizontals as a part of stiff floor slabs. Walls strengthened in this way become tough and capable to accept the alternative horizontal dynamic displacements.

If horizontal element are not stiff in their own plane, floor slabs and ceilings are being strengthened by impregnation with a thin, lightly reinforced, concrete slab or incorporating horizontal bracings, connected with the vertical stiffeners.

Figure 6. Position of largest tensile stresses Figure 7. FE model redesigned building

The foundation structure is confined with the foundation collar, connected by anchors and in which the vertical stiffening elements are anchored.

Testing of dampers

Produced dampers for DC 90 system (Fig. 8) were tested by variable loading on MTS servo-hydraulic closed-loop machine (Fig. 9) in Military Technical Institute (VTI), Belgrade7. Obtained hysterezis loop diagram is given in Fig. 10.

170

Figure 8.Typical dampers of DC 90 system Figure 9. Damper testing on MTS machine

Figure 10. Hysteretis loop diagram obtained by testing of dampers

Analysing of diagrams force-displacement one can conclude that clasic low cycle fatigue is in question. During operation, in displacement control, permanent plastic defor-mations appear, or in another words, with increasing number of cycles material weakeneds and maximum force reduces: when force decreases for the same deflection, material is weaker. For smaller displacement, number of cycles needed for failure is greater8. This is not classical failure of the sample, this is above all weakening of the sample.

Testing of retrofit and non-retrofit objects

Experimental testing on non-retrofit and retrofit object in Mionica, Serbia (Fig. 11), performed by Institute for earthquake engineering and engineering seismology (IZIIS)9, Skopje, by methods of ambient and forced vibrations, have shown that dominant frequencies of built objects are within the range of 6-8 Hz, while after retrofit stiffness raises up to approximately 35%.

171

Figure 11. Experimental testing of building

TESTING by QUASIDYNAMIC LOADING of HOLLOW BLOCKS WALL

Figure 12 presents testing by quasidynamic loading of experimental two store wall, made of hollow blocks 19x19x25, framed by girders, as classical solution, (left), as strengthened according to DC 90 system (midlle) and during testing. after few cycles(right): for displacement 10 mm of console top point, extended flaws of 8-10 mm appeared in the first area from the point of force application. In the area to the wall constraint, cracks typical for dominant bending stresses appeared10.

Initial shear stiffness was Es-start = 60/6 = 10 kN/mm; after few cycles for deflection of about 30 mm, it decreased to Es-end = 9/6 = 1.5 kN/mm. This multiple decrease of wall stiffness can be considered as a collapse of the wall and whole building.

Figure 12. Two-store wall model, before retrofit, after retrofit and during testing

Shear stiffness at testing of strengthened - retrofitted wall was initially Es = 20/3.5 = 5.7 kN/mm, and later it slightly decreased, mostly due to yield of lower anchorage tie, and then it rised up again. That leads to the conclusion that anchorage must be done properly, and that quality control of performed works must be done.

Wall behaves much tougher at cyclic loading. For loading of 60 kN, for top displacement of 25 mm far less cracks occurred, with openings 2 to 3 mm.

Tested stiffener has carrying capacity for larger deformation, shows sufficient durability and probably can, as part of the vertical stiffening frame, preserve the building during increasing deflections and keep it from collapse, making the building safer during

172

earthquake, especially if building is built without girders. Described testing was performed in the Institute for material testing (IMS), Belgrade.

SHAKING TABLE TEST OF A wall MODEL Exerimental research of models (1/10) of masonry walls strengthened by DC 90

system was performed with shaking table (Fig. 13) in the Dynamic Testing Laboratory of Institute of Earthquake Engineering and Engineering Seismology (IZIIS)11 in Skopje, Macedonia. Three models have been constructed in scale 1/10, with length of 30 cm, height of 25 cm and thickness of 3.5 cm, made of gitter bricks, of plane burned bricks and of plane dried bricks. Each type of model was made as conventional and as strengthened by DC 90 System.

Model 1. Hollow brick Model 2. Burned brick Model 3. Dried brick Figure 13. Shaking table test of a wall model

The single component shaking table has been used to test the models under harmonic excitation within the frequency range of 1.0-100 Hz and amplitude range (0-10) g. The idea was to compare dynamic behavior of traditional and strengthening method of construction for the same bricks type, and also to compare the effect of brick type on dynamic behavior of the models.

The testing program consists of several phases: - Definition of resonant frequencies - Definition of elastic response of the models, comparing non-strengthened and

strengthened models - Determination of limit state and fracture mechanism Result of testing are presented in Table 1. The calculated stiffness should be

related to the scale factor 1/10 to obtain the actual stiffness of the real wall element. The results have clearly shown increasing of stiffnesss of walls by f the elements of System DC 90.

Table 1. Calculated lateral stiffness of the models, kN/cm

Type of the model Strengthened model Non - strengthened.model Model 1- hollow brick 35.53 24.75 Model 2- burned brick 19.95 10.49 Model 3- dried brick 6.48 4.40

173

From numerical analysis performed during this study and from experimental results it can be concluded that System DC 90 is powerful tool for engineers to solve problems of retrofitting of damaged structures. Since this system is cheap, fast and applicable outside, it can be applied properly for earthquake regions, as it is done in Kolubara region, Serbia, and in Algeria.

EXAMPLES OF APPLICATION OF DC 90 SYSTEM

Thanks to your attractive performance, DC 90 System became very popular. It has been successfuly applied on 4 continents: America, Africa, Asia and. Europe, Most interesting applictions will be shortly presented.

Wall construction in hall of Bornuda Hydroelectric Power Station, Quebec, Canada

The wall construction of the engineering hall at Bornuda Hydroelectric Power Station in Quebec, Canada highlights some basic properties of applied “Mionica+” type damper. The designed damper construction was tested at VTI, Belgrade, and as presented wall model testing at IZIIS, Skopje12. Numerous testing of more than 50 specimens were conducted at Civil Engineerinf Faculty, Ljubljana, Slovenia, and VTI, Belgrade.

The estimated seismic loading of 0.20 g and frequent appearance of minor quakes at the area of large hydroelectric objects in Quebec region, Canada, motivated the “HydroQuebec” company to analytically approach the problem of damage risk estimation as far as company’s objects are concerned. Numerous detailed tests of materials and constructions, model testing and numeric analyses were conducted to cover the topic. The numeric analyses were made by Canadian, Britain and Indian expert teams, dynamic model testing was executed by means of the vibrating platform at IZII, Skopje, and dynamic testing of objects by ambient vibration method was made in Canada. All these events precede the damper model testing at VTI, Belgrade

Numeric modeling Numeric modeling of damper performance designed for wall construction of

engineering hall at ˝Boauharnois˝ Hydroelectric Power Station in Canada is made on the base of following elements of stress – strain curve ( – )13.

a. Initial elasticity module to linear elastic limit Eo=20.000 kN/cm2 (that is valid for strain values up to = 0.0011 and stress values up to = 22 kN/cm2.

b. Elasticity module at yielding limit E1 = 4.333 KN/cm2 (up to = 0.0060 and = 26 kN/cm2). As number of cycles and frequency increase (strain/sec), the elasticity module decreases due to accumulated strain. For strain range from 0.0011 to 0.0060 stress values of initial phase vary from 22 to 26 kN/cm2, it is about yield limit. It can be presented by any approximate curve. However, for simplification this is done by means of COSMOS software as a bilinear curve.

174

c. Elasticity module (actually, it is a curve slope in plastic regiime operation of the damper) E2 = 120 kN/cm2 (for > 0,006 and = 26-32 kN/cm2). As number of cycles and frequency increase, the elasticity module increases.

The values of E1 and E2 elasticity module depend on: - cycle number; - accumulated strain; - values of accumulated strain through time. The Coffin-Manson relation is given as:

pNf = C1 1p fN C

where p is the cyclic plastic strain range (accumulated strain), Nf is the number of cycles to failure, and C1 and are material constants. The exponent usually varies between 0.5 to 0.7, and depends on whether torsion or axial loading is in question. The constant C1 is defined from damper serial test.

The Palmgren-Miner cumulative damage rule proposes that:

n/ Nf = (n1/Nf1) + (n2/Nf2) + ........ = 1.0 1

1 0k i

i i

n ( t )D( t ) .N

where n1 is the number of cycles at the stress or strain range level 1, and Nf1 is the number of cycles to failure at that stress or strain range, etc. The problem of low-cycle fatigue and operation of damper in large strain range up to 5% can be presented by numeric methods, but bilinear relation would be quite enough for the analysis with elementary work through time improvements.

General information on applied DC90 Damper performance

Bilinear stress - strain curve of the damper (the approximation of damper operation can be presented by bilinear stress–strain curve, it is valid for the ascendant curveand E2 curve). The slope of the ascendant curve decreases depending upon the number of cycles and accumulated strain according to the Coffin-Manson law (the function of two coefficients: the number of cycles Nf and the value of accumulated strain p).

If it is necessary to limit the displacement, for example, by 5 mm that corresponds to the maximum strain of 5%, then it is possible, accordingly to calculate the average slope of the ascendant curve. The slope of the E2 curve for this type of damper is 3-10% of the ascendant curve slope, obtained by experiments.

The consideration of frequency effect additionally complicates the diagram. DC 90 Damper works in plastic area at 100 mm weakening length. So, the pipe,

having a size of 16x1 mm, should be weakening by two longitudinal cuts of 7 mm long. The pipe surface is P = 0, 50 cm2. The surface of the weaken cross-section is P -

2x0.7x0.1=0.36 cm2. As damper design prevents lateral bending and local buckling by means of external and internal elements built into the pipe (micro reinforced polymer concrete and pipe covered by aluminum foil). The overall weakening length of the damper should be covered by plastics. It is ca 100 mm. To simplify the matter one can consider the average values of strain and stress for overall weakening length, i.e. the length that works in

175

plastic zone. Besides, to sustain the fatigue (low cycle fatigue) the surface of the pressed part should be removed to exclude the possibility of any local or surface cracks (damages).

The damper is made of flexible structural steel of high ductility produced by “US Steel” company (Smederevo, Serbia). All the dampers should be made of the steel of same composition of guaranteed mechanical characteristics. There is a variety of steel compositions with different tensile properties, but no one is beyond the guaranteed value.

The initial part of curve slope (E – elasticity module) should be at standard level, as well as yielding limit. But in this case the E2 curve, as well as total strain increase thatprovides damper effects (behaviour) at low cycle fatigue. The deformation is controlled by distance control elements (rings). The lead addition affects the reduction of stroke and brittle fracture as well as damper behaviour and work

Anyway, the above described design is supposed to provide high strain (accumulated strain) during numerous large earthquakes.

The object of the research is to learn how to manage the materials crystal grid disarrangement and to arrange it at the length of 100 mm. In real products the disarrangement is concentrated on the small areas and can’t be controlled at plastic hinge area. Such method of the deformation control can be used in design made of reinforced concrete (frames), steel or other systems.

Residency of Finland Ambassador in Algeria, Africa The detailed research of construction condition, as well as a non linear dynamic

analysis and seismic strengthening of the object were conducted to meet the requirements of Ministry of Foreign Affaires14. Intensive technological research was made during realization of the object.

The realization of the object was aimed to investigate the technological process in aspect of promotions humanization, economical realization and modifying or simplifying construction solutions of the system in this work.

Principle features of the object

Object dimensions: 27.00x17.75x13.02m Number of floors: basement, ground-floor, first floor, second floor, tower. Walls: made of stone d = 80cm. Inter-storey structure: Steel bracings I180 with bow of bricks (h = 40cm) Foundation: stone wall, d=80cm

176

The view of the object during repairing action is presented in Fig. 14.

Figure 14. View of residency of Finland ambassador in Algeria under retrofitting

Technology and process features

Weight of steel elements: 3552 kg Total number of dampers-absorbers is 41 pieces, consisting of 18 pieces of type

“Algeria”, 19 of type “Mionica” and 4 of 30 (4). Total number of displacement compensators: 29 Total number of wall connectors: 59 Duration of cracks and splits repairing - from 21.11. to 28.11.2006, duration of

object strengthening – from 06.12.2006 to 30.12.2006. Total time spent on cracks and splits repairing: 553 hours. Total time spent on construction strengthening: 2280 hours. Working hours: 10 hours per day with eventual breaks caused by rain. Engaged experts staff included 2 civil engineers, 1 interpreter, 4 metal structure

experts, 6 experts in erection and construction, 1 assistant, the cook. Equipment. Classic pipe scaffolding, movable aluminum scaffolding, diamond

saw for wall cutting “Stihl”, drilling hand tools “Bosch” and “Hilty”, injecting pump, electric arc welding equipment, concrete mixers, dozers and digital weighing machines, equipment for control and measurements and necessary small instruments and appiencies.

Technologies applied on the object

Vertical wall bracings with dampers. Horizontal tension of inter-storey structure by means of displacement

compensators - time deformations. Vertical tension of the tower by means of damper and displacement compensators

through time. Wall connectors at the positions of wall conjunction to preserve wall integrity and

avoid separation.

177

The conclusions made after technological research and analysisThe technological problems of high noise and huge amount of dust during wall

cutting is solved applying hydraulic machines with diamond saws and dust aspirators. Alternative solution is moistering the surface during wall cutting.

The problem in construction, significant wall destruction after cutting procedure, particularly refered to “Algeria” type of diagonal bracings, 100 mm, had been solved by applying all bracing members (vertical, horizontal, diagonal) made of filled steel elements (pipes) of square or round cross-section of minimal dimension 20 mm.

The technology of concrete pressing should be used. The concrete must have an adequate consistence (WK).

Design solutions The system should be based on the following assumption: 1. All the members (vertical, horizontal, diagonal) have filled square or round

cross-section. The construction members can be built inside or outside the walls with specially designed details which provide the connection between walls and construction elements.

2.The connection details should be typified and unified. 3.The life span of all steel elements is of great importance. The problem will be

studied by the experts in technology and structural integrity and life. All the conclusions and recommendations gathere in this activity were taken into consideration during the Azerbaidjan President Residence rehabilitation project.

Azerbaidjan President Residence, baku, Azerbaidjan, Asia “SERBAS” Company, Baku, applied the DC 90 System technology to provide

seismic strengthening and protection of the Azerbaidjan President Residence building in Baku15. In Fig. 15 are designed the location of applied retrottit action.

The results of the technological research conducted in Algeria contributed much to technology promotion and elimination of all unfavourable observations during realization of project in Baku.

(konceptualno resenje)

-1.03

600 2600 1280 2600 860

250 2202600 640 880 1080 2600 1080 957

2622600 850

2451902600 1330 2560

170

501

499180

930 2385 875170

900990114025151270

3131

2980

1550

3020

570

268

0.00

-1.42

1.05

1.15

0.07

4011569

1200

5798 67

0723

21

4974 1580

93

247

1240 1510 710 880 1600 1550 1610740

210

580 1500 1640

245

4600 1530 720 1000 600 3320 612

900760

0.25

730

6257

1500

738

4738

860

2900

6318

1550

630

8498

6328

1500

670

4380

3344

775 17

8924

0043

1090

052

0091

031

20

550332023201530505090053515003140155031901510380

3450

2904

2993

650 1350 1385 2760 9050 900 8920 3300 550

1070

2895

1400

2370

1430

5200

902

3110

895

1225

728

1620276013851350850204

324 3535

300

1000 1902 1220300 159

3861 920

400

2360 1070 2930 1070 2270

210

862

250

811 2070 1730

1611716060

6432

5784

1378713753

7603

6976

5562

4322

7101

11631

10530

10787

8317

-0.15

6330

2420

R339

4

6780

2566

3347

542

950

598

250

383

1580

270

240

816

2202

2106

327 1243357

856 25

0

8498

2781 1080 920490

1560 1730 2610 1560 2600 882 799 3800420

0.00

-0.20

3657

525

100

871

536

540

560

3147

800

3277

500

400 2360 1070 2930 1070 2270210

862 250

1001

1160

950

1237

4300

1090

1970

4936

2310

1381

1319

570

960

560

250 286

3088

1505

3038

855

380

1520

1530

400

3793

3320

10402631

1119

35541550 7701260

1540410

4553

6976

7117

7576

8273

1053653

25981 333

25876 20

605

320 1340 11002800

2800

6770

10252

10455

400

1580

350

39 22

1511870

1151

9

974

30 02

287 8102622

377

850

5109

2266900

1061

4492

1069

6781

886

2421

850

880

1580

970

3401160

920

2451

4558

358124 2

32 7

856

250

2701580

240

1910

280

356 6

873

2680

4043

2417

973210

890210

1070

320

511

1420

730

1379

790

2940

500 600 600 5724284

1853

32 10740

310

994

3 657 90

5 0011 2 0

51 07 9 0

3 6 0 2860 1390 1970 159 013 40

1 51 0

5229544631

7546 7638

115974 0

282

1987

2084

2061 865 2030

1652

2053508

1618

681

351 334

1400

1210

1360

3 281197

8021180

38 0

370 2796 314

0.00

? 1:100.

120 6

5

640 2930 1070 2280

210

900

162 69

826

3719

8

3195

1107 1530 996

3339

159

2565

1651

800

1112

1160

1163

697

1935

901

1370

830

960

436

901 337781

2595

760 1751181

P I

P I

5

A

6 7 8 9

4

1

B

V

G

3

R - II

P III

R - II

P III

2

- VI

R - VI

R - V

R - V

P IV

P IV

3800

DJ

DY9,

5

3500

Y9,

A

.X J

.X D,J

J

.Y7,

5

.X 6,5

polozaj horizontalnog utezanja

polozaj vertikalnih ukrucenja

L E G E N D A

B<3,2 mTIP "K"

B>3,2 mTIP "D"

vert

ikal

no u

kruc

enje

vert

ikal

no u

kruc

enje

vert

ikal

no u

kruc

enje

vert

ikal

no u

kruc

enje

damper "BAKU"damper "BAKU"

damper "BAKU"

"D"

"D""D"

.XY 1

2795

XY 1`

3170

.

XY 2.

3550

.XY1-2

XY

1-23 860

3307

.

.

5453

4976

XY5-

6

X B,5-6

.

Y5,

A-B 5285

.

6496

X B,8

5640

Y

XY1

XY2

XY7 12

3 4

6

5

8

97

10

11 12 13

XY8

XY3

XY4

XY5

XY6

XY9

XY10

XY11

XY12

XY13XY14

XY15

XY16

XY17

XY18

XY19

XY20

XY21

XY22

XY23

XY24

XY25

XY26

14

15

1617

18

19

20

21

22 23 24 25 26

27

28 29 30 31 32

3334 35 36 37 38

3940

41

42 43

44 45

46 47 48 49 50 51 52 53 54 55

8093

30080

800 800

800 800

parket 2,5 cmAB ploca 6,0 cmdrvene grede 30/8 cmdrvene grede 30/8 cmdascana oplata 2,5 cmplafon 2,0 cm

?????? V-V .

? 1:100.

0.00

5.90

0.00

5.90

0.00

5.90

0.25

4.58

5.485.48

0.00

3.70

3.95

6.45 6.45

7.35

16.68

11.55

15.80

15.5815.20

18.90

10.25

0.00

5.90

6.45

12.50

13.0513.05

17.06

23.10

23.73

26.12

27.60

28.41

32.27

34.36

36.38

"D"

damper "BAKU""D"

damper "BAKU"

"D"damper "BAKU"

verti

kaln

o uk

ruce

nje

verti

kaln

o uk

ruce

nje

verti

kaln

o uk

ruce

nje

H=

6,45

m

B<3,2 mTIP "K"

B>3,2 mTIP "D"

vertikalno ukrucenje verical bracings(konceptualno resenje) conceptual solucion

"D"damper "BAKU"

vertikalno ukrucenje X B,8

verti

kaln

o uk

ruce

nje

.

"D"damper "BAKU"

verti

kaln

o uk

ruce

nje

"D"damper "BAKU"

"D"damper "BAKU"

"D"damper "BAKU"

.

vertikalno ukrucenje X B,5-6

"D"damper "BAKU"

"D"

damper "BAKU"

verti

kaln

o uk

ruce

nje

verti

kaln

o uk

ruce

nje

"D"

damper "BAKU"

"D"

damper "BAKU"

"D"damper "BAKU"

"D"damper "BAKU"

"D"

damper "BAKU"

"D"damper "BAKU"

"D"damper "BAKU"

"D"

damper "BAKU"

"D"

damper "BAKU"

"D"

damper "BAKU"

damper "BAKU"

"D"

"D"damper "BAKU"

damper "BAKU"

.

vertikalno ukrucenje XY 1-2

Figure 15. The locations of applied retroffitting on the Azerbaidjan President Residence

178

Some examples of application of DC 90 System in Europe.

In Serbia, Montenegro and Slovenia many retrofitting projects have been realized on different objects, mostly on the family houses (Fig.11). In January 2001 DC 90 System Innovation center accepted an order for experimental strengthening of the six different objects damaged by earthquake in Kolubara region (Fig. 4). The objects are situated in the area of Ljig place.

After detailed experimental verification the technology was promptly apply to rehabilitate 350 objects in six municipalities in trussed (critical) Kolubara region in Serbia. The experimental reconstruction of house (property Lazi ) and object strengthening by applying DC 90 System technology are presented in Fig. 16.16

Figure 16. Experimentally reconstructed house, Lazi property (left) and strenghten object (right)

PROJECT ”PROHITECH” The project ”PROHITECH” is a Euro-Mediterranean project aimed to provide an

earthquake protection of historical objects. It is leaded by Prof. F. M. Macovani from Naples University, who invited the Innovation Centre for Seismic Engineering, Belgrade, to participate in research and cooperation programs. Other participants in the projects are institutes and universities from Israel, Ljubljana, Naples, Timisoara, Athens and Barcelona.

The DC 90 System technology was analyzed and tested in six WP projects ”PROHITECH”. The participants of the project are from Italy, Greece, Portugal, Morocco, Romania, Macedonia, Belgium, Slovenia, Turkey, Israel, Egypt and Algeria.

As far as the research is under way the results are presented without any analysis or comment. That type of damper was tested within “PROHITECH” Euro-Mediterranean project with participation of twelve European and Mediterranean countries and fourteen Institutes, Faculties and Research Centers.

DISCUSSION It is well known for a long time that people always wish to preserve their buildings

from destructions caused by earthquake as beyond any doubt the most priceless heritage. The process of creation, innovation and implementation of the invention in how to save or retrofit the integrity of building enabled to apply with succss new damper system all over

179

the world (America, Africa, Asia and Europe). The invention as human creation is generated and developed in accordance with the needs of its author at first and other users afterwards.

It is reasonable here to cited the reflexion about his topic of great scientist Nikola Tesla, probably the most brilliant inventor born. “Invention is a crown of intellect. The development of human kind is substantially depended on the invention, as most important product of the creative brain. Final goal of human kind is to master the nature by intellect and expoatation of its power for mankind needs”. Nikola Tesla recognised well that science can’t be realised by mathematics alone, since the facts appropriate for demonstration by symbolised process are minor and of less significance compared to great truth gathered by experience.

It is not possible to imagine development of mechanics without knowledge of substance essence of matter and physical fields (electrical, magnetic, gravity), mot only on macro, but also at micro and nano level, and also beyond. “Beyond” is always present. It is difficult to explain how at the end of infinity there is also next point, and how it is possible to divide infinitesimal value on two, so is impossible to predict development of mechanics and its un foreseeable capacity. In that sense we are prone to think that presented simple innovation is a contribution in mastering the nature. In same category belongs the development and implementation of new materials and structures, in this moment ended by nano structures.

New design of dampers will ask for mew, more tougher material, and in the scope of science and invention, this topic is steadily present.

CONCLUSION. Considering the performed researches and investigation it became clear that the

system for object seismic strengthening needs to be innovated continuously. The innovations should cover the areas of design of new systems, technology, and numeric modeling as well as.

The analysis at levels of crystal grid disarrangement and inter-atoms connections weakening, and even beyond that, determines the materials behaviour in non-elastic zone are f major importance. It can stimulate the discovery of completely different materials or structural members that react properly at seismic loads.

As far as bridge construction is concerned the investigations should be directed by numeric analysis that can provide necessary damper performances (stress, strain and force, displacement relations).

For special objects (nuclear power stations) the investigation should be made in the same direction according to the design schemes, necessary damper performances should be obtained by the numeric analysis.

The prior numeric and experimental damper behaviour investigations should be continued to guarantee the adequate numeric model invention.

It is undoubtedly that further innovations will give numerous original and bettre solutions of constructions and damper device designed for different building structures. The technological research of the process conducted at experimental and exhibited objects is one of the strategies of the “DC 90 System” Innovation Centre for Seismic Engineering.

180

After completion and equipping the laboratory for model innovation investigations and organizing the international research coperation the step with permanent creation of innovation and inventions willbe assured.

The loss of fracture ductility during the low-cycle fatigue process was investigated from the viewpoint of the existence of small surface cracks. The relationship between the growth of small cracks and the residual fracture strength during the low-cycle fatigue of 70/30 brass was investigated. The crucial cause for the loss of fracture ductility was elucidated on the basis of microscopic observations. The results are summarised as follows:

1. The low-cycle fatigue process in an annealed medium carbon steel (0.46% C steel) was almost 100% dominated by the growth process of a single crack. In an extreme case, microcrack initiation was observed on the surface of a plain specimen during the firststress cycle.

2. The previous history of fatigue practically has no effect on fatigue damage and hardly effects the subsequent growth of very small surface cracks.

3. The fracture ductility loss during the low-cycle fatigue was investigated from the aspect of small surface cracks existence rather than historical comprehension of so called fatigue damage phenomenon. If the surface of the fatigued specimen is removed by machine turning and thereafter by electro-polishing to exclude the possibility of any surface cracks there is no sign of the fracture ductility loss. In case of plain specimens the fracture ductility loss becomes obvious only when the relative number of cycles (N/Nf0) exceeds 0.6. On the other hand, in case of holed specimens the critical number of cycles (N/Nf0) is lower in comparison with the plain specimens. Taking into consideration so obtained experimental evidence the conclusion has been derived that the fracture ductility loss during low cycle fatigue damage is caused by the existence of fatigue cracks on the specimen surface.

4. After the material constants C1 and are determined the “DC 90 System” Damper behaviour obeysin satisfactory way both, the Coffin-Manson law and the Palmgren-Miner rule.

REFERENCES:

[1] Z Petraskovic, Seismic Strengthening and protection of objects, Monograf Sistem DC 90, Belgrade, 2005.

[2] Z.Petraskovic, D Šumarac, M. An elkovi , S. Miladinovi , M.Trajkovi ,Retrofitting Damaged Masonry Structures by Technology DC 90, Journal of the society for structural integrity and life, Belgrade, 2005, p. 59-71.

[3] Patent in USA No.10/555,131 from 31.10.2005, patent in Australia No. AU 2003254327A1 FROM 2004.11.23.

[4] Petraškovi , Z., Miladinovi , S., Šumarac, D., Technology of seismic strengthening of masonry structures by applying vertical ties and diagonals with seismic energy absorber “System dc 90”, International conference on earthquake engineering, Parallell Session, Topic: Retrofit of structures, p T6-9, August-september 2005.

[5] D Šumarac, Z.Petraskovic, M. Maksimovi , S. Miladinoi , I.Džuklevcki, N. Trišovi , Seismic Retrofit of masonry structures applzing vertical braces with dampers

181

Sistem DC 90 and newly designed wall buildings, Internacionalni nau ni skup, Žabljak Crna Gora, 2006, p. 373-381.

[6] The earthquake response, Institute IZIIS, Skoplje, Makedonia, 2005, p. 13-33. [ 7] D. Šumarac, Z.Petraskovic, M. Maksimovi , S. Miladinoi , J.Petraškovi ,

Structure Retrofit for residental house of Finlands Ambassador in Algier, Internacionalni nau ni skup, Žabljak Crna Gora, 2006, p. 367-373.

[8] Z. Petraskovic, Ž.Petraškovi , from the anty-seismic dc 90 damper invencion to its implementacion all over four continents, Internacionalni nau ni skup, Žabljak Crna Gora, 2008, p. 433-439.

[9] Tashkov, Lj., Manic, M., Petrashkovich, Z., Folich, R., Bulajich, B.: Experimental verification of dynamic behavior of “System DC 90” under seismic conditions, Belgrade 2003.

[10] Taškov Lj, Mani M, Shaking table test of a brick-masonry models in scale 1/10, strengthened by DC 90 System, Institute of Earthquake Engineering and Engineering Seismology, University" Ss. Cyril and Methodius", Skopje, Republic of Macedonia, Skopje, May 2004

[11] 20. Tashkov, Lj., Manich, M., Petrashkovich, Z.: Vibroplatform testing of brick- masonry models strengthened by System DC 90 in 1:10 ratio, JGDK Symposion, Vrnyachka Banya, 29.09. – 01.10.2004.

[12] Mazzolani, F., Petraskovich, Z.: Sixth Fram work Program, Priority FP6-2002-INCO-MPC-1, E rthquake Protection of Historical Buildings by Reversible Mixed Technologies PROHITECH, WP6, Naples, 2004-2007.

[13] Petraskovich Z., Sumarac, D., Miladinovic, S., Trajkovic, M., Andjelkovic, M., Trisovic, N.: Absorbers of seismic energy for damaged masonry structures, Alexandropoulos,ECF 16, World Association for Structure Integrity. 2006.

[14] Lj. Taškov, L.Krstavska, Z., Exparimental testing and strenthening of president palace in Alzir by DC 90 System, Institute of Earthquake Engineering and Engineering Seismology, University" Ss. Cyril and Methodius", Skopje, Republic of Macedonia, Skopje, May 2005.

[15] Lj. Taškov, L.Krstavska, Z. Petraskovic, Exparimental testing and strenthening of president palace in Baku by DC 90 System, Internacionalni nau ni skup, Žabljak Crna Gora, 2008, p. 475-481.

183

Zoran Petraškovi 125

HISTEREZISNO PONAŠANJE KONSTRUKCIJE ELI NIH DAMPERA U POLJU ZAMORA ZEMLJOTRESNIM OPTERE ENJEM VRLO MALOG BROJA CIKLUSA

Rezime:Niskocikli ni zamor se razmatra teoretski i eksperimentalno pomo u koncepta mehanike loma. Posebna se pažnja posve uje takozvanom zamoru sa vrlo malim brojem ciklusa, u skladu sa istraživanjima Jean Lemaitre Razumevanje ponašanja gra evinskih konstrukcija, na dejstva zemljotresa je prakti no nemogu e bez poznavanja ponašanja materijala i elemenata konstrukcije u procesu nisko-cikli nog zamora. Na in propagacije inicijalnih prslina i na in narušavanja me u-atomske veze materijala usled cikli nog dejstva, uz poznvavanje na ina naruša-vanja stabilnosti elemenata konstrukcije bitno determinišu ponašanje konstrukcije kao celine u zemljotresnim uslovima. Analiziraju se i utvr uju parametri koji definišu histerezisni ponašanje eli nog Dampera. Klju ne re i: histeresis, mehanika loma, damper, zamor

HYSTERESIS BEHAVIOR OF STEEL DAMPERS CONSTRUCTION IN FATIGUE ZONE UNDER EARTHQUAKE LOADING OF EXTREMLY LOW NUMBER OF CYCLES

Summary: Low cycle fatigue can be revealed theoretically or by experimental evidence by means of fracture mechanics concept. Special attention is devoted to so called low cycle fatigue according to Jean Lemaitre investigation. It is practically impossible to understand the behavior of the construction structures under the earthquake loading without adequate comprehension of material and construction members’ behavior under low cycle fatigue. The propagation of initial cracks caused by low-cycle fatigue and the disturbance of the stability of the interatomic relations of the material define the behavior of the construction in whole under earthquake loading. All the parameters that define the hysteresis behavior of the steel damper should be established and analyzed. Key words: hysteresis, fracture mechanics, damper, fatigue

1 lan SAIN, dipl. gra . inž., SISTEM DC 90, Beograd

184

UVOD

Fenomen niskocikli nog zamora karakteriše mali broj ciklusa i velike amplitude dilatacija u plasti noj oblasti. Visokocili ni zamor karakteriše veliki broj ciklusa, obi no preko 105 i male amplitude dilatacija u elesti noj oblasti.

Niskocikli ni zamor karakteriše broj ciklusa koji se obi no kre e u granicama izme u 102 i 104. Nivo napona se kre e oko granice te enja u elasti nom i plasti nom delu. Od velikog je interesa za zemljotresno inžinjerstvo razmatrati ponašanje materijala u polju vrlo malog broja ciklusa, kao što i govore istraživanja od Dufailly i Lemaitre [1]. Broj ciklusa se kre e od svega par ciklusa do oko 100. Plasti ne dilatacije su vrlo velike kao i plasti ni rad. Ovaj fenomen je vrlo zna ajan za potrebe konstruisanja metalnih dampera u konstrukcijama izloženih zemljotresnim uticajima. Tada se locira i kontroliše mesto zamora materijala. Ni jedna gra evinska konstrukcija ne može opstati bez trošenja energije u plasti noj oblasti. U koliko bi se taj fenomen želeo izbe i sa namerom da konstrukcije rade samo elasti no kao posledicu imali bi smo predimenzionsane i skupe konstrukcije. Dobro poznavanje fenomena zamora u malom broju ciklusa je od presudnog zna aja za opstanak konstrukcija u zemljotresnim uslovima, (11), (12).

ZAMOR MATERIJALA U VRLO MALOM BROJU CIKLUSA I PRIMENA NA METALNIM DAMPERIMA U ZEMLjOTRESNOM OPTERE ENJU TEORETSKE ANALIZE I EKSPERIMENTALNI REZULTATI TESTA ISTRAŽIVANjA NA MODELIMA.

Manson-Coffin zakon iz 1955:

f pN C (1)

je sa nekoliko parametra definisao ponašanje materijala na zamor u odnosu na broj ciklusa. Forula (1) je definisana slede im parametrima:

- Nf je broj ciklusa zamora - p je amplituda plasti ne akumulirane dilatacije i - C i su konstante materijala [2], [3].

Dijagram na slici 1c je logaritamski i prikzuje zavisnos broja ciklusa i akumulirane dilatacije. Ispitivani uzorci su oblika i dimezija datih na na slici 1a. Uzorci su oslabljeni rupama razli itih pre nika i dubina na osnovu sl.1b. i to 40 m, 100 m i 200 m. Ova ispitivanja koja su obavili Japanski istraživa i poslužila su za definisanje stepena površinske obrade Dampera na delu t.z.v. pasja kost .

185

c)

Sl. 1. a), b) Uzorci sa prikazanim dimenzijama rupa za testiranje uzroka površinske inperfekcije i c) Dijagram zavisnosti akumulirane dilatacije ( p) i broja ciklusa (Nf).

Uticaj stepena površinske obrade Dampera u pojasu oslabljenja (pasja kost) je eksperimentalno dokazan i sa dijagrama se može jasno uo iti pad broja ciklusa na uzorcima sa ve im stepenom po etne inperfekcije slika 3a.

Testiranja su obavljanja tokom inoviranja konstrukcije dampera u više Instituta: VTI Zarkovo, IMS Belgrade, IZIIS Skopje i na Gra evinskom Fakultetu u Ljubljani, slike 2a i 2b, (13), (14), (15), (18), (20). Eksperimentalni rezultati za veliki broj ciklusa Nfodgovaraju zakonu Manson-Coffin.

Sl.2. Damperi: a) Na test mašini b) Više tipova Dampera neposredno pre tesata

Na Sl.2a i 2b je prikazana test mašina i serija dampera za testiranje .Histrezisni dijagram Sila pomeranje za Damper tipa Mionica+ je dat na sl 3b. Na slici 3a su prikazane rupe razli itih dijametara i dubina koje su na injene na uzorcima pre testa.

Force vs. displacement

-250

-200

-150

-100

-50

0

50

100

150

200

-5,0 -4,0 -3,0 -2,0 -1,0 0,0 1,0 2,0 3,0 4,0 5,0

Displacement [mm]

Forc

e [k

N]

Sl.3.a)

Propagacija prslina na uzorcima b) Dijagram sila-pomeranje za Damper

186

Na sl.3a. Prikazan je na in oslabljenja uzoraka rupama pre nika 40, 100 i 200 m. Ovim uzorcima je dokazan uticaj stepena površinske obrade na brzinu akumulaciju dilatacija.

Dijagram sila pomeranja sl.3.b. prikazan je za uzorke ispitivane sa sl.2b. Dijagram prikazuje dvostepeni na in rada Dampera. Prvi stepen do momenta maksimalnog akumuliranja dilataja-do kolapsa i potom rad dampera u kontroli deformacija sa prihvatanjem zna ajne sile.

U tabeli 1, sl.3c. prikazana je karaketrizacija zamora u odnosu na broj ciklusa i veli inu akumulirane dilatacije.

Na slici 4 i dijagramu(Nr, p) je prikazano bitno odstupanje zakononitosti Manson-Coffin kod malog broja ciklusa ispod 100.

Sl.3c. Tabela 1

Sl. 4. Dijagram broj ciklusa-akumulirana dilatacija (Nr, p)

Broj ciklusa do loma

Nivo napona Odnos dilatacijap/ e

Odnos energije Wp/ We

Visokocikli ni zamor >105 < y ~ 0 ~ 0

Niskocikli ni zamor 102do 104 y do u 1 do 10 1 do 10

Zamor sa vrlo malim brojem ciklusa

1 do 100 Ve e od u 10 do 100 10 do 100

187

TESTIRANJE DAMPERA TIPA “BRIDGE“ =1000mm U VOJNO TEHNI KOM INSTITUTU VTI U BEOGRADU

Potrebno je bilo inovirati novu konstrukciju metalnog histerezisnog ure aja-Dampera koji bi kontrolisao velika pomeranja i zna ajnije sile koje se jaljaju kod mostovskih konstrukcija. Ovi tipovi Dampera se ugra uju u vrhu mostovskih stubova i povezuju glavne mostoske nosa e (gornji stroja) sa stubovima (donji stroj). Testiranje u Institutu VTI u beogradu je prikazano na slici 5.

Sl.5. Testiranje DAMPERA tipa “BRIDGE“ =1000mm u Vojnom Institutu VTI u Beogradu.

Iz dijagrama sila-pomeranje vidi se dvostepeni rad ure aja sa oja anjima oko +_55mm. I kontrolom deformacija, SL.6. Takodje se uo ava trostepeni rad dampera. U elasti noj oblasti, u oblasti kontrole deformacija i u post-kolapsnoj oblasti sa kontrolom deformacija i sile.

Sl.6. Dijagram Sila-pomeranje a novu konstrukciju dampera tipa MOST.


-400

-300

-200

-100

0

100

200

300

400

-80.0 -60.0 -40.0 -20.0 0.0 20.0 40.0 60.0 80.0

Displacement [mm]

Forc

e [k

N]

188

Na slici 7. prikazan je deo dijagrama koji se odnosi na post-kolapsni rad i na kontrolu deformacija konstrukcije dampera Tipa Most. I posle kolapsa u fazi II stepena rada Damper kontroliše pomeranje definisano razmakom prstenova i prihvata zna ajne sile u oba smera što e se videti na dijagramu Sl.9

Sl.7. Histerezisni dijagram u postkolapsnom periodu

Slede i dijagram prikazuje zavisnost pomeranja od broja ciklusa, gde se jasno uo ava post-kolapsno dejstvo i kontrola pomeranja.

Sl.8. Dijagram broj ciklusa pomeranje

Na Sl.9. je prikazan vrlo interesantan dijagram koji prikazuje zavisnost sile i broja ciklusa sa jasno izraženim postkolapsnim dejstvom i prihvatanjem zna ajnih sila.


-400

-300

-200

-100

0

100

200

300

-80.0 -60.0 -40.0 -20.0 0.0 20.0 40.0 60.0 80.0

Displacement [mm]

Forc

e [k

N]

Displacement vs. No of cycles

-80

-60

-40

-20

0

20

40

60

80

0 10 20 30 40 50 60 70 80

No of cycles

Dis

plac

emen

t [m

m]

189

Sl.9. Dijagram broj ciklusa-sila

Vrlo interesantno pove avanje akumuliranja energije sa porastom broja ciklusa i u post kolapsnom deformacijski kontrolisanom delu dijagrama, Sl.10., Posle 46-og ciklusa vidan je nagli rast energija na osnovu pove anih kontrolisanih deformacija u okviru eli nih prstenova-grani nika.

Sl.10.Dijagram Energija-broj ciklusa

ANALIZA HISTEREZISA I NISKOCIKLI NOG ZAMORA KOD METALNIH DAMPERA ZA PRIMENU U ZEMLJOTRESNIM USLOVIMA GEOMETRIJSKA ANALIZA HISTEREZISNIH DIJAGRAMA

Na slede im slikama dati su osnovni uobi ajeni parametri za definisanje geometrije histerezisnog ponašanja (dijagram sila-pomeranje), Sl.11 i Sl.12.,(5), (6).

Energy vs. No of cycles

0

50000

100000

150000

200000

250000

0 10 20 30 40 50 60 70 80

No of cycles

Ener

gy [k

Nm

m]

Max and min force vs. No of cycles

-400

-300

-200

-100

0

100

200

300

400

0 10 20 30 40 50 60 70 80

No of cycles

Forc

e [k

N]

190

Sl.11. histeresis dijagram Sl.12. Elementi krive k, yleld,ratio, exp

Osnovni parametri koji definišu krive su: • Krutost elementa- k; • Granica elasti nosti – yield;• Krustos posle granice elasti nosti – ratio k;• Aproksimacija krive koja spaja dve prave linije-exp.

VAN GEOMETRIJSKI PARAMETRI DEFINIŠU ZAMOR MATERIJALA I BEZ NJIH JE NEMOGU E RAZUMETI HISTEREZISNO PONAŠANJE MATERIJALA, Sl.13.

Hystereses loop depends on : • p – akumulirana plasti na dilatacija, • Nf – broj ciklusa do loma i • p/t-brzina promene dilatacije

Sl.13. Histerezisni dihagram , , sa prikazom uzlazne (optere uju e) i silazne (rastere uju e) krive

Ramberg-Osgood formula definisanja histerezisnog rada eli nih elemenata izloženih cikli nom van elasti nom optere enju.:

Osnovna-skeletna kriva – uzlazna i silazna kriva

1(RpR

ypy

( )RpR r-1) osnovna kriva

RpRiR

ypyy

221

( )2Rp

RiR r silazna kriva

191

RpRiR

ypyy

221

( )2Rp

RiR r uzlazna kriva

Ove relacije su validne samo za mala pomeranja i veliki broj ciklusa bez uzimanja u obzir inperfekcije geometrije i lokalne stabilnosti elemnta koji se optere uje. Za prakti nuupotrebu kod konstruisanja Dampera DC 90 bilo je potrebno prona i novu primereniju zavisnost koja opisuje histerezisno ponašanje i uzima u obzir zamor materijala.

Na Konstrukciji Dampera sistema DC 90 tipa Mionica prikaza e se parametri koji definišu zamor i karakterišu ure aj-Damper, Sl.14., (7), (8).

Sl.14.Šematski prikaz konstrukcije Dampera tipa Mionica sa osnovnom histerezisnom krivom

Parametri kojima se definišu zamorsko ponašanje Dampera u visokoplasti nom cikli nom naprezanju sa kontrolom deformacije (plasti nosti) su:

- Pre nik Ø (procenta % redukcije osnovnog popre nog preseka, iskustveno i eksperimentalno to je minimum 20% redukcije)

- Dužina dela sa redukovanim popre nim presekom “pasja kost (odgovara maksimalno dopuštenoj dilataciji od 10%, što je iskustveno i eksperimentalno odre eno i potvr eno za odre ene vretste konstrukcijskog elika-materijala koji se koriste za te potrebe)

-Površinska obrada elementa “pasja kost .-Elemenati za obezbe enje lokalne i globalne stabilnosti i klizanja , posebno za

fazu pritiska (betonsko jezgro, aluminijumski limovi, olovni prstenovi), - C i –konstante materijala i konstrukcije Dampera i - Brzina promena deformacija (dilatacija u vremenu).

Posebna injenica da Manson-Coffine zakon ne defiše adekvatne odnose za broj ciklusa manji od 100 i spe ifi nosti konstrukcije Dampera DC 90 zahtevaju definisanje parametra koji mogu realnije opisati stvarno ponašanje konstrukcije na velike udare sa malim brojem ciklusa do kolapsa.

192

Histerezisna kriva se definiše sa: Van geometrijske-materijalne karakteristike vezane za zamor materijala u skladu

sa zkonom Manson-Coffine (Zamor konstrukcije materijala vrlo malim brojem ciklusa dinami kog naprezanja)

• p – akumulirana plasti na dilatacija, • Nf – broj ciklusa do loma i • p/t-brzina promene dilatacije i

Geometrisko prikazivanje hsiterezisne krive F ( , )

Faze rada Dampera: 1. < y visoko cikli ni zamor, 2. y < < u nisko cikli ni zamor, 3. close to u zamor sa vrlo malim brojem ciklusa

4. > (5-10%) u ovom polju naprezalja-amplituda dilatacija damper radi sa kontrolisanim pomeranja što omogu avaju sigurnosni prstenovi i olovni prstenovi, (16).

5. Posle kolapsa dampera dijagonale rade u elasti noj oblasti i njihov kolaps se o ekuje zbog gubitka stabilnosti ili u manjem broju slulajeva zbog zamora materijala (slu aj spre enog izbo avanja i izvijanjem ili ugradnjom dijagonala u zidove koji onemogu avaju gubljene stabilnosti, što je eksperimentalno testovima u Institutu IMS i dokazano.)

Sa dve linije (deo prave linije-duž i kružni luk, kao deo kruga radijusa R) mogu eje definisati histerezisni dijagram, F ( , ) sa aspekta geometrije a sa aspekta materijala uzima se u obzir zamor materijala.

Dve krive se mogu za potrebe numeri ke analize definisati slede im relacijama: 1. fi( , ) =b+k , (line) and (i=1…n), gde je b-otse ak na y-osi a k nagib ka x-osi

Prava linija i 2. fi( , ) ( - 1)2+( - 1)2=R2 (i=1…n), gde je ( 1, 1)-koordinate centra

kruga a R-radijus kruga

Zamor materijala se definiše Manson-Coffine zakonom ( , Np, ), uzimaju i u obzir istoriju akumulirane dilatacije i nivo trenutnog naprezanja koji je iskazan trenutnom dilatacijom , kao i brzinu promene dilatacije p/t.

ZAKLJU AK

Zamor materijala sa vrlo malim brojem ciklusa od svega nekoliko do 20, što je karakteristika za optere enje konstrukcija izloženih udaru zemljotresa, definiše velika akumilirana plasti na deformacija. To je od velikog zna aja za konstruisanje dampera koji se ugra uju u konstrukcije sa dominantnim dejstvom zemljotresa ili drugih ekcesnih optere enja udara (eksplozije i sl.). Eksperimentalne analize pokazuju da Manson-Coffin zakon ne prikazuje dobro zavisnost akumulirane deformacije i broja ciklusa sa broj ciklusa malji od 10. Prakti no je nemogu e razumeti i shvatiti ponašanje gra evinskih konstrukcija u polju dejstva zemljotresa ili drugih udra (eksplozie i sl.) bez analize zamora materijala u

193

malom broju ciklusa. Lokalna i globalna stabilnost je tako e istovremeno zna ajna jer kolaps vitkih eli nih elemenata vrlo esto pre dolazi zbog gubljenja stabilnosti nego zbog lokalnog loma ili je me usobno povezan. Brzina promena deformacija (dilatacija u vremenu) je tako e parametar koji je nezaobilazan za razumevanje ove pojave. Da bi se posebnim ure ajem-Damperom moglo kontrolisati i upravljati udarom, moraju se odrediti njegove karakteristike.Karakteristike-parametri Dampera su; Pre nik oslbljenja, t.j. procenat % redukcije osnovnog popre nog preseka, Površinska obrada dela dampera koji se zamara “pasja kost, postavljanje elemenata za obezbe enje lokalne i globalne stabilnosti, posebno za fazu pritiska (betonsko jezgro, aluminijumski limovi, olovni prstenovi) dužina redukovanog popre nog preseka dela dampera, i koeficienti materijala-Dampera C i .

194

REFERENCES

Dufailly, J. and Lemaitre J. (1995), “Modeling Very Low Cycle Fatigue”, Int. J. Damage Mechanics, 4, pp. 153-170. Manson, S.S. (1954), “Behaviour of Materials under Conditions of Thermal Stresses,” N.A.C.A.. Tech. Note, 2933. Coffin, L.F. (1954), “A Study of the Effects of Cyclic Thermal Stresses in a Ductile Metal”, Transactions of the A.S.M.E., 931, pp.76. Šumarac, D. and Kraj inovi D. (1990), Elements of Fracture Mechanics, Scientific Book, Belgrade (In Serbian). Z Petraškovic, Seismic Strengthening and protection of objects, Monograph System DC 90, Innovation Centre Belgrade for Earthquake Engineering, Belgrade, 2005. Z.Petraskovic, D Šumarac, M. An elkovi , S. Miladinovi , M.Trajkovi , Retrofitting Damaged Masonry Structures by Technology DC 90, Structural integrity and life (IVK), Belgrade, Vol. 2, 2/2005, p. 59-71. Patent in USA No.10/555,131 from 31.10.2005, patent in Australia No. AU 2003254327A1 FROM 2004.11.23. Petraškovi , Z., Miladinovi , S., Šumarac, D., Technology of seismic strengthening of masonry structures by applying vertical ties and diagonals with seismic energy absorber “System dc 90”, International conference on earthquake engineering, Parallell Session, Topic: Retrofit of structures, p T6-9, August-september 2005. D Šumarac, Z.Petraskovic, M. Maksimovi , S. Miladinoi , I.Džuklevcki, N. Trišovi ,Seismic Retrofit of masonry structures applying vertical braces with dampers Sistem DC 90 and newly designed wall buildings, Internacionalni nau ni skup, Žabljak Crna Gora, 2006, p. 373-381. The earthquake response, Institute IZIIS, Skoplje, Makedonia, 2005, p. 13-33. D. Šumarac, Z.Petraskovic, M. Maksimovi , S. Miladinoi , J.Petraškovi , Structure Retrofit for residental house of Finlands Ambassador in Algier, Internacionalni nau ni skup, Žabljak Crna Gora, 2006, p. 367-373. Z. Petraskovic, Ž. Petraškovi , from the anty-seismic dc 90 damper invencion to its implementacion all over four continents, Internacionalni nau ni skup, Žabljak Crna Gora, 2008, p. 433-439. Tashkov, Lj., Manic, M., Petrashkovich, Z., Folich, R., Bulajich, B.: Experimental verification of dynamic behavior of “System DC 90” under seismic conditions, Belgrade 2003. Taškov Lj, Mani M, Shaking table test of a brick-masonry models in scale 1/10, strengthened by DC 90 System, Institute of Earthquake Engineering and Engineering Seismology, University" Ss. Cyril and Methodius", Skopje, Republic of Macedonia, Skopje, May 2004 20. Tashkov, Lj., Manich, M., Petrashkovich, Z.: Vibroplatform testing of brick- masonry models strengthened by System DC 90 in 1:10 ratio, JGDK Symposion, Vrnyachka Banya, 29.09. – 01.10.2004. Mazzolani, F., Petraskovich, Z.: Sixth Framework Program, Priority FP6-2002-INCO-MPC-1, Earthquake Protection of Historical Buildings by Reversible Mixed Technologies PROHITECH, WP6, Naples, 2004-2007.

195

Petraskovich Z., Sumarac, D., Miladinovic, S., Trajkovic, M., Andjelkovic, M., Trisovic, N.: Absorbers of seismic energy for damaged masonry structures, Alexandropoulos,ECF 16, World Association for Structure Integrity. 2006. Lj. Taškov, L.Krstevska, Z., Exparimental testing and strenthening of president palace in Alzir by DC 90 System, Institute of Earthquake Engineering and Engineering Seismology, University" Ss. Cyril and Methodius", Skopje, Republic of Macedonia, Skopje, May 2005. Lj. Taškov, L.Krstevska, Z. Petraskovic, Exparimental testing and strenthening of president palace in Baku by DC 90 System, Internacionalni nau ni skup, Žabljak Crna Gora, 2008, p. 475-481. La recherche des vibrations ambiantes, Institut IZIIS, Skoplje, Macedonie -CGS, Algérie, lzir, La recherche des vibrations forcées- le séisme artificiel , Institut IZIIS, Skoplje, Macedonie -CGS, Algérie, Alzir, Le rapport du contrôle d’entreprise des travaux , CTC, Ain Defla, Algérie D. Šumarac, Z.Petraškovi , Demage control and repair for security of buildings, ARW NATO-Science for Peace and Security Series, Portorož, 2008.

197

Milenko Stankovi 1 i Sr an Stankovi 226

ŽIVOT BEZ STRAHA OD ZEMLJOTRESA-POTREBA-VIZIJA-IMPERATIV Rezime:

"Priroda se samo mo nima pokorava. Nemo ne prezire i uništava." GeteRad nastoji promovisati i afirmisati potrebu kontinuirane me unarodne saradnje nau no-stru nih institucija u oblasti prirodnih rizika. Cilj je razmje-nom znanja i iskustava, stvoriti povoljan ambijent, zaštiti ljude i materijalna dobara, savladati strah od zemljotresa, tj. život u Banjoj Luci u ini izvjesni-jim i ugodnijim. Profesionalna je obaveza da nau ne spoznaje iz ove oblasti uklju imo u obrazovni proces, tj. u inimo ih dostupnim stru noj javnosti, gra a-nima, studentima i nastavnicima. Za o ekivati je da e Konferencija o zemljo-tresnom inžinjerstvu27

na initi zna ajan iskorak i dati jasan reper budu ih istraživanja u graditeljstvu. Nove spoznaje u oblasti prirodnih rizika su potreba, vizija i imperativ, za kreiranje humanijeg model života - bez straha od zemljotresa, uz evidentnu potrebu za inoviranjem graditeljskih pravila iz obla-sti seizmološkog inžinjerstva. O ekujemo da e pravovremeno i adekvatno informisanje gra ana o rezultatima u injene prevencije na zaštiti od zemljo-tresa, doprinijeti afirmaciji života.Ukazuju i na potrebu kontinuiranog obrazovanja i prevencije, skre emo pažnju na injenicu da su graditeljstvo, kultura i prirodni procesi postali nerazmrsivo

povezani na brojne i neo eki-vane na ine. Klju ne rije i: život, zemljotres, graditeljstvo, potreba, vizija, imperativ

LIFE WITHOUT FEAR OF EARTHQUAKE-NEEDS-A VISION-AN IMPERATIVESummary:

Nature obeys only the powerful. Weak , despised and destroyed." GoetheThe work seeks to promote and affirm the need for continued international cooperation between scientific and professional institutions in the field of natural risks. The aim is with exchange knowledge and experiences to create a favorable environment, the protection of people and material goods, overcome the fear of

1 . , . . ., , -, , . . .

2 , ,26-28.10.2009. .

.- ,

,.

198

earthquakes, life in Banja Luka make safe and comfortable. Professional obligation is scientific knowledge in this field include in the educational process, make them available to the professional public, citizens, students and teachers. It is expected that the Conference on seismic engineering make a significant step forward and provide a clear benchmark for future research in architecture. New knowledge in the field of natural risks are the needs, vision and imperative, to create a more humane model of life - without fear of earthquakes, with the evident need for innovation in the field of construction rules Seismological engineering. We expect that the timely and adequately informing citizens about the results made in the prevention of earthquake protection, will contribute to the affirmation of life. Pointing to the need for continuous education and prevention, we draw attention to the fact that the architecture, culture and natural processes have become inextricably linked to the numerous and unexpected ways.Key words: energy, building, ecology, need, vision, imperative

UVOD

"Nikad priroda ne kaže jedno, a mudrost drugo". (D. Julius Juvenalis)Gra ani Banjaluke i okruženja su prije etrdeset godine doživjeli pani an strah od

nepoznatog, kada je proradila iskonska snaga zemlje, uz jako podrhtavanje tla, magnitude 7°, 8° i 9° stepeni po MCS skali. Dva siva oktobarska dana (26. i 27.) bila su dovoljna da izmijene razvojni tok grada Banjaluke i njene regije. U tom sukobu prirode i ovjeka 27. oktobra 1969. u 9 sati i 11 minuta stao je normalan život u gradu.

Slika 1. Sat na Trgu Krajine u Banjoj Luci, danas podsje a nas i opominje

Ti su dani neizbrisivo utisnuti u srca svih gra ana Banjaluke. (1) Na sre u,pobijedio je radni elan, ponos, prkos, humanost i solidarnost ljudi u okruženju, obnovljena je Banjalu ka regija. I danas, etrdeset godina kasnije, mnogi hroni- ari i gra ani mjere vrijeme do “zemljotresa” i poslije njega. injenice govore da su i naši pretci znali za opasnost koja im prijeti pri naseljavanju, ali je, o igledno, bogato prirodno okuženje biloprevelik izazov, pa su ove prostore gusto naselili. Prvi udar zemljotresa gra ani Banja Luke su do ekali potpuno nespremni, bez kodeksa ili pravila za bezbjedno življenje. Nedostajali

199

su tehni ki normativi, zakonska regulativa za organizovanje i sprovo enje racio-nalne i ekonomski opravdane seizmi ke preventive. Strah od ponovnog rušila- kog dejstva zemljotresa rezultirao je intenzivnim izu avanjem ove pojave od strane doma ih stru njaka, ali i kroz me unarodnu razmjenu znanja i iskustva.328 U okviru inžinjersko-seizmoloških prou avanja terena i izrade karte seizmi ke mikrorejonizacije date su preporuke za urba-nisti ko i arhitektonsko projekto-vanje, tj. pružena mogu nost projektantima konstrukcija da djeluju preventivno, sprije e štetne posljedica ove pojave i pravilno odrede dinami nekarak-teristike konstrukcija novih gra evina. Od eminentnog skupa, povodom etrdeset godina od zemljotresa u Banjoj Luci, se o ekuje da inspiriše nau no-stru nu javnost da istražuje i unapre uje spoznaje iz oblasti graditeljstva i seizmološkog inžinjerstva, tj. ponudi jasne koncepcije prevencije. Konferencija e poslati jasnu poruku organima vlasti i gra anima, da zajedni kim djelovanjem stvore povoljne zakonske okvire, izrade i donesu provodljiva podzakonska akta i odluke, sprovedu racionalne i ekonomski prihvatljive seizmi ke preventive, što e život sa zemljotresom približiti svim gra anima i time ga u initi manje stresnim.

PREDMET I METODE RADA

“Što god se mijenja, ono ve jest, a ono što god jest, ono ima svoj po etak”. (Vuk Karadži )

Seizmološko inžinjerstvo je aktuelna i zna ajna tema u oblasti graditeljstva na istraživanoj teritoriji. Rad analizira ovaj prirodni rizik kao osebnost urbanog identiteta grada i njegove regije.(2) Sagledavaju i periodi nost ove pojave, pokušava se do i do injenice: Da li postoje propusti (svjesni i nesvjesni) u razvoju grada, izradi planske i

tehni ke dokumentacije? Da li pravilno i dosledno primjenjujemo propise o seizmi kojzaštiti? Moramo svi postati svjesni da propusti u zaštiti od zemljotresa imaju nesagledive posljedice na ljude i materijalna dobra. Ovom prilikom sugešemo široj javnosti potrebu kontinuirane kontrole procesa prevencije od prirodnih rizika. Iskustvo429 nas je pou ilo da izostanak sveobuhvatnog sagledavanja i preventivnog djelovanja u prostoru vrlo esto rezultira propustom, parcijalnim i ishitrenim rješenjem problema. Burna prošlost, ratna zbivanja, ubrzan razvoj i tranzicija name u opravdano pitanje: Da li je intenzivan razvoj grada rezultirao djelimi nim previdom preventive od zemljotresa? Zakonska regulativa je precizna, svaka gra evina se posebno valorizuju i štiti od zemljotresa. Šta se dešava u praksi, kakva je kontrola sistema-seizmi ke zaštite, tj. kvalitet primjenjenih materijala i ko ih kontroliše. Sama injenica da nije inovirana karta seizmi ke mikrorejonizacije do danas, dovoljno govori o stanju u oblasti preventive. Smatramo da je profesionalna obaveza podsticati stu nu javnost, ali i sve gra ane, da daju aktivan doprinos zaštiti od prirodnih katastrofa, tj. omogu iti im ugodniji život, bez straha od zemljotresa. Istorijske, ambi-jentalne, kulturne i umjetni ke vrijednosti Banjaluke mame, prizivaju, ispirišu i podsti ukreativna rješenja, ali prirodni zakoni opominju da ih uskladimo s kapacitetom prirode. U

3 , -, , 1971.

4 „... ,”

, , ..

200

radu je korišten istorijski i uporedni metod, kako bismo iz raspoložive dokumentacije i stanja na terenu, ponudili viziju razvoja grada u budu nosti. Autori sugerišu aktivno u eš esvih kreativnih snaga za kontinuiranu i adekvatnu kontrolu svih procesa prevencije, te animiranje gra ana na aktivan i kreativan dijalog – život sa zemljotresom.

REZULTATI

"Zemlja uvijek vra a sa kamatom ono što je dobila." Ciceron, I v.p.n.e. Litosfera530 je dio geološke sredine, medij u kojem ovjek temelji sve graditeljske

aktivnosti. Ona ima svoje, vrlo stroge prirodne zakone po kojima funkcioniše. Ne uskladi li ovjek sopstveni razvoj sa prirodnim zakonima, graditeljske aktivnosti prerastaju u

rušila ke, bez obzira da li su trenutne, povremene ili postepene. Razvoj se pretvara u katastrofu koja uništava graditelja. Priroda je na taj na in zatvorila krug po svojim “zakonskim i podzakonskim aktima”. Zapravo, priroda je na taj na in zaštitila sebe od nesavjesnog graditelja, koji nije bio svjestan injenice da zaštitom prirode u stvari štiti sopstveni opstanak te obezbje uje opstanak budu im pokoljenjima. ovjek u litosferi nalazi skoro sve resurse neophodne za opstanak. Njegova želja da što više tih resursa iskoristi za napredak i prosperitet. To vrlo esto rezultira injenicom da ne vodi dovoljno ra una o kapacitetu prirode. Sve što nije prirodno, nije ni dugovje no, zato želje i potrebe ovjek mora uskladiti sa mogu nostima medija koji eksploatiše (litosfere), ali i zaštititi resurs koji iskoriš ava (površinsku i podzemnu vodu, mineralne sirovine, poljoprivredno zemljište, šume, tlo kao gra evinski medij i drugo). Ako se ne poštuje put “trajnog razvoja” te ne uspostavi ravnoteža izme u potreba ljudi i prirodnih mogu nosti, zatvori emo krug na štetu ovjeka, a životna sredina osta e bez resursa te ne e biti u prilici da održava sopstvenu

vrstu ili e, u najblažem slu aju, ovjek morati promijeniti sopstveno stanište. Litosfera kao eko-sistem permanentno je izložena prirodnim promjenama

izazvanim endogenim procesima (iz unutrašnjosti Zemlje) i promjenama izazvanim egzogenim procesima (atmosfera, hidrosfera). Zajedni ko ime i jednih i drugih su “procesi”, oni su trajni, što zna i da su i promjene trajne. Endogene procese (vulkanizam, tektonika, potresi631izmjena stijena itd.), ovjek ne može regulisa-ti. Izgradnju treba prilagoditi uslovima terena u kojem vladaju ovi procesi. Treba se upravljati prema prirodnim rizicima, jer se ne mogu potpuno otkloniti endogeni uticaji. Vrlo je važno, ali i neophodno, da se u procesu planiranja, u timski rad uklju e stru njaci iz oblasti geologije, tj. dati planerske i projektantske smjernice šta, gdje i kako graditi. Stru njaci ne mogu dati smjernice bez principa upravljanja rizicima, kada se vagaju umnošci vjerovat-no a i

5 “ , .,

. 1200°C. , -

.” ,, , , 2001.

, .6 , , , , 1998,

,,

,

30 , , 1999.

201

posljedica. Time bi do sada uo ene greške i propuste u prostoru prevazi-šli, otklonili ili se od njih pravovremeno zaštitili. Egzogene procese (raspa-danje stijena, erozija, denudacija, dejstvo površnih i podzemnih voda, klizišta i drugo) ovjek može regulisati (mijenjati) inžinjerskim radom,732 što je zadatak planera. Bilo bi dobro kada bi ovjek takve promjene vršio samo u slu ajevima kada se ne može promijeniti mjesto gra enja ili kada je regulisanje jeftinije od primjene specijalnih konstrukcija zgrada i gra evina prilago enih uslovima egzogenih, geodinami kih procesa. Detaljnim upoznavanjem zemljotresa – priro-dne pojave - nepogode prvo su stru njaci prevazišli strah od nepoznatog, zatim su pristupili edukaciji stanovništva “za život bez straha od zemljotresa”. Sve što se nau ilo o seizmologiji i zemljotresnom inžinjerstvu ugra eno je u zakonsku regulativu. Me unarodna saradnja stru nih institucija iz ove oblasti, uz razmjenu znanja i iskustava, stvori eadekvatan ambijent zaštite ljudi i materijalnih dobara; omogu i e savaladavanje straha od prirodne nepogode te život u budu nosti u initi sadržajnijim i kvalitetnijim. Potpisom Sporazuma o saradnji na Ministarskoj konferenciji o seizmologiji u zemljotresnom inži-njerstvu u Beogradu833podržane su regionalne aktivnosti, od kojih se o ekuje proširenje sopstvenih vidika, saznanja, poboljšanje uslova žiljenja sa ovom pojavom, kao i obe-zbje enje adekvatne prevencije. Promjene litosfere izazvane su i tehnogenim aktivnostima(rad ovjeka), koje su najmasovnije u urbanim prostorima. Tehnogeni uticaji mogu biti: direktni, indirektni, planski ili stihijski. U svakom slu aju, mijenjaju geološku sredinu promjenom naponskih stanja u stjenskim masivima, režimu površinskih i podzemnih voda, reljefa, biosfere... Temeljeli se na nedovoljnom poznavanju uslova geološke sredine, obi no se izazivaju ekcesne situacije, koje za sobom povla e:

- Pove anje troškova za promjenu namjene prostora; - Porast troškova izgradnje planiranih gra evina; - Kontinuirane troškove sanacije tokom cijelog perioda eksploatacije gra evine; - Nadoknadu štete za ošte ene gra evine, jer se takva ošte enja naj eš e progla-

šavaju elementarnim nepogodama. Dobit od izgradnje gra evine na taj na in pretvara se u trošak. Promjene geološke

sredine najmasovnije su u urbanim naseljima, u rubnim podru jima gradova. Životni prostor urbanih naselja i njihova životna sredina na istraživanoj teriroriji preoptere eni su uprostornom i u ekolškom smislu. Geološka sredina ima ograni en kapacitet, te je neophodno na osnovu odgovaraju ih studija utvrditi podru ja koja: - moraju biti zašti ena i isklju ena iz daljnje izgradnje; - ograni enog kapaciteta i intenziteta koriš enja; - traže specijalni tretman, uz navo enje uslova njihovog koriš enja;- mogu da se koristiti za izgradnju.

ovjek gradi, ali i razgra uje. Samim tim, mora na i na ina da zaštiti ono što je sagradio, te da oplemeni, za svoje potrebe, ono što je razgradio, jer je sve to njegova životna sredina. Prepusti li zaštitu ovjekove sredine drugome, pa bilo to i prirodi, siguran je gubitnik.

7 , , ,, 2009.

8 29.9.2003. ..

202

Slika 2 i 3. Ministarska konferencija o seizmologiji u zemljotresnom inžinjerstvu u Beogradu, u esnici i potpis sporazuma o saradnji govore o jasnoj svijesti u Republici

Srpskoj o potrebi obrazovanja ljudi za život sa zamljotresom.

DISKUSIJA

"Lakše je razbiti atom nego predrasude." Ajnštajn Na strateškom mjestu, uzvišenju uš a rje ice Crkvene u Vrbas, lociran je

kompleks utvrde, koja dominira okruženjem. "Kastel" je arhitektonska baština grada Banjaluke, mjesto njenog osnivanja, okosnica razvoja - urbani embrion i posebna duhovna vrijednost.

Drevne kule i zidine, ali i pet zgrada Muzeja Bosanske Krajine u "Kastelu", nisu odoljele udaru razornog zemljotresa. Arhitektonskoj baštini grada u kompleksu "Kastel" nanesena ogromna šteta. (3) Analizom sa današnje distance, ovaj doga aj ima suštinski, ali i istorijski zna aj. U danima poslije zemljotresa kulturni život u Banjoj Luci bio je "zamro".Pozorište i dom kulture preselili su se u veliki šator ispred "Kastela". Tu su organizovane priredbe koje su zna ajno doprinijele normalizaciji života u gradu na Vrbasu. Prostori oko tvr ave "Kastel" koriš eni su za pružanje gostoprimstva visokim gostima. U poljskoj kuhinji, koja je locirana u tvr avi. Tu je pripreman obrok za ambasadore velikog broja država, koji su u jesen 1969. posjetili razrušeni grad, nude i pomo , svojih zemalja. Solidarnost, humanast, ali i stvarne ljudske potrebe usmjerile su razvoj tamo gdje je logi no, istorijski provjereno, sigurno, na mjesto po etka razvoja u "Kastel" i oko njega.Smatramo ovo klju nim momentom, kada se spontalno kulturni život i gradske manifestacije preselile u "Kastel" (može se re i silom prilika vratile u svoje istorijsko središte).

Evidentno je tokom proteklog civilizacijskog razvoja da ovjek gradi, ali i razgra uje.934 Na primjeru razvoja grada poslije zemljotresa ovjek nije uspio da o uva ono što je sagradio, ali to prihvata u kriti nom trenutku katastrofe - "Kastel". Ono što je zabrnjavaju e leži u injenici da urbanisti i gradska uprava tada nisu izvukli poruku iz te katastrofe, tj. propustili istorijski trenutak da ožive i oplemene prostor "Kastela", za potrebe grada. Naprotiv, spletom okolnosti i ishitrenih aktivnosti, svjesno ili nesvjesno, oni su razgradile kompleks "Kastel" i pretvorile ga u izolovano ostrvo. Trenutno stanje u "Kastelu" potvr uje da je u praksi zašti eno kulturno dobro slobodno tretira, od strane urbanista, arhitekata, pa i zaštitara. Ukazuje na hitnu potrebu ponovnog aktiviranja kompleksa, vizionarskim i planskim aktivnostima iz idejnog projekta, upoznaju i širu doveli u opasnost gra ane.

9 " " 1944.

203

Arhitektonsko-gra evinski fakultet Univerziteta u Banjoj Luci je u okviru projekta TEMPUS (Trans-European mobility scheme for university studies) CD_JEP-15012/2000 BiH Archicur organizovao seminar na temu "Odnos: rijeka-tvr ava-grad" u kompleksu "Kastela".1035 (4) Cilj ovih aktivnosti bio je ponuditi smjernice za rješavanja me usobno povezanih funkcionalnih cjelina u srcu grada Banjaluke. Izlažu i svoja vi enja prisutnih problema i potreba u esnicu seminara ukazali su na potencijale "Kastela", kao i njegov odnosu na rijeku i grad. Predavanja doma ih autora i gostiju ukazivala su na zna aj i opravdanost tradicije, autohtonih na ina gradnje, regionalnih osobenosti i potreba za zdravim životom, odnosno potrebi uspostave harmonije prirodne i izgra ene cjeline. Izlaganjem nekoliko primjera uspješnog planiranja i projektovanja iz Evrope, nastojao se stru noj javnosti i gra anima prikazati stvarala ki proces projektovanja, od ideje do materijalizacije. Radionice na temu "rijeka-tvr ava-grad" ukazivale su na neke ekstremne reakcije. Tokom seminara izdvojile su se etiri radionice. Prva radionica je preispitivala stavove ljudi na licu mjesta i bilježila njihove reakcije. Druga radionica je imala suprotan pristup od prve. Agresivnim navo enjem i usmjeravanjem ljudi da posjete tvr avu, skretana je pažnja na neiskorišteni potencijal u okruženju da bi pobudili svijest o zna ajugraditeljskog naslje a. Tre a radionica je sagledala morfologiju terena, te ponudila jedinstvo unutrašnjosti i spoljašnosti Kastela, kao jasan opredjeljuju i stav, odgovor na zadatu temu. etvrta radionica je putem amaterskog filma prikazala sve orginalne faze koje su se smjenjivale pri sagledavanju problema, svede i ih na ljudska ula. Komparacijom uticaja ljudskih ula raznim sredstvima personifikuju uticaj kulturnog naslje a na ljudsku svijest.

Zna ajno je ista i da seminar, predavanja i radionice nisu ostale samo na nivou akademske rasprave i teoretskih dostignu a, jer su mnogi studetni nastavili istraživanja o Kastelu. Usledio je zatim raspis Me unarodnog konkursa za izradu idejnog urbanisti ko-arhitektonskog rješenja remodelacije i revitalizacije tvr ave "Kastel" u Banjoj Luci.1136

Prvo nagra eni rad1237 za izradu idejnog urbanisti ko arhitektonskog rešenja remodela-cije i revitalizacije tvr ave Kastel u Banja Luci postupio je u skladu sa sugestijama i preporukama Konkursne komisije i gra ana, te izradio Idejni projekat remodelacije i revitalizacije tvr ave Kastel u Banjoj Luci po etkom 2008. Idejnim projektom predvi enoje fazno rješavanje kompleksa, sa devet prostornih cjelina, i to: Hotel sa prate im sadržajima; Kulturni centar; Uslužni centar; Turisti ki informa-tivni centar; Gradski trg i javna garaža; Sportski centar; Gradski vrt; Priobalje; Grad-ski muzej i edukativni centar.

Slu ajnost, solidarnost, ciljana aktivnost ili istorijski zna ajan doga aj da izrada Glavnog projekta remodelacije i revitalizacije tvr ave Kastel u Banja Luci po ne baš u ovom trenutku Konferencija povodom Banjalu kog zemljotresa, etrdesetoj godini poslije.

10 31. 5. 2004. , .11 7.8. 6.12.2006. .

, .12 "01234", " " . , .

13. 20. 2007. ., ,

( : , , ,, , , ,

, , , ,. , . , . .).

204

O ekujemo da gradska uprava ubrzati zapo ete aktivnosti i aktivno doprinijeti integrisanju istorijskog centra sa gradskim središetem danas.

Preispitivanjem postoje eg na ina koriš enja zašti enih prostora i unošenjem novih atraktivnih sadržaja u zonama, komunikacijama i gra evinama, može se unaprijediti vizuelno - estetske atrakcije, aktivirati izvorne vrijednosti centra grada po mjeri njegovih stanovnika i posjetilaca. Estetska uklopljenost, vremenska slojevitost i modularna povezanost sklopova iz razli itih epoha pove avaju atraktivnost i privla e korisnike u grad, koji afirmiše profesionalne kriterijume, savremene spoznaje i prakse isti e osobenost, uva i unapre uje sopstveni identitet .

ZAKLJU AK

"Prava priroda, vrijednost i cena ovekovog života na zemlji odre uju se u osnovi, njegovim geografskim položajem, to zna i njegovim odnosom prema suncu i odnosom

sunca prema njemu". (Ivo Andri , Znakovi pored puta)

Prou avaju i seizmi ke pojave i njihove posljedice po gra ane Banja Luke, ciljano informišemo šira javnost o potrebi izrade i usvajanja novih pravila u graditeljstvu, kako bismo osigurali što sigurnije gra enje u ovom seizmi ki aktivnom podru ju, tj. gra ane oslobodili straha od zemljotresa, ali i aktivno uklju ili u kontrolu svih procesa.

Osobenost ovog rada je injenica da u periodu katastrofe, ali i sumiranja rezultata ove pojave, ponovo se vra amo neprolaznim vrijednostima, istorijskom središtu. Navedeni primjer vrednovanje nasle a grada, na primjeru "Kastel," nudi mogu i ostvariv model zaštite i o uvanja, uz rješavanje uo enih i skrivenih problema.1338. Nove funkcionalne cjeline u kompleksu, fazna izgradnja i višenamjensko koriš enje, znak su i motivacija gra anima u zaštiti javnog interesa i dobar primjer, kako timski rad i sinhronizovane aktivnosti svih aktera u prostoru mogu ostvariti napredak. Ure enje kompleksa zahtjeva kontinuiran i kontrolisan proces predvi anja – programiranja – planiranja - projektovanja i realizacije uz u eš e svih aktera i javnosti na partnerskim principima.

Spoj prirodnog okruženje i ambijentalnih karakteristikama arhitektonskog nasle a, ideja su vodilja za uspješan koncept i dizajn kompleksa. Ponu eni sadržaji (unutar zidina), moraju osigurati dinamiku doga ana, otkrivanja, uklju ivanja, ali i kreiranja novih scena, baziranih na komunikativnosti, prijatnosti, uz korištenje sinergetskih efekata. Gra ani Banjaluke aktivno su se uklju ili tokom izložbe Konkursnih rješenja i s nestrpljenjem o ekuju realizaciju ovog Projekta.

Briga o ljudima, afirmacija kompleksa, ali i jasna poruka planerima i gra anima,da su integralnost i predvidivost, nezaobilazni parametri u kreiranju života grada. (5) Trasiranje puta za budu nost je izazov, kako o uvati "duh mjesta," tj. ponuditi novu viziju,specifikum našeg prostora, razviti kulturni prostor u "duhu vremena," primjeren Banjoj Luci "po mjeri prirode i ovjeka-bez straha od zemljotresa."

13 , , , ,, , ,

, , .

205

LITERATURA: (1) Stankovi , M. 2007, edicija "Iskustva graditelja," knjiga pod nazivom "Harmonija i

konfli-kti u prostoru," Izdava Arhitektonski fakultet, štampa GrafoMark Laktaši, Banjaluka ISBN 978-99938-616-7-6, (str. 1-302)

(2) Stankovi , M. 2004, "Prostorno - teritorijalno održiv razvoj i LEAP," IzdavaKnjižev-na zadruga Republike Srpske, ISBN 99938-33-19-3, Banjaluka (str. 1-164)

(3) Karabegovi B. 1974. (urednik), Banjaluka pet godina poslije zemljotresa, Glas,Banjaluka, (kultura str. 189-202)

(4) Tempus, "BiH Archicur," 2004-2008, Zbornik radova sa seminara projekta na temu: Odnos: Rijeka-Tvr ava-Grad, Arhitektonsko-gra evinski fakultet, Univerziteta u Banjoj Luci, ISBN 978-99938-616-9-0, Banjaluka (str. 1-140).

(5) Stankovi , M. i dr. 2008, Monografija nau no istraživa kog projekta, "Urbana i graditelj-ska obnova grada Banjaluke u duhu održivog razvoja-uvodna razmatranja," Arhitektonsko-gra evinski fakultet univerziteta u Banjaluci, ISBN 978-99938-6616-8-3 (str. 1-475)

207

Lidija Krstevska139, Ljubiša Živkovi 240

ISPITIVANJE OBJEKTA "NOVA BANKA" U BANJA LUCI METODOM AMBIJENTALNIH VIBRACIJA

Rezime:U radu su prikazani rezultati eksperimentalnog ispitivanja poslovnog objekta "Nova Banka" u Banjaluci, Republika Srpska, Bosna i Hercegovina. Osnovni cilj ispitivanja bio je odre ivanje dinami kih karakteristika konstrukcije: rezonansne frekvencije, tonovi oblici vibracija i koeficijenti prigušenja. Objekat je mešoviti konstruktivni sistem - AB okviri i zidovi, i sastoji se iz dva krila odvojena dilatacijom iznad nivoa temelja. Imaju i u vidu dobivene rezultate evidentno je da oba dela konstrukcije (krila) vibriraju intenzivnije za razli ite frekvencije. Klju ne re i: ambijent vibracije, frekvencija, tonovi oblici vibracije, prigušenje

IN SITU TESTING OF "NOVA BANKA" IN BANJA LUKA BY AMBIENT VIBRATION MEASUREMENTS Summary:

In this paper presented are the results from experimental in-situ testing of Nova Banka in Banja Luka, Republika Srpska, Bosnia and Hercegovina. The main objective of the testing was to define the dynamic characteristics of the structure-resonant frequencies, mode shapes of vibration and damping coefficients. The building is a reinforced concrete mixed system of frames and shear walls and it consists of two wings, separated by a construction joint. Considering the obtained results it is evident that the two parts of the structure vibrate separately more intensively at different frequencies.

Key words: ambient vibration, frequency, mode shape, damping

1 Assoc. Prof. PhD., Institute of Earthquake Engineering and Engineering Seismology, University Ss. Cyril and Methodius, 1000, Skopje, Republic of Macedonia, Salvador Aljende 73. E-mail: [email protected] 2 Mr Sci student, Institute of Earthquake Engineering and Engineering Seismology, University Ss. Cyril and Methodius, 1000, Skopje, Republic of Macedonia

208

1 INTRODUCTIONExperimental in-situ testing of "Nova Banka" in Banja Luka, Republika Srpska,

Bosnia and Hercegovina, was performed by ambient vibration measurements in September 2008 within the cooperation between "Projektinvest" d.o.o. Banja Luka and the Institute of Earthquake Engineering and Engineering Seismology, University "St. Cyril and Methodius", Skopje, Republic of Macedonia. The main objective of the testing was to define the dynamic characteristics of the structure - resonant frequencies, mode shapes of vibration and damping coefficients.

2 DESCRIPTION OF THE STRUCTURE "Nova Banka" is located in the centre of Banja Luka, Fig.1. The building is

designed and constructed as a reinforced concrete mixed system of frames and shear walls and it consists of two wings - part A and part B, separated by a construction joint d=20cm which starts above the foundation slab. Part A has dimensions in plan 7.0 x 26.7m while part B is 18.9 x 12.2m in plan and has one level more than part A. The characteristic plan of the building and vertical cross section is given in Figs. 2 and 3.

Figure 1 - Nova Banka in Banja Luka

Figure 2 - Nova Banka in plan

construction joint d=20cmPART A

PART B

209

Figure 3 - Vertical cross section of the structure

3 TESTING PROCEDURE AND APPLIED EQUIPMENT

Nova Banka was tested by ambient vibration testing method. This is widely applied and popular full-scale testing method for experimental definition of structural dynamic characteristics. It is based on measuring the structural vibrations caused by the ambient. As ambient forces can be treated the wind, the traffic noise or some other micro-tremor and impulsive forces like wave loading or periodical rotational forces of some automatic machines. The method is very fast and the relatively simple procedure can be performed on a structure in use, without disturbing its normal functioning.

The ambient vibration testing procedure consists of real time recording of the vibrations and processing of the records. The initial test is the dynamic calibration test. During this test all sensors (seismometers) are placed on the same position in the same direction and the signals are recorded simultaneously and Fourier spectra obtained. Resonant frequencies of the structure can be preliminary defined using the dynamic calibration tests, but the final definition of the natural frequencies is possible after obtaining the mode shapes of vibration. After this calibration test, the seismometers are placed at different levels and different points of the structure, but at the same direction, for simultaneous recording. This is necessary for obtaining the mode shapes of vibration. One point is chosen as a reference one, usually at the highest level of the structure. The duration of the recording should be long enough to eliminate the influence of possible non-stochastic excitations which may occur during the test.

For recording of structural vibrations caused by some ambient excitation, a system of seismometers, amplifiers and recorders is used. The seismometer measures the velocity and it has limitations in frequency and amplitude range. The signal from the seismometer through special cables is transmitted to the signal conditioning system which eliminates the

210

effect of higher frequencies. During the ambient vibration measurements of Nova Banka, four seismometers Ranger type (a), Kinemetrics product, were used and the measured signals were amplified by four channel Signal Conditioner (b) also Kinemetrics product. The amplified and filtered signals from the seismometers were than collected by high-speed data acquisition system (c) which transforms the analogue signals to digital. PC and special software for on-line data processing has been used to plot time history and Fourier amplitude spectra of the response at any recorded point (d). The equipment used for the measurements is presented on Fig. 4.

Figure 4 - Equipment for measuring ambient vibrations

For post-processing and analysis of the recorded vibrations in all measuring points ARTeMIS software was used. This software is based on the Peak Picking technique and Fre-quency Domain decomposition and has possibilities for good graphical presentation of the obtained data. The geometry of the structure generated by this software is given in Fig. 5.

Figure 5 - Geometry of Nova Banka generated by ARTeMIS software

a bcd

PARTaB PART A

211

4 TEST SET-UP

Nova Banka was tested measuring vibrations in 63 points, including the reference point R on the highest level (terrace) on part A. 67 tests were performed, including the dynamic calibration tests. During the data acquisition process, 16 averages were considered. The time duration of each particular record was 90 seconds and the sampling frequency was 200 samples/sec.

The measurements were performed in both orthogonal (transversal) directions X and Y at selected points on different levels for both wings of the building - part A and part B. Presentation of measured points in plan is given in Figure 6, while spatial distribution of measuring points and measured directions is given in Fig. 7 together with position of the seismometers in some points.

Figure 6 - Disposition of measuring points in plan

Figure 7 - Spatial presentation of measured directions and points on the structure with photos showing the seismometers

X

Y

R

212

5 EXPERIMENTAL RESULTS

The peak picking of the dominant frequencies of Nova Banka is presented in Fig. 8, while in Table 1 given are the values of these frequencies as well as the corresponding damping coefficients.

Figure 8 - Peak-picking of the dominant frequencies of Nova Banka

Table 1 – Dominant frequencies for Nova Banka

Mode Frequency (Hz)

Damping (%)

Mode 1 2.34 3.1 Mode 2 3.42 2.2 Mode 3 5.27 2.4 Mode 4 8.30 3.1 Mode 5 12.1 1.6 Mode 6 15.43 0.40

The frequency of 2.34Hz is the first resonant frequency of the structure, which belongs to part (wing) A of the building, direction Y. The shape of vibration at this frequency is given on Fig. 9.

The frequency of f=3.42Hz is the most expressed frequency for the structure and at this frequency the shape of vibration is complex, with translational movement in diagonal direction, Fig. 10.

The third dominant frequency is f=5.27 Hz and it belongs to translational vibration of part (wing) B - X direction, Fig. 11.

213

Figure 9 - Shape of vibration for frequency f=2.34Hz


214


The next three frequencies dominating on the spectrum - f=8.30Hz, f=12.1 Hz and f=15.43Hz belong to the higher modes.

6 CONCLUSIONS - Dynamic testing of Nova Banka in Banja Luka has been performed by means of ambient vibration method with main objective to define its dynamic characteristics. - The measurements have been performed in both orthogonal directions on both structural wings of the building - part A and part B. - Total of 67 tests have been performed giving the comprehensive experimental data related to dynamic properties of the structure. - The investigated frequency range was 0-25 Hz. - The natural frequencies are very well expressed and their values are the following: f=2.34 Hz; f=3.42Hz; f=5.27Hz; f=8.30 Hz ; f=12.1Hz; f=15.43Hz - The damping coefficients are in range 1.6 to 3.1 % of the critical damping. - Comparing the mode shapes of the structure, it is visible that the two parts of the structure vibrate separately more intensively at different frequencies. Even there is a construction joint between the two wings A nad B, there is slight interaction between them during vibration caused by the pavement connection at the floor levels.- The concentration of stresses during vibration caused by earthquake will be probably at the central shear walls, near the staircases, due to the lateral movement of the ends (wings) of the structure. - As a general conclusion, it can be pointed out that obtained experimental data represent very good and comprehensive base for verification of the numerical model of the structure and evaluation of its seismic behaviour.

REFERENCES 1 In situ testing of "Nova Banka" in Banja Luka by ambient vibration measurements

/ L. Krstevska// Report IZIIS 2008-70

215

. 141, 242

:

.,

“ ”..

Mathematica FORTRAN, .

ABAQUS.

: , ,

STOHASTIC VIBRATIONS OF PLATE IN BENDING USING THE FINITE STRIP METHOD

Summary:

This paper presents modelling of plate structures bending response due to arbitrary dynamic loading. Equations of motion are solved by transformation to normal coordinates and the “step by step” method is used to obtain the solution of modal equations. Total structure response is obtained with modal superposition. Algorithm described here was implemented in Mathematica and FORTRAN and the result is program MKTSV. Code verification is done by comparing the results from commercial FEM software ABAQUS.

Key words: finite strip method, earthquake, dynamic analysis

1 . , . . , - ,2 . . , - ,

216

1

. -,

. -.

-.

( ). ( )

. 4 -

.

2

x y z t ( . 1). w y

.

1 –

Ritz- ,

217

1 1( ) ( )

r c

m k kmm k

f Y y xN q , (2.1)

, . 1,

1 2 3 41

r

wm wmm

w Y N N N N q , (2.2)

:2 3 2 3

1 2

2 3 2 33 4

1 3 / 2 / , / 2 / / ,

3 / 2 / , / / ,

N x b x b N b x b x b x b

N x b x b N b x b x b (2.3)

:- :

( ) sin( / ), , 2 ,....,m m mY y y a m ; (2.4)

- :( ) sin( / ) sinh( / ),

sin( ) (4 1), 3.9266,7.0685,....., ;sinh( ) 4

m m m m

mm m

m

Y y y a y am (2.5)

- :( ) sin( / ) sinh( / ) cos( / ) cosh( / ) ,

sin( ) sinh( ) (2 1), 1.8750,4.6940,....., ;cos( ) cosh( ) 2

m m m m m m

m mm m

m m

Y y y a y a y a y am (2.6)

- :( ) sin( / ) sinh( / ) cos( / ) cosh( / ) ,

sin( ) sinh( ) (2 1), 4.7300,7.8532,....., ;cos( ) cosh( ) 2

m m m m m m

m mm m

m m

Y y y a y a y a y am (2.7)

:

1 1 2 2T

wm m m m mw wq , (2.8)

.

3

(2.4-2.7), .

N

:

218

( )t t t tMq Cq Kq F , (3.1)

M, C K N x N , , ;tq , tq , tq N x 1 ,

F(t)

( ) gt u tF MI , (3.2)

I N x 1 ( )gu t .

: .

,.

., “ ”.

.C ,

. ,2nn n nc . n

n– .

0t tMq Kq . (3.3)

, ( ) i tt eq q , (3.3)

12

10,K M I q , (3.4)

. (3.4) N- ,

N 1 2, ,...., ,.....n N N

1 2, ,..., ,...,n Nq q q q .

( 1/n n ). FORTRAN-

, Mathematica.

(3.4)

219

1 2 ... ...n Nq q q q . (3.5)

(3.4) , ., ,

. .,

. -

.:

M̂Tn n nq Mq (3.6)

nq . ,ˆ nq

ˆˆ M-1/ 2n n nq q , (3.7)

ˆ , .ˆ ˆT M I 2ˆ ˆT K I ,

I .:

ˆ ˆ ˆˆ ˆ ˆ, , ,q q q q q q

(3.1) :

ˆ ˆ ˆ ˆ ˆ ˆ ˆˆ ˆ ˆ ( )T T T T tM q C q K q F

2 ˆˆ ˆ ˆ2 ( )T tq q q F , (3.8)

, .ˆ ˆ 2T C I .

(3.8) N.

,:

ˆ ˆ ˆˆ ˆ ˆ, , .q q q q q q (3.9)

.

[2]. .

0.005 ,.

220

n-ˆR( ) ( )Tt tF ,

2ˆ ˆ ˆq 2 q q R ( ), 1,2,... .n n n n n n n t n N (3.10)

, ,, ,

.R( )t

( . 2).

2 – ,

Dt

1R( ) 1 R Ri it ttt t

, (3.11)

t .

1 1 2 1 3 1 4

1 1 2 1 3 1 4

1 1 2 1 3 1 4

ˆ ˆ ˆq( ) ( )q ( )q ( ) R ( ) R ,

ˆ ˆ ˆq( ) ( )q ( )q ( ) R ( ) R ,

ˆ ˆ ˆq( ) ( )q ( )q ( ) R ( ) R .

i i i i

i i i i

i i i i

t A t A t A t A t

t A t A t A t A t

t A t A t A t A t

(3.12)

1. (3.12) (3.10). (3.12) t ,

:

1 1 2 1 3 1 4

1 1 2 1 3 1 4

1 1 2 1 3 1 4

ˆ ˆ ˆq ( ) q q R R ,

ˆ ˆ ˆq ( ) q q R R ,

ˆ ˆ ˆq ( ) q q R R .

i i i i i

i i i i i

i i i i i

t A A A A

t A A A A

t A A A A

(3.13)

221

1 -

2 20 1 0

2 3 1 2 4 1 5 0 6 222

7 5 6 8 5 9 10

1 / 1 2 / 1

1/ /

/ 2

D

D

D D D

a t a a

a t a a a a a a a a a

a a a a a a a

9 10

9 10

1 2 3 1 2 3 42

4 5 6 72

( ) sin( ) ( ) ( ) ( ) ( ) ( ) ( )

( ) cos( ) ( ) ( ) ( ) ( ) ( ) ( )1 1( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )

1( ) ( )

tD D

tD D

D

S t e t S t S t C t S t a S t a C t

C t e t C t C t S t C t a C t a S t

A t C t S t A t S t A t a a t a S t a C t

A t a a t a S t a8 5 6

7 2 3 4 8 6 7 8 92 2

10 11 3 4 12 7 82 2

1( ) ( ) ( ) ( ) ( ) ( )

1 1( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )

1 1 1( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )

D

D

C t A t C t S t A t S t

A t a a S t a C t A t a a S t a C t A t C t S t

A t S t A t a S t a C t A t a S t a C t

1 12A A . , 12 9 .

10-6

. 200 50 0.01 , . 100

1 . Fourier-

Duhamel- .,

.

4 –

[1] -. 3.

222

3 –

223

- ABAQUS, .

-:

215 , 10 , 0.12 , 31500000 / , 0.0.L m B m t m E kN m 0.05.

. - Düzce 21.11.1999. ( . 4) 7.2 [3].

4 – Düzce

5 - ABAQUS-

6 - -

.,

, .- ( )

(wmin=-7.69cm) ABAQUS- 300

.

224

, ( . 7 8).

7 - , 0.12t m

8 - , 0.05

,

,

.

1 Finite strip method in computational engineering / . . // Faculties of civil engineering: Subotica, Belgrade and Budapest, 1997,

2 Static and Dynamic Analysis of Structures /E.L.Wilson// Computers & Structures, 2002.

3 http://peer.berkeley.edu/

225

Mihail Garevski143 Veronika Sendova244 and Blagojce Stojanoski345

REKONSTRUKCIJA PRAVOSLAVNOG SABORNOG HRAMA "SV. BOGORODICA" U SKOPLJU

Rezime:Pravoslavna saborna crkva “Sveta Bogorodica” datira iz 1835 godine, 1943 godine je izgorila i nakon toga srušena do temelja. 2003 godine Skopska mitropolija donosi odluku da rekonstruiše crkvu. Na osnovu definiranog i usvojenog izgleda obnovljenog hrama, IZIIS, Skopje, je izradio Glavni projekat za rekonstrukciju crkve, koji sadrži definisanje konstruktivnog sistema i fundiranje objekta iznad postoje ih temelja, kao i detaljnu stati ku i dinami ku analizu konstrukcije sa verifikacijom stabilnosti i kapaciteta nosivosti i deformabilnosti. Klju ne re i: rekonstrukcija postoje ih spomenika, analiza seizmi ke sigurnosti, troslojni nosivi zidovi

RECONSTRUCTION OF THE ORTHODOX CATHEDRAL CHURCH OF THE VIRGIN MARY IN SKOPJE

Summary: “St.Bogorodica” Orthodox Church, dating back to 1835, was damaged in 1943 fire and utterly demolished later on. In April 2003 Macedonian Orthodox Church carried out a decision for reconstruction of St.Bogorodica Church. According to the defined and adopted appearance of the renovated Church, IZIIS, Skopje has prepared the Main Project for Church Reconstruction. It includes definition of the basic structural system for reconstruction, definition of the foundation conditions over the existing foundations and their consolidation as well as detailed static and dynamic analysis of the structures with verification of their stability and ultimate bearing capacity. Keywords: reconstruction of existing monuments, analysis of seismic safety, three-layered bearing walls

1 Prof. Dr. Institute of Earthquake Engineering and Engineering Seismology, IZIIS, P.O.Box 101, 1000 Skopje, Republic of Macedonia 2 Prof. Dr, IZIIS, Skopje, Republic of Macedonia 3 Ass. Researcher, M.Sc., IZIIS, Skopje, Republic of Macedonia

226

1. INTRODUCTION The orthodox cathedral church of the Virgin Mary, (fig.1) constructed by the

most eminent Macedonian and Balkan builder Andrea Damjanov, was put on fire in 1943 and later torn to the ground. On the remains of the foundation walls of the church, a youth hostel was constructed in 1962 and was ruined in during the Skopje earthquake in 1963.

Figure 1. Air shot of the church of the Virgin Mary

In April 2003, based on the initiative launched by the established Board for Renovation of the Church of the Virgin Mary in Skopje, the reconstruction of the church was entrusted to the Institute for Protection of Cultural Monuments of the City of Skopje. Based on the decisions made, a working group composed of eminent experts from among all the necessary profiles was established to elaborate a project on the reconstruction of the church of the Virgin Mary upon the existing foundation, (Fig.2) with full consideration of its authentic appearance.

Figure 2. Remains of the authentic foundation walls

Based on the defined and adopted appearance of the renovated church, the Institute of Earthquake Engineering and Engineering Seismology – IZIIS, Skopje elaborated the

227

main project on the reconstruction of the church of the Virgin Mary containing definition of the principal structural system for the reconstruction of the church and its belfry, the conditions for founding the renovated church on the existing foundation and its consolidation as well detailed static and dynamic analysis of the designed structures with verification of their stability considering the ultimate bearing capacity.

2. DEFINITION OF THE PRINCIPAL STRUCTURAL SYSTEM FOR RECONSTRUCTION OF THE CHURCH OF THE VIRGIN MARY IN SKOPJE

The main project on the structure of the church of the Virgin Mary - Skopje has been elaborated in accordance with the Preliminary Project that was adopted by the Investor. The church of the Virgin Mary will be reconstructed in its authentic shape and position. The church is of the type of a three-nave basilica with outline proportions of 36.70 m/26.40 m, with an elevated central nave with a storey height of 11.90 m in respect to the side naves with a storey height of 8.47 m (Fig. 3).

Figure 3. Ground floor plan and cross section

The adopted variant of foundation of the structure involves circular reinforced concrete caissons with a diameter of 80 cm and a length of 5.0 m that are to be constructed at level – 2.00 m on both sides of the existing foundation walls, i.e., below the columns in the central nave and the paraccleseion. The existing foundation walls and the caissons are to be structurally connected by reinforced concrete foundation beams. At level +0.00 m, a floor reinforced concrete slab with a thickness of 16 cm is anticipated to be constructed.

The main characteristic of the structural system arising from the priorities regarding the appearance of the church on one hand and its seismic safety on the other, is

228

the solution of the bearing fasade massive walls with a total thickness of 1.16 m. In the interior, these are to be constructed of solid bricks in cement lime mortar and to have a thickness of 38 cm, while on the exterior, these are to be constructed of broken and dressed stone in cement lime mortar and to have a thickness of 53 cm. In the central part, a continuous reinforced concrete wall with a thickness of 25 cm is anticipated to be constructed. To connect the reinforced concrete and the masonry part of the wall, 2 reinforced concrete cramps are anticipated to be incorporated at each square metre, in accordance with the details on the reinforcement.

The central colonnades consist of a total of 16 reinforced concrete round columns with a diameter of 45 cm, connected to a system of reinforced concrete arches. The floor structures are anticipated to be constructed as reinforced concrete flat slabs with a thickness of 12 cm at all the three levels. All the walls constructed of stone and brick end with reinforced concrete belt courses with corresponding dimensions at all the platforms.

3. STRUCTURAL ANALYSIS Such a designed structure has been analyzed by consideration of gravity and

seismic loads in compliance with the valid regulations and the proposed European regulations using two methods:

1. Static and equivalent seismic three-dimensional analysis of the structure by using the finite element method used as a basis for proportioning all the structural elements;

2. Analysis of the bearing and deformability capacity of the structure composed of designed and proportioned structural elements;

3. Dynamic analysis of the structure involving analysis of the structure with consideration of the nonlinear response spectrum related to the Petrovats earthquake, N-S component, and nonlinear dynamic analysis of the church structure applying the method of concentrated masses.

1. Based on the defined structural system, static and equivalent analyses have been performed by using the finite element method using the computer software package SAP 2000. The reinforced concrete slabs and the bearing walls have been modeled by using SHELL elements, while reinforced concrete columns and beams have been modeled by 3D FRAME elements. 3D TRUSS elements have been used to model the steel ties, (fig. 4).

229

a) Southeast view b) Southwest view

Figure 4. Three-dimensional mathematical model of the church of the Virgin Mary

2. The methodology used for definition of the bearing capacity of the structure in the form of ultimate storey shear force that compared to the equivalent seismic force yields the safety factor against failure, represents a procedure that is widely applied in equivalent static analysis of masonry structures at IZIIS. Applying this methodology and considering the already proportioned structural elements, the bearing and deformability capacity of the structure of the church of the Virgin Mary has been analyzed. The results obtained from the analysis are in the form of storey Q- diagrams and output digital results for each individual wall and the integral structure separately for both the orthogonal directions. From the obtained results (Table 1), it can be considered that each individual wall has a sufficient bearing and deformability capacity whereat the safety factor at occurrence of the first cracks and particularly failure of all the storeys is greater than 1 (Fj > 1, Fu > 1).

230

Table 1: Summary results on the church of the Virgin Mary

storey stiffness

Ki (kN/cm)

x-x y-y

ultimate bearing capacity

Qu (kN)

x-x y-y

safety factor

Fu=Qu/S

x-x y-y

storey 3 8782 12257 4462 4018 2.52 2.27

storey 2 30289 38217 11157 9183 2.06 1.70

storey 1 37492 50952 17020 23381 1.87 2.57

3. The main dynamic parameters of the structure have been defined in the form of fundamental periods and mode shapes as a constituent part of the three-dimensional analysis of the church structure by using SAP2000 programme package, (Tx = 0.1012 sec and Ty = 0.1406 sec). The structure has a higher stiffness in the longitudinal direction X-X where the massive bearing walls have greater lengths.

Within the three dimensional analysis of the church structure performed by using the SAP2000 programme package, dynamic analysis has been carried out using the nonlinear response spectrum related to the Petrovats earthquake, N-S component, which has been selected as the most unfavourable one. The analysis has been carried out for the design level of input acceleration (amax = 0.34 g). For the needs of the analysis, the linear spectrum has been reduced by reduction coefficients R depending on the period and the adopted level of ductility of the church structure. From the obtained results it can be concluded that, due to the high stiffness of the structure, relatively small deformations due to the nonlinear acceleration spectrum are obtained, i.e., this load case is not referent for the analysis of the structure.

Applying modeling with concentrated masses that assumes concentration of distributed structural characteristics at the characteristic levels, a nonlinear dynamic analysis has been performed by application of a corresponding storey hysteretic model obtained by summing up the elastic-plastic characteristics of each of the bearing walls, whereat the bearing capacity of each of them has been limited to the lower value of bending or shear bearing capacity. These input data are, in fact, the results from the analysis of the bearing and deformability capacity of the church structure (Table 1). Three different types of earthquakes (Petrovets N-S, 1979, Ulciw Olimpik N-S, 1979 and El Centro, 1940) with input acceleration of amax = 0.34 g and 0.40 g and return periods of 475 and 1000 years, respectively, have been applied.

It has been concluded that the Y-Y direction is logically more flexible wherefore higher ductility is obtained in this direction. It can be concluded that such designed structure possesses sufficient bearing and deformability capacity, i.e., that its dynamic behaviour complies with the established design criteria. Ductility of <1.5 for a design earthquake with a return period of 475 years, i.e., <2.0 for the maximum expected earthquake with a return period of 1000 years have been obtained.

231

4. CONCLUSIONS

The church of the Virgin Mary possesses sufficient bearing and deformability capacity in respect to the designed level of seismic protection; The church of the Virgin Mary satisfies the design safety criteria; It is generally concluded that the designed structure of the church of the Virgin Mary completely satisfies the prescribed requirements and criteria for such type of structures of special importance.

The reconstruction of the church of the Virgin Mary over the existing foundation walls and in compliance with the solutions anticipated with the main project started in the summer of 2004. After a year, the finishing works on the roof structure are being carried out (Fig. 5).

Figure 5. Construction of the church of the Virgin Mary

ACKNOWLEDGEMENT IZIIS and the participants in this project would like to extend their gratitude to the

Macedonian Orthodox Church – the Skopje diocese and the Institute for Protection of Cultural Monuments of the City of Skopje for the entrusted task and the continuous cooperation throughout the realization of this project.

REFERENCES [1] Garevski, M., Sendova, V., Stojanoski, B. (2004), Main Project for the Structure of

the Church of the Virgin Mary, Skopje, IZIIS Report 2004-20, Skopje

233

Veronika Šendova146Predrag Gavrilovi 247, Blagoj e Stojanoski348Goran Jeki 449

INTEGRIRANI PRISTUP SANACIJE I SEIZMI KOG OJA ANJA MUSTAFA PAŠINE DŽAMIJE U SKOPLJU

RezimePoštuju i moderne principe zaštite kulturno istoriskih spomenika, kao što su primena novih materijala i tehnologija, reverzibilnost i nevidlljivost, izražen je projekat o sanaciji i oja anju Mustafa Pašine Džamije u Skoplju. Koncept oja anja, radi postizanja projektovanog nivoa seimi ke zaštite, selektiran je na osnovi: (i) istraživanja uslova tla, (ii) istraživanje karakteristika ugra enih materijala, (iii) istraživanje osnovnih dinami kih karakteristika objekta i (iv) prethodna eksperimentalna istraživanja modela džamije u razmeru. Klku ne re i: istoriski spomenik, sanacija i oja anje, zemljotresna zaštita, novi materijali i tehnologije

INTEGRATED APPROACH TO REPAIR AND SEISMIC STRENGTHENING OF MUSTAFA PASHA MOSQUE IN SKOPJE

Summary: A project on repair and strengthening of Mustafa Pasha’s mosque has been elaborated respecting the modern requirements in protection of historical monuments, i.e. application of new technologies and materials, reversibility and invisibility of the applied technique. The concept of structural strengthening and repair aimed at reaching the designed level of earthquake protection has been selected based on: (i) investigations of the soil conditions, (ii) investigations of the characteristics of the built-in materials, (iii) investigation of the main dynamic characteristics, and (iv) previous experimental investigation of the mosque model. Key words: historic monument, repair and strengthening, earthquake protection, new materials and techniques

1 Prof. Dr, IZIIS, Skopje, Republic of Macedonia 2 Emeritus Professor, IZIIS, Republic of Macedonia 3 Ass. Researcher, M.Sc, IZIIS, Republic of Macedonia 4 Ass. Researcher, IZIIS, Republic of Macedonia

234

1. INTRODUCTION The repair and/or strengthening of historical monuments in seismic regions is

highly dependent on the earthquake conditions to which they have been exposed in their past history and the ground motion to which they are expected to be frequently exposed in future, as well as the materials and methods used for their construction. Due to these reasons, it will be of importance that repair and/or strengthening, as a part of preservation, conservation and restoration of historical monuments be planned based on detailed studies of the following factors: seismic hazard; local soil conditions and its dynamic behaviour under earthquake loading; dynamic properties of the structural systems and their strength and deformability and the dynamic response of the structure under the expected ground motions.

Considering that the above factors are of main importance for the determination of the earthquake response of historical monuments, as well as the fact that seismic analysis cannot be performed using seismic design codes for modern buildings, determination of the criteria, methods and techniques for strengthening and the process of restoration and preservation should be based on detailed studies with consideration of the cost effectiveness of the alternative solutions.

3. IZIIS' EXPERIENCE IN PROTECTION OF CULTURAL HERITAGE

Within the frames of the IZIIS’ research activities, in addition to seismic design and protection of modern structures, particularly noteworthy is also the experience gathered in the field of protection of structures pertaining to the cultural historic heritage. During a period of more than 30 years of activities in this field, the Institute has realized important scientific research projects involving experimental and analytical research, field surveys of historic structures and application of knowledge during earthquake protection of important cultural historic structures and monuments.

Within the frameworks of the scientific research projects realized at the Institute in the period 1990-2000 for the purpose of development of appropriate methods for repair and strengthening of the Byzantine monuments in general, and particularly the Byzantine churches located within Macedonia, shaking table testing of a church model in a realistic geometrical scale was performed for the first time in the world, [1]. 1:2.75 scaled model of St. Nikita church was constructed and tested on the seismic shaking table in the IZIIS laboratory in its original state, strengthened state by use of "ties and injection", and as a base isolated model. The knowledge gained in this way is unique and incomparable and hence necessary for seismic strengthening of individual important cultural-historic structures where it is important to have an insight into the effect of the interventions upon the authenticity of the monument.

After realization of these projects, IZIIS became partner of the Republic Institute for Protection of Cultural Historic Monuments which enabled direct application of the gained knowledge in actual conditions and for specific historic monuments. Presented further are the two most characteristic examples of application of the developed methodology by implementation of vertical and horizontal strengthening elements.

1. Reconstruction and seismic strengthening of the St. Athanasius church, [3]:On August 21, 2001, during the armed conflict in R. Macedonia, the monastic church of St.

235

Athanasius in Leshok experienced strong detonation, which resulted in its almost complete demolition, (Fig. 1). From structural aspects, there have been two approaches taken in the attempt to renovate and reconstruct the structure, (Fig.2). Based on the performed detailed analysis of the structure, (i) solution for repair and strengthening of the existing damaged part of the monastic church and (ii) solution for seismic strengthening of the ruined part of the church to be reconstructed were made.

Image 1. The Church after detonation Image 2. The church after reconstruction

2. Reconstruction of the St. Pantelymon church in Ohrid, [2]: In the process of conservation and rebuilding of the St. Panteleymon church, Ohrid, (Fig. 3) having in mind the importance and specific nature of the structure representing a historic monument classified in the first category and a structure of a particular national interest, it was necessary to design a building structure that will satisfy the stability conditions in the process of application of the conservation principles regarding shape, system and identification of materials. Seismic strengthening was provided in accordance with the previously developed and verified methodology, (Fig.4).

Image 3. The church during reconstruction Image 4. The rebuilt St. Panteleymon church

236

4. INTEGRATED APPROACH TO REPAIR AND STRENGTHENING OF HISTORIC MONUMENTS

The decisions for repair and strengthening of a historic structure should be tho-roughly clarified and justified in advance because of the existence of some unusual aspects (compared to modern structures) that have to be taken into account. The characteristic structural entity, the variety of the built-in material, the complex history of successful modifications done in the past, as well as the degree of deterioration makes each historic structure a case for itself. Only a team of experts of different profiles, (architects, archaeologists, art historians, conservators and other profiles), who are completely competent in their fields but sufficiently flexible to accept the arguments of the others can successfully protect a historic structure. Nevertheless, protection of structures in seismically active regions is mainly a task to be done by a civil engineer - a structural engineer.

Extensive research activities have been performed by IZIIS for the purpose of evaluation of a procedure for repair and strengthening of valuable historic monuments. Such a procedure is based on conventional understanding of retrofitting, although, in our concepts, there are also techniques, which are based on the idea of structural control. As a result of several decades of gathering of experience, it can be said that an integral approach to seismic protection of extraordinarily important cultural historic structures has been adopted by the Institute. This approach, first of all, complies with all the restoration and conservation requirements set in a number of international documents and declarations, as well as procedures and legislative regulations for high category structures. This integrated approach to repair and seismic strengthening of historic monuments should encompass the following:

o Definition of expected seismic hazard; o Definition of soil conditions and dynamic behaviour of soil media; o Determination of structural characteristics along with the bearing and

deformability capacity of existing structures; o Definition of criteria and development of a concept for repair and/or

strengthening; o Design of structural methods, techniques, materials and types of excitation; o Determination of the response of repaired and/or strengthened structures and

verification of their seismic stability; o Definition of field works, execution and inspection.

Although the above stated seems to be the "normal procedure", it is the only way of providing high quality in protection of cultural heritage. This task is certainly much more than simply listing of what is to be done since it requires a lot of knowledge and efforts.

4. REPAIR AND SEISMIC STRENGTHENING OF MUSTAFA PASHA MOSQUE IN SKOPJE

Mustafa Pasha's Mosque is one of the biggest and the best preserved monuments of the Ottoman sacral architecture in Skopje and the Balkan. The building style belongs to the early Constantinople period at the beginning of the second half of the 15th century. The structural system of the mosque consists of massive peripheral walls in both orthogonal

237

directions with a thickness of about 170 cm constructed partially of hewn stone and brick. The walls of the building are original in principle. They are constructed with two faces between which there is an infill of lime mortar and pieces of bricks and stone.

The catastrophic Skopje earthquake of 1963 inflicted damage to the mosque structure that dominantly affected the central dome and the domes of the porch, the east facade and the minaret. In 1968, these damages were repaired by injection of cement mortar based mixtures as well as incorporation of RC belt courses, (Fig.5).

Image 5. Mustafa Pasha Mosque after the 1963 Skopje earthquake

Today, Mustafa Pasha’s Mosque represents a cultural historic monument of an extraordinary importance for the city of Skopje and Republic of Macedonia. As such, it is under the protection of the Law on Protection of Cultural Heritage and it is categorized as a structure belonging to the first category. The authors of this paper had the opportunity and the challenge to design a system for seismic strengthening of the structure, (financed by the Turkish foundation TIKA) for the needs of the conservation project on repair of Mustafa Pasha Mosque prepared by the Foundation of the University of Gazi in cooperation with the Ministry of Culture and Tourism of Turkey, the Ministry of Culture of Republic of Macedonia and the National Conservation Centre in Skopje and realized in 2006. Respecting the modern requirements in the field of protection of historical monuments, as is the application of new technologies and materials, reversibility and invisibility of the applied technique, the authors have decided to choose a concept of repair and strengthening involving the use of composite materials. In realization of this project, the established integrated approach has thoroughly been respected. The concept of structural strengthening and repair aimed at reaching the designed level of earthquake protection has been selected based on:

(i) investigations of the characteristics of the built-in materials, (i) investigation of the main dynamic characteristics, (ii) shaking table testing of the mosque model; (ii) investigations of the soil conditions; (iii) detailed geophysical surveys for definition of geotechnical and geodynamic models of the site.

238

4.1. INVESTIGATION OF THE CHARACTERISTICS OF THE BUILT-IN MATERIALS AND THE MAIN DYNAMIC CHARACTERISTIC

In the course of 2006, within the bilateral scientific project with the Yildiz University, Turkey, the dynamic characteristics of the Mustafa Pasha Mosque structure were investigated by use of the ambient vibration technique, [4]. The obtained natural periods of the structure were Tx=3.0s and Ty=3.2s, whereas the fundamental period of the minaret structure is T=1.04s. Within this project, the mechanical characteristics of the built in materials were also investigated through testing of samples of material taken from the structure. These data are particularly important and necessary for correct analysis of the seismic stability of the monument.

Image 6. Damages to the mosque model Image 7. Strengthening of the model

4.2. SHAKING TABLE TESTING OF THE MODEL OF MUSTAFA PASHA MOSQUE

Experimental shaking table tests on a model of the Mustafa Pasha Mosque were carried out in IZIIS within the frames of PROHITECH project “Earthquake Protection of Historical Buildings by Reversible Mixed Technologies”, [5]. The 1:6 scaled model has been subjected to the effect of a series of earthquakes that caused damage (Fig. 6). Then the model was strengthened by application of CFRP elements and subjected to iterative tests (Fig. 7). The results from the investigation, although obtained for small geometrical scale, have shown that the system is efficient, which was the starting point in making the decision about the concept of strengthening of the prototype.

239

4.3. INVESTIGATIONS OF THE SOIL CONDITIONS To define the bearing capacity of the local soil and identify the foundation level of

the structure, geomechanical drilling (3 sondages down to the depth of 8.00 m) was carried out in June 2007. These geomechanical investigations enabled sufficient definition of the lithological composition of the soil, definition of the geotechnical soil profiles below the structure as well as definition of the physical-mechanical characteristics of the present materials. It has been also concluded that the structure is founded at level –4.00 m on a layer of semi-bound to well compacted sandstone of high bearing capacity.

4.4 DETAILED GEOPHYSICAL SURVEYS FOR DEFINITION OF GEOTECHNICAL AND GEODYNAMIC MODELS OF THE SITE

The main purpose of these investigations has been to define the seismic parameters for evaluation of the seismic stability of the structure. Namely, for structures of extraordinary importance according to the valid technical regulations, it is necessary to define the seismic input concretely for the site of the structure in order to perform correct dynamic analysis. The investigations have been carried out in compliance with the latest achievements in the field of earthquake engineering, [6]. The main concept of the applied procedure is to consider the expected earthquake effect through a probabilistic approach, including also the local soil effects through nonlinear dynamic analysis of a representative geotechnical model. Table 1 shows the expected values of maximum acceleration for different return periods, whereas Table 2 shows the defined seismic design parameters through input acceleration and characteristic spectra for the design and maximal earthquake. For such defined input parameters, the characteristic records whose frequency content covers the frequency range of interest have been selected in the dynamic analysis.

Table 1 - Maximal acceleration (in g) for different return periods Return period (years) DAF 50 100 200 500 1000

Bedrock 1.00 0.13 0.19 0.25 0.27 0.36 Foundation level 1.35 0.176 0.257 0.338 0.365 0.480

Table 2. Seismic design parameters Serviceabilitylife (years)

Level of seismic risk (%)

Earthquake Max. acceleration amax(g)

30-40 Design 0.34 100 and more 10-20 Maximum 0.39

240

4.5. CONCEPT FOR SEISMIC STRENGTHENING OF MUSTAFA PASHA MOSQUE

Based on these investigations and the defined seismic parameters, as well as detailed analysis of the seismic stability of the structure, the solution of structural strengthening has been accepted, (Fig. 8, 9), that complies thoroughly with the conservation principles for repair and strengthening of cultural historic monuments, [7]. It consists of incorporation of strengthening elements in the process of conservation of the architecture of the structure and in accordance with the existing project on conservation of the structure, with the main purpose of providing integrity to the structure as well as simultaneous behaviour of the bearing walls at the corresponding levels.

Image 8. Strengthening of Mustafa Pasha Mosque, (facade)

241

Image 9. Strengthening of Mustafa Pasha Mosque, (cross section)

DOME: After removal of the cement mortar layer over the dome placed in 1968, the following is anticipated to be carried out: (1) coating of the formerly constructed reinforced concrete ring in the base of the main dome with an injection mixture based on lime mortar and (2) placement of a CFRP strip in a layer of epoxy glue along the perimeter of the dome base within a width of 2.9 m. Then the entire dome is anticipated to be externally coated with a protective layer of lime mortar and covered in accordance with the project on the conservation of the architecture, (Fig.9-A).

TAMBOUR, BEARING WALLS: After cleaning of all the joints on the outside with a depth of 8 – 10 cm, it is anticipated to place CFRP bars of defined mechanical characteristics (tensile strength of ft = 1800 – 2000 MPa) in an epoxy mortar layer and connect them in the vertical joints. Then, it is planned to fill the joints with pointing lime mortar in accordance with the project on conservation of the architecture, (Fig.9- B,C,D,E,G,H).

FOUNDATION STRUCTURE: The solution consists of construction of a reinforced concrete wall with a thickness of d=25 cm along the perimeter of the foundation walls, on the external side, below the terrain level and down to the foundation level. From conservation reasons, this reinforced concrete wall will be physically separated from the existing foundation walls by a polyurethane coating for the purpose of separating the concrete from the existing stone masonry. To ensure interaction between the newly designed RC wall and the existing foundation structure, the solution anticipates placement of anchors made of chrome steel to an alternating length according to the reinforcement details, in previously formed openings filled with epoxy mortar (mixture of quartzite sand and epoxy) in accordance with the enclosed scheme, (Fig.9-F).

Existing RC belt course

Existing RC belt course

242

In the entire part of the structure extending above the terrain level, it is anticipated to remove the cement mortar injected in the cracks after the 1963 earthquake. With the project on conservation of the architecture, it is anticipated that these cracks as well as all the cracks detected after the opening of the external joints be injected with lime mortar with defined mechanical characteristics.

5. REALIZATION OF THE PROJECT Strengthening of the mosque structure in accordance with the designed system

started in the fall of 2007 with strengthening of the foundation structure. The solution consists of construction of a reinforced concrete wall with a thickness of d=25 cm along the perimeter of the foundation walls, on the external side, below the terrain level and down to the foundation level. From conservation reasons, this reinforced concrete wall was physically separated from the existing foundation walls by a polyurethane coating for the purpose of separating the concrete from the existing stone masonry. To ensure interaction between the newly designed RC wall and the existing foundation structure, the solution anticipates placement of anchors made of chrome steel to an alternating length according to the reinforcement details, in previously formed openings filled with epoxy mortar (mixture of quartzite sand and epoxy) in accordance with the enclosed scheme, (Fig. 10).

The realization of the project continued with the parallel working on the dome structure and bearing wall which strengthening is still under construction in compliance with the following:

Dome: After removal of the cement mortar layer over the dome placed in 1968, the following is anticipated to be carried out: (1) coating of the formerly constructed reinforced concrete ring in the base of the main dome with an injection mixture based on lime mortar and (2) placement of a CFRP strip in a layer of epoxy glue along the perimeter of the dome base within a width of 2.9 m. Then the entire dome is anticipated to be externally coated with a protective layer of lime mortar and covered in accordance with the project on the conservation of the architecture.

Bearing Walls: After cleaning of all the joints on the outside with a depth of 8cm, (Fig. 11), it is anticipated to place CFRP bars of defined mechanical characteristics (tensile strength of ft = 1800 – 2000 MPa) in an epoxy mortar layer and connect them in the vertical joints. Then, it is planned to fill the joints with pointing lime mortar in accordance with the project on conservation of the architecture.

Image 10. Strengthening of the foundation Image 11. Strengthening of bearing wall

243

In the entire part of the structure extending above the terrain level, it is anticipated to remove the cement mortar injected in the cracks after the 1963 earthquake. With the project on conservation of the architecture, it is anticipated that these cracks as well as all the cracks detected after the opening of the external joints be injected with lime mortar with defined mechanical characteristics.

REFERENCES [1] Seismic Strengthening and Repair of Byzantine Churches / P. Gavrilovic, V. Sendova and W.S.Ginell,// Journal of Earthquake Engineering,, Vol. 3 No. 2, 1999 [2] Consolidation and Reconstruction of St. Panteleymon Church in Ohrid/ P. Gavrilovic, G. Necevska –Cvetanvska, R. Apostolska, // IZIIS Report 2001, IZIIS - Skopje, 2001. [3] Main Project on Repair, Strengthening and Reconstruction of St. Ahtanasius Church in Leshok / V. Sendova, B. Stojanoski, // IZIIS Report 2004, IZIIS – Skopje, 2004. [4] Experimental testing of the historical monuments from / Lj. Tashkov, L. Krstevska, Bilateral scientific research project, Yildiz Technical University, Tureky – IZIIS Skopje, 2006. [5] Shaking Table Testing of Mustafa – Pasha Mosque Model / L. Krstevska, Lj. Tashkov, F. Mazzolani, K. Gramatikov, // EU-FP6 Programme, Project PROHITECH, WP7, 2007. [6] Definition of the seismic parameters for the evaluation of seismic stability of Mustafa Pasha Mosque in Skopje / V. Sesov et al., // IZIIS Report 2007-47, Skopje 2007. [7] Main Project on Repair and Strengthening of the Mustafa Pasha Mosque in Skopje / V. Sendova, B. Stojanoski, P. Gavrilovic// IZIIS Report 2007-41, Vol.1- 3, Skopje 2007.

245

Veronika Šendova150, Zoran Raki evi 251,Predrag Gavrilovi 352, Dimitar Jurukovski453

SEIZMI KA ZAŠTITA VIZANTIJSKIH CRKVI PRIMENOM SISTEMA ZA PASIVNU KONTROLU

RezimePrezentirani su rezultati eksperimentalnih i analiti kih istraživanja za primenu seizmi ke izolacije kod Vizantiskih crkvi. Testovi su pokazali da seizmi ka izolacija nudi zaštitu i sigurnost kulturno istoriskih spomenika, a njena primena bi trebala postati imperativ u njihovoj zemljotresnoj zaštiti. Da bi iskoristili prednost sistema, a istovremeno bili u saglasnosti sa konzervatoskim zahtevima, izvršene su analize za uklju enje sistema viskoznih dampera, radi konroliranja amplitude horizontalnih pomeranja u oba ortogonalna pravca. Kqu ne re i: vizaniske crkve, zemljotresna zaštita, izolacija

RETROFITTING OF BYZANTINE CHURCH USING PASSIVE BASE CONTROL SYSTEM

Summary: This paper presents the results from the experimental and analytical investigations performed to develop a methodology for application of seismic isolation in Byzantine churches. The experimental tests have proved that the seismic base isolation offers safety and protection and that its application should become an imperative in earthquake protection of historic monuments in future. To benefit from the advantages of base isolation and to be in compliance with the conservators' requirements, an attempt has been made to insert passive control viscous dampers to control the horizontal displacement amplitudes in both orthogonal directions. Key words: Byzantine churches, earthquake protection, seismic isolation

1 Prof. Dr, IZIIS, Skopje, Republic of Macedonia 2 Assoc.Prof. Dr., IZIIS, Skopje, Republic of Macedonia3 Emeritus Professor, IZIIS, Republic of Macedonia 4 Emeritus Professor, IZIIS, Republic of Macedonia

246

1. INTRODUCTION

Traditional stone and brick masonry structures, whether or not they are historic monuments, have low ductility, and, due to their stiff and brittle structural components, are usually severely damaged during strong earthquakes. The main reasons for damage or collapse are the lack of ductility of the masonry components, high displacements that the structure cannot afford, and amplification of dangerously high frequencies due to their dynamic behaviour in response to earthquake action making them vulnerable to those harmonies of the ground motion. In order to improve the behaviour of these masonry structures in response to earthquake forces, strengthening measures that emphasize the use of reinforcing steel have been widely adopted in building codes. In this practice, steel reinforcement is placed within piers and walls at critical areas to compensate for the lack of tensile strength and ductility and to increase the stiffness of these elements. Horizontal bands are provided at different levels in order to ensure "box-like" action and to reduce the possibility of "out-of-plane" failures. These strengthening concepts have met with success by greatly decreasing the potential of collapse. However, these methods are quite intrusive, may require the partial disassembly of building elements, can destroy valuable and irreplaceable interior finishes, and can also alter the external appearance of a building. In spite of these interventions, masonry structures can still be cracked during earthquakes of medium or higher intensity. Therefore, alternative design philosophies for seismic strengthening need to be explored for historic masonry structures where the conservation of historic fabric is paramount. One of the main tasks and problems in retrofitting of historic monuments in seismic regions in the process of their reconstruction and protection is to answer the question as to how far we should go as to the level of safety on one hand and the extent of the intervention on the other. Modern approach to protection of cultural heritage should be a multidisciplinary one sticking to the maxim of "minimum intervention - maximum protection", which should serve as a basis for introducing of criteria for protection against earthquake effects considering all the specificities of the site and the expected level of seismic effect as well as the specific characteristics of the monuments, their historic, cultural importance, their structure and characteristics of materials used for their construction.

2. CONCEPT OF BASE ISOLATION AND ITS APPLICATION IN CULTURAL HERITAGE

Seismic base isolation would appear to be an attractive alternate to the widely adopted seismic strengthening techniques discussed above. With base isolation seismic loads can be reduced to the same order of magnitude as the existing strength of the historic buildings and this in turn will decrease and may eliminate altogether the need for intervention above the level of isolation. In this manner, the interior and exterior fabric can be left undisturbed. With base isolation, a major part of the earthquake energy that would have been transferred into the building structure would be absorbed at the base level. Consequently, ductility demand to the structure is greatly reduced; displacement would be strictly controlled by the addition of an appropriate amount of damping within the base isolation system; and the frequency of the isolated structure is decreased to a value below that which dominates in a typical earthquake. So the three damaging criteria discussed above, lack of ductility, high displacements, and amplification of high frequencies are

247

addressed. In addition, the application of the technique of base isolation to increase the seismic safety of monuments with high cultural-historic value is promising where other possible strengthening techniques do not seem to be applicable. The main advantages of implementation of this methodology in the case of Byzantine churches are:

Isolators may give large reductions in the seismic loads for those structures, with short periods and low damping, which are most prone to suffer severe seismic attack if unisolated. Selected isolators may give very large reductions in the seismic loads on secondary structures and on the contents of appropriate structures. The complete intervention (construction work) is done at the foundation level so that only conservation work is done on the existing upper structure. The fresco-paintings on all the walls of the interior of churches are maximally protected against possible damage in the course of structural strengthening of the structures so that the process of fresco-conservation and protection runs independently of the structural intervention. A great number of churches that possess extremely valuable frescoes are quite small in size. Seismic isolation is very efficient and simple for their conservation.

The flaws of the above methodology are the following: The intervention carried out by incorporation of devices for seismic isolation at the foundation level represents a huge task that modifies, in a certain way, the basic concept of the existing structure. This represents a problem in the sense that conservators have to accept this as a possible way of protection in each concrete case. An isolator, which is effective in reducing seismic attacks on a structure, must have features, which result in relatively large isolator displacements. These large displacements, i.e. large movements of the monument structure in its base is, for other reasons, unacceptable for the conservators and the specialist dealing with protection of cultural heritage, although in that way the total structural displacements are little larger than the displacements of the supporting isolator. The isolation devices are modern technological products, which are costly and are produced by specified companies. The cost of seismic protection by application of such devices is higher than that of traditional seismic protection. However, for historic monuments, it is difficult to estimate the cost of such a protection since seismic isolation has so far been done on a very limited number of monuments.

It must be pointed out that application of seismic isolation in historical buildings is recommendable and its wider application should be expected in future if all the requirements of the involved specialists would be satisfied. In addition the cost of the base isolation, due to its advancement, has decreased significantly.

3. EXPERIMENTAL INVESTIGATIONS OF THE BASE ISOLATED CHURCH MODEL

To develop appropriate approaches for repair and strengthening of Byzantine churches, in general, and particularly churches located within Macedonia, research projects on seismic strengthening, conservation and restoration, including seismic isolation of Byzantine Churches in the Republic of Macedonia were realized in the period 1990-2000, (Gavrilovic et al., 1995, Gavrilovic et al., 2001). To experimentally verify the methodologies, a model of the church of St. Nikita, selected as a representative prototype

248

Byzantine church in Macedonia, was constructed to a scale of 1:2.75 and tested on the seismic shaking table in the Dynamic Testing Laboratory of IZIIS in its original, nonretrofitted state (M-SN-EXIST), strengthened state by use of "ties and injection", (M-SN-STR) and as a base isolated model (M-SN-BIC). Taking into account the previously acquired knowledge, the practice and the main assumptions from the aspect of analytical and experimental behaviour of the system, the concept of isolation within the framework of the investigation project is "laminated rubber bearing with stopper elements", (Fig. 1.) with two clearly distinguished elements: A. Laminated rubber bearing element for receipt and transfer of vertical gravity forces and

limited displacement (insulation) in horizontal direction, B. "Stopper element", which is introduced in trying to satisfy the criterion of "limited

displacement", one of the elements of specific behaviour and requirements for historic monuments regarding their protection in seismic conditions.

Image 1 – Seismic isolator in action

For this system and according to the main characteristics of the structure (weight, stiffness and expected-desirable behaviour), a total of 8 bearings with their main characteristics have been specially designed with the characteristics as given in Table 1.

Table 1. Characteristics of seismic isolators Circular cross-section, D=13.1 cm F = 134.7 cm2 Total height H = 15.1 cm No. of rubber elements Ng = 10 No. of steel sheets Nc = 9 Thickness of rubber element Dg = 1.06 cm Thickness of steel sheet Dc = 0.5 cm Sliding modulus of rubber G .20 kN/cm2 Stiffness of an isolator K =2.5 kN/cm Stiffness of a system of 8 isolators K8 =20 kN/cm Mass of the church model m = 29.6 t Expected (designed) period of the model M-SN-BIC T=0.80-1.0 sec

249

Image 2 – Model M-SN-BIC on the shaking table

In accordance with the possibilities and the objectives of the project (to demonstrate the effect of seismic isolation, not to produce isolators) these isolators have been completely manufactured in Republic of Macedonia. A steel structure for connection of the isolators with the church model structure and the seismic shaking table, (Fig. 2) has been designed and constructed for this system.

The model was subjected to the same series of dynamic tests as for the previously tested nonretrofitted model, (M-SN-EXIST) and the model strengthened by use of "ties and injection" (M-SN-STR). However, due to the base energy dissipation characteristics of the base isolated model, the tests were also continued at higher intensities. Detailed experimental investigations were performed for different levels of input acceleration and different earthquake records. The results obtained from the tests on model M-SN-BIC are given in Table 2, while Table 3 shows the comparison among the experimentally obtained results for all the three models. In accordance with the programme, the stopper elements were activated only during the last three tests.

250

Table 2. Response to different intensities of El Centro earth-quake.

* with stopper elements

Table 3. Comparison among the experimental results obtained for the three models.

Output acceleration (in g) for the church models

Earthquake Input acc(g)

M-SN- EXIST

M-SN- STR

M-SN- BIC

lev 1 lev 2 lev 1 lev 2 lev 1 lev 2 El Centro 0.17 0.29 0.55 0.20 0.47 0.08 0.10 El Centro 0.30 - - 0.65 1.10 0.11 0.16 El Centro 0.49 - - 0.91 1.59 0.26 0.39 El Centro 0.54 - - 0.77 1.41 0.35 0.68 El Centro 0.60 - - - - 0.42 0.82 Petrovac 0.19 0.39 0.76 0.27 0.48 0.09 0.15 Petrovac 0.40 - - 0.77 1.36 0.15 0.28 Breginj 0.28 - - 0.20 0.40 0.10 0.16 Breginj 0.38 - - 0.34 0.79 0.14 0.23

From the comparison of the experimental results, the following can be concluded: The results from these tests, especially compared with the results from the previously tested model strengthened by use of "ties and injection", pointed to a decrease of input acceleration in the model structure for 50-60%. It was evident that the output acceleration and the relative displacement through the height of the base isolated model were considerably smaller than those for the strengthened model. There was also a significant difference in the amplification of the dynamic response of the dome structure; in the case of M-SN-BIC, it was negligible. The base isolated model did not suffer damage under low and moderate earthquake intensities, while the damages under the expected accelerations with a return period of 1000 years, (amax=0.54g) were minimal and absolutely allowable and repairable.

Input Foundation Top of the dome

Acc(g)

Disp(mm)

Acc(g)

Disp(mm)

Acc(g)

Disp(mm)

0.17 9.0 0.07 11.0 0.10 12.0 0.30 18.0 0.10 23.0 0.16 35.0 0.40 27.0 0.19 32.0 0.22 40.0 0.49* 33.0 0.24 44.0 0.39 65.0 0.54* 36.0 0.27 48.0 0.68 73.0 0.60* 42.0 0.30 63.0 0.82 97.0

251

Finally, to conceive the efficiency of the applied seismic isolation system, the model was tested under the maximum earthquake intensity, (El Centro, amax=0.60g). Under this intensity, the monument was evidently stable and safe, with the exception of some minor damage.

However, from the detailed analysis of the model response, particularly when tested under high input acceleration, the following can be concluded:

It is a characteristic of the Byzantine churches that they are symmetric in respect to the longitudinal axis (x-x direction). During the experimental tests, the excitation was applied in the transverse direction (y-y) wherefore the model response was characterized by torsion. Under the highest input accelerations, horizontal displacements of up to 63 mm were observed in the middle of the model base (Table 2), i.e., up to 90 mm at the base of the east part of the model (at the apse). In accordance with the geometrical scale to which the model was built, these displacements would amount up to 25 cm in the case of the prototype, i.e., in real conditions. From the discussions held with specialists working in the field of protection of cultural heritage, it turns out that such displacement at the base of the Byzantine churches is unacceptable for the conservators although the relative displacements along the height of the structures in the case of base isolation would be negligible. Consequently, the problem is to be solved in another way.

4. ANALYTICAL INVESTIGATIONS OF BASE ISOLATED CHURCH MODEL M-SN-BIC 4.1. MODELING AND IDENTIFICATION OF THE ANALYTICAL RESULTS WITH THE EXPERIMENTAL ONES

To model the response of the M-SN-BIC model structure, the SAP 2000 computer package was used. Taking into account the complexity and the specific characteristics of the structural system and the materials built in the model of the structure on one hand and the possibilities offered by the programme package on the other hand, an attempt was made to define the church structure model in the most appropriate way by using finite elements, (Fig. 3). A moderately dense network of a total of 2548 nodes and 1604 elements was adopted to encompass the global geometrical characteristics of the model disregarding the inhomogeneity of the material. The reinforced concrete platform, the massive bearing walls and the walls of the tambour were modeled by a total of 1376 SOLID elements, while the vaults and the dome were modeled by a total of 220 SHELL elements. The isolators were modeled by a total of 8 LINK elements of the type of “rubber isolators”. The previously obtained experimental values were used as input data on the physical mechanical characteristics of all these elements.

252

Image 3. SAP2000 model with isolators

While analyzing the model by using SAP 2000, the option for nonlinear time history analysis was used whereat the same input excitations (harmonics and earthquakes) that were applied physically on the shaking table during the experimental tests were used as input time histories. Being complicated for precise analysis, those tests in which the stopper elements were activated in accordance with the programme (Table 2) were omitted from the analyses. The identification of the analytical results with the experimental results was done for all the conducted tests through comparison of the displacements and accelerations obtained at the characteristic points. The results are displayed through comparative presentation of the displacements and the accelerations at the level of foundation and the top of model obtained experimentally and analytically under the El Centro earthquake with amax=0.40g (Fig. 4).

Image 4. Comparison of obtained displacements, (El Centro, amax=0.40g)

Top of the dome

Foundation

253

4.2. ANALYTICAL SIMULATION OF MAXIMAL EARTHQUAKE EXCITATIONS

To obtain the analytical values of the church model response under the maximal possible excitations, the developed analytical model was used whereat the time history of the same excitation but with increased intensity (El Centro, amax = 0.60 g) was used. The results from this analysis spoke for themselves about the increased values of displacement and acceleration in the midst of the model base (dispmax = 86 mm, amax=3431 mm/s2), whereat the relative displacements and the amplification along the height of the church model remained negligible (dispmax

top = 97 mm, amaxtop=3672mm/s2). However, what is

important for the behaviour of the model is the torsion which can be seen through the comparative presentation of the displacements of the isolators L1 and L3 (Fig. 5a), i.e., the difference between the displacements of L1 and L3 during the excitation duration, which amounts to 59 mm at a certain moment (Fig. 5b).

a) Displacement of L1and L3

b) Difference between displacements of L1and L3

Image 5. Analytical results for the isolators L1 an L3, (El Centro, amax=0.60g)

4.3. INCORPORATION OF VISCOUS DAMPERS

To benefit from the evident advantages of base isolation over the other techniques used for retrofitting and also to comply with the requirements of the conservators, an attempt was made analytically to insert passive control viscous dampers to control the horizontal displacement amplitudes in both orthogonal directions. For that purpose, 4 new LINK elements of the “damper” type were inserted at the level of the model base (Fig. 6),

254

two in the longitudinal x-x direction (DX1and DX2 in line with L1) and two in the transverse y-y direction (DY1 in line with L1 and DY2 in line with L3).

Image 6. SAP2000 model with dampers

Numerous parametric analyses were performed during further investigations for the purpose of more precise definition of the necessary characteristics of the dampers. To get an insight into the effect of the dampers upon the maximum horizontal displacements and decrease the effect of torsion, the values of the viscous damping of isolator L3 were varied (from C=12 kNs/m to C=72 kNs/m), whereas those for the isolator L1 were kept constant (C=12kNs/m, C=18kNs/m, C=24kNs/m, C=36kNs/m). The results from the analysis show that, depending on the requirements of the conservators, the horizontal displacements at the base of the model can be reduced up to 40% for total damping in the analyzed direction of CDY1+DY2=48kNs/m or 13% of the critical damping and up to 60% for total damping in the analyzed direction CDY1+DY2=96kNs/m or 26% of the critical damping, whereat there is extensive reduction of torsion (Fig. 7).

Image 7. Time history of the difference between the displacements of L3 and L1 for the model without and with dampers (CDY1=36 kNs/m, CDY2= 60kNs/m)

255

5. CONSLUSIONS The seismic isolation system in historic monuments provides protection through

reduction of the seismic loads and relative displacements, as well as absorption of the hysteretic energy. This has been demonstrated by the experimental test, whose results point to complete control of the behaviour of the base isolated church model.

To satisfy the "limited displacement" criterion, viscous dampers can be inserted in the base of the historic monuments by previously defining their location and characteristics. In addition to the considerable reduction of deformations at the base (even up to 60%), the analytical investigations have demonstrated the effect of the dampers as to controlling and reducing the torsional effects.

The performed experimental and analytical investigation has undoubtedly proved that the new technology of seismic base isolation of historic monuments by installation of dampers offers absolute safety and protection and that its application should become an imperative in earthquake protection of historic monuments in future.

References [1] Earthquake Protection of Byzantine Churches Using Seismic Isolation/ Gavrilovic

P., Sendova V. & Kelley S. // Macedonian – US joint research & PHARE Cultural Development Program, Report IZIIS-2001-59, 2001

[2[ Seismic Strengthening, Conservation and Restoration of Churches Dating from Byzantine Period in Macedonia / Gavrilovic P., Ginell W. & Sendova V.,// Joint research project, IZIIS - Skopje, GCI -LA; Reports IZIIS 500-76-91, 92-71, 94-68, (1991-1995)

[3] The Application of Seismic Isolation for the Retrofit of Historic Buildings/ Kelly, S. J// EERC, University of California at Berkeley. 2000

[4] Analytical Modeling of Dynamic Behavior of a Frame Structure with and without Base Control System/ Rakicevic Z., D. Jurukovski and P. Nawrotzki // Paper ID-181, 4WCSCM, San Diego, 2006

257

Roberta Apostolska154 Golubka Necevska-Cvetanovska255 Zdravko Bonev356

Elena Vasseva457 Dylian Blagov558 and Julijana Cvetanovska659

METODA SPEKTRA KAPACITETA ZA OCENU SEIZMI KOGPONAŠANJA AB ZGRADA NA DEFORMABILNOM TLU

Rezime:Metoda spektra kapaciteta (CSM) je pouzdana alatka za ocenu seizmi kog ponašanja konstrukcije. U radu je prikazan seizmi ki odgovor 3D AB konstrukcije sastavljene od nosivih zidova dobijen primenom CSM sa uklju enim deformacijama tla. Kao rezultat push-over analize dobivena su ciljna pomeranja i faktor ponašanja. Kao rezultat fleksibilnosti tla analizirani 3D sistem je fleksibiniji, seizmi ki zahtevi su redukovani, a ciljna pomeranja su pove ana. Sistem je manje disipativan i njegovo ponašanje može biti zna ajno razli ito. Uticaj tla kombiniran sa neregularnoš u je zna ajan naro ito posle rušenja nekih od zidova.Klju ne re i: metoda spektra kapaciteta, push-over analiza, ciljno pomeranje, fleksibilnost tla

CAPACITY SPECTRUM METHOD FOR SEISMIC PERFORMANCE OF RC BUILDINGS INCLUDING SOIL FLEXIBILITY

Summary:Capacity Spectrum Method (CSM) is a reliable tool to predict seismic performance of structures. Seismic response of 3D RC wall systems using CSM which is consider for soil flexibility is presented in the paper. Calculated target displacement and behaviour factor as results from push-over analysis is elaborated. Due to the soil flexibility 3D wall system became more flexible, seismic demands are reduced and the target displacements are increased. The structure is less dissipative and the design performance could significantly changed. The influence of the soil conditions combined with irregularity is very important after collapse of some walls happened. Key words: capacity spectrum method, push-over, target displacement, soil flexibility

1 Assoc. professor, D-r, grad.civil eng., UKIM-IZIIS, Skopje, Macedonia; email:[email protected] 2Professor, D-r, grad.civil eng., UKIM-IZIIS, Skopje, Republic of Macedonia 3 Assoc. professor, D-r, grad.civil eng., UACG, Sofia, Bulgaria 4 Assoc. prof. D-r, grad. civil eng., BAS, Sofia, Bulgaria 5 D-r, grad. civil eng., UACG, Sofia, Bulgaria 6 Grad. civil eng., Student on postgraduate studies, UKIM-IZIIS, Skopje, Republic of Macedonia

258

1 INTRODUCTION

The Capacity Spectrum Method (CSM), [1] by means of a graphical procedure, compares the capacity of the structure with the demands of earthquake ground motion on the structure. The capacity of the structure is represented by a nonlinear force-displacement curve, sometimes referred to as a pushover curve. The recent advent of performance based design has brought the nonlinear static push-over analysis to the forefront as one of the most simplified procedure for evaluation of structural capacity. The base shear forces and roof displacements are converted to equivalent spectral accelerations and spectral displacements, respectively, by means of coefficients that represent effective modal masses and modal participation factors. These spectral values define the capacity spectrum. The demands of the earthquake ground motion are represented by response spectra. A graphical construction that includes both capacity and demand spectra results in an intersection of the two curves that estimates the performance of the structure due to earthquake action.

Capacity Spectrum Method is a reliable tool to predict seismic performance of structures subjected to design earthquake. The CSM is applicable to a variety of uses such as a rapid evaluation technique for a large inventory of buildings, a design verification procedure for new construction of individual buildings, an evaluation procedure for an existing structure to identify damage states, and a procedure to correlate damage states of buildings to amplitudes of ground motion.The great advantage of the method is its relative simplicity and opportunity to visualize Newmark’s postulates being related with peak responses of both purely elastic and elastic perfectly plastic systems. As is recommended by the new generation of seismic resistant design codes the method can be used for evaluation of seismic demands and for capacity assessment of newly designed or existing building structures. This procedure is implemented into Eurocode 8 [2] and enables calculation of behaviour factor and peak seismic response of the structure.

During the passed couple of years the Capacity Spectrum Method is under active development. The main goal is to extend the method with inclusion a variety of relevant factors that may influence the structural behavior.

It is already recognized that one of the factor that can significantly affect the seismic response and performance of structure is soil flexibility. Bonev [3] reported the possibility to apply Capacity Spectrum Method to soil-foundation-structure problem and how the method should be generalized including soil influence.

The subject of interest in this paper is 3D RC structure being designed as a wall system. It is shown in FEMA 450 [4] that the wall systems are much more sensitive to soil deformations because the most stiff elements - RC walls dictate internal force distribution. In this paper only linear soil properties are taking into account being represent by unit foundation modulus (Winkler's constant). Due to the footing flexibility the target displacements are increased and the global ductility of the structure is reduced. It is shown that for soft soils the design performance of the structure may remain completely elastic. In countrary, for stiff soils seismic demands are large enough to develop significant inelastic deformations. It is found that the influence of accidental eccentricity is much essential after collapse of the maximum loaded walls. Considering the analysis of the obtained numerical results some important conclusions is presented in the paper.

259

2 NUMERICAL MODEL AND ANALYSIS METHOD The influence of foundation flexibility on the capacity curve and on capacity spectrum method as a whole is studied on 3D RC building structure with RC walls as a primary bearing elements.The model is consisting of RC walls, columns and slabs and its general view is presented in Figure 1.

Figure 1 - General 3D view of the mathematical model and structural layout

The numerical model used in calculations is defined with compliance of the assumptions listed below:

Floor slab is treated as a rigid diaphragm in its own plane. The membrane stiffness of the floors is practically infinitely large and the slab may move horizontally as absolutely rigid body. On the other hand the slabs distribute the seismic loads between the walls. The vertical loads are carried by shear walls and columns. Lateral loads are carried by the shear walls only. Slab to column connection is not designed as moment resisting. It is assumed that columns are pinned at both ends and could bear only vertical loads. Shear walls are modelled by vertical frame elements. The potential locations of plastic hinges are considered at each floor level. The structure is symmetric in plan with respect to X- and Y- axis. Bending stiffness of the slab is taken into account only to obtain the vertical loads distribution between the vertical elements – walls and columns. Single footing under each wall is used. The foundation is supported by soil with vertical resistance. The elastic soil properties are implemented by the unit foundation modulus (Winkler’s constant). Loading pattern used for pushover analysis in both X- and Y- directions has the shape of inverted triangle and implies linear force distribution in elevation. Forces are applied in CM for each floor level. The calculation of each spring stiffness implies that only rocking motion of the footing is considered.

e

eyCRCMM

260

Torsion effects due to different disposition of CM with respect to CR are taken into account. The influence of accidental eccentricities is accounted for as a source of torsion. Axial forces are remaining constant during the lateral pushover analysis and plastic hinge properties once determined after application of the vertical loads are kept the same. The potential location of the first plastic hinge is at the base of the shear wall (in the

centre of the plastic zone). The plastic zone approach is based on distributed plasticity model. The typical constitutive relationship is moment-curvature. Distribution of elastic/inelastic curvature for a simple wall element is shown in Fig. 2.

Figure 2- A single wall designed as dissiptive wall

Capacity spectrum method is used as evaluation tool where the use of design demand spectra is recommended. A version of the method based on elastic demand spectrum is available in Annex B of Eurocode 8 [2]. The global behavior of the structures is indicated by the base shear – top displacement relationship. For the mathematical model of RC building structures a series of nonlinear push-over analysis were carried out using SAP2000 computer program, [5]. Lateral load is increased until collapse prevention state is reached. Two simplified and independent analyses in each X- and Y- direction are carried out. Two equivalent single degree of freedom systems are used. After that the capacity spectrum method is applied in both orthogonal directions in order to calculated values of behaviour factors, performance points and target displacements with including influence of the foundation flexibility, [6,7].

3 RESULTS FROM ANALYSIS The numerical model described above is subjected to monotonically increasing vertical

and horizontal loads. Six values of unit foundation modulus are used in calculations: fixed base (infinitely large modulus), 60 000, 50 000, 40 000, 30 000 and 20 000 in kN/m3 metric units. After completion of vertical loading procedure the horizontal loading pattern is applied. The effects of accidental torsion are studied considering eccentricity of 15% (large eccentricity and irregular structure) and 0% (regular structure). Selected numerical results are presented further. More detailed information can be find in [7].

M

zLp/2

L

Lp

d

(z) = el(z)

z

x

z

261

Figure 3, (3a and 3b) illustrate that global ductility demand is reduced with increasing the footing flexibility and the initial (elastic) stiffness is reduced due to flexibility. At the same time the base shear strength is relatively slightly influenced by the footing flexibility.

0

100

200

300

400

500

600

700

0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1

Roof Displacement, [m]

Bas

e Sh

ear,

[kN

]

rigid base

c=60000

c=50000

c=40000

c=30000

c=20000

Figure 3a - Capacity curves in X-direction (ecc. 0%) obtained for different Winkler's constants

0

100

200

300

400

500

600

700

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

Roof Displacement, [m]

Bas

e Sh

ear,

[kN

]

Figure 3b - Capacity curves in Y-direction (ecc. 0%) obtained for different Winkler's constants

Results form the investigations show that the largest values for the behaviour factor could be achieved if the fixed base is considered, (fig. 4). The smallest target displacements are observed in the same case. If the foundations are flexible (fig. 5 and 6) target displacement is increased but the behaviour factor decreases. When the soil is soft and structure reaches the target displacement the global behaviour of the structure may remain completely elastic. This mode of deformation implies that the soil fails before yielding happens in structure. Safe design solutions could be provide if soil deformations are taken into account in capacity curves.

262

0

0.5

1

1.5

2

2.5

3

0 0.02 0.04 0.06 0.08 0.1 0.12

u*, [m]

S d, V

/m, [

m/s

2 ]

0

0.5

1

1.5

2

2.5

3

0 0.02 0.04 0.06 0.08 0.1u*, [m]

S d, V

/m, [

m/s

2 ]Figure 4 - Capacity spectrum method applied to fixed-base structure: (a) in X-direction; (b) in Y-direction.

0

0.5

1

1.5

2

2.5

3

3.5

4

0 0.05 0.1 0.15 0.2 0.25

u*, [m]

S d, V

/m, [

m/s

2 ]

0

0.5

1

1.5

2

2.5

3

3.5

4

0 0.05 0.1 0.15 0.2 0.25

u*, [m]

S d, V

/m, [

m/s

2 ]

Figure 5 - Capacity spectrum method applied to structure with flexible foundations (UFM 60 000 kN/m3)-(a) in X-direction; (b) in Y-direction.

0

0.5

1

1.5

2

2.5

3

3.5

4

0 0.1 0.2 0.3 0.4 0.5

u*, [m]

S d, V

/m, [

m/s

2 ]

0

0.5

1

1.5

2

2.5

3

3.5

4

0 0.1 0.2 0.3 0.4

u*, [m]

S d, V

/m, [

m/s

2 ]

Figure 6 - Capacity spectrum method applied to structure with flexible foundations (UFM 20 000 kN/m3)-(a) in X-direction; (b) in Y-direction.

q=10u*

t,0=0.064mu*

t,15=0.067m

q=10u*

t,0=0.074mu*

t,15=0.075m

(a)

(b)

q=4.25u*

t,0=0.18mu*

t,15=0.19m

q=4.25u*

t,0=0.16mu*

t,15=0.22m

(a)

(b)

q=4.25u*

t,0=0.23mu*

t,15=0.23m

(a) q=4.25

u*t,0=0.22m

u*t,15=0.23m

(b)

263

4 CONCLUSIONS Considering the analysis of the numerical results presented above the following conclusions

could be made: The foundation flexibility caused by soil deformations influence essentially the capacity curves. The global ductility factor is seriously being reduced due to significant increase of elastic part of deformations. Participation of soil deformation in overall structural deformation is significant with tendency to become more essential in case of plastic soil deformations and foundation uplift. Target displacements are increased if the soil is becoming softer. The behaviour factor however shows decreasing tendency for weaker soils which is not on the safety side. Safe design solutions could be expected if soil deformations are taken into account in capacity curves. The influence of accidental torsion effects is small considering the elastic behaviour of the structure. More important influence is observed when some plastic hinges yield and when some walls are collapsed. The capacity curves are sensitive to accidental torsion when wall elements yield or collapse occurs. The global structure strength is relatively independent of soil stiffness and accidental torsion effects. It is concluded that wall systems are sensitive to flexible soil conditions in a large extent. Better results for the structure could be expected if pile foundations are used to decrease the effect of soil deformations.

ACKNOWLEDGEMENTThe authors are indebted to Ministry of Education and Science of R. Macedonia, National

Science Foundation at Ministry of Education and Sciences of Bulgaria, (project grant No. BM-6/2006) and University of Architecture, Civil Engineering and Geodesy – Sofia, Bulgaria, (project grant No. BN-84/2008).

264

REFERENCES

[1] Freeman, S., Development and use of Capacity Spectrum Method, Paper No. 269, Proc. of the 6th U. S. National Conference of Earthquake Engineering, Seattle, Washington, 1998.

[2] EN 1998: 2004 Eurocode 8: Design of structures for earthquake resistance. Part 1: General rules, seismic actions and rules for buildings. [3] Bonev, Z. et al., Behavior factor evaluation accounting for the elastic foundation, Proc.

of the International Conference on EE "Earthquake Engineering in the 21st Century, IZIIS 40 EE-21C, 27th August-1st September, 2005, Skopje/Ohrid, Macedonia.

[4] Fema 450 (2003) NEHRP Recommended Provisions for Seismic Regulations for New Buildings and other Structures, Commentary C7A.

[5] Wilson and Habibullah. SAP 2000 - Structural Analysis Programme. CSI, Berkeley, California, 2006.

[6] Necevska-Cvetanovska G., Vasseva E., Bonev Z., Apostolska R. et al., "Reduction of seismic vulnerability of RC building structures based on EC8-Application in Bulgaria and Macedonia", Joint Macedonian-Bulgarian project, IZIIS Report 2008-54.

[7] Blagov D., V. Georgiev and Z. Bonev, Influence of Flexible Foundations on the Design Response of Buildings with Accidental Eccentricity, Proceedings of the 5-th European Workshop on Irregular and Complex Structures (EWICS), 16-17 September 2008, Catania, Italy.

265

Golubka Necevska-Cvetanovska160 Roberta Apostolska261, Nataša Mir i 362

and Julijana Cvetanovska463

KONSTRUKTIVNE MERE ZA POBOLJŠANJE SEIZMI KOGPONAŠANJA BEZGREDNIH KONSTRUKTIVNIH SISTEMA

Rezime:Bezgredni konstruktivni sistem za razliku od tradicionalnog AB ramovskog sistema je seizmi ki povredljiviji. U cilju evaluacije seizmi kog ponašanja bezgrednog sistema, ura ene su komparativne analize izme u pet razli itih modela konstruktivnih sistema i referentne ramovske konstrukcije. Istraživani su efekti predloženih konstruktivnih mera na dinami ke karakteristike i na kapacitet nosivosti i deformabilnosti bezgredne kon-strukcije. Rezultati analize pokazuju da projektovane konstruktivne mere poboljšavaju mali kapacitet nosivosti i pove avaju otpornost i krutost ovih sistema. U papiru su prikazani selektirani rezultati analiti kih istraživanja. Klju ne re i: bezgredna plo a, seizmi ko ponašanje, obodne grede, konstruktivne mere

STRUCTURAL MEASURES FOR IMPROVING SEISMIC PERFORMANCE OF FLAT-SLAB BUILDING STRUCTURAL SYSTEMS

Summary:Flat-slab structural system is more vulnerable under seismic events than traditional RC frame system. To evaluate the seismic performance of the flat-slab structural system, comparative analyses have been made between five different models of structural systems and the referent frame structure. The effects of structural modifications upon the dynamic characteristics as well as upon the bearing and deformability capacity of the flat-slab structure have been investigated. Results from the analysis show that the design structural measures improve small bearing capacity of the system and increase its strength and stiffness. Selected result from the analysis are presented in the paper. Key words: flat-slab, seismic performance, perimeter beams, structural measures

60 Professor, D-r, grad. civil eng., UKIM-IZIIS, Skopje, Macedonia, email:[email protected] 61 Assoc. professor, D-r, grad. civil eng., UKIM-IZIIS, Skopje, Republic of Macedonia 62 M. Sc., grad. engineer architect, Skopje, Republic of Macedonia 63 Grad. civil eng., Student on postgraduate studies, UKIM-IZIIS, Skopje, Republic of Macedonia

266

1 INTRODUCTIONIn design and engineering practice, the selectively defined design of space, design of

structure, speed and efficiency of realization represent an extraordinarily important factor for the Investor. This assertion is supported by the fact that the flat-slab RC system has lately been increasingly imposed as a more acceptable and more attractive structural system in the world and in Macedonia as well. What is rational and optimal for these flat-slab structures is that they enable simple design, pure and clear space with absence of beams (the role of the beams is transferred to the RC floor slab), faster construction and time saving.

The system consists of columns resting directly on floor slabs for which sufficient strength and ductility should be provided to enable sustaining of large inelastic deformations without failure. The absence of beams, i.e., the transferring of their role to the floor RC structure which gains in height and density of reinforcement in the parts of the hidden beams, the bearing capacity of the structural system, the plate-column and plate-wall connection, all the advantages and disadvantages of the system have been tested through long years of analytical and experimental investigations. For the last 20 to 30 years, the investigations have been directed toward definition of the actual bearing capacity, deformability and stability of these structural systems designed and constructed in seismically active regions.

The paper displays part of the results from analyses of six types of structural systems for a prototype of a residential building in Skopje for the purpose of defining the seismic behaviour and resistance of flat-slab structural systems [1].

2 ANALYSIS OF SEISMIC RESISTANCE OF RC FLAT-SLAB STRUCTURAL SYSTEM 2.1 GEOMETRICAL CHARACTERISTICS OF THE ANALYZED

STRUCTURAL SYSTEMS

To evaluate the seismic behaviour and resistance of a flat-slab RC system, analyses of a typical prototype of a residential building in Skopje with B +GF +4 + A have been carried out (Fig. 1).

Figure 1- Characteristic plan and cross-section

267

For the chosen prototype of the residential building six types of structural systems have been analyzed. Geometrical characteristics of each of these structural systems are presented in the table 1 and figure 2.

Table 1- Geometrical characteristics of the analyzed structural systems Type of structural system plate

[cm] columns

[cm] beams [cm]

perimeter beams [cm]

RCwalls

Frame M1 14 60/60 40/40 No No Purely flat-slab M2 20 60/60 No No No Purely flat-slab M3 25 60/60 No No No Flat-slab strengthened by a perimeter beam

M4 20 60/60 No 40/40 No

Flat-slab strengthened by RC walls

M5 20 60/60 No No Yes

Flat-slab strengthened by perimeter beam and RC walls

M6 20 60/60 No 40/40 Yes

Figure 2- Characteristic plans – purely flat-slab system and flat-slab system strengthened by perimeter beams and RC walls

2.2 SEISMIC AND DYNAMIC ANALYSIS AND RESULTS FROM ANALYSIS

To evaluate the seismic behaviour and resistance of the flat-slab structural system, comparative analyses have been made between the models of structural systems M2, M3, M4, M5 and M6 and the referent frame structure – model M1. The effects of the designed modifications upon the dynamic characteristics as well as upon the bearing and deformability of the flat-slab structure have been investigated.

The analyses have been performed by using the finite element method and the SAP2000v10.0.9Advanced computer programme [2]. The 3D mathematical model of each of the analyzed structures has been formulated by discretization of the bearing system into

268

finite elements. The vertical loads have been defined in accordance with the valid national technical regulations and the purpose of the structures.

Seismic analysis has been carried out in compliance with the national regulations for design of high rises in seismically prone areas, [3]. The horizontal loads have been defined in the form of a design spectrum of acceleration in accordance with Eurocode 8, [4] scaled in such a way that it generates the total shear force at the base to the amount of 10% of the weight of the structure.

Dynamic analysis has been carried out for selected structural systems (model M1, M2 and M4) exposed to the effect of the El Centro earthquake with amax=0.32g.

The results obtained from the analyses of different structural systems are presented in the form of: dynamic characteristics (periods and mode shapes), maximal displacements and relative storey drifts in both orthogonal directions, time histories of absolute displacements at the top as well as bearing capacity and deformability of the selected structural systems (model M1 and model M2), [5].

Presented further are some of the results obtained from the ample analytical investigations. The first mode shape of vibration of the structural system of model M1 and model M2 is given in Fig. 3.

Figure 3- First mode shape for model M1 and model M2

Table 2 shows the maximal values of moments due to vertical loads above support and in the middle of the span for the analyzed models of structural systems. The distribution of maximal moments under vertical loads over the plate of the second story of models M1 and M2 is presented in Fig. 4.

Table 2 - Comparison of maximal moments due to vertical loads in the plate at the second story

Type of structural system Maximum moments [kNm] Mmax, support Mmax, middle

spanFrame M1 -4.55 3.90 Purely flat-slab M2 -25.2 11.2 M3 strengthened with a perimeter beam M4 -19.8 8.8 M3 strengthened with perimeter beam and RC walls M6 -21.6 9.6

T1=0.767sec T1=0.991sec

269

Figure 4 - Maximal moments in the plate at the second storey for model M1 and model M2

The time histories of displacements at the top of the structure for model M2 and model M4 are presented in Figure 5.

Figure 5 - Time histories of maximum displacement

2.3 ANALYSIS OF RESULTS After the performed analytical investigations, comparative analyses have been

performed for: the fundamental period of vibration (T1), the maximal horizontal displacements in both orthogonal directions (Table 3), the time histories of displacement as well as bearing and deformability capacity for selected structural systems.

The results have shown that the purely flat-slab system has a greater fundamental period and greater displacements in respect to the frame system. The occurrence of torsion in the first mode shape is also characteristic. The best behaviour has been exhibited by model M6 whose fundamental period is less than that of the frame system, with reduction of displacements of 40%. The relative storey displacements (Fig. 6) show the same tendency.

d=5.114cm d=3.352cm

270

Table 3 - Comparison of periods and maximal displacements of the analyzed models Type of structural system plate T1 Max. displacements [cm]

[cm] [sec] X-Xdirection

Y-Ydirection

Frame M1 14 0.767 2.725 2.662 Purely flat-slab M2 20 0.998 3.522 3.416 Purely flat-slab M3 25 0.794 2.743 2.676

M3 strengthened with a perimeter beam

M4 20 0.789 2.786 2.752

M3 strengthened with RC walls M5 20 0.956 1.970 2.592 M3 strengthened with perimeter

beam and RC walls M6 20 0.740 1.719 2.310

Fig. 7 shows the comparative results from the analysis in nonlinear range referring to bearing and deformability capacity of structural systems carried out for models M1 and M2. The obtained results show that the strength and stiffness capacity of the flat-slab system are lower for 38% in respect to the frame system.

Figure 6 - Relative storey displacements in x direction [cm]

4 CONCLUSIONS AND RECOMMENDATIONS The purely flat-slab RC structural system is considerably more flexible for

horizontal loads than the traditional RC frame structures which contributes to the increase of its vulnerability to seismic effects. The critical moment in design of these systems is the

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

1

3

5

7

M3 0.17 0.3a1 0.335 0.309 0.26 0.199 0.134

M2 0.313 0.635 0.708 0.656 0.542 0.4 0.268

M1 0.29 0.535 0.56 0.5 0.399 0.278 0.163

1 2 3 4 5 6 7

271

slab-column connection, i.e., the penetration force in the slab at the connection, which should retain its bearing capacity even at maximal displacements. The ductility of these structural systems is generally limited by the deformability capacity of the column-slab connection.

To increase the bearing capacity of the flat-slab structure under horizontal loads, particularly when speaking about seismically prone areas and limitation of deformations, modifications of the system by adding structural elements are necessary.

The realized investigations have shown that the flat-slab structural system with well defined modifications can exhibit a favourable and rational factor of behaviour compliant with Eurocode 8 and can thus be treated as a system with acceptable seismic risk. The modification with certain structural elements improves the low bearing capacity and deformability of the system and leads to more adequate seismic behaviour of the purely flat-slab structure.

Figure 7- Comparison of bearing capacity and deformability

Displacement

Shear force [kN]

Model M-1 Model M-2

272

REFERENCES o Mircic N. (2006). "Seismic Resistance of Flat-slab RC Structural Systems

of High Rise Buildings“, Master Thesis, IZIIS, „Ss. Cyril and Methodius“ University, Skopje.

o Wilson and Habibullah. SAP 2000 - Structural Analysis Programme. CSI, Berkeley, California, 2006.

o Rulebook on Technical Norms for Construction of High-rises in Seismically Prone Areas (1981)

o EN 1998: 2004 Eurocode 8: Design of structures for earthquake resistance. Part 1:

General rules, seismic actions and rules for buildings. o Necevska - Cvetanovska G., Petrusevska R. (2000)."Methodology for

Seismic Design of R/C Building Structures", Proc. of the XII-th World Conference of Earthquake Engineering, (12WCEE), New Zealand, February 2000.

273

164 265

2

2

. -

.-

, 2 .

.: 2 , ,

POSSIBILITIES FOR 2D GEODETIC NETWORKS UTILIZATION FOR SEISMIC EVENTS DETERMINATION

Summary In this paper accuracy of one 2D geodetic network form are being analyzed and possibilities for utilization of crustal movements determining. Reliability and possibilities for of geodetic networks point movements determination are of special interest in this paper. Geodetic networks are analyzed in dependence on geodetic instruments and measurement methods which are used. The results of utilization classical terrestrial geodetic methods, GPS methods and their combination are analyzed. Also, short review of the up-to date utilization of geodetic methods in seismic events research is given. Key words: 2D geodetic networks, reliability, measurement methods

1 . , . . . , , [email protected] . . . ., „ “ , [email protected]

274

1

. 20 .

.

.

.

.

.

2 2

2 ( , . , ., 1984, , ., 1981, 1982, 1985 1990).

.

. 2. 2 16

( ) 25km x 25 km. 2 10000 km2. 1 2

.

1 –

275

3 2

( ) .

.

.. .

, , 5 km.

..

.

.

. ( )

.

..

., ( 2

).

..

.

4

.

.. (W.F. Caspary,1987

. , 1986):

276

(1)

(1)

:- – ,- – ,- – ,- – - – .

2,

. ().

. (W.F. Caspary,1987 .

,1989). .

( . , 1989; . . . ,1994)

.

.

. 2

.. -

,.

( )( . , . . ,1994) :

(2) :

- ,- ,- ,- .

.

277

5 2

.,

. ., 2

. 2 10000 km2,

100 km 100 km. 25 25 km.

.

2

.. .

2 :- ,- ,-

,- ,-

.

1. .

1 – 2

uii rii a b P D P D

1 105.3 mm 90.9 mm 0.32 / 0.68 / 16.8 ppm 2 10.0 mm 8.1 mm / 0.33 / 0.67 1.4 ppm 3 9.8 mm 8.0 mm 0.01 0.32 0.99 0.68 1.4 ppm 4 19.3 mm 15.7 mm / 0.33 / 0.67 2.7 ppm 5 18.5 mm 15.2 mm 0.02 0.31 0.98 0.69 2.6 ppm

.

. 2 3

2 .

278

.

,.

:

- ,

;

- ,

;

- ,

- ,

.

..

6 2

. 2 .

. 2 -

. 2

.

.

279

1( ) / . , . ,// , 1984,

2 II, I / . // , 2. , 1981,

3 II, II / . // , 2. , 1982,

4 III / . // , 1985, 5 , / . // ,

1990, 6 / . , ., ., //

, 1994, 7 / . // ,

1989, 8 / . // , 1986, 9 / . // , 2005, 10 Concepts of network and deformation analysis / W.F. Caspary// School of Surveying

The University of New South Wales, Kensington, Australia 1987 11 Geodetic strain in peninsular Italy betwen 1875 and 2001/ Hunstad, I. et al.: //,

vol.30,No.4,1181, doi: 10.1029/2002GL016447,2003

281

Prof Božidar Pavi evi 1 66

PROSTORNO-URBANISTI KO PLANIRANJE KAO KLJU NIASPEKT INTEGRALNOG UPRAVLJANJA SEIZMI KIMRIZIKOM

Rezime:U radu se pledira na neodložnoj potrebi rehabilitacije i uvažavanja svojstvenih doma ih iskustava na tretiranju i smanjenju seizmi kog rizika, ste enih nakon razornog Crnogorskog zemljotresa, 1979. Nažalost, tako voluntaristi ki igno-risanih kroz novodonijeti PPR Crne Gore i krajnje kontraverzni novi republi kiZakon o ure enju prostora i izgradnji objekata (2008). Elaborira se stav o nužnosti hitnog prevazilaženja postoje eg stanja i zakonskog uvodjenja zdravog sistema prostorno-urbanisti kog planiranja, uklju ivo donošenje posebnog zakona za utemeljenje sveobuhvatne nacionalne strategije za integralno upravljanje seizmi kim rizikom, a time i za racionalno upravljanje prostorom. Apostrofira se aspekt regionalne saradnje u tretiranju relevantne problematike, uklju ivo i domenEurocode 8.Klju ne rije i:Seizmi ki rizik, mitigacija, urbanísti ko planiranje, Lex Specialis.

LAND USE PLANNING AS THE KEY ASPECT OF INTEGRATED SEISMIC RISK MANAGAMENT

Summary:Some effects of the disastrous earthquake that struck Montenegro in 1979 are presented, and the need to learn from experience gained in this catastrophic earthquake is emphasized. The position that the existing situation should be improved through coordination of all relevant approaches is advocated. The adoption of a special law, on which the strategy of an integrated seismic risk management would be based, is proposed. The regional cooperation, as well as adoption Eurocode 8 NPDs, on the national level is considered significant. Key words: Seismic risk, mitigation, land-use planning, Lex Specialis, codes.

1Profesor (Mostovi, Aseizmi ko projektovanje, Aseizmi ko planiranje;šef Katedre za zemljotresno inženjerstvo), u penziji; bivši Gen. direktor RZUP-a, Titograd (u vrijeme i nakon Crnogorskog zemljotresa 1979); ekspert UNEP-a; Koordinator Programa SEISMED&UNEP/MAP Priority Action: "Land-Use Planning in Mediterranean Earthquake zones", PAP/RAC, Split; osniva i bivši predsjednik CDZI-a&YUZI-a, odnosno CAZI-a; Po asnipredsjednik CAZI-a; M: EERI, US Forum, EAEE; etc. E-mail: [email protected]

282

1 UVODNE NAPOMENE I IZAZOVI 1.1 RIJE (za) UNAPRIJED

Kada je, u pitanju vremenski kontekst odnosno termin održavanja ove vrlo zna-ajne me unarodne Konferencije sticajem okolnosti, može se konstatovati da je rije o

krajnje dobrodošlom i višestruko izazovnom trenutku. Naime, termin održavanja ovog skupa koincidira upravo sa svojevrsnim razme em obilježavanja 45-te godišnjice velikog Skopskog i teku e 30-te godišnjice razornog Crnogorskog zemljotresa, 1979.– kao doga aja me aša u razvoju zemljotresnog inžinjerstva i zaštite od zemljotresa ne samo na našim prostorima nego i sa najširim uticajem i zna ajem.

Ovaj prvi - sa specifi nim odrazom na za etak jugoslovenskog i evropskog zemljotresnog inženjerstva kao i na odlu no razjašnjenje uloge arhitekture u praksi aseizmi kog projektovanja u zgradarstvu, a ovaj drugi - na prepoznavanju klju ne uloge prostorno-urbanisti kog planiranja (rije ju, integralnog planiranja održivog razvoja) u pogledu efektivne kontrole i smanjenja seizmi kog rizika.

1.2 NAŠ “ŽIVOT” SA ZEMLJOTRESOM vs SEIZMI KI “BRISANI PROSTOR”

Pri datim okolnostima - autoru se ini dodatno zna ajnim i potencijalno šire izazovnim što se ovaj skup dešava i u situaciji kada se Crna Gora (valjda ne i zemlje iz njenog najbližeg okruženja) nalazi pod autogenim iracionalnim pristupom/pritiskom nekriti kog uklanjanja svih vrsta tzv. barijera biznisu267 – naro ito onih u domenu koriš enja prostora. Sve to, pod izgovorom - radi “otvaranja vrata” stranom kapitalu (alias, radi što bržeg pribli-žavanja Evropskoj Uniji, itd.). Naravno da se kao glavna barijera pri tome, ispostavlja i ozna ava njen dosadašnji sistem prostornog i urbanisti kog planiranja (iako svojevremeno meritorno verifikovan kao sistem najbolje svjetske prakse), ak - ne samo kao takav nego i kao pojmovna kategorija, uopšte.

Otuda, navodno - po ugledu na praksu nekih susjednih zemalja (bliskih ulasku u Evropsku uniju), po skra enom i hitnom postupku donosi se aktuelni zakon tako eskra enog naziva: "Zakon o ure enju prostora i izgradnji objekata".

Na ovaj na in je, kona no, ta jedna od glavnih prepreka biznisu tj. prethodno postoje isistem urbanisti og planiranja (apostrofiraju i pored ostalog njegovu okosnicu – GUP, Generalni urbanisti ki plan) decidno uklonjena, a samo planiranje, eksplicitno i generalno prepušteno investitorima – biznismenima, "budu i da oni najbolje znaju šta im treba", a država e tek zakonski sprovoditi "monitoring stanja u prostoru"... (SIC!)

Može se re i da je ovim i ovakvim republi kim zakonom ( ijim su stupanjem na snagu prestali da važe dosadašnji: Zakon o planiranju i ure enju prostora; Zakon o gra evins-kom zemljištu; Zakon o izgradnji objekata; te Zakon o urbanisti koj i gra evinskoj inspekciji), kaogod i prethodno donijetim novim Prostornim planom Republike - izveden potpuni otklon ne samo od jednog prethodno utemeljenog i trajno aktuelnog koncepta integralnog upravljanja prostorom odnosno fizi kim i održivim razvojem Republike, nego i od svih elementarnih iskustava ste enih nakon Crnogorskog zemljotresa od 1979. Ina e, svojevremeno sublimiranih kroz dalje nazna ene dokumente, istraživanja, studije i projekte

2 Nažalost, sve to uz pre utnu i grubu zamjenu teza – s obzorom da se nare eni relavantni stavovi i uslovi Evropske Unije eksplicitno odnose na vješta ke i administrativne prepreke trgovini (tj. jedinstvenom evropskom tržištu).

283

- a posebno kroz Projekat UNDP-YUG/79/104 (Prostorni Plan Republike i GUP-ovi opštinskih centara Crne Gore). I to, uz striktno isticanje zaštite od zemljotresa i zaštite okoline - kao vode ih komponenti planiranja toga razvoja.

Odustaju i od zalaženja u zna enje u injenog otklona i njegov odraz na fakti ko urušavanje odnosno potiranje i same ideje o integralnom upravljanju prostornim razvojem Republike (jer, “prostorom upravlja svako ko ga mijenja”) – neminovno i nezaobilazno ostaje sasvim otvoreno (za upu ene i dramati no zabrinjavaju e) pitanje upravljanja seizmi-kim rizikom, ukupno i kroz njegove klju ne pojavne oblike i aspekte. Podrazumijevaju i

pri tome sve nazna ene integrativne aspekte i opcije smanjenja seizmi kog rizika - pa u odre enom smislu, i apostrofiraju i one koji se odnose na oblast aseizmi kog prostorno-urbanisti kog planiranje , projektovanja i izgradnju objekata kao i infrastrukturnih sistema (life lines).

Ovo zadnje dodatno poprima na zna aju - specifi no u okolnostima predstoje e trans-pozicije Eurocode 8 u nacionalno zakonodavstvo (kaogod i izrade korespodentnog Nacio-nalnog aneksa, NDPs), s obzirom na evidentno uvriježeno poimanje o njegovoj samo-dovoljnosti tj. kao dovoljnom iskazu mjera kontrole ukupnog seizmi kog rizika. Što, naravno, nije ta no i ne može biti prihvatljivo ni u kom slu aju – a posebno ne za seizmogene zone sa nivoom hazarda kakvome je izložena teritorija Crne Gore odnosno Južnog Jadrana, uklju ivo i šire okruženje – odnosno i Balkanski region u cjelini. Ina e, o stavu autora u vezi sa navedenim (eksplicitno iznijetim i na Eurocodes Workshopu, Brisel, februara 2008.) bi e više rije i na drugom mjestu.

Ergo, pri nazna enom stanju stvari a smatraju i to više izazovnim i prigodnim priliku održavanja tako zna ajnog skupa kao što je BE40CE, te obillježavanja teku e 30-te godišnjice razornog Crnogorskog zemljotresa od 1979., osnovna intencija i profesionalno-eti ka posve enost ovog autora okrenuta je - uslovno govore i, prema nastojanju rehabili-tacije ciljeva nacionalne strategije za smanjenje seizmi kog rizika proizašle nakon velikog Crnogorskog zemljotresa (korespodentno oja ane aktuelnim napretkom u projektima, studijama i istraživanjima realizovanim u me uvremenu), a ije se polazne osnove i koncept izlažu u nastavku rada.

Uz sve prethodno navedeno, ini se vrlo o iglednim da je pri datoj situaciji neophodan urgentni korektivni zaokret na nivou države, ostvariv jedino hitnim donošenjem posebnog zakona o sveobuhvatnoj zaštiti od zemljotresa - Lex Specialis (ve ako ne nadomještaj za ve u injene hendikepe - a ono barem za nužno i interventno predupre enje za njihove potencijalno dalekosežne posledice). Nime, time bi se u suštini i jedino efektivno moglo posti i nadilaženje postoje eg stanja i takore i otvorenog “brisanog prostora” za seizmi kirizik, naro ito kod primorskog i najviše ugroženog regiona, proizvoljno prepuštenog neprihvatljivom i sve ve em seizmi kom riziku. Najzad, zapravo bi se iskazala odgovorna javna svijest o trajno prisutnoj visokoj seizmi koj ugroženosti zemlje i na jedini cjelishodan na in obezbijedilo legislativno utemeljenje adekvatne nacionalne strategije za efekasno integralno upravljanje seizmi kim rizikom tj. identifikovali, koordinirali i harmonizovali raznorodni sektorski pristupi i procedure3.68 Time bi se kreirali neophodni uslovi za zdrav i dugotrajno ordživ razvoj zemlje.

3 Novousvojeni Zakona o zaštiti i spašavanju (2007) predstavlja komplement nacionalnoj strategiji upravljana seizmi kim rizikom kao njen dodatni ali ne i sastavni dio. Naime, kada je u pitanju upravljanje seizmi kim rizik om (izuzimaju i odre ene aspekte pripremljenosti) njegov domašaj treba smatrati ograni enim na upravljanje katastrofom.

284

2 POLAZNE OSNOVE I KONCEPTI PLANIRANJA

2.1 OSVRT NA NEKE EFEKTE I ISKUSTVA IZ CRNOGORSKOG ZEMLJOTRESA OD 1979. GODINE

Kao što je poznato, katastrofalni Crnogorski zemljotres od 15. aprila 1979.godine bio je najja i zemljoters koji se ikada dogodio u Evropi. Imao je intenzitet I=IXo/Xo stepeni MCS skale i magnitudu M = 7 (prema Richter-u) - i bio pra en velikim brojem serija jakih naknadnih udara, izme u njih je bio najsnažniji onaj što se desio 24. maja i imao intenzitet VIIIo MCS skale, odnosno magnitudu M = 6,10 (prema Richter-u).

Ovaj zemljotres je prouzrokovao ogromne štete na cijelom Crnogorskom primorju i velikom podru ju jednog broja opština kontinentalnog dijela Crne Gore (Slika 1). Ina e,kao što je dobro poznato, zemljotres je zahvatio teritoriju veli ine od preko 50.000 km2 bivše Jugoslavije, uklju ivo i Dubrova ku regiju u Hrvatskoj. Tako e, u isto vrijeme pogodio je i podru je Skadra i Leša u Albaniji.

Što se ti e posljedica u odnosu na Republiku Crnu Goru sa aspekta ukupnih gubitaka i šteta, izme u ostalog, od naro itog zna aja je ukazati na sljede e:

- da je, sre nim sticajem okolnosti, u ovom zemljotresu izgubljen samo 101 ljudski život u Crnoj Gori i 35 u Albaniji,

- da je, me utim, preko 100 hiljada ljudi bilo ostalo bez krova nad glavom, - da je ukupni obim šteta, kako direktnih tako i indirektnih, iznosio ne manje od 4,

5 milijardi tadašnjih USA dolara (uz konverziju danas vrijednih mnogostruko više), što je inilo oko 4 godišnja bruto nacionalna dohotka Crne Gore za 1979.godinu, odnosno

približno 10% ukupnog bruto nacionalnog dohotka tadašnje SFR Jugoslavije. Pri nazna enom, ini se krajnje primjerenim podsjetiti da je prevazilaženje tako

dramati nih posljedica bilo mogu e ostvariti prevashodno uz solidarnu pomo bivših jugoslovenskih republika – ispoljenoj kako u neposrednoj i urgentnoj post-zemljotresnoj fazi tako i u kasnijoj fazi obnove i izgradnje postradalog podru ja.

U prethodno nazna enom kontekstu može se sagledati zna aj i uloga okvirnog Programa neposrednih i daljih postzemljotresnih aktivnosti na prevazilaženju efekata ovog zemljotresa, spontano pokrenutog i pripremljenog od strane Republi kog zavoda za urbanizam i projektovanje (RZUP), za vladine i druge nadležne organe SR Crne Gore. Ina e, kako je to ve šire poznato, ovim programom bile su trasirane osnovne post-zemljotresne mjere i aktivnosti, kao što su bili identifikovani i glavni faktori i segmenti opšte seizmi ke sigurnosti, na kojima se imaju temeljiti kako sve pomenute mjere i aktivnosti na sanaciji i obnovi postradalog podru ja tako i svi planovi rekonstrukcije i daljeg fizi kog razvoja u Crnoj Gori.

Svakako da prva globalna aktivnost koju iz toga programa treba izdvojiti - jeste Projekat utvr ivanja i klasifikacije ošte enja objekata sa ocjenom njihove upotrebljivosti.Projekat je izveden prema metodologiji pripremljenoj od strane RZUP-a Titograd, u saradnji sa IZIIS-om, Skopje. Ovaj projekat, kroz iju realizaciju je obuhva ena inspekcija i pregled preko 64.000 objekata (razli ite namjene, vrste materijala i tipova konstrukcije), predstavlja jedinstven poduhvat u svjetskim razmjerama. Ina e, cio poduhvat je sproveden uz angažovanje preko 660 inženjera, arhitekata i tehni ara iz itave ondašnje Jugoslavije -

285

uz vo enje i organizaciju cijele operacije od strane RZUP-a, u tijesnoj koordinaciji sa nadležnim državnim organima iz pojedinih republika odnosno opštinama sa postradalog podru ja.

Slika 1- Diskretna raspodjela ošte enja objekata u Crnogorskom zemljotresu 1979, (predstavljena po opštinama)

Veoma zna ajnu podršku pružila je i me unarodna zajednica, iz koje se za ovu priliku posebno ukazuje na tehni ku pomo Ujedinjenih nacija, realizovanu preko njenog Programa za razvoj (UNDP) odnosno njenih korespodentnih specijalizovanih agencija. U navedenom kontekstu i naro ito za ovu priliku treba apostrofirati osobito onu tehni ku pomo UNDP-a koja je bila orijentisana ne samo ka rehabilitaciji pogo enog podru ja nego i na savremeni - i po mnogo emu pionirski pristup integralnom planiranju dugoro nog prostornog i urbanisti kog razvoja Republike, ije su vode e komponente bile smanjenje seizmi kog rizika i zaštita okoline (UNDP/UNCHS Project YUG/79/104: Prostorni plan Republike i generalni urbanisti ki planovi opštinskih centara Crne Gore). S tim u vezi, svakako, treba ista i da su produkti iz ovog poduhvata (odnosno projekata pokrenutih za njegovu podršku - kako na nacionalnom tako i na regionalnom nivou) u me unarodnim okvirima tj. kroz sistem UN bili naišli na široku afirmaciju, te preuzimani kao obrazac planiranja i zaštite održivog razvoja u podru jima izloženim zemljotresu posebno u regionu Mediterana.

U prethodno navedenom smislu, izme u ostalog, ini se veoma ilustrativnim izdvojiti programe regionalne saradnje (ostavrene kroz realizaciju pomenutog kao i drugih regionalnih UNDP projekata), apostrofiraju i Stalni koordinacioni komitet za smanjenje seizmi kog rizika Balkanskih zemalja (Balkan PCC); tako e i Kooperativni program za smanjenje seizmi kog rizika u regionu Mediterana (SEISMED), i dr.

Sa pravom se može ustvrditi da su iskustva ste ena nakon ovog zemljotresa svojevremeno prerasla i bila sublimirana u jedan svojstven, suštinski autenti an i savremen koncept sistema integralnog planiranja održivog razvoja - uz potenciranje i njegove zaštite upravo sa aspekata smanjenja seizmi kog rizika. Time su, može se slobodno ustvrditi, bile istovremeno uspostavljene temeljne i programske osnove (u najve oj mjeri i danas aktu-

286

elne) za izgradnju savremene i cjelovite strategije upravljanja seizmi kim rizikom ne samo u Crnoj Gori nego i šire.

2.2 NEKI VAŽNIJI POSTZEMLJOTRESNI PROJEKTI I ISTRAŽIVANJA SA CILJEM SMANJENJA SEIZMI KOGRIZIKA

2.2.1. Me unarodni postzemljotresni projekti Nacionalnog i regionalnog karaktera

U datom kontekstu – uz ve apostrofirani UNDP Project YUG/79/104 (Prostorni plan Republike i generalni urbanisti ki planovi Crne Gore), valja podsjetiti i na ostale projekte iz okvira tehni ke pomo i Ujedinjenih nacija realizovane na regionalnoj osnovi i uz u eš e cijelog niza specijalizovanih agencija iz sistema UN-a koji su primarno služili kao naposredna podrška prethodno navedenom projektu YUG/79/104. Ali, istovremeno bili orijentisani i na diseminaciju i promociju njegovih rezultata odnosno iskustava na cio region Balkana, a kasnije i Mediterana. U tom smislu, dakle, posebno se apostrofiraju sljede i projekti me unarodnog karaktera.

- Prostorni plan Republike i generalni urbanisti ki planovi Crne Gore (UNDP/UNCHS/UNDRO Project YUG/79/104);

- Me unarodni konsultativni odbor za pitanja obnove i rekonstrukcije podru jaCrne Gore postradalog u zemljotresu od 1979. (UNDP/UNCHS/UNDRO Project YUG/79/003);

- Smanjenje seizmi kog rizika u regionu Balkana (UNDP/UNESCO Project RER/79/014);

- Izgradnja objekata pod seizmi kim uslovima u regionu Balkana (UNDP/UNIDO Project RER/79/015).

- Prostorno-urbanisti ko planiranje u zemljotresnim zonama Mediterana (UNEP/MAP Program PAP/RAC/83/6: Land-Use Planning in Mediterranean Earthquake Zones”) Sintezni produkt ovog projekta sabran je u njegovom finalnom izvještaju “Seismic Risk Reduction in the Mediterranean Region”, MAP Technical Reports Series No 17, UNEP/PAP/RAC, Athens/Split, 1987. Kasnije, ovaj projekat je prerastao u tzv. Projekat SEISMED: “Kooperativni program za smanjenje seizmi kog rizika u regionu Mediterana”, sa sjedištem u enovi, Italija. Ina e, participacija Crne Gore, SFR Jugoslavija, u ovom projektu, odnosno programu, prekinuta je sa nastupanjem njenog raspada, da bi – ipak, zadnjih godina u eš e Crne Gore bilo nastavljeno u izvjesnim domenima, preko JU “Morsko dobro”.

2.2.2 Svojstvena iskustva sa nivoa prošlog PPR Crne Gore i GUP-ova (UNDP Project YUG 79/104) i doma a istraživanja sa ciljem smanjenja seizmi kog rizika

U predmetnom kontekstu odnosno neposrednoj vezi sa iznijetim, u kontekstu širih jugoslovenskih prilika, smatra se posebno primjerenim i cjelishodnim ovom prilikom posebno podsjetiti i ukazati na istrajna i dugogodišnja nastojanja koja su u nazna enom smislu tokom IDNDR 1996.-2000. (Me unarodna decenija za smanjenje prirodnih nepogoda), odnosno tokom minulog perioda, inili Gra evinski fakultet Univerziteta Crne Gore (i sam osnovan nakon ovog zemljotersa, 1980.) i Republi ki seizmološki zavod, zajedno sa JUZI-em, kao sljedbenikom Jugoslovenskog društva za zemljotresno

287

inženjerstvo osnovanog nakon Skopskog zemljotresa 1963. (ina e prvom / najstarijom asocijacijom te vrste u Evropi).

a) Metodološki i razvojno planski aspekti Kao što je to ve pomenuto, polaze i od poznatih i naprijed navedenih okolnosti,

programsko-metodološki pristup u tretiranju komponente smanjenja seizmi kog rizika na nivou cjelovitog projekta UNDP-YUG/79/104 (Prostorni plan Republike i generalni urbanisti ki planovi opštinskih centara), bio je formiran tokom izrade prve faze Prostornog plana Republike, nazvane Osnove Plana. Metodološko-planerski pristup, dat u vidu zaokruženog koncepta ugra enog i kao posebno poglavlje “Seizmi ki hazard i kontrola seizmi kog rizika”, bio je konsekvetno verifikovan i koriš en u daljim fazama realizacije ovog Projekta u cjelini, tj. kako pri izradi Prostornog plana Republike, tako i svih generalnih urbanisti kih planova.

Pri tome, može se re i da je tamo izloženi pristup i koncept djelotvorno ispunio svoje obje vode e funkcije: koordiniraju u i usmjeravaju u - kako u odnosu na sam Prostorni plan Republike, tako i u odnosu na sve ostale prostorne i urbanisti ke planove nižeg reda, a posebno generalne urbanisti ke planove opštinskih centara.

Ina e, izrada Prostornog plana Republike (PPR) bila je zasnovana na prethodnoj izradi preko 30 baznih studija koje su pokrivale odre ene posebne sektore razvoja. Izme uostalih, ovim programom bila je obuhva ena i posebna Studija vulnerabiliteta za potrebe PPR (IZIIS, Skopje / RZUP, Titograd).

b)Istraživanje i analiza dogo enog odnosno o ekivanog seizmi kog vulnerabiliteta Ve pomenuta i izuzetno vrijedna dokumentaciona osnova dobijena iz programa

inspekcije, utvr ivanja i klasifikacije pregledanih objekata pored preliminarnih ciljeva: utvr ivanje upotrebljivosti ošte enih objekata i procjene prouzrokovanih ekonomskih šteta - neminovno je poprimila i proširenu namjenu nau no-istraživa kog karaktera.

Prethodno nazna ena istraživanja izvršena nakon zemljotresa od 1979. (kaogod i nivo šteta koje je prouzrokovao), odnosno seizmi ka aktivnost i u estalost pojave zemljotresa visokog intenziteta koja se može o ekivati na cijeloj teritoriji Republike, a posebno u njenom primorskom regionu, evidentno i rigorozno potvr uju injenicu da eljudi i njihova imovina , ako i sva društvena dobra - biti permanentno izloženi dejstvu velikog broja manjih i srednje jakih zemljotresa te sa velikom vjerovatno om i dejstvu razornih zemljotresa velike magnitude, sli ne magnitude katastrofalnog zemljotresa od 15. aprila 1979.

c) Neki specifi ni uslovi i ograni enjaUspostavljanje, obezbje enje i sprovo enje šire društvene politike blagovremene

prevencije kroz odgovaraju e planiranje (uklju ivo nazna ene i druge relevantne aspekte prostornog i urbanisti kog planiranja, izgradnju investicionih objekata i realizaciju drugih razvojnih projekata), svakako da podrazumijeva potrebu komplementarnog pra enja, podrške i osiguranja adekvatnom zakonodavnom, urbanisti kom, tehni kom i drugom regulativom (uklju ivo relevantne norme i standarde), pogotovu u domenu i u vezi sa koriš enjem i politikom koriš enja zemljišta. U tom smislu, i na koherentnoj osnovi, bile su date i odgovaraju e smjernice za sprovo enje Prostornog plana Republike.

Dakle, strategija zaštite od zemljotresa tretirana je kao dio strategije razvojnih ciljeva. Kako svaki plan prostornog razvoja, pa i Prostorni plan Republike, definiše obrazac ("Plan struktura") prostorne distribucije razvojnih programa i slike o ekivanog stanja ovjekove sredine u nekom datom vremenskom periodu (u konkretnom slu aju do 2000.

288

odnosno 2025. godine), tako i proces njegovog uspješnog sprovo enja mora biti zasnovan na strategijskom konceptu potrebnih prostorno-ekonomskih razvojnih operacija. Ovo, uz definisanje vode ih faktora toga razvoja i njihove me uzavisnosti.

Otuda, kod izrade Prostornog plana Republike - nametao se itav niz posebnih zadataka i ograni enja, kako je dato u Odjeljku 2.3.2 ovog rada .U vezi sa prethodnim stoji, može se re i, najvažnija strategijska obaveza prostornog planiranja - da definiše podru je tih konflikata, karakter samih koflikata, kao i da formuliše razli ite mogu nosti razvojnih obrazaca i kriterija za njihovo prevazilaženje i razrješavanje.

2.3 KONCEPT KONTROLE I SMANJENJA SEIZMI KOG RIZIKA KROZ PROSTORNO I URBANISTI KO PLANIRANJE

2.3.1. Opšti osvrt i premise

Sve prethodno navedene injenice i podaci konsekventno potvr uju i od ranije poznata saznanja da region Balkana predstavlja seizmi ki najaktivnije i najugroženije podru je Evrope, kao i injenicu da u tom okviru teritorija Crne Gore i neposrednog okruženja ima eksponiranu poziciju sa izrazito visokim nivoom hazarda.

Pri svemu, ne može a da se ne ozna i paradoksalnom injenica da, i pored toga što je inicijativa da UN proglase IDNDR (pokrenuta na VIII Svjetskoj konferenciji za zemljotresno inženjerstvo, San Francisko, 1984.) bila dobrim dijelom podstaknuta i iskustvima iz Crnogorskog zemljotresa od 1979.godine – niti na nivou same Crne Gore niti na nivou Jugoslavije – nijesu preduzimane nikakve mjere i akcije predvi ene okvirom me unarodnih aktivnosti tokom ove Decenije, proglašene Rezolucijom Generalne Skupštine OUN br. 44/236.

Me utim, s obzirom na okolnosti i prirodu stvari, ini se svrsishodnim ukazati na i danas aktuelne glavne faktore i aspekte seizmi ke sigurnosti i njihove komponente u kontekstu iskustava ste enih nakon zemljotresa 1979., tj. kao što slijedi:

- Formulisanje i razvoj opšte politike smanjenja seizmi kog rizika, zasnovane na integralnom pristupu (uklju ivo razvoj institucionalnih sposobnosti zemlje i državnih programa za predvi anje, upozoravanje, prevenciju i ublažavanje posljedica, sa posebnim naglaskom na urbane sredine);

- Uspostavljanje i razvijanje multidisciplinarnog pristupa i sveobuhvatnog koncepta upravljanja seizmi kim rizikom (uklju ivo sicio-ekonomske aspekte, aseizmi ko projektovanje objekata, prostorno planiranje,urbanisti ko planiranje i projektovanje, nau no istraživa ki rad u oblasti zemljotresnog inženjerstva), i td.;

- Revizija postoje ih i izrada novih prostornih i urbanisti kih planova, uz razvijanje svih relevatnih aspekata smanjenja seizmi kog rizika kao integralnog dijela ovih planova;

- Identifikacija elemenata seizmi kog rizika, istraživanje i utvr ivanje vulnerabiliteta ovih elemenata (apostrofiraju i vulnerabilitet postoje ih zgrada i drugih izgra enih struktura, definisanje prihvatljivog nivoa seizmi kog rizika, kao i obezbje nje potrebne seizmi ke sigurnosti kod postoje ih objekata);

- Identifikacija ekonomskih dobiti od mjera i akcija ublažavanja posljedica seizmi kog hazarda, zasnovana na identifikaciji post-zemljotresnih ekonomskih posljedica (uklju ivo štete po osnovu izgubljenih života i povrije enih, koštanje ošte enja, gubitak proizvodnje i tržišta, troškove odnosno zahtjeve po osnovu osiguranja, kao i koštanje izgubljene dinamike razvoja);

289

- Uspostavljanje sistema sveobuhvatne pripremljenosti na zemljtores, uz njegovo permanentno unapre ivanje i ja anje (uklju ivo kontigentno planiranje i planove za vanredno okolnosti, obebje enje uslova za preživljavanje stanovništva neposredno nakon katastrofe, programe i uslove za obnovu i izgradnju poslije zemljotresa, i dr.);

- Razvijanje kolektivne svijesti u društvu, u odnosu na seizmi ki rizik (uklju ivoobrazovanje u cilju ublažavanja posljedica, informisanje javnosti, obuku za ponašanje u slu aju katastrofe, i dr.).

U takvom kontekstu treba snažno naglasiti naro it zna aj koji se mora pridavati potrebi adekvatne institucionalne i kadrovske izgra enosti i obaviještenosti, odnosno kompetentnosti raznih subjekata involviranih o svim relevantnim podru jima.

Pri svemu, iako se to podrazumijeva, valja ukazati da neophodan nivo i obim odgovaraju e obaviještenosti i kompetentnosti nazna enih subjekata - koji, u principu, obuhvata kako opšti koncept i nivoe odgovornosti, tako i sistem specifi nih mjera i procedura korespodentan pripadaju em – po pravilu, specijalisti kom podru ju.

2.3.2. Specifi ne mjere na nivou Prostornog plana Republike

Prostorno i urbanisti ko planiranje raspolaže (po prirodi stvari, a i po zakonskoj definiciji) svojstvom, snagom i mogu nostima da usmjerava sve razvojne projekte i druge akcije u odnosu prema prirodnoj sredini – i to na takve na ine kod kojih e se ili izbje isam hazard ili kod kojih e se primijeniti odgovaraju e ekonomski opravdane mjere zaštite, i/ili koji e usmjeravati na izbore manje povredljivih prostornih razvojnih šema sa nižim i razli itim nivoima hazarda, odnosno seizmi kog rizika.

U suštini, jedino je tako i mogu e na obuhvatan na in formulisati strategiju zaštite na seizmi ki i druge prirodne hazarde, a koja bi uz odgovaraju u ocjenu prihvatljivih troškova za razli ite preventivne mjere postala sastavni dio racionalnog planiranja u cjelini.

Uspostavljanje, obezbje enje i sprovo enje šire društvene politike blagovremene prevencije kroz odgovaraju e planiranje (uklju ivo nazna ene i druge relevantne aspekte prostornog i urbanisti kog planiranja, izgradnju investicionih objekata i realizaciju drugih razvojnih projekata), svakako da podrazumijeva potrebu komplementarnog pra enja, podrške i osiguranja adekvatnom zakonodavnom, urbanisti kom, tehni kom i drugom regulativom, (uklju ivo relevantne norme i standarde), pogotovu u domenu i u vezi sa koriš enjem i politikom koriš enja zemljišta. Dakle, strategija zaštite od zemljotresa mora biti tertirana kao dio strategije razvojnih ciljeva.

Kod izrade novog Prostornog plana Republike – propra enog odgovaraju im strategijskim konceptom njegove realizacije, kako je to bilo i kod izrade prethodnog PPR-a, pored ostalog stajao je i bio sugerisan itav niz posebnih zadataka i dilema, kao što su:

- Dalje istraživanje hazarda, odnosno veli ine i karaktera kao i drugih parametara zemljotresnog dejstva neophodnih pri detaljnom urbanisti kom planiranju i projektovanju,

- Ograni avanje tehni kih ekonomskih i društvenih posljedica katastrofe shodno o ekivanom vulnerabilitetu zgrada, objekata vitalne infrastrukture i drugih društvenih dobara odnosno izgra ene sredine,

- Definisanje društveno i ekonomski usaglašenih ograni enja i troškova kao kriterijuma zaštitnih mjera i prihvatljivog nivoa seizmi kog rizika,

- Uspostavljanje i razmatranje sistema i mahanizma pripremljenosti na zemljotres, itd. Isto tako, otvarao se i odre en broj specifi nih pitanja i konflikata, relativno trajnog

zna enja, me u kojima su najvažniji:

290

- Konflikt izme u nivoa hazarda i atraktivnosti izgradnje i razvoja na odre enom podru ju,

- Konflikt izme u ekonomije obima i decentriralizovanog modela prostornog razvoja; - Konflikt izme u ekonomskog koštanja i cjelishodnosti u odnosu na druge društvene i

politi ke kriterije, - Konflikt izme u trenutnih i dugoro nih pogodnosti i koristi, - Konflikti izme u vlasni kih, grupnosvojinskih i društvenih interesa.

U vezi sa tim stoji, može se re i, najvažnija strategijska obaveza prostornog i urbanisti kog planiranja da definiše podu je tih konflikata, karakter samih konflikata kao i da formuliše razli ite mogu nosti razvojnih obrazaca i kriterija za njihovo prevazilaženje.

Ina e, u ovom razmatranju, samo su nazna eni neki osnovni elementi pristupa i ilustracija problema – podrazumijevaju i nastojanje da njihov odgovaraju i tertman bude sproveden, shodno stavovima Studijske osnove odnosno i Finalnog izvještaja Projekta SS-AE69, kako kod izrade novog Prostornog plana Republike, tako i kroz reviziju i izradu generalnih urbanisti kih planova pojedinih opština, prevashodno.

2.3.3. Glavni principi i elementi planiranja

a) Opšte. Planske mjere za mitigaciju seizmi kog rizika po prirodi i opštem konceptu,u suštini su komplementarne ili se ak poklapaju sa opšte prihva enim pravilima regularnog oblikovanja ovjekove okoline. Naime, ve ina postoje ih dosljedno primijenjenih standarda i regulative iz planiranja i projektovanja djeluje istovremeno i u prilog kontrole vulnerabiliteta i mitigacije rizika, dok isto tako ve ina planskih mjera na kontroli hazarda i vulnerabiliteta podupire najsavremenije pristupe i ideje na oblikovanju okoline i formiranju funkcionalne strukture naselja i gradova.

Dakle, ne postoji opšte opre nosti izme u zahtjeva da se istovremeno planira i dobar i siguran grad, iako uvijek preostaje mnoštvo posebnih konfliktnih situacija koje se moraju razrješavati i prevazilaziti.

U svakom slu aju može se opravdano rezimirati da e biti prihvatljiv samo takav sistem prostornog i urbanisti kog planiranja, koji e na sveobuhvatan na in, kombinovati i razvoj i mitigaciju rizika i to kroz bilo koji nivo i/ili obrazac tzv. planskih struktura.

Pri tome je razumljivo da planske mjere za mitigaciju rizika do izvjesnog stepena variraju u zavisnosti od prirode i nivoa seizmi kog rizika, zatim, od nivoa prostornog i urbanisti kog planiranja, a isto tako je i njihova prakti na primjenljivost u odre enoj mjeri zavisna od društveno-ekonomskih i drugih uslova koji se odnose na dato podru je odnosno zemlju u cjelini.

b) Mjere za prilago avanje hazardu. Glavni planerski koraci odnose se, po pravilu, na manipulisanje izme u zona sa razli itim nivoima hazarda i razvojnih programa sa razli itim nivoima osjetljivosti. Obadvije grupe ovih nivoa trebalo bi da za svako podru je stoje u inverznom odnosu.

U ovom pogledu, odnosna strategija na nivou Prostornog plana Republike nalazi svog neposrednog odraza (kao rezultat istraživanja seizmi kog hazarda i njegovog sukobljavanja sa postoje im razvojnim obrascima, trendovima i ciljevima), kroz diferenciranje na:

- podru ja koja se mogu smatrati kontraverznim, gdje su razvojni trendovi u konfliktu sa prirodom i nivoom hazarda,

69 vidjeti Odjeljak 2.4

291

- podru ja ograni enim ili zatvorenim za odre ene vrste razvoja i - podu ja gdje dati razvojni programi imaju prioritet i gdje e biti primijenjene sve

potrebne mjere za mitigaciju rizika (zone sa izraženim prednostima, potencijali budu egrazvoja, prioriteti kroz obnovu i rekonstrukciju postoje ih naselja, neki hidrološki i drugi projekti i sl.).

Pri svemu tome treba ista i da se na nivou generalnih urbanisti kih planova postoji šira mogu nost, ali i ve a odgovornost za ostvarenu interpretaciju zoniranja hazarda kako u svrhu definisanja namjene zemljišta, tako i za funkcionalno tertiranje naselja.

S tim u vezi, ve najjednostavnija klasifikacija zona prema seizmi kom hazardu (relativno): na najbolje, srednje i najlošije, pruža planerima odgovaraju e smjernice na koja podru ja bi se moglo locirati najvažnije i najosjetljivije razvojne programe i koja bi podru ja mogla biti zatvorena za razvoj. U tom pogledu postoji ve dovoljno planerskog i prakti nog iskustva, (uklju ivo uslove zahtjeva za potrebe tzv. “pripremljenosti na zemljotres”), o tome kakvi se obrasci me uzavisnosti izme u funkcionalnog i zoniranja hazarda trebaju primjenjivati, i dr.

c) Distribucija kao sredstvo kontrole vulnerabiliteta. Nasuprot hazardu, iji nivo zavisi od prirodnih uslova i sila, nivo vulnerabiliteta (povrjedljivosti) ljudskih naselja zavisi prevashodno od rezultata ovjekove aktivnosti, od primijenjenih rješenja kroz obrasce prostornog i urbanisti kog planiranja, aseizmi kog projektovanja odnosno seizmi kesigurnosti objekata, primijenjenih materijala i tehnologije, funkcionalnih programa i – najzad, ali ne i najmanje važno, od primijenjenih mjera preventivne zaštite i mitigacije.

U vezi sa ovim može se re i da su koncentracija i gustina dva klju na razvojna elementa koja se definišu na svakom nivou urbanisti kog planiranja, predstavljaju i bitne faktore njihove ekonomske implikacije. U podru jima podložnim jakim zemljotersima (što i jeste slu aj sa ve im dijelom teritorije Crne Gore), ova dva aspekta razvoja, po pravilu, direktno uslovljavaju kako veli inu same katastrofe tako i njene dalje posljedice.

Kako su obadva elementa, i koncetracija i gustina, izvedeni iz obrasca distribucije – to su planski koncepti prostorne distribucije stanovništva, zgrada i drugih društvenih dobara i aktivnosti upravo od direktnog uticaja na nivo seizmi kog rizika.

d) Glavne mjere kontrole na nivou generalnih urbanisti kih planova. Aspekt koncentracije i gustine, kao što je ve napomenuto, na lokalnom nivou planiranja izražavaju se kroz opredjeljivanje namjene površina i funkcionalni zoning. Ovaj zoning, posebno za urbana naselja, fiksira specifi ne funkcije za svaku oblast (kao što stanovanje, školstvo, trgovina, industrija, zdravstvo, rekreacija, itd.), i to u okvirima izvršenog mikrorejoniranja na seizmi ki hazard. Pored predvi enih i propisanih funkcija za svaku oblast ovaj, zoning treba tako e da definiše intenzitet koriš enja prema svakom izdvojenom elementu funkcije urbanog zemljišta (dozvoljena gustina, odnos izgra enog dijela prema ukupnoj površini podru ja, fiksiranje minimalnog iznosa otvorenih površina u okviru svake lokacije, dozvoljena visina zgrada, karakter, i vrsta konstrukcija otpornih na zemljotres, vrste materijala i dr.).

U svakom slu aju, Generalni urbanisti ki planovi naseljenog mjesta predstavljaju zakonske dokumente sa punom i obavezuju om snagom primjene za svakog od subjekata planiranja razvoja na tom podru ju.

e) Kontrola vulnerabiliteta kroz detaljne urbanisti ke planove i urbanisti keprojekte. Sa stanovišta mjera kontrole vulnerabiliteta detaljni urbanisti ki planovi

292

predstavljaju najvažniju sponu izme u prostornog/urbanisti kog planiranja i odgovaraju ih projektantskih mjera. Ovi planovi – uklju ivo urbanisti ko tehni ke uslove koji iz nih proizilaze, u suštini prevode namjenu površina i funkcionalni zoning iz generalnih urbanisti kih planova u realan urbanizovani prostor.

S tim u vezi sve elemente ovog nivoa planiranja treba smatrati jednako važnim sa stanovišta njihove osjetljivosti i vulnerabiliteta. Ovo se isti e stoga što iskustvo pokazuje da su planeri odnosno projektanti, do sada, glavnu ako ne isklju ivu pažnju posve ivali otpornosti i sigurnosti pojedina nih zgrada i objekata.

Me utim, treba imati u vidu zna aj svakog od elemenata u kontekstu njihove ukupne funkcionalne korelacije, a posebno u uslovima nakon zemljotresa, i to polaze i od tzv. urbanisti kog interijera (ulice, skverovi, pješa ke staze i drugo), pojedinih specifi nih gradskih sadržaja, do problematike vezane za pravilno aseizmi ko projektovanje objekata i tretiranje njegovih osnovnih principa, i to naro ito u pogledu zasnivanja poželjnih i primjerenih konstruktivnih oblika.

Sasvim posebna situacija u zaštiti od posljedica zemljotersa nastaje u odnosu na seizmi ku sigurnost ve izgra ene sredine tj. postoje e objekte, kulturno-istorijske spomenike kao i stara kulturno-istorijska gradska jezgra i stare gradove, gdje se mora sou iti sa dosta ograni enim mogu nostima intervencije u jednu ve vrsto zasnovanu i postoje u fizi ku strukturu.

2.4. Osvrt na tretman i doma a istraživanja vulnerabiliteta i prihvatljivog seizmi kogrizika u okviru Projekta UNDP-YUG/79/104, kao i Projekta GTZ-UCG&RZUP, 2006

Kao što je evidentno, iskustva iz Crnogorskog zemljotresa 1979. donijela su mnoge po svemu specifi ne i izuzetno zna ajne rezultate, izražene posebno kroz nove multidisciplirane pristupe – kako u mitigaciji seizmi kog rizika, tako i u postzemljotresnom upravljanju. Pri tome, polaze i ak i od same definicije i klasifikacije zna enja osnovnih pojmova (odnosno tehni kih termina) za Hazard, Rizik, Vulnerabilitet, i dr, kao i njihovog uzajamnog odnosa (R=H xV).Uklju ivo temeljna istraživanja i adekvatne metodologije za utvr ivanje vulnerabiliteta – i to kako zgrada tako i vitalne infrastrukture, odnosno korespodentnog rizika.

Apstrahuju i ostale veoma zna ajne medjunarodne/regionalne i doma e/savezne projekte i studije realizovane u postzemljotresnom periodu, za ovu priliku se posebno izdvaja bazi na Studija povredljivosti i prihvatljivog seizmi kog rizika za potrebe PPR Crne Gore (Projekat UNDP-YUG/79/104). Ina e, ova obimna Studija bila je realizovana sa 1984, od strane IZIIS/Skoplje i RZUP-a Titograd (uz u eš e i ovog autora). Programski ciljevi ove Studije direktno su korespondirali kako sa ciljevima i metodologijom tadašnje izrade PPR i GUP-ova Crne Gore, tako i sa ciljevima ranije pomenutog regionalnog UNDP/UNESCO Projekta RER/79/014 (Smanjenje seizmi kog rizika u regionu Balkana).

Kao što je to ve iskazano, polaznu osnovu za ovo istraživanje predstavljala je ranije spomenuta inspekcija i pregled 64 000 ošte enih objekata, uz njihovu klasifikaciju prema dogodjenom odnosno opaženom vulnerabilitetu. Pri tome je analizirano ukupno 40 000 objekata (sa težištem na primorske Opštine i Cetinje) uz njihovu diferencijaciju prema namjeni, tipu konstrukcije, spratnosti, materijalu, tipu temelja, i uslovima temeljnog tla. Ova obimna Studija (od šest tomova), sažeta u Završni elaborat (iz dvije knjige) – bila je prilagodjena potrebama za prakti no koriš enje njenih rezultata od strane direktnih korisnika, u prvom redu nosilaca izrade i sprovodjenja GUP-ova i drugih urbanisti kih

293

planova i projekata. Naime, opšte prihva eno je da prostorni i urbanisti ki planovi – naro ito generalni, GUP-ovi (a takodje i detaljni, DUP-ovi) po svojoj statutornoj prirodi i zakonskoj snazi, u suštini predstavljaju polaznu i klju nu osnovu za smanjenje seizmi kog rizika a time i za izgradnju i sprovo enje odgovaraju e efektivne strategije za upravljanje seizmi kim rizikom – i to kako sa nivoa Republike tako i na lokalnom nivou.

Ne upuštaju i se u analizu efektivnog koriš enja i poštovanja rezultata ove Studije tokom perioda važenja i sprovodjenja prethodnog PPR-a (posebno kroz nare eni»tranzicioni period«) mora se ozna iti krajnje indikativnom koincidencija nekih vrlo relevantnih okolnosti. Sa jedne strane, da prilikom izrade novousvojenog PPR-a ista nije bila uopšte tretirana od strane obra iva a Plana, niti dostupna nosiocu i obradjiva uprojekta tzv. Studijske osnove SS-AE, (Projekat GTZ/UCG&RZUP, 2006., realizovanom za potrebe upravo njegove izrade), kaogod i - sa druge strane, da se kroz upravo donijeti novi Zakon o uredjenju prostora i izgradnji objekata vrši ukidanje GUP-ova ( ime se suštinski ukida svako sistemski potavljeno i realno promišljano urbanisti ko planiranje).

Otuda, u odnosu na opšti kontektst ovog razmatranja, a imaju i u vidu injenicu da GUP-ovi predstavljaju klju ni instrument upravljanja prostorom i za realizaciju osnovnih smjernica sa nivoa PPR-a, kaogod i nosioca sistema urbanisti kog planiranja u cjelini, sasvim je o igledno da se na ovaj na in derogiraju prethodno ostvareni kapitalni rezultati istraživanja o ekivanog vulnerabiliteta i prihvatljivog seizmi kog rizika, posebno kod najugroženijih (primorskih) opština. Štoviše, time se kompromituje svaka mogu nost racionalnog pristupa smanjenju i kontroli tzv. urbanog seizmi kog rizika, uklju ivo i korespodentno integralno i efektivno planiranje adekvatne pripremljenosti na zemljotres, kaogod i postzemljotresnih operacija i aktivnosti na obnovi i rekonstrukciji postradalog podru ja.

Nažalost, to sve i pored smjernica i koncepata datih kroz prethodno navedeni projekat Studijske osnove SS-AE, specifi no kroz tzv. posebnu Sektorsku studiju (SS-AE) 4.12, posve enu upravo seizmi kom riziku kao i riziku od ostalih prirodnih hazarda i akcidenata. Dakle, izme u ostalog, zasnovane u cjelini na pristupu i konceptu „multihazardnog zemljotresnog inžinjerstva i integralnog upravljanja seizmi kim rizikom“.

3. ZAKLJU NI OSVRT Na samom kraju - iako se to ve i po sebi ini evidentnim, vode i motiv ovog rada

mogao bi se izraziti kroz osje aj i uvjerenje autora da 40-ta godišnjica zemljotresa u Banja Luci kao i teku a 30-ta godišnjica Crnogorskog zemljotresa, predstavljaju obavezuju u i neponovljivu priliku za jedan dodatni podsticaj ozbiljnijem i adekvatnom tretiranju zemljotresne opasnosti i aseizmi koj zaštiti održivog razvoja zemlje. I to, zasnovanoj na sveobuhvatnom pristupu - ali i po svim relevantnim integrativnim aspektima, shodno njihovoj prirodi i domašaju.

U prethodno navedenom kontekstu, iskustvo iz Crnogorskog zemljotresa, zasnovano na vanrednim rezultatima citiranih fundamentalnim me unarodnih i nacionalnih projekata (realizovanih u bivšoj SFRJ, kao i regionima balkana i Mediterana) te prete ih istraživanja – nedvosmisleno je dovelo od saznanja o klju noj ulozi prostornog i urbanisti kog planiranja u smanjenju seizmi kog rizika. I to kao pretpostavci sine qua non za racionalno i integralno upravljanje prostorom - a time i seizmi kim rizikom.

Me utim, za puno razumijevanje i ostvarenje te uloge i uspostavljanje zdravih koncepata, uz prepoznavanje relevantnih inilaca i utvr ivanje primjerenih procedura, prvi i

294

osnovni preduslov predstavlja odgovaraju e definisanje statusa i koherentnog sistema prostornog i urbanisti kog planiranja.

Na doma em nivou, kaogod i u okruženju, pri svemu nazna enom na po etkuovog rada, taj status bi morao biti radikaln rehabilitovan. Na širem nivou, a time i za doma e uslove izvjesna skorašnja mogu nost bi se mogla otvoriti kroz izradu Nacionalnog aneksa (NDPs) za Eurokod 8. Ovo bi se najprikladnije moglo posti i, s jedne strane kroz zajedni ki i kooperativni nastup svih zemlja Balkanskog regiona, kao i – s druge strane, ako se ima u vidu upravo zaklju eni Memorandum o razumijevanju Evropske komisije i GRF Davos (Global Risk Forum), s obzirom na njegovu krajnje primjerenu aktivnost i misiju posve enu upravo istim ciljevima i sa sli nim ako ne i istim pristupom.

REFERENCES 1 Pavi evi , B.S. / Aseizmi ko projektovanje i upravljanje zemljotresnim rizikom (

Aseismic Design and Earthquake Risk Management, Univerzitet Crne Gore, Podgorica, “Obod”, 2001, Cetinje. 2 Pavi evi , B.S. / Seizmi ki hazard i kontrola seizmi kog rizika. Prostorni plan SR Crne

Gore, Osnove Plana (UNDP/UNCHS Project YUG/79/104), RZUP-Titograd, UNCHS-Nairobi, UNDRO-Ženeva, 1983, Titograd. 3 Petrovski, J. / B.S. Pavi evi / Methodology on Vulnerability and Seismic Risk Analysis

Applied in the Studies of the Coastal Region of SR Montenegro, Yugoslavia. Yugoslav National Report, WG B and C, UNDP/UNESCO Project, RER/79/014 (Earthquake Risk Reduction in the Balkan Region), IZIIS/Skopje-RZUP/Titograd, 1982, Titograd. 4 Pavi evi , B.S., editor: Seismic Risk Reduction in Mediterranean. Priority Action ”Land

Use Planning in earthquakes Zones”, MAP Technical report No.17(245p.), UNEP/MAP, Athens and PAP/RAC, 1988, Split. 5 Strategy from Yokohama and Plan of actions for Safer World, Final Document, UN-

World Conference on Natural Disaster Reduction, 1994,Yokohama/Japan 6 Hays, W., B. and J. Mohammadioun: Seismic Zonation, Monograph. The IDNDR

Activity in conjunction with XIth ECEE, Paris, Ouest Editions, 1998, Paris. 7 Global Cooperation in Seismic Disaster Mitigation - A Story of WSSI , The WSSI

Board of Directors WSSI (World Seismic Safety Initiative), 12th WCEE, 2000, Aukland. 8 UNEP/UNESCO Project RER/88/004 (Permanent Coordination Comitee for

Earthquake Risk Reduction in the Balkan Region, PCC): Report on the seventh Session of PCC. IZIIS, Skopje, 1992, Skopje. 9 Meguro, K./ Yoshimura, M./ Integrated information system for total Disaster

Management, Institute of Industrial science, University of Tokio, Bulletin of ERS Centre No.37, 2004. 10 Oliveira C.S. / A.Roca / X.Goula / Assesing and managing eartquake risk. Springer,

2006. 11 Pavi evi B. / R.Zejak: Izvještaj o u eš u na Workshop-u Evropske komisije

“EUROCODES: Background and Applications”, Brisel, 18-20 februar 2008. Institut za standardizaciju Crne Gore, Gradjevinski fakultet UCG & CAZI, 2008, Podgorica. 12 Aman J.W. / Integrated Disaster Risk Managament and Disaster Resilience Capacity

Building, Global Risk Forum GRF Davos, IDRC Changdou 2009, July 2009, Changdou, China.

295

Prof Božidar Pavi evi 170, Jadranka Mihaljevi 271

GENERALISANI KONCEPT INTEGRALNOG UPRAVLJANJA ZEMLJOTRESNIM RIZIKOM U SAVREMENIM USLOVIMA

Rezime:

U radu se pledira na rehabilitaciji svojstvenih iskustava na smanjenju seizmi kogrizika ste enih nakon Crnogorskog zemljotresa 1979. S tim u vezi naglašava se potreba razvijanja savremene nacionalne strategije za sveobuhvatno upravljanje seizmi kim rizikom, u skladu sa aktuelnim me unarodnim strategijama i novim iskustvima. Apostrofiraju se njegovi relevantni integrativni aspekati, isti u i klju-nu ulogu prostorno-urbanisti kog planiranja. Poseban osvrt dat je na zna aj etike

u zemljotresnom inženjerstvu i tretiranju seizmi kog rizika. Klju ne rije i: Hazard, seizmi ki rizik, mitigacija, integralno upravljanje, etika.

GENERAL CONCEPT OF INTEGRATED SEISMIC RISK MANAGAMENT

Summary:

The article deals with the rehabilitation of authentic expirience in seismic risk mitigation gained after the 1979 Montenegro earthquake. The ephasize is given on evolvement of comprehensive national seismic risk managament strategy based on current intenernational strategies and new experiences. Importance of basic relevant integrative aspects and the specific importance of land-use planning are explined in more details. Role of ethic aspects in this field is stressed out. Key words: Seismic risk, mitigation, integrated management, vulnerability, ethics.

1 Profesor (Mostovi, Aseizmi ko projektovanje, Aseizmi ko planiranje;šef Katedre za zemljotresno inženjerstvo), u penziji; bivši Gen. direktor RZUP-a, Titograd (u vrijeme i nakon Crnogorskog zemljotresa 1979); ekspert UNEP-a; Koordinator Programa SEISMED&UNEP/MAP Priority Action: "Land-Use Planning in Mediterranean Earthquake zones", PAP/RAC, Split; osniva i bivši predsjednik CDZI-a&YUZI-a, odnosno CAZI-a; Po asni predsjednik CAZI-a; M: EERI, US Forum, EAEE; etc. E-mail: [email protected]

2 Jadranka Mihaljevic, dipl.inž. gra , kosultant za zemljotresno inženjerstvo, Seizmološki zavod, Vlada Crne Gore

296

1. UVOD

1.1 ZNA AJ I ODRAZ ISKUSTAVA IZ CRNOGORSKOG ZEMLJOTRESA 1979. NA DEFINISANJE FILOZOFIJE I UTVR IVANJA SEIZMI KOG RIZIKA

Kao što je evidentno i upu enima dobro poznato, iskustva iz Crnogorskog zemljotresa 1979. donijela su mnoge po svemu specifi ne i izuzetno zna ajne rezultate, izražene posebno kroz nove multidisciplirane pristupe – kako u mitigaciji seizmi kog rizika, tako i u postzemljotresnom upravljanju. Pri tome, polaze i ak i od same definicije i klasifikacije zna enja osnovnih pojmova (odnosno tehni kih termina) za Hazard, Rizik, Vulnerabilitet, i dr, kao i njihovog uzajamnog odnosa (R=H xV)3.72 Uklju ivo temeljna istraživanja i adekvatne metodologije za utvr ivanje vulnerabiliteta – i to kako zgrada tako i vitalne infrastrukture, odnosno korespodentnog seizmi kog rizika.

Apstrahuju i ostale veoma zna ajne medjunarodne/regionalne i doma e/savezne projekte i studije realizovane u postzemljotresnom periodu, za ovu priliku se posebno izdvaja bazi na Studija povredljivosti i prihvatljivog seizmi kog rizika za potrebe PPR Crne Gore (Projekat UNDP-YUG/79/104). Ina e, ova obimna Studija bila je realizovana sa 1984. godine, od strane IZIIS/Skoplje i RZUP Titograd (uz u eš e i ovog autora). Programski ciljevi ove Studije direktno su korespondirali kako sa ciljevima i metodologijom tadašnje izrade PPR i GUP-ova Crne Gore, tako i sa ciljevima ranije pomenutog regionalnog UNDP/UNESCO Projekta RER/79/014 (Smanjenje seizmi kogrizika u regionu Balkana).

Polaznu osnovu za ovo istraživanje predstavljala je ranije spomenuta inspekcija i pregled 64 000 ošte enih objekata, uz njihovu klasifikaciju prema dogodjenom odnosno opaženom vulnerabilitetu. Pri tome je analizirano ukupno 40 000 objekata (sa težištem na primorske Opštine i Cetinje) uz njihovu diferencijaciju prema namjeni, tipu konstrukcije, spratnosti, materijalu, tipu temelja, i uslovima temeljnog tla. Ova obimna Studija (od šest

3 Klasifikacija i zna enje nekih termina u ovoj oblasti upotrijebljeni su prema konceptu i definiciji usvojenim od strane UNDRO-a (Ženeva, jula 1979.) i UNEP (Najrobi, januara 1980.), a prihva enim i kroz UNDP/UNESCO Project RER/79/014 – Projekat Smanjenje seizmi kog rizika u regionu Balkana (Herceg Novi, aprila 1981. i Skoplje, januara1983.).Usvojena formula glasi: R = V x X, (gdje su R – Rizik; V – Vulnerabilitet/povrjedljivost; H – Hazard).

U vezi sa tim tako e su usvojene i sljede e definicije:

Hazard (prirodni) ozna ava vjerovatno u pojave zemljotresa ili nekog drugog potencijalnog rušila kog prirodnog fenomena, a u okviru specifi nog vremenskog perioda i na odre enom prostoru.

Vulnerabilitet ozna ava stepen gubitka nanesenog datom elementu rizika ili skupu takvih elemenata, zbog pojave prirodnog fenomena date magnitude, a izražava se u skali od 0 (bez štete) do 1 (totalni gubitak).

Elementi rizika ozna avaju stanovništvo, zgrade i druge gra evinske objekte, ekonomske aktivnosti, javne servise suprastrukturu, infrastrukturu i druge elemente izložene riziku na datom prostoru.

Specifi ni rizik ozna ava o ekivani stepen gubitka izazvan pojavom odre enog prirodnog fenomena i predstavlja funkciju zavisnu i od prirodnog hazarda i od vulnerabiliteta.

Rizik ozna ava o ekivni broj izgubljenih života, povrije enih osoba, ošte enja i gubitaka na imovini, poreme aja privredne aktivnosti usljed odre enog prirodnog fenomena i sl., te prema tome, predstavlja proizvod specifi nog rizika i elemenata izloženih riziku.

297

tomova), sažeta u Završni elaborat (iz dvije knjige) – bila je prilagodjena potrebama za prakti no koriš enje njenih rezultata od strane direktnih korisnika, u prvom redu nosilaca izrade i sprovodjenja GUP-ova i drugih urbanisti kih planova i projekata. Naime, opšte prihva eno je da prostorni i urbanisti ki planovi – naro ito generalni, GUP-ovi (a takodje i detaljni, DUP-ovi) po svojoj statutornoj prirodi i zakonskoj snazi, u suštini predstavljaju polaznu i klju nu osnovu za smanjenje seizmi kog rizika a time i sprovo enjeodgovaraju e efektivne strategije za upravljanje seizmi kim rizikom – kako sa nivoa Republike tako i na lokalnom nivou.

1.2 ANALIZA URBANOG RIZIKA

Ne upuštaju i se u detaljniji prikaz ovih studija i dobijenih rezultata, zasnovanoih kako na podacima o ogromnom fondu ošte enih zgrada tako i zna ajnog dijela vitalne infrastrukture, može se re i da je time bila uspostavljena relavantn podloga za definisanje adekvatnih metodologija i iza utvr ivanje tzv. urbanog rizika.

Imaju i u vidu propulzivan napredak na ovim podru jima ostvaren prethodnih godina u zemljama regiona, prikladno je nazna iti jedan od modela artikulisan tokom realizacije ve pomenutih projekata, slika 1. ataširan sistemima vitalne infrastrukture.

Slika1 - Metodologija za utvr ivanje vulnerabiliteta i prihvatljivog rizika kod vitalnih infrastrukturnih sistema

U vezi sa navedenim modelom može se uspostaviti i bliska analogija sa metodologijom za utvr ivanje urbanog rizika. Pri tome, (kaogod i kod sistema vitalne infrastrukture za sebe) analiza urbanog rizika mora obuhvatiti razmatranje svih elemenatarizika – ljude, materijalne (zgrade, infrastrukturu, vitalne sisteme, arhitektonsko naslje e,prirodne resurse, itd.) i/ili nematerijalne (kulturne, društveno-ekonomske, spomeni ke,idr.), ali tako e i funkcionalne relacije izme u tih elemenata, urbanih aktivnosti (proizvo-dnja, potrošnja, razmjena), odnosno gradske uprave kao i relacije kroz veze sa mjestima iz okruženja, itd.

298

Dakle, dijagnoza vulnerabiliteta mora pokriti cio urbani sistem tj. sve konstitutivne elemente - pojedina no i kao homogenizovane grupe, ali i sistem kao takav - sa njegovom strukturom, komponenatama, unutrašnjim funkcionisanjem kaogod i spoljnjim vezama.

N.B. Kada je pak rije o karakteru pouka i ukupnih iskustava ste enih nakon Crnogorskog zemljotresa 1979. – koji su bili i ostali od izuzetnog zna aja ne samo za doma e okvire ve i širu me unarodnu zajednicu, o njihovom odrazu može se posredno zaklju iti i kroz sljede e Poglavlje 2. Me utim, ono što – s tim u vezi, valja i ovdje ista ijeste sticaj okolnosti da je tada po prvi put promovisana nezaobilazna uloga prostorno-urbanisti kog planiranja u mitigaciji zemljotresnog rizika. Isto tako, pri tome je (uz izvanrednu tehni ku pomo Ujedinjenih Nacija) prakti no ostvaren pionirski pristup integralnom upravljanju seizmi kim rizikom.

2. PROGRAMSKA USMJERENOST I CILJEVI SAVREMENE STRATEGIJE ZA SMANJENJE SEIZMI KOG RIZIKA2.1. STRATEGIJA IZ JOKOHAME I PLAN AKCIJE ZA BEZBJEDNIJI VIJET U BUDU NOSTI (UN WR, YOKOHAMA / JAPAN, 1994.)

Na polovini UN IDNDR (Me unarodne decenije za zaštitu od prirodnih katastrofa, 1990-2000)4, 73pod pokroviteljstvom UN - održana je Svjetska konferencija o smanjenju rizika od prirodnih katastrofa (Jokohama, Japan, maj 1994). Tom prilikom Konferencija je usvojila Završni dokument pod naslovom ”Strategija iz Jokohame i Plan akcije za bezbjedniji svijet u budu nosti – smjernice za prvenciju prirodnih katastrofa, pripremljenost i mitigaciju (ublažavanje)”.

Usvojeni pristup, intencije i filozofija prevencije i pripremljenosti na prirodne katastrofe, u prvom redu na seizmi ki hazard – komuniciraju sa pionirskim doprinosom, pristupima i konceptima ustanovljenim i realizovanim kroz Projekat izrade Prostornog plana Republike i generalnih urbanisti kih planova Crne Gore (UNDP/UNCHS/UNDRO Project YUG/79/104), izveden nakon Crnogorskog zemljotersa od 1979.godine.

Završni dokument iz Jokohame strukturno (i krajnje instruktivno) pored ostalog, obuhvatio je sljede e aspekte: (1) Principi, (2) Strategija i (3) Plan akcije.

U odnosu na Principe definisane u ovom dokumentu u cjelini, posebno se isti usljede i stavovi i smjernice:

- Preventiva i pripremljenost za prirodne katastrofe su od primarnog zna aja za smanjenje i otklanjanje posljedica prirodnih katastrofa;

- Preventivu i pripremljenost za prirodne katastrofe treba smatrati integralnim aspektima razvojne politike, uz planiranje na nacionalnom, regionalnom i lokalnom nivou, te bilateralnim, multilateralnim i globalnim uslovima;

4 IDNDR je bila proglašena od strane UN na predlog WCEE (San Francisko, 1984 godine), podstaknut u prvom redu efektima Crnogorskog zemljotresa 1979. kao i zemljotresa El Asnam/Alžir, 1980. Prema njenim glavnim smjernicama sve zemlje lanice UN bile su dužne da donesu svoje nacionalne programe. Pri tome, mandat za pripremu takvog programa na nivou SFRJ imali su SHMZ, Beograd i IZIIS, Skoplje uz u eš e SDSGJ – iji je rad bio prekinut nakon raspada SFRJ. Ina e, nakon formiranja SRJ - na njenom nivou bio je nastavljen rad na pripremi i donošenju takvog programa,tako e od strane SHMZ u saradnji sa JUZI-em (i uz u eš e ovog autora).

299

- Zaštita životne sredine kao komponente održivog razvoja u skladu sa otklanjanjem siromaštva je od primarnog zna aja u sprje avanju i ublažavanju prirodnih katastrofa. U odnosu na Strategiju, specifi no u vezi sa procjenom aktuelnog stanja u oblasti

smanjenja katastrofa, izme u ostalog, posebno su apostrofirani sljede i stavovi:

- Svijest o potencijalnim koristima od smanjanja katastrofa još uvijek je ograni ena na usko specijalizovane krugove, i još uvijek nije uspješno prenešena na sve sektore društva, naro ito ne na one koji donose odluke, kao ni na širu javnost;

- Programi obrazovanja i obuke za one ljude i subjekte koji su profesionalno uklju eni i/ili obavezni, kao i za širu javnost, nijesu dovoljno razvijeni niti postaknuti, sa težištem na na ine i sredstva smanjivanja katastrofa. Tako e, potencijali informativnih medija, nau na zajednica, privreda, javni i privatni sektor u cjelini - nijesu, uopšte ili bar ne dovoljno, mobilisani;

- Iskustvo je pokazalo da (iako to nije bio dio zadataka Decenije), koncept smanjenja katastrofa treba proširiti da obuhvati i druge prirodne situacije – uklju ivo ekološke i tehnološke katastrofe i havarije kao i njihove me usobne odnose koji mogu imati zna ajan uticaj na društvene, ekonomske, kulturne i druge sisteme životne sredine, naro ito u zemljama u razvoju.

2.2. DOKUMENTI SVJETSKE KONFERENCIJE ZA SMANJENJE KATASTROFA (UN WCDR, 2005., KOBE/JAPAN).

Generalna Skupština Ujedinjenih nacija na svom 58. zasijedanju od 23. decembra 2003. godine donijela je Rezoluciju o Me unarodnoj strategiji za smanjenje katastrofa (A/RES/58/214: International Strategy for Disaster Reduction) kao i korespodentnu Rezoluciju o prirodnim katastrofama i vulnerabilitetu/povredljivosti (A/RES/58/214: Natu-ral Disaster and Vulnerability). U okviru ovih rezolucija Generalna skupština je isto-vremeno donijela odluku o sazivanju (druge) Svjetske konferencije o smanjenju katastrofa (WCDR, Kobe-Hygo, Japan, januar 2005.) Ina e, neki od glavnih dokumenata te Konfe-rencije su: Hyogo Declaration & Hyogo Framework for Action 2005-2015 (Building the Resiliance of Nations and Communnities to Disasters).

3. BAZI NI KONCEPTI I PREDUSLOVI 3.1. OKVIRNI STRUKTURNO-INTEGRATIVNI PRISTUP

Podrazumijevaju i mogu nost i druga ijeg konceptualnog pristupa odnsono i druga ijih interpretacija, ovaj autor smatra da je strategiju i sistem integralnog upravljanja seizmi kim rizikom (odnosno smanjenje neprihvatljivog sezmi kog rizika) pogodno najsažetije opisati i okarakterisati putem korelativne sprege i korespodentnog ukrštanja funkcionalih skupova koje ine: (a) Strukturno-tematska matrica PPI/STAPLE, i (b) Integrativni aspekti/opcije smanjenja seizmi kog rizika.

Ad (a): Strukturno-tematska matrica PPI/STAPLE. Pojedine komponente u navedenom matri nom sklopu, ozna ene preko po etnih slova njihovog naziva - u datom kontekstu, imaju zna enja prema navodu kako je dato u Tabeli.

300

- Politika (koja konstituiše politi ke težnje i ciljeve u smanjenju ljudskih, materijalnih i ekonomskih gubitaka kao i nematerijalnih šteta od budu ihzemljotresa – i to kako na nacionalnom tako i na regionalnom odnosno lokalnom nivou);

- Planiranje (koje obuhvata i podrazumijeva formulisanje aktivnosti i programa sa raznih podru ja – putem razli itih oblika i nivoa, orijentisanih na prevo enje politika smanjenja seizmi kog rizika u potreban postupak sprovo enja odnosno realizaciju); i

- Implemantacija (odnosno sprovo enje, koje uklju uje donošenje potrebne legislative, zakona, propisa, pravilnika, standarda, uzansi i dr.; pripremu specifi nih planova i mjera predvi enih u planovima i programima višeg reda za smanjenje seizmi kog rizika; uspostavljanje odgovaraju eg institucionalnog sistema za utvr-

ivanje i ja anje mehanizama potrebnih za primjenu i sprovo enje donijetih planova i programa).

Tabela 1 - Glavne faze, PPI: PPI / STAPLE Politika Planiranje Izvršavanje Socijalni * * * Tehni ki * * * Administrativni * * * Politi ki * * * Legislativni * * * Ekonomski * * *

Opšti aspekti, STAPLE: Shodno svakoj od prethodno navedenih glavnih faza - kao imanentni i od naro itog zna aja, izme u ostalih, posebno se izdvajaju sljede e aspekti: Socijalni, Tehni ki, Administrativni, Politi ki, Legislativni i Ekonomski.

Ad (b): Integrativni aspekti i opcije smanjenja seizmi kog rizika. Kao osnovni integrativni aspekti smanjenja seizmi kog rizika, tretiraju i ih istovremeno kao komponente jedinstvenog integralnog sistema upravljanja seizmi kim rizikom, u kontekstu ciljeva ovog razmatranja, mogu se uslovno ozna iti sljede a podru ja:

- Utvr ivanje zemljotresnog hazarda, - Utvr ivanje seizmi kog rizika i njegovog prihvatljivog nivoa, - Aseizmi ko projektovanje i izgradnja objekata i infrastrukturnih sistema, - Prostorno-urbanisti ko planiranje u seizmi kim uslovima, - Mitigacija seizmi kog rizika (legislativno-institucionalni aspekti, i sl., - Pripremljenost na zemljotres, u širem i savremenom zna enju,- Upravljanje zemljotresnim rizikom, kao i - Integrisani informacioni sistem sa bazom podataka o prostoru i izgra enoj sredini

(GIS), za totalno upravljanje seizmi kim rizikom (kao i rizikom od drugih priro-dnih hazarda).

301

Slika 2 - Upravljanje seizmi kim rizikom, uklju ivo i postzemljotresni odgovor

U navedenom kontekstu, pod pojmom upravljanja seizmi kim rizikom podrazumijevaju se sve aktivnosti na predzemljotresnom planiranju i izvršenju. Pri tome ovdje nijesu eksplicitno uklju ene operacije spašavanja i drugih prate ih vanrednih intervencija na saniranju urgentnog stanja, kaogod ni na postzemljotresnoj obnovi i rekonstrukciji. Me utim, prema prirodi stvari i one su bitan institucionalno-organizacioni aspekt ukupnog sistema zaštite od prirodnih katastrofa i tehnoloških havarija na svim nivoima društvene zajednice (od države do lokalne samouprave, uklju ivo ulogu gra ana odnosno javnosti), Slika 2.

3.2. POLITIKE ZA SAVREMNO UPRAVLJANJE SEIZMI KIM RIZIKOM

Politika seizmi ke sigurnosti konstituiše politi ke težnje u smanjenju ljudskih, materijalnih i ekonomskih gubitaka. U suštini ona predstavlja odre ene planove, pravila, stru nu i profesionalnu praksu ili, pak, druge na ine koji djeluju sa snagom zakona a usmjereni su na ispunjenje cilja za smanjenje seizmi kog rizika kroz mitigaciju, pripre-mljenost, emergentni odgovor, obnovu i rekonstrukciju. Uklju ivo potrebne odgovaraju emjere i regulativu koje se odnose na sve vjerovatne vanredne okolnosti koje mogu zadesiti zemljotresu izloženo - odnosno od njega postradalo podru je.

Sve te mjere i regulativa treba da omogu e zajednici planiranje za neizbježne zemljotrese, tj. da kontroliše predvidive posljedice ali i da preduprijedi one nepredvidive, uklju ivo razna i mogu a “iznena enja”. O igledno, pod navedeni generalni okvir politike seizmi ke sigurnosti potpada i opšta javna politika u ovoj oblasti, kaogod i svi ostali relevantni aspekti i segmenti (sintetski spregnuti putem tzv. strukturno-tematske matrice PPI/STAPLE), uz njihovo adekvatno rekognosciranje i tretiranje kroz navedena inte-grativna podru ja.

Ina e, u kontekstu savremenog pristupa upravljanju seizmi kim rizikom polaznu osnovu moraju predstavljati: (a) identifikovanje i utvr ivanje samog hazarda - uz odgovaraju e seizmi ko zoniranje (Slika 2); kao i (b) prepoznavanje globalnog procesa utvr ivanja seizmi kog rizika - uz uspostvljanje odgovaraju eg konzistentnog sistema za njegovo smanjenje.(Slika 3) - korespodentno integrativnim aspektima navedenim u Odjeljku 3.3

3.3. NA INI I OPCIJE ZAJEDNICA ZA SMANJENJE NEPRIHVATLJIVOG SEIZMI KOG RIZIKA

Ovi na ini i opcije tretiraju i ih u dosta generalisanoj formi i korespodentno prethodnim odjeljcima, mogu se najsažetije predstaviti kao što slijedi:

Usvajanje postoje eg i osvajanje novog znanja. Ovo je neophodno da bi neka zajednica mogla stvarano i uspješno upravljati svojim seizmi kim rizikom. Druga ije re e-

302

no, ona mora biti sposobna da mijenja i prilago ava javnu politiku u ovoj oblasti i to na bazi nau nih, tehni kih, politi kih i zakonskih usaglašenosti. Ove, pak, po prirodi stvari treba progresivno da evoluiraju sa vremenom - budu i da je zajednica prinu ena da živi sa zemljotresima, i to uz stalno pove anje rizika od tog hazarda. Pri tome sasvim je razu-mljivo da upravljanje seizmi kim rizikom može biti uspješno, najprije, ako je zasnovanao na iskustvima svoje zemlje odnosno ( ak još i više) na iskustvima drugih naprednijih zemalja.

Mitigacija. Mitigacija (ublažavanje) obuhvata itav niz razli itih aspekata uklju ivo politike, legislativne mandate, stru nu i profesionalnu praksu i osposobljenost, kao i razna društvena, strukturna i nestrukturna prilago avanja - orijentisana i projektovana za zaštitu, smanjenje i svo enje na najmanju mogu u mjeru seizmi kog rizika odnosno efekata zemljotresa na odre enu zajednicu. Kategorije mitigacionih mjera i regulacija, onako kako se tretiraju u zadnjih dvije dekade, uklju uju: (1) propise i standarde za izgradnju objekata i urbanisti ko planiranje; (2) prostorno-urbanisti ko planiranje i upravljanje; (3) aseizmi ko projektovanje objekata; (4) seizmi ku dijagnozu i rehabilitaciju postoje ih objekata; (5) kontrolu i zaštitu propisnog sprovo enja radova; (6) predvi anje, javnu svijest i planiranje; (7) planiranje obnove i rekonstrukcije u post-zemljotresnim uslovima, uz planiranje daljeg razvoja; (8) fianasiranje i (9) osiguranje.

Slika 3 - Seizmi ko zoniranje kao okvir konzistentne veze izme u utvr ivanja rizika i upravljanja tim rizikom

Pripremljenost. Uloga pripremljenosti je da olakša predvi anje i prevazilaženje o ekivanih i neo ekivanih posljedica zemljotresa. Ona, tako e, uklju uje cio niz politika, zakonskih mandata, profesionalnog znanja i društvenih prilago avanja i angažovanja obuhvataju i sve potencijalne u esnike procesa - od pojedinaca, preko raznih organizacija i zajednica, do nivoa vladinih institucija odnosno države. Podrazumijeva se da se na takvom predvi anju i scenarijima zasnivaju i donose korespodentni planovi za urgentni odgovor koji treba da uslijedi odmah nakon zemljotresa, kaogod i planovi potreba za kasniju obnovu i rekonstrukciju. Ovi planovi, po pravilu, donose se po dnevnoj vremenskoj dinamici tj. za prve dane odmah nakon zemljotresa a zatim sukcesivno - po nedjeljama, mjesecima i godinama. Pripremljenost na zemljotres, svakako, podrazumijeva i uklju uje obavezno

303

stvaranje odgovaraju ih instiucionalnih kapaciteta, osposobljenih i opremljenih shodno povjerenim zadacima.

Opcije za olakšanje predvi anja odgovaraju ih aktivnosti kao i za neophodnu pripremljenost na katastrofu uklju uju: (1) javnu svijest; (2) scenarije zemljotresa; (3) predvi anje same pojave i njene posljedice; (4) obuku; (5) pregled i klasifikaciju ošte enih objekata, uz ocjenu njihove useljivosti i upotrebljivosti; (6) metodologiju za procjenu prouzrokovanih šteta; (7) post-zemljotresna istraživanja; (8) ugrentni odgovor; (9) traganje i spašavanje; (10) predplaniranje za obnovu; (11) predplaniranje rekonstrukcije i izgradnje.

Opcije za stvaranje novih institucionalnih kapaciteta, izme u ostalog, uklju uju naro ito: (a) komisije za seizmi ku sigurnost; (b) relevantne istraživa ke centre i (c) odgovaraju e zemljotresne konzorcije; i dr.

Emergentni (urgentni) odgovor. Ovaj odgovor korespondira u svemu sa pristupom usvojenim kroz pripremljenost, uz proširenje na obezbje enje hitnih službi za djelovanje odmah nakon zemljotresa, tj. shodno i postoje em “Republi kom zakonu o zaštiti i spašavanju”.

Pri tome, glavne opcije za obezbje enje hitnih službi uklju uju: (1) pomopojedincima i organizacijama u okviru njihovih domova i radnih mjesta, te pomo zajednici na spašavanju i zaštiti života kao i zaštiti imovine; (2) alokaciju resursa, zadataka, kao i uvr ivanje vremena za obezbje enje kontinuiteta u sinhronizovanom funkcionisanju organizacionih struktura i procedura u okviru date zajednice; i (3) utvr ivanje integralnog obrasca komunikacije i povezivanja pojedinaca i organizacija angažovanih na traganju i spašavanju, kao i me u drugim hitnim službama.

Obnova i rekonstrukcija. Pojam obnove i rekonstrukcije tako e u svemu kore-spondira sa pristupom i premisama nazna enim u odnosu na pripremljenost, uz evidentnu orijentaciju na ponovno uspostavljanje klju nih službi i funkcija, kao i na krajnje osmišljen pristup obnovi i izgradnji postradalog podru ja prema vremenskoj dinamici iskazanoj u nedjeljama, mjesecima i godinama što slijede nakon zemljotresa.

Pri tome, podrazumijeva se, moraju biti uklju ene u proces i odgovaraju e mjere predostrožnosti i mitigacije, kako bi se zaustavili odnosno preduprijedili mogu i dalji gubici, posebno usljed efekata after-šokova.

Ina e, glavne akcije i rješenja za restauraciju lokalnih službi, kao i za obnovu i izgradnju nakon zemljotresa, uklju uju odgovaraju e i blagovremeno strategijsko planiranje prije zemljotresa, zasnovano na harmonizovanom i sinhronizovanom multi-legislativnom pristupu.

Integrisani informacioni sistem za totalno upravljanje rizikom. Pripremljenost društvenih zajednica i njihovih gra ana na katastrofu ne može biti efektivna ako svi oni prethodno ne mogu razumjeti kako takve situacije uopšte izgledaju. To uklju uje sve lanove društva, a prije svega: politi are, javne zvani nike, istraživa e, ljude iz mas-medija,

profesionalce iz raznih struka i oblasti, javnosti itd. Baziraju i se na prošlom iskustvu, uklju ivo i ono iz Crnogorskog zemljotresa od 1979., bi e potpuno jasno da nedostatak sposobnosti ljudi da zamisle situaciju potencijalnog zemljotresa predstavlja jedno od najvažnijih pitanja za pravovremenu pripremu adekvatnog odgovora, kako onog urgentnog za vrijeme samog doga aja tako i nakon njega, tj. obnove i rekonstrukcije postradalog podru ja.

304

3.4. OSVRT NA DRUGE AKTUELNE KONCEPTE I NEKE SPECIFI NE PROJEKTE

Ovaj osvrt, s obzirom na okolnosti, može imati tek provizoran – eventualno i podsticajan karakter (tj. naro ito u odnosu na neposredno doma e okruženje).

Može se re i da su, na globalnom nivou, vrlo važnu ulogu odigrale razne supranacionalne organizacije kao što su Evropska (EAEE) i Svjetska (IAEE) asocijacija za zemljotresno inženjerstvo. Takodje, a posebno kada je rije o analizi i utvrdjivanju seizmi kog hazarda na globalnom nivou, nose u ulogu u razvijanju brojnih projekata imale su Medjunarodna asocijacija za seizmologiju i fiziku unutrašnjosti zemlje (IASPEI), Evropska seizmološka komisija (ESC), idr. Iz toga okvira, kao svojstven i veoma zna ajan,može se izdvojiti Projekat „Globalni seizmi ki hazard „ (GSHAP, Gobal Seismic Hazard Project).

Kada su, pak, u pitanju neki drugi aspekti integralnog upravljanja seizmi kim rizikom, svakako da izmedju niza manje/više u Svijetu razvijenih i poznatih projekata, treba izdvojiti: Projekat HAZUS, proizveden od strane FEMA (USA); kao i Projekat RADIUS, razvijen od strane UN-ISDR,uz namjenu za gradove zemalja u razvoju.

Ina e, inicijativa za razvijanje odredjene saradnje evropskih zemalja na ovom podru ju – posebno onih iz regiona Mediterana, bila je pokrenuta kroz Projekat: Program prioritetnih akcija (UNEP/MAP – PAP/RAC, Split), odnosno njegov Kooperativni program tzv. SEISMED, enova 1986. Otuda, je kasnije došlo i do ideje da se u jedinstven koncept Evropske regulative za gradjevinarstvo (Stuctural Eurocodes), uvede i seizmi karegulativa tj. EC 8 (Eurocode 8: Projektovanje objekata za zemljotresnu otpornost ). Tako e, i do (sada ve uveliko proklamovanog ) sporazuma o široj saradnji zemalja medi-teranskog regiona.

Izmedju brojnih EU projekata realizovanih tokom minulog šestogodišnjeg ciklusa (nažalost bez zna ajnijeg u eš a Crne Gore, ako se izuzme Projekat DPPI) valja spomenuti i takve kao što su: Euro Seis Test; Euro Seis Mod; Euro Seis Risk, PROCHTECH, ISARD, kao i Risk-UE.

Pri svemu, kao izuzetno zna ajan doprinos Medjunarodnoj strategiji za smanjenje rizika od prirodnih katastrofa (UN WCDR, Kobe/Japan 2006) kao i za Evro-Mediteransku saradnju na zaštiti od rizika, apostrofira se Sporazum Savjeta Evrope »EUR –OPA Major Hazard Agreement«, sa širokom mrežom korespodentnih Evro– Mediteranskih centara rasporedjenih po svim zemljama lanicama Savjeta Evrope.

Tako e, ali za naše uslove i konkretnu priliku kao posebno zna ajan, treba navesti i apostrofirati Projekat »Harmonizacija mapa seizmi kog hazarda za zemlje zapadnog Balkana«, iji je nosilac koordinacije Republi ki seizmološki zavod, Podgorica. Za poželjeti je i o ekivati da to bude tek po etak obnove dalje i šire saradnje uklju enih zemalja kao i drugih relevantnih me unarodnih subjekata na smanjenju seizmi kog rizika u našem regionu.

Na kraju, ali ne i zadnje po zna aju – naprotiv, jesu uloga i kontinualna dragocjena aktivnost GRF (Global RiskForum) Davos, Švajcarska. Iz tog konteksta (sa velikim uvažavanjem injegovoh prethodnih aktivnosti i poduhvata) ini se osobito zna ajnim u ovoj prilici ukazati na dva sasvim skorašnja postignu a i to: (1) GRF Davos IDRC Changdu, 2009 (International Desaster Reduction Conference), Changdu, 13-15 July, 2009, China, kao i (2) upravo potpisani Memorandum o razumijevanju zaklju en sa Evropskom komi-sijom.

305

4. ETI KI ASPEKTI I UPRAVLJANJE ZEMLJOTRESNIM RIZIKOM

4.1. NEKA OPŠTA RAZMATRANJA Osvrt na osnovne obrasce primijenjene i inženjerske etike.U ovom osvrtu

pojmovi eti ki i moralan koriste se kao sinonimi, uz povezanost njihove definicije sa onim principima na kojima se mora temeljiti odlu ivanje kada ljudi treba i/ili moraju primjereno postupiti i ispravno djelovati u odre enim posebnim situacijama.

U suštini, iako esto zanemarivana, etika je veoma prisutna u zemljotresnom inže-njerstvu i važna u procesu smanjenja seizmi kog rizika. Svi u esnici u tom procesu, kada god se eti ki izazovi javljaju ostaju sami pri suo avanju sa njima, pa odluke koje tada donose bivaju uglavnom zasnovane na intuiciji i iskustvu, odnosno njihovoj li noj i ukupnoj kompetentnosti.

Imaju i u vidu multidsciplinarnost zemljotresnog inženjerstva to etiku u njegovom kontekstu valja posmatrati i tretirati široko: razjašnjavaju i i osvjetljavaju i razlike u shvatanjima, razmatranjima, obavezama i ograni enjima, tj. uz suo avanje i su eljavanje raznih stru njaka i donosilaca odluka u oblasti smanjenja seizmi kog rizika. Ina e, mnogo je raznih disciplina uklju eno u ovu oblast, kao što su: gra evinsko i drugo inženjerstvo, arhitektura, seizmologija, geologija, ekonomija, društvene nauke, prostorno-urbanisti ko planiranje, javna politika, itd. Tako e, lanovi ove tzv. zajednice zemljotresnog inženjerstvasu razni stru ni konsultanti, vladini službenici, istraživa i, profesori, poslovni investitori i vlasnici, stru njaci u osiguranju, projektanti, planeri, preduzetnici odnosno izvo a i, itd. I dok svi oni imaju svoje razli ite profesionalne obaveze, ono što bi moralo da im bude zajedni ko jeste odre eno specijalizovano znanje o zemljotresnom hazardu, relevantnim aspektima seizmi kog rizika, te prate im rizicima i posljedicama, na ini razumijevanja i analize rizika, kao i putevi izlaska na kraj sa njima kroz inženjerstvo, upravljanje rizikom, javno informisanje, legislativu, kaogod i kroz ukupnu razvojnu politiku.

Priroda ve ine ovih podru ja je takva da ona nose karakter unaprijed teško odredljive izvjesnosti pa - otuda, uklju uju nesigurnost i prosu ivanje, uskla ivanje razli itih interesa koji esto zahijevaju donošenje beskompromisnih (ponekad i kontraverznih politi kih) odluka, itd. Potrebno je da svi involvirani subjekti a posebno profesionalci moraju imati na umu promjenljivu prirodu seizmi kog rizika, kako njegovog permanentnog uve anja - tako i na ina njegovog smanjenja. I da li te brze promjene, kao i naše saznajne mogu nosti, u tom pogledu mijenjaju i uve avaju naše obaveze?

Kona no, da li su eti ke dimenzije iste za svaku od involviranih disciplina, odnosno kategorija odgovornosti? Da li inženjer i geolog mogi imati istu moralnu odgovornost u zaštiti javnog dobra kakvu ima, na primjer, ministar? I obratno, da li neki gradona elnik odnosno administrator i/ili službenik - nadležan za planiranje i izgradnju uopšte može da ima istu individualnu odgovornost za svoj rad, kakvu ima jedan kompetentni akademski istraživa i profesionalni specijalista?

Pristupi eti kom donošenju odluka. Model eti kog ponašanja. Donošenje ispravnih i dobrih eti kih odluka nije uvijek lak zadatak. Ina e, genralno uzevši, model eti kog ponašanja sadrži etiri osnovne komponente: (1) moralnu osjetljivost; (2) moralno prosu ivanje; (3) moralnu obaveznost; (4) moralnu hrabrost.

Naravno, ne upuštaju i u bilo kakvo detaljnije razmatranje ovdje navedenih komponenata, treba ukazati da izostanak ili prenebregavanje makar i jedne od ovih

306

karakteristika kompromituje eti nost u datom slu aju, odnosno obara model eti kog ponašanja u cjelini. Uklju ivo, podrazumijeva se, i same donosioce odluka iz sfere javne odgovornosti.

4.2. RELACIJE PREMA SEIZMI KOM RIZIKU a) Utvr ivanje i razotkrivanje rizika. Jedan od najvažnijih problema u širokom

podru ju smanjenja seizmi kog rizika jeste identifikovanje, razumijevanje i prihvatanje rizika. U stvari, rizik je životna injenica. Nikada ga se ne može u potpunosti eliminisati. Štoviše, naše znanje o ovom riziku – iako suštinsko, ustvari je još uvijek nekompletno i sa karakterom zna ajne nedostatne nesigurnosti. Upravo ova karaketristika nesigurnosti, obilježava svaku odluku u vezi sa zemljotresima, izuzev onih koje se donose nakon vedogo enog zemljotresa.

Šta znamo o zemljotresnom riziku, koliko smo sami sigurni o tome šta znamo, kako interpretiramo raspoložive podatke i postoje e injenice, kako pristupamo rješavanju datog problema i uz koliko zasnivanja na našoj profesionalnoj pripremljenosti i osposobljenosti, našem obrazovanju i kulturi, elementu vremena povezanom sa onim što znamo, šansama koje želimo zadobiti – sve su to aspekti i dimenzije istog problema, tj. nesigurnosti. Znanje predstavlja srž onoga što profesionalni poslenici iz raznih oblasti zemljotresnog inže-njerstva rade, i ono je to koje u razli itim formama može kreirati obaveze i izazvati razli ite eti ke i moralne dileme, kako za pojedince tako i za zajednicu.

b) Prihvatljivi rizik. Kada je rizik jednom utvr en i razotkriven moramo se suo iti sa pitanjem (ina e, detaljnije razmatranim na drugom mjestu): koliko visok rizik može biti tolerisan i po koju cijenu, odnosno koji je nivo rizika prihvatljiv? Naravno, pri tome se ne može govoriti o nekoj jasnoj cijeni koju treba pripisati vrijednosti ljudskih života, ali koštanje razli itih nivoa ošte enja na zgradama i drugim strukturama, kaogod i koštanje odgovaraju ih nivoa sigurnosti objekata i mitigacije rizika, može i treba biti procijenjeno. Tako e je važno napomenuti da razni ljudi, pogotovu neposerdno zainteresovani, imaju razli ite procjene i stavove o tome šta predstavlja prihvatljiv rizik za jednu te istu situaciju.

c) Profesionalne odgovornosti. Svi koji rade na podru jima povezanim sa smanjenjem sezimi kog rizika imaju svoje profesionalne odgovornosti. Neke profesije, kao što su planeri, državni lokalni administratori, nadležni zvani nici zaduženi za izgradnju, kao i drugi vladini službenici, imaju široke odgovornosti budu i da su zaduženi da služe opštem javnom dobru. Ove odgovornosti je ponekad teško uravnotežiti, jer su interesi specifi nih grupa u zajednici ili društvu, veoma esto u konfliktu. Drugi stru njaci, odnosno profesionalci, kao što su gra evinski inženjeri, seizmolozi, geolozi i arhitekte imaju direktne zakonske odgovornosti sa nivoa pojedina nih projekata. Oni se, tako e, ponekad mogu na i u me usobnom konfliktu. Jedan inženjer, na primjer, može biti suo en sa balansiranjem sukobljenih interesa investitora i njegovog poslodavca. Uopšte uzevši, profesionalne odgovornosti možemo razvrsat, u sljede e kategorije: javnost i njeni organi(dakle, nosioci vlasti odnosno javnih ovlaš enja „stakeholders”); klijenti; poslodavci; konkurenti; i kolege. Valja podrazumijevati da svaka zemljotresna profesija ima odre enuodgovornost u svakoj od navedenih kategorija.

d) Uloga i odgovornosti profesionalnih asocijacija i udruženja. Mnogostrana priroda zemljotresne problematike i upravljanja seizmi kim rizikom, neminovno podra-zumijeva i potencira postojanje odre enih odgovornosti takozvanih profesionalnih udru-ženja i asocijacija, a koje se razlikuju od individualnih obaveza i akcija. Rade i kroz

307

profesionalne organizacije, kaogod i kroz razna tijela iz okvira javne politike – obrazovana na nivou pojedinih zajednica (državne odnosno lokalne) mogu se zahtijevati i adresirati i takve šire odgovornosti kao što su priprema i sprovo enje potrebnih i relevantnih propisa ili, pak, izmjene odgovaraju e druge legislative. Dodatno, profesionalna asocijacija za zemljotresno inženjerstvo trebalo bi da može preuzeti vo stvo u promociji boljeg razumijevanja ukupne problematike upravljanja seizmi kim rizikom i, naro ito u promociji razumijevanja kako narasla multikulturna priroda zemljotresnih profesija može i mora uticati na izbor alternativnih akcija i rješenja, odnosno na njihovu primjerenu optimizaciju.

Tako e, jedna druga osobita odgovornost ovih asocijacija (odnosno zajednice) posebno u doma im uslovima, morala bi biti pokretanje i vo enje rasprave o eti kim implikacijama brzog razvoja ve nazna enih novih aspiracija i trendova u preduzetništvu (situiranih u krajnje neregularne i, tobože, neizbježno propratne uslove tzv. “društva u tranziciji”) - posebno u oblastima izgradnje objekata odnosno projektovanja i prostorno urbanisti kog planiranja.

3. ZAKLJU NI REZIME I PREPORUKE

Na samom kraju - iako se to ve i po sebi ini evidentnim, vode i motiv ovog rada mogao bi se izraziti kroz osje aj i uvjerenje autora da 30-ta godišnjica Crnogorskog zemljotresa od 1979. godine predstavlja obavezuju u i neponovljivu priliku za jedan novi po etak u tretiranju i zaštiti održivog razvoja zemlje – i to po više osnova. Prije svega i posebno u razvijanju djelotvorne kampanje za svojevrsnu rehabilitaciju ukupne javne i društvene svijesti o injenici da je ona trajno upu ena na tzv. “život sa zemljotresom”. Otuda – i na njeno primjereno i adekvatno državno organizovanje, u skladu sa ve opšte prihva enim principima i savremenom strategijom efektivne paraseizmi ke zaštite. Ina e, generalno sublimiranim kroz tzv. Strategiju iz Jokohame i Plan akcije za bezbjedniji svijet u budu nosti – smjernice za prevenciju prirodnih katastrofa, pripremljenost i mitigaciju.

Uz nepretencioznu konstataciju da su iskustva ste ena nakon Crnogorskog zemljotresa 1979. predstavljala relevantan prilog takvoj strategiji – nadati se da i ovaj referat može dodatno inovirati i oja ati platformu za produktivan doprinos njihovoj reafirmaciji u aktuelnim uslovima. Otuda, treba smatrati neophodnim konzistentan pristup donošenju i sprovo enju odgovaraju eg okvirnog programa prioritetnih akcija i aktivnosti, i to kako na nacionalnom tako i na regionalnom nivou.

Ina e, na nacionalnom nivou kao inicijalno klju ni, izdvajaju se slede i elementi i komponente takvog programa.

Legislativni i razvojno upravno aspekti: a) Donošenje sistemskog Lex Specialis: „Zakona o zaštiti od zemljotresa“ (alias,

Zakona o integralnom upravljanju seizmi kim rizikom), kao preduslova za sveobuhvatno smanjenje i kontrolu seizmi kog rizika.

b) Rekognosciranje relevantne sektorske zakonske i tehni ke regulative (koordinirane sa nivoa prethodno navedenog zakona kao hijerarhijski najstarijeg), uz donošenje nove i/ili uz harmonizaciju postoje e.

c) Uspostavljanje koherentnog sistema upravnog organizovanja (na državnom i lokalnom nivou), primjerenog prirodi i jedinstvu koncepata upravljanja prostorom i zatitom životne sredine tj. u kontekstu proklamovanih principa održivog razvoja i tzv. ekološke države.

308

Institucionalni okviri a) Osnivanje Državne agencije /Centra za zaštitu od zemljotresa, sa ulogom i

ingerencijom promocije i nadzora nad primjenom profesionalnih i regulativnih odredbi definisanih zakonom i odgovaraju im propisima.

b) Konstituisanje ili formiranje odgovaraju e republi ke institucije za prostorno-urbanisti ko planiranje i životnu sredinu (uz napuštanje pogubne prakse tržišnog tretiranja razvojnih prostorno-urbanisti kih planova).

c) Rekognosciranje i uspostavljanje harmonizovanog sistema/mreže relevantnih institucija – strateških nosilaca razvojnih aktivosti i operativnih odgovornosti u oblastima upravljanja prostorom i seizmi kim rizikom.

Neka posebno aktuelna pitanja i sugestije a) Promovisanje ekspertnosti i istraživanja, sa insistiranjem na adekvatnoj

edukovanosti i profesionalnoj kompetentnosti svih nosilaca javne odgovornosti (uklju ivo tzv. stake holders ).

b) Uspostavljanje strategije za obezbje enje potrebne zemljotresne sigurnosti postoje ih objekata – kroz njihovu sistematsku seizmi ku evaluaciju, rehabilitaciju i oja anje.

c) Istraživanje i razvijanje autenti nih metodologija za utvr ivanje urbanog vulnerabiliteta i prihvatljivog sezmi kog rizika, kao i vulnerabiliteta i prihvatljivog rizika kod vitalnih infrastrukturnih sistema (life lines).

d) Razvijanje me unarodne saradnje i promovisanje obnove stalne saradnje balkanskih država u kontekstu ranijih UNDP projekata orjentisanih na smanjenje seizmi kog rizika u regionu Balkana, kaogod i u kontekstu teku ih aktivnosti na transpoziciji Eurokodava u nacionalno zakonodavstvo a naro ito pri uzradi nacionalnih aneksa ( NDPs) uz Eurokod 8 .

*Najzad, valja podrazumijevati da razmatranja i sugestije – iznijete u odnosu na

ostale aspekte opšte strategije integralnog upravljanja seizmi kim rizikom na nivou Republike (tretirane kroz pojedine odjeljke izloženog rada) treba na elno smatrati sastavnim dijelom i ovog rezimea odnosno datih preporuka.

309

REFERENCES

1 Pavi evi , B.S. / Aseizmi ko projektovanje i upravljanje zemljotresnim rizikom ( Aseismic Design and Earthquake Risk Management, Univerzitet Crne Gore, Podgorica, “Obod”, 2001, Cetinje.

2 Pavi evi , B.S. / Seizmi ki hazard i kontrola seizmi kog rizika. Prostorni plan SR Crne Gore, Osnove Plana (UNDP/UNCHS Project YUG/79/104), RZUP-Titograd, UNCHS-Nairobi, UNDRO-Ženeva, 1983, Titograd.

3 Petrovski, J. / B.S. Pavi evi / Methodology on Vulnerability and Seismic Risk Analysis Applied in the Studies of the Coastal Region of SR Montenegro, Yugoslavia. Yugoslav National Report, WG B and C, UNDP/UNESCO Project, RER/79/014 (Earthquake Risk Reduction in the Balkan Region), IZIIS/Skopje-RZUP/Titograd, 1982, Titograd.

4 Pavi evi , B.S., editor: Seismic Risk Reduction in Mediterranean. Priority Action ”Land Use Planning in earthquakes Zones”, MAP Technical report No.17(245p.), UNEP/MAP, Athens and PAP/RAC, 1988, Split.

5 Strategy from Yokohama and Plan of actions for Safer World, Final Document, UN-World Conference on Natural Disaster Reduction, 1994,Yokohama/Japan

6 Hays, W., B. and J. Mohammadioun: Seismic Zonation, Monograph. The IDNDR Activity in conjunction with XIth ECEE, Paris, Ouest Editions, 1998, Paris.

8 UNEP/UNESCO Project RER/88/004 (Permanent Coordination Comitee for Earthquake Risk Reduction in the Balkan Region, PCC): Report on the seventh Session of PCC. IZIIS, Skopje, 1992, Skopje.

9 Meguro, K./ Yoshimura, M./ Integrated information system for total Disaster Management, Institute of Industrial science, University of Tokio, Bulletin of ERS Centre No.37, 2004.

10 Oliveira C.S. / A.Roca / X.Goula / Assesing and managing eartquake risk. Springer, 2006.

11 Hyogo Declaration & Hyogo Framework for Action 2005-2015, UN World Conference on Natural Disaster Reduction (WCDR, Kobe – 2005, Hyogo, Japan.

12 Aman J.W. / Integrated Disaster Risk Managament and Disaster Resilience Capacity Building, Global Risk Forum GRF Davos, IDRC Changdou 2009, July 2009, Changdou, China.

311

Violeta Mir evska1 Vladimir Bi kovski2

BEM – REŠENJE HIDRODINAMI KOG PRITISKA

Rezime:Fenomen interakcije fluid-konstrukcija za prvi par tretiran od Westergaard-a, koji je dao fizi ko objašnjene fenomena i teoretsko rešenje koje bazira na odre eneaproksimacije i pretpostavke ime se fenomen uproš ava i onda se može analiti kirešiti. Postoji sofisticiraniji na in rešavanja ovog problema koriš enjem metode grani nih elemenata. Metod se koristi da bi definirao intenzitet i distribuciju HDP u zavisnosti od manifestovanja akceleracija na licu brane i oblik kanjona. Koriste iteoretske principe BEM BEL3-ver.1 softwer razvijen u IZIISU-u, a rezultati su uspore eni sa rešenjem Westergaard-a. Kljucne reci: Hidrodinamicki pritisak, metod granicnih elemenata, lucna brana

USE OF BEM IN SOLVING FLUID – STRUCTURE INTERACTION

Summary: This phenomenon of fluid-structure interaction has been treated for the first time by Westergaard. He gave a physical and mathematical explanation to this phenomenon, based on certain assumptions by which the phenomenon is simplified in order to be solved analytically. There is a more sophisticated approach to solving this phenomenon by using the boundary element method. This method is used to define the intensities and the distribution of the hydrodynamic pressures depending on the exhibit accelerations at each moment of the dynamic response of the dam – fluid system as well as the geometrical shape of the canyon. Using the theoretical principles of the boundary element method, the BEL3-ver.1 computer program has been developed in IZZIIS, and the solution has been compared to the classic Westergaard’s one. Keywords : boundary element method, hydrodynamic pressure, arch dam

________________________________ 1.Assoc. Prof. Dr., (IZIIS),Skopje, Republic of Macedonia 2. Prof.Emeritus, (IZIIS), Skopje, Republic of Macedonia

312

1. INTRODUCTION

When the dam-water fluid is exposed to strong dynamic excitation, there is interaction between both media. The inertial forces that occur when the wavy fluid stroke the dam are known as hydrodynamic forces. The intensity of the hydrodynamic forces and pressures depends on the geometrical shape of the dam and the canyon as well as the deformability characteristics of the dam. This phenomenon has been treated for the first time by Westergaard. He gave a physical and mathematical explanation to the phenomenon, based on certain assumptions simplifying the phenomenon, in order to be solved analytically. Later, it has been treated by other authors as Zangar, Napetvaritze and Clough. They have applied the principle of added masses for solving HDP that was originally proposed by Westergaard. Their solutions are based on the assumptions that the water fluid is ideally incompressible and that the motion of the fluid in the reservoir is a stationary process. Since the inertness of the water mass is considerably higher in respect to the ground motion, the velocity of the water particles is characterized by relatively small amplitudes wherefore this motion is treated as a stationary wavy motion. However, unlike the fluid motion, the motion of the dam body is considerably more expressed and with higher values of total accelerations and relative velocities. However the real structures are deformable and the dam is not ideally rigid body as it is treated by Westergaard. The manifested total accelerations at certain zones of the dam are characterized by a variable value and are defined as a sum of the relative accelerations and the ground motion accelerations. There is a more sophisticated approach in solving this phenomenon by use of BEM. The advantage of using BEM is the possibility to take under consideration the geometrical shape of the dam and the canyon as well as the deformability characteristics of the dam since it essentially affect the intensity of the HDP. Using the theoretical principles of the boundary element method, the BEL3-ver.1 computer program has been developed in IZZIIS, and the solution has been compared to the classic Westergaard’s one.

2. MODEL OF BOUNDARY ELEMENTS

The main equation by which the stationary type of motion of the incompressible fluid is solved is the Laplace’s differential equation of the second order amended by the functions of the boundary conditions.

02

2

2

2

2

2

zW

yW

xW

),,( zyx (1)

2),,,(),,(

ntzyxW

nzyxW

1),,,( tzyxWW (2)

The Laplace’s differential equation can be solved by the application of the method of final difference method, finite elements method and boundary elements method. In this work, the solution of boundary element method is presented. The Laplace's differential equation for the stationary motion of an ideally incompressible

fluid (1) is expressed through function ),,,( tzyxWW and represents distribution of

313

hydrodynamic pressure in the domain. The solution of equation (1) is not possible without definition of the functions of the boundary conditions. There are two different types of boundary conditions as follows:

Boundary conditions of the “essential type”, when the values of the hydrodynamic pressure function ),,,( tzyxWW are defined in part of contour 1 of the domain, Fig. (1).

Boundary conditions of the “natural type”, when the values of the derivation of the hydrodynamic pressure

function are defined in part of contour 2 in the domain, Fig. (1). The derivation is in the direction of the normal to the surface is as follows:

ntzyxW

nzyxW ),,,(),,(

(3)

Figure 1: Boundary conditions at the BE model

The Laplace’s differential equation contains derivations of the hydrodynamic pressure of the second order. To apply the BE method, it is necessary to transform the derivations of the second order into derivations of the first order for decreasing the order of the Laplace’s differential equation. Applying the method of weight residuals upon the Laplace’s differential equation and the corresponding functions of the boundary conditions, and applying several mathematical transformations, equation 4, which is suitable for application of the boundary element method, is obtained.

22

11

)()(21 d

npWp

nWd

npWp

nWWm

(4)

Expression (4) represents a typical equation referring to one discrete point M located on contours 1 or 2. The digital form of equation (4) is obtained when the boundary areas of the type of 1 and 2 are represented as an assemble of small boundary areas, which represent, for us boundary elements.:

dBEnpWp

nWdBE

npWp

nWW

NEL

i

NEL

em )()(

21

2111 (5)

314

3. Analysis results Comparison is made between both, BEM and Westergaared solutions. The

comparison serves to some extend for verification of the solution obtained by the application of the BEL3D-ver1 program. The computation is achieved by application of the first 1000 terms of the infinite row of sine and cosine functions that constitute the Westergaard’s solution, using the simple program Westgard-IZIIS. Westergaard’s solution treats a rigid and ideally straight dam and assumes that the dam-fluid system manifests a stationary wavy motion of low speed, whereat the intensities of hydrodynamic pressure are given in function of the period of the harmonic motion of the fluid, fig(2). For this case, the applied acceleration is 1g. In order to compare Westergaard’s solution with the corresponding solution obtained by application of BEM fig (3), it is necessary, the BE model to be as much as it is possible close to the assumptions on which the Westergaard’s solution is based. For that reason applied is unit acceleration at all points of the dam, and the dam is treated as a rigid one. Furthermore, assumed are the following dimensions of the BE model, H=100m B = 100 m and L = 1000 m. The length of 1000m is sufficient to approach the assumption of infinitely long reservoir. However the width of 100m is not correct to simulate the 2D effect of the Westergaard solution. So we could discuss on this issue, below. As the period of the sinusoidal excitation is smaller, the length ( VsTl ) of the generated stationary wave of motion is shorter which result in higher mobility of the wave motion and bigger HD pressures on the dam face. According Westergaard solution there is a critical case when for the certain value of the dam height and the period of the sinusoidal excitation the effect of resonance in the water occur which results in extremely high pressures. In this case, for H=100m, the effect of resonance occur at T=0.269 sec. However, as the period of the sinusoidal excitation is bigger, the length of the generated stationary wave of motion is higher which result in inert wave motion and smaller HD pressures on the dam face. The BEM solution, treating the 3D effect and under assumption of rigid dam body, is closer to the Westergaard solution in case of inert wave motion. If the value of the model width gets small than 100m the solution would be much closer to the Westergaard.

Figure 2: Westergaard’s solution of HDP, H=100m and acc=1g in function of the vibration period (T) of the system dam-fluid - obtained by using the Westergaard-software-IZIIS

315

Figure 3: Hydrodynamic pressure [kN/m2]in a dam H=100m B=1000m L=1000m and acc=1g obtained by using the BEL3D –ver 1 software

Presented below is a discrete model of boundary elements and the intensities of developed hydrodynamic pressures calculated by applied unit acceleration at all point on the upstream face of the dam treating it as a rigid and using the BEL3D program, fig (6). The generation of the boundary elements of the extrados of the arch dam, both banks, the reservoir bottom, the water mirror and the end of the reservoir is automatic, is by direct extraction of the coordinates of the boundary elements from the general FE model, fig (4), which is generated for computation of the stress-strain state of the arch dam using the ARCH–ver2–IZIIS program. On figure (5) presented is BE mesh at the dam extrados face and both banks while on figure (6) presented is BE mesh at the reservoir bottom and the at the water mirror. The BE model is used

316

Figure 5: BEM model for solving dam-fluid interaction by use of software BEL3D –IZIIS boundary elements at the dam extrados face and both banks

Figure 6: BEM model for solving dam-fluid Figure 7: Hydrodynamic pressure[kN/m2]interaction by use of software BEL3D –IZIIS upon the arch dam. Hw=70m acc=1g boundary elements at the reservoir bottom (effect of the dam shape upon and water mirror pressure intensity)

4. Conclusion

The comparison of both solutions serves for verification of the solution obtained by the BEL3D software. Given is explanation for the difference in the results obtained by use of both methods.

References [1] Water Pressure on Dams During Earthquakes“,Westergaard H. M ,Trans. Am.

Soc. Civ. Eng., 98, 418-33, [2] Study on Boundary Element Method and Hydrodynamic Pressures , Mircevska,

V., Bickovski, V.,” Software BEL3D ver. 1, IZIIS, Skopje, 2007

317

Zoran Raki evi 1, Aleksandra Bogdanovi 2, Dimitar Jurukovski3

ASEIZMI KO PROJEKTOVANJE ELI NIH RAMOVSKIH KONSTRUKCIJA SA DODATNIM PRIGUŠENJEM

Rezime:

Koliko treba da bude ukupni kapacitet prigusuvanja viskoznih prigusivaca, da bi se zadovoljio EK8. Da se odgovori na ovo pitanje i da se demonstrira koncept projektiranja konstrukcije sa dodanim prigusivanjem, hipoteticka pet spratna celicna knstrukcija projektovana je prema propisima EK3 i EK8 za zabrzanja od 0.2g. Napravljena je serija analiza pri cemu je menjana velicina prigusivanja od 10-30% i ulazno zabrzanje do 0.6g. Bice demonstrirano da sa dodanim prigusivacima po spratovima mogu da budu isprojektovane bolje i sigurnije upravljane konstrukcije od sezmickih dejstava.

Kljucne re i: Celicne konstrukcije, kontrolirane konstrukcije, dodato prigusivanje.

ASEISMIC DESIGN OF STEEL FRAME STRUCTURES WITH ADDED DAMPING

Summary:

What should be the total damping capacity of these devices in order to meet the EC8 design requirements? To give an answer to this question and demonstrate this concept of structural design with added damping a hypothetical five storey three bay steel frame structure is designed for ground acceleration of 0.2g and 0.4g based on EC 3 and EC 8 regulations. A series of analysis have been performed varying the viscous damping of the dampers from 10-30%, as well as increasing the peak ground motion up to 0.6 g. It will be demonstrated that by adding of devices at the floor levels more advanced and safer structure could be designed and the response of the structure could significantly be improved.

Keywords : steel structures, controlled structures, added damping

1 Assoc. Prof. Dr. Institute of Earthquake Engineering and Engineering Seismology (IZIIS), Skopje, Republic of Macedonia, Zoran Rakicevic 2 Research Assistant, Phd student, Institute of Earthquake Engineering and Engineering Seismology (IZIIS), Skopje, Republic of Macedonia, Aleksandra Bogdanovic 3 Prof. Emeritus Institute of Earthquake Engineering and Engineering Seismology (IZIIS) Skopje, Republic of Macedonia, Dimitar Jurukovski

318

1. INTRODUCTION The behaviour of structural systems when responding to dynamic loads is mainly

associated with their ability to dissipate the kinetic and the potential energy through hysteretic and viscous mechanisms of the structures. Vibration of structures and associate forces could be reduced and controlled through damping of the structure. The damping compensates for structural nonlinearity through which the external energy should be dissipated (absorbed). Also, the effect of damping can be has influence on the vulnerability of the structure, particularly that of the non-structural elements, which means that the overall cost for retrofitting is also decreased.

In general, the damping in steel structures consists of inherent-viscous damping, which is usually low (1%-5% of critical), hysteretic damping through nonlinear behaviour of structural elements and damping as a result of added different energy dissipation systems.

According to Eurocode 8 (EC8) requirements, earthquake resistant steel buildings shall be designed in accordance of two concepts: a) low-dissipative structural behaviour and b) dissipative structural behaviour. In concept a), the action effects may be calculated based on an elastic global analysis without taking into account a significant non-linear material behaviour and this concept is recommended for designing of steel structures in low seismicity regions. In concept b) the capability of parts of the structure, so called “dissipative zones”, to resist earthquake actions through inelastic behaviour is taken into account. Structures designed in accordance with this concept belongs to structural ductility classes medium or high, which correspond to increased ability of the structure to dissipate energy in plastic mechanisms of the main structural elements.

Contrary to this approach, for proper seismic design, the amount of hysteretic energy dissipated by the structure has to be minimized, which means that additional damping has to be introduced in the structure.

Possibility of introduction of additional energy dissipating mechanisms into the structure, either passive or semi-active, which should be designed to consume a portion of the input energy, reduces the damage to the main structure caused by hysteretic dissipation. What should be the total damping capacity of these devices and their contribution in the overall effective damping of the structure in order to meet the EC8 design requirements, such as total top displacement, inter-story drifts, stress level in the main columns, beams, diagonal elements etc. represents a crucial question.

In this paper is presented a set of data by which it will be demonstrated that by adding of devices at the floor levels more advanced and safer structure could be designed and the response of the structure exposed to time dependent loads could significantly be improved. Comparative analysis of moment resisting frame and the same frame with added dampers, exposed to time dependent loads in linear and non-linear range is conducted. As much as the added damping is increased the structural response is improved resulting in a decrease of storey drifts and hysteretic damping demand towards its vanishing and linear structural response. Through an amount of added damping, it is possible to control the nonlinearity and storey drifts to a required level defined in the Eurocode 8.

319

2. DESIGN AND ANALYSIS OF HYPOTHETICAL STRUCTURETwo hypothetical 5 stories steel frame structures, with three bays, have been

designed as Moment Resisting Frame (MRF) according to EC8 and EC3 requirements for ultimate limit states and serviceability limit states, taking into account design response spectrum calculated for soil type B, PGA=0.20g, and PGA=0.40g damping of 2% and for two different q-factors, q=2 and q=4. So, total three steel frame structures have been designed. For all structures the bay length is 6.0 m and storey height is 3.0 m. The SAP2000 computer program has been used for modelling and optimizing of the structural sections during the design phase and preliminary linear response history analysis only. The masses representing the weight at each floor level, including the weight of the beams and the columns and a portion of live load (24%), are concentrated at the beam-column joints. The total storey mass is 62.10 t. Beams and columns are modelled as frame elements with specified end length offsets and rigid-end factors, typically taken as 0.7. For such a model, at the final design phase it was found that the period of the first mode of vibration for the 5 storey structures designed for PGA=0.2g are T1=1.00 s and T1=1.38 s, designed for q=2 and q=4 respectively. For the 5 storey structure designed for PGA=0.4g and q=4 the first period of vibration T1=1.011 s

Modelling and analysis of the designed steel frame structures for the purpose of non-linear response history analysis was done using computer program NONLIN-Pro, Ref [4], in which the analysis engine is the DRAIN-2DX computer program. Both steel frame structures are designed as 2D structures, using several types of elements that are available in DRAIN-2DX program. Namely Beam and columns are modelled as plastic hinge beam-column elements Type 02, taking into account the axial-flexural interaction for columns. The columns were modelled using the built-in yielding functionality of the DRAIN-2DX program, wherein the yield moment is a function of the axial force in the column. All beams are modelled to respond linearly, since for beams with flexural yielding that is independent of axial force, it better to explicitly model the hinges using simple connection element Type 04 as explained ad in the following context. The plastic hinges, located 30 cm away from the column face, in all beams are modelled using zero length rotational connection elements (springs), which means that 100% of the inelastic rotation is assumed to occur in the rotational plastic hinges. It has to be pointed out that DRAIN-2DX does not have the capability to model loss of strength after first yielding, so it is assumed bilinear, inelastic moment-rotation behaviour for the spring having 3% post yielding stiffness ratio. Since a very significant portion of the total story drift of a moment-resisting frame may be due to deformations that occur in the panel zone region of the beam-column joint, the panel zones are modelled using an approach developed by Krawinkler, Ref [4]. In DRAIN-2DX it consists of a “frame” of Type 02 beam-column elements connected at the four corners by compound nodes. The upper left compound node utilizes a rotational Type 4 spring to represent the panel zone web stiffness and strength. The upper right compound node utilizes a Type 04 rotational spring to represent column flange contributions. The other two compound nodes are simple flexural hinges. So, each panel zone was modelled using 12 elements and 12 nodes, and has 28 degrees of freedom. Also, in each panel zone 20 mm thick doubler plate has been taken into account. Both rotational springs were modelled assuming bilinear, inelastic moment-rotation behaviour for the spring having 3% and 1% post yielding stiffness ratio for flange component and for panel component, respectively.

320

Figure 1: Mathematical models for non-linear response history analysis

The mathematical model with viscous dampers (VISC) is presented in Figure 1. The main structural elements beams, column and panel zones are the same as for MRF model. The viscous dampers are located in the middle bay along the height of the model and are modelled using two inelastic truss bar elements Type 01 in parallel. Both springs are modelled to respond linearly having very low stiffness, k = 0.01 kN/cm, with that difference that one has very high beta value (element stiffness proportional damping factor), and the other has beta value zero. The product of the stiffness and the beta value is equal to the desired damping coefficient, C. The use of a very low stiffness is consistent with the behaviour of a viscous fluid damper which has a near zero storage stiffness (if excited below its cut off frequency).

The diagonal viscous dampers have linear force versus velocity relationship, so the damper force is expressed as:

dj j djF C u (1)

where Cj is the damping coefficient for the damper at level j, while dju is the relative velocity between the damper joints along its axis. The added viscous damping, for each separate vibration mode can be calculated according to: Ref [5]

2 2

12

cos,

4

j j rjjm

vm rj jm j mi im

i

CT

m (2)

where, Tm undamped natural frequency of the m-th mode, cos is the damper inclination angle, im are the ordinates of the m-th undamped mode, rj is the modal drift and mi is the

321

storey mass. For the analysis the Cj coefficient is the same for all dampers along the height of the structures and it is derived from Eq. (2) in order to introduce a predetermined additional damping in the first fundamental mode (10, 20 and 30% of the critical for each analysed case respectively.

Table 1: Earthquake records used in analysis Earthquake Registration PGA (g) Name Taiwan 1999 Chi-Chi CHY028 NS 0.82 CHI Imperial Valley 1940, California El Centro Copmp180 0.35 ELC Erzincan 1992 Turkey Mudurlugu N279 0.51 ERZ Loma Prieta 1989 Santa Cruz Mountains CSMIP 4725 Comp 0 0.37 LOP Hyogo-Ken Nan-Bu , Kobe Japan 1995 Kobe University N-S 0.28 KOB Montenegro 1979 Petrovac-Oliva NS 0.45 PET Montenegro 1979 Ulcinj-Albatros EW 0.22 UAL Montenegro 1979 Ulcinj-Olimpic NS 0.29 UOL San Fernando 1979 OWNER 0241 Comp 360 0.25 SAN Northridge 1994 Sylmar LADWP 306 Comp S38E 0.75 SYL

Modelling of inherent damping of 2% of critical in the first fundamental mode was done through mass proportional damping only, for both MRF and VISC models. The values of stiffness proportional damping factor for all elements in both models, excluding the viscous damping elements, are taken to be zero.

Ten different time history records of real earthquakes, given in Table 1, taking the first 40 s of shaking duration have been used in the analysis. The earthquakes have been chosen based on their frequency content and elastic response spectra, as well as taking into account dynamic properties of the four designed MRF structures.

Non-linear response history analysis has been conducted on three structures two 5 storey, MODEL05_q2 and MODEL05_q4, and one 5 storey, MODEL05A_q4, where “q2 or q4” means which q-factor have been used in the design process. Further the response history analysis was done for constant time step dt = 0.001, for scaled earthquake time histories to PGA=0.2g, 0.4g and 0.6g, and all model have been analysed as MRF(0% of added viscous damping), while the first two models, MODEL05_q2 and MODEL05_q4 have been analysed for 3 values of added viscous damping, 10%, 20%, 30% of critical (VISC). Since for all 3 structures the first natural mode of vibration is dominant and has approx 80% participation in the structural response the damping coefficients for the viscous dampers are obtained from the condition to introduce additional damping 10, 20, and 30% of critical damping in the first mode of vibration. This means that for one structure total number of analyse are 120, and the total conducted analyses for all structures is 360.

3. ANALYSIS RESULTS Having in mind above, analysis results will be presented and discussed for

maximum storey drifts, base shear forces, top displacements and top accelerations and for 5 selected earthquakes (CHI, ELC, ERZ, PET and SAN) from Table 1.

322

In general, for all selected five earthquakes scaled to 0.2g the storey drifts for the 5 storey structures (MODEL05_q2 and MODEL05_q4) are within the storey drift limit for elastic analysis defined in EC8 (0.01h = 3.0 cm), for MRF and VISC models. The exception from this is the case for MODEL05_q4 exposed to SAN earthquake scaled to 0.2g for MRF and VISC model with 10% of added viscous damping only. For MODEL05_q2, in the case for MRF, for all 5 earthquakes scaled to 0.4g all storey drifts are larger up to 1.5 times than EC8 limit. The inertial base shear force is within the range of 24% - 35%, the top acceleration is between 0.83g to 1.03g, while the top displacement is in the range 15.50 cm – 17.21 cm.

In Figure 2 are presented the envelopes of maximum storey drifts, along the height, as line graphs, as well as the extreme values of inertia base shear force, top displacement and top acceleration in tabular form, for MODEL05_q2 exposed to PET earthquake scaled to 0.2g and 0.4g for MRF case (0 added damping) and VISC case (10, 20, 30 % added damping).

Figure 2: Maximum storey drifts, base shear force, top displacement and top acceleration for MODEL05_q2, Petrovac earthquake scaled to PGA=0.2g and 0.4g

In Figure 3 are presented the envelopes of maximum storey drifts, along the height, as line graphs, as well as the extreme values of inertia base shear force, top displacement and top acceleration in tabular form, for MODEL05_q4 exposed to ELC earthquake scaled to 0.2g and 0.4g for MRF case (0 added damping) and VISC case (10, 20, 30 % added damping).

For the same model, in the case for VISC, all storey drifts are smaller than EC8 limit, and for added 20% of damping the storey drifts are between 1.5 cm and 2.5 cm. For the same added damping the inertial base shear force is within the range of 18% - 27%, the top acceleration is between 0.62g to 0.79g, while the top displacement is in the range 8.32 cm – 10.74 cm.

For MODEL05_q4, in the case for MRF, for all 5 earthquakes scaled to 0.4g all storey drifts are larger up to 2.4 times than EC8 limit. The inertial base shear force is within

323

the range of 20% - 24%, the top acceleration is between 0.73g to 0.94g, while the top displacement is in the range 13.83 cm – 21.12 cm. In the case for VISC and for added 20% of damping, the storey drifts are smaller than EC8 limit for CHI, ELC and PET earthquake. For ERZ and SAN earthquake the drifts from first to third storey are above the EC8 limit, up to 1.5 times at second storey. The inertial base shear force is within the range of 17% - 24%, the top acceleration is between 0.64g to 0.81g, while the top displacement is in the range 8.98 cm – 17.30 cm.

Figure 3: Maximum storey drifts, base shear force, top displacement and top acceleration for MODEL05_q4, El Centro earthquake scaled to PGA=0.2g and 0.4g

Figure 4: Maximum storey drift at 4-th storey, for MODEL05_q2 and MODEL05_q4, for 5 selected earthquakes scaled to PGA=0.2g, 0.4g and 0.6g and for added damping of 20%

In Figure 4 are presented the envelopes of maximum storey drifts, for the 4-th storey, for all 5 selected earthquakes scaled to 0.2g, 0.4g and 0.6g, for the MODEL05_q2 and MODEL05_q4 with added 20% of viscous damping. It can be seen that only for CHI and SAN earthquakes scaled to 0.6g the 4-th storey drift is slightly above the EC8 limit.

324

DR

IFT

(cm

)

CHI ELC ERZ PET SAN CHI ELC ERZ PET SAN

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

q=2

0.2g0.4g0.6g

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

q=4

0.2g0.4g0.6g

4.50

Figure 5: Maximum storey drift at 2-th storey, for MODEL05_q2 and MODEL05_q4, for 5 selected earthquakes scaled to PGA=0.2g, 0.4g and 0.6g and for added damping of 20%

Figure 5 shows the envelopes of maximum storey drifts, for the 2-nd storey, for all 5 selected earthquakes scaled to 0.2g, 0.4g and 0.6g, for the MODEL05_q2 and MODEL05_q4 with added 20% of viscous damping. As presented, for q=2, for CHI and SAN earthquakes scaled to 0.6g the 2-nd storey drift is above the EC8 limit; while for q=4 for all five earthquakes scaled to 0.6g the obtained drift for the second storey is above the EC8 limit.

DRIFT (cm)

STO

RE

Y

0 0.5 1 1.5 2 2.5 3 3.5 40

1

2

3

4

5

0 0.5 1 1.5 2 2.5 3 3.5 4 4.50

1

2

3

4

5

0102030EC8PGA = 0.2g PGA = 0.4g

BSF (%)TD (cm) TA (g)

0 18.66 12.19 0.82

10 12.79 6.58 0.36

20 12.27 5.17 0.35

30 11.80 4.33 0.34

BSF (%) TD (cm) TA (g)

0 34.64 14.28 1.14

10 24.82 12.28 0.71

20 24.06 9.93 0.69

30 23.46 8.56 0.69

PGA = 0.2g

PGA = 0.4g

Figure 6: Maximum storey drifts, base shear force, top displacement and top acceleration for MODEL05A_q4, Petrovac earthquake scaled to PGA=0.2g and 0.4g

Figure 6 compares the envelopes of maximum storey drifts, along the height, as line graphs, as well as the extreme values of inertia base shear force, top displacement and top acceleration in tabular form, for the studied steel frame structure designed for 0.4g

325

exposed to Petrovac earthquake scaled to 0.2g and 0.4g for MRF case (0 added damping) and the structure designed for 0.2g VISC case (10, 20, 30 % added damping).

In general, for the model designed to 0.4g, for all selected five earthquakes scaled to 0.2g the story drifts for the five story structure (MODEL05A_q4) are within the story drift limit for elastic analysis defined in EC8, for MRF and VISC model. For the same model in the case for MRF, for all 5 earthquakes scaled to 0.4g, all story drifts are larger up to 1.7 times than EC8 limit. The inertial base shear force is within the range of 23%-37%, the top acceleration is between 0.84g to 1.14g, while the top displacement is in the range 14.28cm – 19.74cm.

In the case for VISC all storey drifts are smaller than EC8 limit, and for added 20% of damping the storey drifts are between 1.5 cm and 2.5 cm. For the same added damping the inertial base shear force is within the range of 18% - 27%, the top acceleration is between 0.62g to 0.79g, while the top displacement is in the range 8.32 cm – 10.74 cm.

STO

RE

Y

0 0.5 1 1.5 2 2.5 3 3.5 40

1

2

3

4

5

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

1

2

3

4

5

PGA = 0.2g PGA = 0.4g

0102030EC8

DRIFT (cm)

BSF (%) TD (cm) TA (g)0 18.39 12.39 0.64

10 10.49 5.97 0.45

20 10.43 4.61 0.41

30 10.13 3.87 0.38

BSF (%) TD (cm)TA (g)

0 30.25 14.61 0.97

10 20.41 10.65 0.89

20 20.16 8.98 0.81

30 19.74 7.87 0.75

PGA = 0.4g

PGA = 0.2g

Figure 7: Maximum storey drifts, base shear force, top displacement and top acceleration for MODELL05A_q4, El Centro earthquake scaled to PGA=0.2g and 0.4g

In Figure 7 are presented the envelopes of maximum storey drifts, along the height, as line graphs, as well as the extreme values of inertia base shear force, top displacement and top acceleration in tabular form, for the studied steel frame structure designed for 0.4g exposed to El Centro earthquake scaled to 0.2g and 0.4g for MRF case (0 added damping) and the structure designed for 0.2g VISC case (10, 20, 30 % added damping).

It has to be pointed out that both structures with added viscous damping of 20%f and 30% for all 5 presented earthquakes when scaled to 0.2g responded in linear range. For the input level twice as much as designed one, 0.4g, both structures with added viscous damping of 20% and 30% respond approaching the linear limit, slightly above it or with minor damages. MRF structures, MODEL05_q2 and MODEL05_q4, for the input of 0.4g collapsed or suffer heavy damages which are not economically repairable. For the

326

MODEL05A_q4 MRF structure, for the input of 0.4g the structure is not in the high damage limit, which is not case for the 0.6g where the structure suffers from heavy damages and collapsed. This point out that the added 20% of viscous or any other type of damping, by introducing passive or semi-active devices, in existing or new structures, compensates for the structural nonlinearity by which the external energy should be dissipated (absorbed) and could withstand earthquake which is twice as much as designed one. Also, the effect of added damping can be expressed through decreasing of vulnerability of the structure, particularly that of the non-structural elements, which leads to a decrease of the overall cost for retrofitting.

CONCLUSION Possibility of introduction of additional energy dissipating mechanisms reduces

the damage of the structure caused by hysteretic dissipation. By analytical simulation it was demonstrated that the optimal damping capacity of these devices in order to meet the EC8 design requirements, such as inter-storey drift is around 20%. It was demonstrated also, that the structures with added dampers that can introduce at least 20% of damping can withstand earthquakes with intensities twice as much as designed one, having no or minor damages. It was also noticed that when the dampers has the same mechanical properties along the height of the structure the storey drift decreases as the height increases. This means, by proper optimization of damping devices more favourable reduction of storey drifts could be achieved, which leads to lower cost solution for the damping devices. Also, the effect of added damping can be expressed through the decrease of the vulnerability of the structure, particularly that of the non-structural elements, which leads to a decrease of the overall cost for retrofitting.

Unfortunately, although this technology has been proven, both analytically and practically, to be very effective still it is not covered by the new European Codes for earthquake resistant design. On the other hand, USA and Japan have advanced in this field by making provisions, guidelines and recommendations.

327

REFERENCES [1] EN 1998-1 Eurocode 8: Design of structures for earthquake resistance – Part 1: General rules, seismic actions and rules for buildings, December 2004. [2] EN 1993-1-1 Eurocode 3: Design of steel structures – Part 1-1: General rules and rules

for buildings, May 2005. [3] NEHRP Recommended Provisions for seismic regulations for new buildings and other structures, FEMA 450, Part1: Provisions and Part2: Commentary – 2003 Edition. [4] NEHRP Recommended Provisions: Instructional Materials, FEMA 451B – June 2007. [5] Development and Evaluation of Simplified Procedures for Analysis and Design of Buildings with Passive Energy Dissipation Systems Ramirez, O.M., Constantinou, M.C., Kircher, C.A., Whittaker, A.S., Johnson, M.W., and Gomez, J.D., , Technical Report MCEER-00-0010, SUNY, Buffalo, 2000. [6] Passive Energy Dissipation Systems in Structural Engineering Soong, T.T., and Dargush, G.F.,John Wiley & Sons, 1997. [7] Supplemental Energy Dissipation: State-of-the-Art and State-of-the-Practice, Engineering Structures 24,Soong, T.T., and Spencer, B.F. Jr, pp 243-259, 2002. [8] From Ductility Demand to Flexibility and Damping Demand, Iemura, H., , Proc. Of 3rd

WCSC, Vol. 1 pp 85-94, Como, Italy, 2002. [9] Optimum Design of Passive Controlled Steel Frame Structures, Rakicevic, Z., and Jurukovski D.IZIIS Report 2001-59, Skopje, Republic of Macedonia, 2001. [10] Effectiveness of Viscous Damping in Controlling Storey Displacement, Rakicevic, Z., and Jurukovski D., Proc. Of 3ECSC Vol. 2 S6-60 – S6-63, Vienna, Austria 2004. [11] Optimal Damping Capacity of Steel Frame Structures, Rakicevic, Z., Jurukovski D and Zlateska A., Proc. Of 4ECSC St Petersburg, Russia, September 8-12 2008. [12] Earthquake resistant design needs structural control, Rakicevic, Z., Jurukovski D and Zlateska A., 14WCEE, Beijing, China, October 12-17 2008.

329

Viktor Hristovski14, Marta Stojmanovska25, and Mihail Garevski3

1 : (XLAM)

:

(XLam) .

. 2

, ,2005

: , , ,

PART 1: PROPOSED CONSTITUTIVE RELATIONSHIPS FOR CONNECTIONS OF XLAM PANELS Summary

In this paper, new computational constitutive relationships for a semi-rigid type joint applied in cross-laminated (XLam) panels have been developed and implemented into a computer program. Friction has also been taken into account as an important modeling parameter. In Part II a comparative numerical study has been performed using quasi-static test results from experiments on panels performed in the Laboratory of the Faculty of Civil and Geodetic Engineering in Ljubljana, Slovenia 2005. Keywords: constitutive relationships, friction, semi-rigid joints

1 PhD, Professor, University “Ss.Cyril and Methodious”, IZIIS Skopje, Macedonia 2 MSc, Research Assistant, University “Ss.Cyril and Methodious”, IZIIS Skopje, Macedonia 3 PhD, Professor, University “Ss.Cyril and Methodious”, IZIIS Skopje, Macedonia

330

1 INTRODUCTION XLam structures (made of cross-laminated wooden panels) are becoming widely

accepted constructive systems in Europe due to their obvious advantages (see Ceccoti et al. 2006) making them one of the best choices among timber-based systems, particularly, when speaking of their application in seismic prone areas. However, research on the behavior of these systems under seismic loadings for the sake of establishing practical design rules, which will improve the existing Euro-codes (5 and 8), is still needed. On the other hand, the computational constitutive modeling of new materials and connections has always been a challenging task. In this paper, an attempt to propose constitutive relationships for connections between panels and foundations has been made.

Generally, the behavior of timber buildings consisting of massive wooden wall panels during earthquakes is influenced by a number of factors. Usually, if the panel is treated as a rigid one compared to the connections, the following types of the panel's response behavior can be expected: 1. Pure rocking behavior; 2. Mixed rocking and shear behavior and 3. Pure shear behavior (see Dujic et al. 2005). Which of these types of behavior will occur, depends on several factors. First, the axial load level is one of the most important factors influencing the racking strength of the panel elements and also determining their response behavior mode. The geometry and the ratio between the height and the width of the panel element can also be crucial. If the panel width is much greater than the panel height, the panel will behave closer to the shear mode. The existence of openings can reduce the panel stiffness, contributing to a kind of a bending mode of behavior, different than the previously mentioned modes. In reality, for lower XLam buildings (which is the usual case) it can be assumed that the dominant mode of behavior of the panels will be pure rocking or mixed (rocking and shear) behavior, since the self-weight will not constrain the rocking degree of freedom.

Usually, in practice the panels are designed to be much stiffer and stronger than the connections in order that the seismic energy be dissipated through the link devices, rather than through the panel that could lead to brittle failure mode. For this reason, the XLam panel is treated to behave as a linear-elastic orthotropic material that is usually a very realistic assumption. Regarding the connections, semi-rigid types of anchors have been investigated (see Dujic et al.2006) and inelastic constitutive relationships for separation and sliding have been proposed. These types of anchors are usually installed at a few specified places along the contact surface (in our examples, 4 for each panel element) and the remaining area of the contact zone between the wood and the concrete contributes to the panel racking strength by friction force. Observing the contact zone, it can be concluded that the greater vertical load is applied, the greater friction force is obtained. However, the friction force can exist only in cases where the contact between the panel and the foundation is fully established, i.e. where there is compression state of stresses. For the case of separation, the friction forces are equal to zero. In this study, the friction has been treated using the foregoing mentioned assumptions, formulating it to behave in elastic-plastic manner with variable limit of the full friction force depending on the actual normal force during the seismic response and predefined friction coefficient.

The proposed constitutive relationships have been implemented into the FELISA/3M software package (Hristovski and Noguchi 2002). In Part II their verification, using the tests performed in Laboratory of the Faculty of Civil and Geodetic Engineering in Ljubljana, Slovenia 2005 (Dujic et al. 2005, Dujic et al. 2006) on massive cross-laminated wooden panels with a semi-rigid type of anchorage system, have been made.

331

2 PROPOSED CONSTITUTIVE RELATIONSHIPS FOR TIMBER, ANCHORS, CONTACT AND FRICTION BETWEEN PANELS AND FOUNDATIONS

2.1 CONSTITUTIVE RELATIONSHIPS FOR THE TIMBER MATERIAL

The behavior of the timber is assumed to be orthotropic linear-elastic, in spite of the fact that for shear forces near the ultimate racking capacity, the panel could crack in the vicinity of the connections. However, we can justify the linear-elastic assumption by considering these failure modes as local ones, treating the local cracked zones as parts of connection devices. On the other hand, having in mind the seismic design strategies based on strong panels and weak connections, the post-cracking timber behavior is not of practical interest at the moment but probably it could be of some academic interest in the future.

In this study, the following four independent constants (properties) have been adopted for the wooden panel material in the orthotropic plane-stress situation, according to the experimentally obtained ones (Dujic et al. 2005). The orientation of the material axes is given in Fig. 1, direction 1 is parallel to the x axis, and direction 2 is parallel to the y axis.

Fig. 1. Orientation of the material axes

From the elasticity theory, the basic constitutive relations are: D

(2)

where D is the elastic constitutive matrix. For orthotropic case, it can be found (with E1/E2=n and G2/E2=m) that:

22

2

2

22

2

100010

1nm

nnn

nED

(3)

22 /900 cmkNEE y

21 /445 cmkNEEx

neglectedEz2

2 /50 cmkNG25.02 (1)

332

where

xy

y

x

are in-plane stresses, and

z

xy

y

x

are strains.

2.2 PROPOSED CONSTITUTIVE RELATIONSHIPS FOR THE ANCHORS

The real behavior of anchors under pulling-out and slippage was investigated and described by the basic tests performed in Ljubljana 2005 (Dujic et al. 2005). The experimental setup configuration (Fig. 2) consisted of small panels with semi-rigid anchors fixed to the panel by 4 mm nails with a length of 40 mm and to the foundation with two bolts M12 (Fig. 3). The pullout (tension) tests were performed with boundary conditions to simulate anchor's behavior in the case of pure panel rotation, while shear tests were carried out to simulate anchor's behavior in the case of pure shear deformation.

Fig. 2. Basic tests on small Xlam panels with semi-rigid anchors a) Tension Test; b)

Shear Test

Fig. 3. The considered semi-rigid type of anchor fixed to a wooden panel by nails and to concrete foundation by bolts

a)b)

333

The behavior of the anchors is described by the specially proposed computational constitutive relationships for both tangential and normal directions, as approximations of the basic tests. In tangential direction, for shear force - slip modelling, a bi-linear envelope with a descending branch has been proposed (Fig. 4) in both positive and negative directions. Both directions are symmetrical. The accumulated slippage is the principal physical parameter for the control of the cyclic rules proposed within the model. The idea is that the connection in tangential direction cannot support any shear force unless the absolute value of the actual slip is greater than the absolute value of the previous accumulated slip, as experimentally observed (see the basic experimental diagram in Fig. 6). In normal direction, for tension, a kind of a hook-type of behavior is proposed for the anchor, i.e., the real working diagram is approximated with a linear-elastic-plastic-softening diagram according to the basic test on tension (see basic experiment diagram in Fig. 7), however in compression it is assumed to be linear-elastic (Fig. 5). It is important to emphasize that both the proposed constitutive models contain descending lines, which are necessary to model the post-peak behavior of the system (for cases where timber cracking does not occur and consequently does not have influence on the overall structural post-peak behavior). The cyclic rules for tangential and normal direction are given in Fig. 4 and Fig. 5, respectively.

1

2

34 5

6 7

11

s [m]

12

1415

1617

13

T [k

N]

Tu

Ty

-Ty

-Tu

sy su s0

-sy-su-s0

y

y

u

u

Fig. 4. Proposed computational constitutive relation shear force - slip for considered anchors

1 2

34

5

6

u[m]

N [k

N]

Ny

uy u0

y

Tens

ion

(+)

Com

pres

sion

(-)

(separation)(sinking)

Fig. 5. Proposed computational constitutive law axial force - separation of considered anchors

334

The connection device itself has been modelled using a standard link element, with zero length and consisting of two points, each having two degrees of freedom, translations in horizontal and vertical direction (Fig. 8).

-16

-14

-12

-10

-8

-6

-4

-2

0

2

4

6

8

10

12

14

16

-30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30

Shear Slide [mm]

Shea

r Fo

rce

on A

ctio

n [k

N]

times

Fig. 6. Experimentally obtained constitutive behavior shear force – slip (cyclic quasi-static test) of semi-rigid types of anchors(nails 4.0/40mm)given in Fig.3

-4

-2

0

2

4

6

8

10

12

14

16

18

20

0 1 2 3 4 5 6 7 8 9 10 11 12

Up-lift of the Panel [mm]

Ten

sion

For

ce o

n A

ctio

n [k

N]

Fig.7. Experimentally obtained constitutive behavior tension force-separation (cyclic quasi-static test) of semi-rigid types of anchors(nails 4.0/40mm)given in Fig.3

Fig. 8. Formulation of the link element for connections as well as for gap/friction

335

For the element stiffness matrix in a global coordinate system, the following can be written:

3232

2121

3232

2121

ssssssssssssssss

K e

where

223

2

221

cossin

sincoscossinsincos

sn

sn

sn

KKsKKs

KKs (5)

Parameter nK represents the tangent stiffness in axial direction; sK is tangent

stiffness in tangential direction. represents the directional angle of the element related to the global x-axis (see Fig. 8). The stiffness parameters nK and sK , together with the stress-update (in terms of element forces T and S) which are the necessary input for the Newton-Raphson iterative procedure for solution of nonlinear system equations, can be obtained from the previously defined constitutive laws (as first derivations of the constitutive law functions).

2.3 PROPOSED CONSTITUTIVE RELATIONSHIPS FOR CONTACT AND FRICTION

For the sake of computational simplicity, the first constitutive and numerical FEM models implemented by the authors in analysis of XLam panels and structures took into account the friction only as a constant parameter along the contact zones which did not change during the system response (Dujic et al. 2006, Hristovski and Stojmanovska 2005). For this purpose Dujic at al. (2006) used SAP2000 (2004) software package and Hristovski and Stojmanovska (2005) used FELISA/3M program (Hristovski 2003). However, using this approach, although the correlations with the experimental tests in relation with the numerically obtained pushover force-displacement curves were satisfactory, the pullout response and consequently the rocking response of the walls was underestimated, leading toward a conclusion that a pure sliding failure mode mechanism occurred, which was not true. One of the conclusions after this previous research was that the applied simplified friction model overestimated the friction force at places with small axial force (or even tension) and underestimated the friction force at places where intensive compression occurred (particularly at the points of rotation). Hence, in this research, a new friction model has been considered, taking into account the relation between the variable (actual) axial force and the friction force using the Mohr-Coulomb friction model. In this way, although the degrees of freedoms of pullout and sliding have been separated in the FEM analysis (using the assumption of uncoupled equations), these two degrees of freedom have practically been connected by the Mohr-Coulomb's model. In the numerical implementation, a uni-axial elastic-perfectly plastic relation has been assumed (Fig. 9), however with a variable yielding shear force Ty , depending on the actual axial force N on compression, as follows:

336

NfTy (6)

where f is a friction constant depending on the contact surfaces. Of course, for cases when the axial force is in tension, Ty will be equal to zero. For the sake of simplicity, in Fig. 12, the constitutive law is presented for the case when Ty is constant. However, Tychanges, according to Eq. (6), and in that case, the constitutive law will be much more complex. Also, although the perfect-plastic behavior is the real nature of friction, for numerical reasons only, a narrow elastic part, having some real value of elastic stiffness has been added herein.

In the already mentioned previous investigation, the friction constant was assumed to be equal to f = 0.3 (Hristovski and Stojmanovska 2005) and f = 0.7 (Dujic et al. 2006) for the concrete-wood contact. The best is if this coefficient can be determined experimentally. However, since the considered experimental tests (Dujic et al. 2005, Dujic et al. 2006) have not covered this parameter, in this research preliminary parametric analyses have been performed in order to obtain the most realistic value for the friction coefficient f. These analyses have showed that the best fitting with the experimental results has been achieved with the friction coefficient of f = 0.1. Therefore, this value for the friction coefficient has been adopted in this research.

Apart from friction, which acts tangentially on the contact zones, the separation and sinking have also been modelled assuming "gap" behavior of the contact. It means that, for the case of compression force acting on the contact, the stiffness becomes infinite (the numerical model assumes some finite value) and the behavior is linear-elastic. However, for the case of tension, the contact does not bear any force and consequently, separation occurs (Fig. 10). This model for separation has been applied along the whole contact zone together with the friction model using one link element for both degrees of freedom. The two degrees of freedom (sinking or separation in the "gap" normal model and slippage in the tangential, friction model) are connected via axial force acting on the contact, as explained previously (see Eq. 6).

T (friction)

S(slippage)

Ty

Ty

Fig. 9 Constitutive relationship for friction (Friction force – slippage) , assuming elastic-perfect-plastic behavior

Using these models for contact and friction, more realistic simulation of the panel behavior has been obtained. The more correct modelling of the friction has contributed toward emphasizing the rocking response of the panel, behaving as a rigid body rotating

337

around one point (especially for the solid wall specimens, without openings), together with some amount of sliding, which was observed by the experimental tests. The numerical FEM models and the obtained results are presented in the third section.

Compression (sinking) Tension (separation)

N

T (friction)

NCom

pres

sion

Te

nsio

n

SeparationSinking

Fig. 10 a) Constitutive relationship for contact zone in normal direction (axial force – separation , and axial force – sinking), assuming "gap" behavior, b)

Contact zone

2.4 SOFTWARE IMPLEMENTATION OF THE PROPOSED CONSTITUTIVE MODELS

The software implementation of the proposed constitutive models has been done using the general computer program for structural analysis FELISA/3M developed by Hristovski (2003). Newly coded subroutines written in Fortran 95 programming language for stress update to follow the material constitutive laws for anchors and contact zones have been incorporated into the program. These newly added subroutines have been designed to work together with the link elements that already existed within the FELISA/3M finite element library. The basic role of these subroutines is to calculate the stresses in each link element for each incremental step of the nonlinear analysis, based on the calculated displacements at the structural level. Also, the stiffness coefficient Kn and Ks (see Fig. 8), which are necessary for assembling the new tangent stiffness matrix of the system for the next iteration or step, are the output of these subroutines. Then, based on the calculated stresses, the unbalanced forces are calculated. The nonlinear procedure is iterative and based on the Newton-Raphson scheme, using the displacement control option, which is suitable for pushover analyses in order to obtain the descending branch of the force-displacement diagrams. This is consistent with the conducted cyclic quasi-static testing procedures on the panels (Dujic et al. 2005), where the displacement control was also used. These experimental tests have been used for verification of the proposed models. The results of the analyses using the FELISA/3M software package are presented in the next chapter. For the dynamic analyses, the force control procedure has been used, which was found to be more suitable for this case. The results from the dynamic analyses of a realistic 2D three-story wooden structure are presented in the fourth section.

a)

b)

338

CONCLUSIONS New computational constitutive relations for a semi-rigid type of connections

applied in XLam panels have been developed and verified using the results from the previously preformed experimental investigation in Laboratory of the Faculty of Civil and Geodetic Engineering in Ljubljana, Slovenia 2005 (Dujic et al. 2005, Dujic et al. 2006).

Based on the obtained experimental results and using the FELISA/3M computer package, finite element models have been developed. Special numerical constitutive relations, describing the connections behaviour in normal and tangential direction, have been used. Having the obtained hysteretic responses approximation numerical models were defined taking into consideration the curve hardening as well as the softening. Additionally in order to achieve a more realistic simulation of the panel’s behaviour, constitutive relationships for contact and friction have been developed. Uni-axial elastic-perfectly plastic friction model has been applied with friction coefficient between wooden panel and concrete foundation equivalent to 0.1. For a contact zone in normal direction, a model for separation has been employed together with the friction model using one link element for both degrees of freedom. In this way, the rocking response of the panel has been pointed out together with some amount of sliding, as observed during the experimental tests. In Part II the proposed constitutive models will be incorporated into the FELISA/3M program and pushover and dynamic analyses are going to be run.

Acknowledgements This research has been carried out within the bilateral scientific cooperation

between Republic of Slovenia and Republic of Macedonia. The support of the Ministries of Education and Science of both countries is gratefully acknowledged. Also, the authors would like to express special thanks to Dujic, B. and Zarnic, R. for the possibility of using their test results from the Ljubljana tests performed in 2005.

REFERENCES [1] “Seismic Behavior of Multi-Storey XLam Buildings”, A.Ceccotti and M.Follesa Proc. Of the International Workshop on Earthquake Engineering on Timber Structures, 2006, Coimbra, Portugal, pp. 81-95.[2] “Investigation on In-Plane Loaded Wooden Elements Influence of Loading and Boundary Conditions” B.Dujic, S. Aicher, and R. Zarnic, Otto-Graf-Journal (Otto-Graf Institute, MPA University, Stuttgart) 2005, 16: 259-272. [3] “Influence of Openings on Shear Capacity of Massive Cross-Laminated Wooden Walls” .B.Dujic, S. Klobcar, and R. Zarnic, Proc. of the International Workshop on Earthquake Engineering on Timber Structures, 2006, Coimbra, Portugal, pp. 105-118. [4] “Comparative Study of FEM Based Reinforced Concrete Analytical Models and Their Numerical Implementation: Software Package FELISA/3M” V.Hristovski, and H.Noguchi, Proc. of the 1st Fib Congress, Section 13: Failure mechanism and non-linear analysis for practice, 2002, Osaka, Japan, pp. 403-410. [5] “Experimental and Analytical Evaluation of the Racking Strength of Massive Wooden Wall Panels -Preliminary Project Phase” V.Hristovski and M.Stojmanovska, Proc. of the Earthquake Engineering in 21st century Conference (EE21C), 2005, Skopje-Ohrid, Macedonia

339

Viktor Hristovski16, Marta Stojmanovska27, and Mihail Garevski3

2 :-

(XLam). 1 (XLam)

. , 2

,, 2005.

.: pushover , ,

PART 2: PROPOSED CONSTITUTIVE RELATIONSHIPS FOR CONNECTIONS OF XLAM PANELS-FE IMPLEMENTATION

SummaryThis is the second of two companion papers on the constitutive relationships for connections of wooden cross-laminated (XLam) wall panels. In Part 1 new computational constitutive relationships for a semi-rigid type joint applied in XLam panels have been developed and implemented into a computer program. Here in Part II attention turns to an implementation and verification of the proposed constitutive models using quasi-static test results from experiments on panels performed at the ULFGG Ljubljana, Slovenia 2005. The obtained results from the analyses have proved the applicability of the proposed constitutive and FEM analytical models. Keywords: pushover analysis, dynamic analysis, constitutive relationships

1 PhD, Professor, University “Ss.Cyril and Methodious”, IZIIS Skopje, Macedonia 2 MSc, Research Assistant, University “Ss.Cyril and Methodious”, IZIIS Skopje, Macedonia 3 PhD, Professor, University “Ss.Cyril and Methodious”, IZIIS Skopje, Macedonia

340

1 INTRODUCTION

Cross laminated wooden structures are becoming very popular product in Europe. Wooden structures have always had a good reputation when subjected to seismic events, mainly due to the wood’s high strength to weight ratio and its enhanced strength under short term loading. The connections of wooden structures are normally more ductile than the timber parts themselves which results in an overall ductile behaviour of wooden buildings and their good seismic performance. However seismic response of wooden structure is complex issue and still a lot of uncertainties exist and need to be clarified and quantified. Having reliable research results will in no doubt lead to the improvement of the design rules and standards for the seismic design of structure and will provide seismic safe homes. In Part 1 efforts have been made towards development of to the modelling of the constitutive relationships that describe the anchors behaviour in tangential and normal direction and their programming and incorporation into the general program for structural analysis FELISA/3M. As the main source of nonlinearity is actually the connection itself, good correlation between the experimental and finite element results can be expected only by their proper modelling. For their verification, tests performed in Laboratory of the Faculty of Civil and Geodetic Engineering in Ljubljana, Slovenia 2005 (Dujic et al. 2005, Dujic et al. 2006) on massive cross-laminated wooden panels with a semi-rigid type of anchorage system, have been used. The tests have been performed using a special equipment to simulate realistic boundary conditions on the tested panel, which could appear during a real earthquake. From the performed analyses, it can be concluded that the numerical results are closer to the experimental results with applied A type of boundary condition (intended to simulate pure bending behavior mode) compared to other types (B and C, intended to simulate rocking and combined shear-rocking behaviour accordingly), that was normally to be expected. More details about the types of applied boundary conditions (A, B and C) can be found in Dujic et al. (2005).

2 FE MODELLING OF TESTED SPECIMENS: PUSHOVER ANALYSES AND DISCUSSION

As examples for verification of the proposed constitutive relationship described in Part 1, two sets of experimental tests performed in Ljubljana, 2005 (Dujic et al. 2005, Dujic et al. 2006) on massive XLam type wooden panels with semi-rigid types of anchors have been used. The first series of 4 quasi-static cyclic tests were performed on solid panel elements without openings with dimensions and cross-section given in Fig. 1.

The second series of other 4 quasi-static cyclic tests were performed on panel elements with openings as shown in Fig.2. For both series, "A" and "B" boundary condition cases were applied. Although more information about these boundary conditions can be found in Dujic et al. (2005), only the basic definitions will be mentioned herein. The boundary condition case "A" assumes a rocking mechanism, simulated by supporting one edge of the panel by the firm base while the other edge can freely translate and rotate. The boundary condition case "B" assumes restricted rocking mechanism, where one edge of the panel is supported by the firm base while the translation and rotation of the other edge is limited by the ballast translating only vertically without rotation.

341

Fig.1. Finite element model for specimens of series 1 (walls without openings)

Fig.2. Finite element model for specimens of series 2 (walls with openings)

The finite element models for the first and the second series of specimens are also given in Fig. 1 and 2, respectively. For the panel itself, a 2D FE mesh has been used employing iso-parametric plane-stress elements with 4 nodes applying the 2x2 integration rule. For connections (semi-rigid types of anchors) and for modelling gap/friction between the panel and the concrete foundation, the link elements described in the previous chapter

342

have been applied. Here, we should emphasize the basic difference between the conditions under which the tests have been performed and the conditions assumed in the analyses, which is related to the applied constraints on the degrees of freedom. The experimentally simulated "A" and "B" boundary conditions (implemented for the reasons explained in Dujic et al. 2005), lead to more expressed unsymmetrical behavior of the panels (although initially the lack of symmetry for the case of solid specimens without openings was not so much emphasized), resulting in "shifted" cycles in the lateral force – lateral displacement envelope capacity diagram. In order to get a clear insight into the behavior of the proposed constitutive models, in the numerical FEM models, no constraints have been applied, i.e. the experimentally produced boundary constraints have not been numerically simulated. Hence, the FEM models (given in Fig.1 and 2) assume that all degrees of freedom are engaged (translations in x and y) in all the nodal points (excluding the fixed bottom nodal points) allowing each nodal point of the panel to translate in both x and y directions including the rotation of the panel as a rigid body around the axis perpendicular to the x-y plane. Using this assumption, analytical pushover capacity curves have been obtained for both types of panels (without and with openings), which are comparatively presented with the experimental test envelopes in Figs.3 and 4, respectively.

For both cases, a reasonable correlation regarding the obtained stiffness, yielding point and descending parts of diagrams has been obtained. Because of the mentioned differences in the tests and the numerical models, the unsymmetrical response has not been so much expressed in the numerical results. However, the main trend of the panel behavior and especially the failure mechanism have been simulated (see Fig. 3 for panel without openings and Fig. 4 for panel with openings).

-80

-40

0

40

80

-40 -20 0 20 40

Displacement [mm]

Shea

r Fo

rce

[kN

]

FE Analysis

Exp: Sp. 1/1

Fig. 3. Numerically obtained pushover curve for specimens of series 1 (solid walls without openings) and correlation with the envelope curve from the quasi-static experimental test

(specimen 1/1 with "A" boundary condition)

343

-80

-40

0

40

80

-80 -40 0 40 80

Displacement [mm]

Shea

r Fo

rce

[kN

]

FE Analysis

Exp: Sp.2/1

Fig. 4. Numerically obtained pushover curve for specimens of series 2 (walls with openings) and correlation with the envelope curve from the quasi-static experimental test


2 SIMULATION OF THE BEHAVIOUR OF A THREE STORY TIMBER BUILDING SUBJECTED TO REAL SEISMIC EXCITATION: DYNAMIC ANALYSIS

In reality, when subjected to real seismic loading, separate panels built in a multi-story wooden building can manifest different behavior modes depending on the produced boundary conditions at their bottom ends as a result of the history of displacements during the response (see Fig. 5).

u

R

a (t)g

Fig. 5. Realistic response of a building built of wooden XLam panels subjected to seismic loading

344

So, depending on the applied connection devices, the level of the dead (axial) load and the position of the panel in the building, the panel's response can be of a pure shear mode, mixed shear and rocking mode, or pure rocking mode, as discussed in the Introduction. For example, the bottom panel in Fig. 5 can behave in a more expressed shear manner, since the dead load does not allow it to rotate freely as a rigid body. On the other hand, the constraint related to the dead load for the upper panel is not so emphasized, allowing it to rotate as a rigid body. In this case, the level of the manifested shear behavior mode, i.e. slippage deformation between the panels (translation of the upper panel as a rigid body related to the bottom one) depends on the connection device applied. One of the aims of the experimental tests performed on panels in Ljubljana (Dujic et al. 2005) was to investigate the mentioned three behavioral modes of the panels, constructing a special testing equipment for simulation of the boundary cases "A", "B" and "C", described in the previous section. Also, the idea of the tests was to pay attention to the existing analytical methods for design and experimental tests for determination of the racking strength of these types of panels (especially according to Eurocodes), which are mainly based on the assumption of a pure shear behavior of panels.

On the other hand, the constitutive relationships and the FEM based numerical models proposed in the paper should simulate the behavior of real wooden structures subjected to seismic loading, taking into account the discussed behavioral modes automatically, depending on the response history. Therefore, an example of a three-story wooden XLam building structure (see Fig. 6) consisting of panels with openings (having the same geometric and material properties as the series of panels with openings analyzed in the previous section) is presented herein.

q=4.33kN/m’

q=4.33kN/m’

q=4.33kN/m’

Kh

Kv

i

j

i

j

Xloc

yloc

Fig. 6. Finite element model for a realistic three-story building built of wooden XLam panels with openings

345

The structure has been analysed subjected to real time-history acceleration record (El-Centro) with peak ground acceleration of 0.6g. Initially, the structure has been subjected to self-weight and additional dead load on each floor level (4.33 kN/m'), which have been incrementally applied. The connections in the base are modelled as semi-rigid types of anchors, using the proposed constitutive relationships. The vertical connections between the horizontal floor panels and the vertical wall panels have been assumed as self-drilling screws described in Ceccotti et al. (2006) and they have been modelled by non-linear bar (BAR2DN) elements following elastic-plastic Von-Mises mixed-hardening constitutive law. Steel angles with similar properties as considered anchors have been assumed for horizontal connection between the panels and floors. The principal concept of the chosen structural system was based on the strategy for proper dissipation of seismic energy during an earthquake, which assumes energy dissipation first in the base semi-rigid anchors and in the horizontal connections between the walls and the panels, than in the vertical joints, and finally in the panels. The mode shapes have been calculated taking into account the initial stiffness of the connections and the mass due to the self weight and the given dead load . The analytical response in terms of force-displacement hysteretic curve for the first floor is given comparatively with the experimentally obtained force-displacement capacity diagram in Fig. 7.

-80

-40

0

40

80

-60 -30 0 30 60

Displacement [mm]

Shea

r Fo

rce

[kN

]

Dynamic Analysis, El Centro 0.6g

Exp. Envelope: Sp. 2/1

Fig. 7. Correlation of the analytical dynamic response of the first floor (El Centro record amax = 0.6g) with the experimental capacity envelope for the wall panel with openings


The time-history responses are presented in Figs. 8.1 to 8.4 from where it can be concluded that the structure behaves as base-isolated, as a result of the applied dissipation energy strategy.

The results have shown that the inter-story drift of the first floor is in total equal to 15 mm (including the base displacements of 6 mm), which is greater than the inter-story drift between the first and the second floors d12 = 5 mm, and the inter-story drift between the second and the third floor d23 = 5 mm. Probably, the additional door opening contributes to the more flexible behavior of the first floor. However, it is obvious that the main seismic energy dissipation occurs in the base due to the frictional properties of the base contact and the energy dissipative characteristics of the base anchors. Also the absolute acceleration differs very little from the first to the third floor (the structure responds mostly as a rigid body), and the amplification factor for the third floor relatively to the input (base) peak acceleration is approximately fa = amax/ag = 1.4g/0.6g = 2.33.

346

Time-History of Displacements

Time

Displacement

0 5 10 15 20 25 30 35 40

-.005

-.004

-.003

-.002

-.001

0

.001

.002

.003

.004

.005

.006

Node 307x

Fig. 8.1 Displacement time-history response for the basement (in m)

Time-History of Displacements

Time

Disp

lace

ment

0 5 10 15 20 25 30 35 40

-.025

-.020

-.015

-.010

-.005

0

.005

.010

.015

.020

.025

Node 1301x

Fig. 8.2 Displacement time-history response for the third floor (in m)

347

Time-History of Accelerations

Time

Acceleration

0 5 10 15 20 25 30 35 40

-12

-10

-8

-6

-4

-2

0

2

4

6

8

10

Node 401x

Fig. 8.3 Absolute acceleration time-history response for the first floor (in m/sec2)Time-History of Accelerations

Time

Acce

lera

tion

0 5 10 15 20 25 30 35 40

-10

-5

0

5

10

Node 1301x

Fig. 8.4 Absolute acceleration time-history response for the third floor (in m/sec2)

5 CONCLUSIONS The main objective of this research was the development of new computational

constitutive relations for a semi-rigid type of connections applied in XLam panels for the purpose of practical seismic assessment. Generally, during seismic actions, the following modes of wooden panels' behavior can be expected: rocking, mixed rocking and shear, and pure shear behavior. In Part I new computational constitutive relations for a semi-rigid type of connections applied in XLam panels have been developed and verified using the results

348

from the previously preformed experimental investigation in Laboratory of the Faculty of Civil and Geodetic Engineering in Ljubljana, Slovenia 2005 (Dujic et al. 2005, Dujic et al. 2006).

Herein, a comparative numerical study of the panels' behavior has been performed using quasi-static tests results from experiments performed in Ljubljana 2005. The proposed computational constitutive relationships for anchors in both normal and tangential directions have been implemented into the general-purpose software package FELISA/3M (Finite Element Inelastic Structural Analysis Program) program and pushover and dynamic analyses have been run. The provided results have shown a reasonable correlation with the experimental results (especially with the specimens of "A" boundary condition type) simulating the main trend of the panel behavior and the failure mechanism. Although in this research, constitutive models for a semi-rigid type of anchors with specified characteristics have been proposed, the same computational constitutive models can be used for other types of similar connections with provided necessary basic experimental data in terms of force-displacements relationships. In this case, the proper choice of the non-linear parameters (as yielding and ultimate points) necessary for the approximation of the axial force-separation and shear force – slip constitutive diagrams is of crucial importance.

Acknowledgements This research has been carried out within the bilateral scientific cooperation

between Republic of Slovenia and Republic of Macedonia. The support of the Ministries of Education and Science of both countries is gratefully acknowledged. Also, the authors would like to express special thanks to Dujic, B. and Zarnic, R. for the possibility of using their test results from the Ljubljana tests performed in 2005.

REFERENCES [1]“Seismic Behavior of Multi-Storey XLam Buildings”, A.Ceccotti and M.Follesa Proc. Of the International Workshop on Earthquake Engineering on Timber Structures, 2006, Coimbra, Portugal, pp. 81-95.[2] “Investigation on In-Plane Loaded Wooden Elements Influence of Loading and Boundary Conditions” B.Dujic, S. Aicher, and R. Zarnic, Otto-Graf-Journal (Otto-Graf Institute, MPA University, Stuttgart) 2005, 16: 259-272. [3]“Influence of Openings on shear Capacity of MassiveCross-Laminated Wooden Walls”.B.Dujic, S. Klobcar, and R. Zarnic, Proc. of the International Workshop on Earthquake Engineering on Timber Structures, 2006, Coimbra, Portugal, pp. 105-118. [4] “Comparative Study of FEM Based Reinforced Concrete Analytical Models and Their Numerical Implementation: Software Package FELISA/3M” V.Hristovski, and H.Noguchi, Proc. of the 1st Fib Congress, Section 13: Failure mechanism and non-linear analysis for practice, 2002, Osaka, Japan, pp. 403-410. [5]“Quasi-Static and Pseudo-Dynamic Tests on XLAM Wallsand Buildings” Lauriola, M. P. and Sandhaas, C, Proc. of the International Workshop on Earthquake Engineering on Timber Structures, 2006, Coimbra, Portugal, pp. 119-133.

349

Radmila B. ŠALI 18, Mihail A. GAREVSKI29, Zoran V. MILUTINOVI 310

ODGOVOR KONSTRUKCIJA IZOLIRANIH OLOVNO-GUMENIM LEŽIŠTIMA Rezime:

Detaljna analiza uticaja sistema za seizmi ku izolaciju od olovno-gumenih ležišta (LRB) na seizmi ko ponašanje P+7 spratne, ortogonalno skoro simetri ne stambene kule od AB nosivih zidova, je izvršena na osnovu GPS sinhroniziranih merenja ambijent vibracija objekta i ARTeMIS obrade dobijenih signala. LRB sistem od 32 olovno-gumena ležišta je projektovan prema USB-97 odredbama za maksimalno o ekivan zemljotres. Kvantitativna analiza i pore enje nelinearnih odgovora uklještene u osnovi i seizmi ki izolirane konstrukcije je izvršeno za 4 karakteristi nevremenske istorije ubrzanja. U radu su detaljno prezentirani rezultati i zaklju ciizvršenih istraživanja. Klju ne re i: Seizmi ka izolacija, olovno-gumena ležišta, ambijent vibracije, dinami ka analiza.

RESPONSE OF LEAD-RUBBER BEARING ISOLATED STRUCTURE

Summary:A GF+7 storey, orthogonally almost symmetric, shear wall residential tower building has been studied in all details (GPS synchronized ambient vibration measurements, ARTeMIS processing of acquired signals) to clarify the influence of lead-rubber bearings (LRB) seismic isolation system upon its seismic performance. The LRB system consisting of 32 LR bearings has been designed for USB-97 maximum expected earthquake and 4 characteristic earthquake time histories were used to quantitatively define and compare nonlinear responses of fixed-base and LRB isolated structures. The paper presents in all details results and findings of this study. Key words: Seismic isolation, lead-rubber bearings, ambient vibrations, dynamic analysis.

1 Research Assistant, M.Sc., Institute of Earthquake Engineering and Engineering Seismology, Skopje, Macedonia 2 Professor, Director, Institute of Earthquake Engineering and Engineering Seismology, Skopje, Macedonia. 3 Professor, Head RDM-IZIIS, Institute of Earthquake Engineering and Engineering Seismology, Skopje, Macedonia.

350

1 INTRODUCTIONAnalytical and experimental results presented are derived under the NATO

Science for Pease SfP978029 Project “Development of Low-Cost Rubber Bearings for Seismic Protection of Structures in Macedonia and Balkans”, launched in a support to affirmation of seismic isolation concept in the Republic of Macedonia and Balkans as an efficient strategy of building protection against destructive earthquake effects.

The dynamic response of the real seven-story residential building in Skopje (Img. 1a) has been studied in all details for elucidating potential influence of seismic isolation under the maximum expected earthquake actions at a site of construction.

Image 1- Building: 1, Settlement: Taftalidze 1, Skopje

The buildings’ structural system consists of reinforced concrete shear walls in two orthogonal directions (Fig. 1b). The building has ground floor /GF/, seven stories and loft /L/ (GF+7+L). The disposition of RC shear walls is almost symmetrical in both orthogonal directions.

To assure adequate vertical load resistance due to high vertical stiffness, sufficient horizontal flexibility and energy dissipation mechanism the Lead-Rubber Bearing (LRB) isolation system was chosen.

2 DYNAMIC CHARACTERISTICS OF THE REAL STRUCTURE

To identify the parameters of FB mathematical model, the ambient vibration measurements were performed on the real structure.

Ambient vibration measurements were performed by six GPS synchronized TROMINO (Micromed, Italy) seismometers. Two, placed at the top of the building were set as referent ones (Img. 2b). Other four, the mobile set (Img. 2c), had been moved from story to story, from top to the bottom of the building, providing ten minutes 3D measurements at all four corners and each storey.

(a) (b)

351

Image 2 - TROMINO instruments arrangement

Fig. 3 Amplitude spectra’s

The mode shapes, natural frequencies (Table 1) and damping ratios of the real structure have been defined based on Fourier analyses of amplitude spectra’s (Img. 3) of recorded signals, as well as by ARTeMIS (Ambient Response Testing and Modal Identification Software) processing of 34 GPS synchronized 3D signals.

Table 1 - Natural frequencies/comparative results

Ambient vibration testing Modal analysis Frequency fA(Hz) Direction fM (Hz) Direction fM/fA (%)

F1 2.40 Torsion 2.16 Torsion 10.0 F2 3.10 Translation –x 3.10 Translation -x 0.0 F3 3.20 Translation –y 3.15 Translation -y 1.6

(b) (c)

(a)

352

3 MATHEMATICAL MODELING and analysis

3.1 MODELING

Two mathematical models were defined for evaluating and comparison of the response of the real structure: (1) The Fixed Base (FB) model representing dynamic behavior of the real structure; and (2) Seismic Isolated (SI) model representing the dynamic behavior of the structure isolated by LRB seismic isolation system (Img. 4).

Fig. 4 Mathematical models

Dynamic analyses of both models have been performed by ETABS (Nonlinear version 9.0.4). Used shell elements combine membrane behavior and bending of the plates. The finite element model was chosen to satisfy the needs of the analysis.

The total mass for dynamic analyses (structural system plus additional loading of 2.5kN/m2, distributed uniformly on all floor elements) has been defined based on the study of actual building loadings tuned to fit the elastic frequencies of the FB model that were determined by ambient vibration measurements.

Parameters of adopted FB mathematical model, viz dynamic characteristic of the real structure obtained by ambient vibration testing’s, are presented comparatively in Table 4 indicating fear identification of the geometry and material parameters of the principal structural system.

3.2 SEISMIC ISOLATION SYSTEM Seismic isolation system applied consists of 32 LRB, uniformly arranged in the base,

placed at intersections of structural modulus. Three different LRB bearing groups were designed based on the calculated axial forces. Each of these groups was designed according to UBC-97 designing procedure which also includes deformation and stability checks.

(FB) model (SI) model

353

3.3 SELECTED EARTHQUAKE TIME HISTORIES Dynamic responses of FB and SI models have been calculated for four types of real earthquake time histories of different frequency characteristics, scaled to 0.3g, which value is determined based on the detailed site response analyses. The paper presents and discusses results related only to El-Cento (a(x)max=2.942m/sec2, a(y)max=1.809m/sec2).

Image 6 - Scaled El-Centro time histories (0.3g)

3.4 REPRESENTATIVE NODES FOR DYNAMIC ANALYSIS Four representative nodes (Fig. 7) have been selected to elucidate the seismic isolation effects of studied building:

1. node P_682 in the base of the structure; 2. node P_682 on level 400 (fifth floor); 3. node P_682 on level 100 (second floor); and

Table 2 - LRB Design parameters

Image 5 - LRB

Table 3 - LRB Link parameters

354

4. node P_682 on level 700 (loft). The results derived are presented in terms of dynamic characteristics of studied FB and SI models (Table 4), base shear force (Table 5), storey displacements (Table 6), interstory drifts (Table 7), and in (Table 8) as storey accelerations.

Image 7 - Representative nodes

Table 4 - Periods of vibrationFB Model SI Model Mode T (sec) Direction T (sec) Direction

1 0.4637 Torsion 2.38 Translation –y 2 0.3225 Translation -y 2.38 Translation –x 3 0.3178 Translation -x 2.32 Torsion 4 0.1336 Torsion 0.26 Torsion 5 0.0961 Translation -y 0.18 Translation –y 6 0.0943 Translation -x 0.18 Translation –x 7 0.0517 Translation -x 0.99 Torsion 8 0.0467 Translation -y 0.083 Torsion 9 0.0376 Translation -y 0.07 Translation –x

10 0.0267 Translation -x 0.07 Translation –y

Table 5 - Base shear force

Shear force (kN), Direction X Shear force (kN), Direction Y

FB model SI Model Reduction (%) FB model SI Model Reduction (%) max. 12232.30 2890.29 76.4 8533.40 3362.09 60.6 min. 15006.70 3266.24 78.2 9674.37 2768.30 71.4

355

Table 6 - Story displacementsDisplacement (m), Direction X Displacement (m), Direction Y

FB model SI Model Amplification (%) FB model SI Model Amplification

(%)Level 100 0.0046 0.1439 96.8 0.0032 0.1516 97.9 Level 400 0.0128 0.1449 91.2 0.0083 0.1528 94.6 Level 700 0.0197 0.1458 86.5 0.0127 0.1538 91.7

Table 7 - Interstory drifts

Drift (m), Direction X Drift (m), Direction Y

FB model SI Model Reduction (%) FB model SI Model Reduction (%) Level BASE 0.0000 0.0000 / 0.0000 0.0000 / Level ISO / 0.1431 / / 0.1507 / Level 001 0.0021 0.0004 81.0 0.0016 0.0005 68.8 Level 100 0.0025 0.0003 84.0 0.0016 0.0004 75.0 Level 200 0.0027 0.0004 88.9 0.0017 0.0004 76.5 Level 300 0.0027 0.0003 85.2 0.0017 0.0004 76.5 Level 400 0.0028 0.0003 89.3 0.0017 0.0004 76.5 Level 500 0.0025 0.0003 88.0 0.0016 0.0004 75.0 Level 600 0.0024 0.0003 87.5 0.0015 0.0003 80.0 Level 700 0.0020 0.0003 85.0 0.0013 0.0003 76.9 Level 800 0.0016 0.0002 87.5 0.0012 0.0003 75.0

Table 8 - Story accelerations

Acceleration (m/sec2), Direction X Acceleration (m/sec2), , Direction Y

FB model SI Model Reduction (%) FB model SI Model Reduction (%) Level 100 3.106 3.025 2.60 2.026 1.931 4.70 Level 400 6.779 3.034 55.2 3.629 1.932 46.8 Level 700 9.045 2.944 67.5 4.362 1.918 56.0

4 CONCLUSIONS Conclusions resulting from analytical and experimental study of the selected structure entirely verify positive aspects of the seismic isolation on the structural earthquake response.

1. Increase of natural period (Table 4). As result of the increased flexibility of the system, natural period of the structure increased from =0.46sec to T=2.38sec, shifting natural period of the system from the predominant periods of the expected earthquake actions.

2. Reduction of base-shear (Table 5). Reduction of the base-shear force is evident in the model with implemented seismic isolation. The base-shear force under the El-Centro earthquake excitation has been reduced 4.6 in X direction in 3.5 times in Y direction.

356

3. Increase of displacements (Table 6). Increased flexibility of the system led to increase of the total displacements due to the elasticity of the existing isolation. Displacements of the system are concentrated at the isolation plane level. Total displacement at the level 700 under the El-Centro earthquake excitation has risen from 0.0197m to 0.1458m in X direction and from 0.0127m to 0.1538m in Y direction.

4. Reduction of interstory drifts (Table 7). Implementation of the isolation system resulted into the reduction of the interstory drifts to negligible level, so it can be said that they practically do not exist. This reduction enables the structure to behave as almost ideally stiff. In this way the damage risk of the structural and non-structural elements is minimized.

5. Reduction of story accelerations (Table 8). Analysis of SI Model has shown significant reduction of the story accelerations. Acceleration at platform 700 under the El-Centro earthquake excitation have been reduced from 9.045m/sec2 to 2.944m/sec2 (3.07 times) in X direction, and from 4.362m/sec2 to 1.918m/sec2 (2.27 times) in Y direction.

6. Energy dissipation mechanism. Contrasting the classical structure where the energy dissipation mechanism is based on the plastic deformations at certain points of the structure, in the seismically isolated structure energy dissipation mechanism is concentrated at the isolation level enabling simple design, control and eventual repair.

REFERENCES [1] Influence of Lead-Rubber Bearings on the Response of Seismically Isolated Structures

/ Salic, R., Master Thesis, IZIIS, University Ss. Cyril and Methodius, 2007, Skopje, Republic of Macedonia.

[2] Structural Bearings / Eggert, H., Kauschke, W., Earnst & Sohn, 2002. [3] Evaluation of the Proper Functioning of the Rubber Isolators of the Primary School

‘Pestalozzi’ in Skopje Under Strong Earthquake / Garevski, M., Kelly, J.M., 2001, IZIIS-Skopje.

[4] Engineering with Rubber / Gent, Alan N., 2001, Carl Hanser Verlag. [5] Base Isolation of Structures / Kelly, Trevor E., Base Isolation of Structures, Holmes,

Consulting Group Ltd. [6] Seismic Isolation for Earthquake-Rasistant Design / Komodromos, P., 2000,

Massachusetts Institute of Technology. [7] Design of Seismic Isolated Structures / Naeim, F. and Kelly, J., 1999, John Wiley &

Sons.[8] Experimental Dynamic Testing of the First Structure in the World Isolated by Rubber

Bearings / Garevski, M., Kelly, J., Bojadziev, M., 1998, Proceedings of the Eleventh European Conference on Earthquake Engineering, Paris.

[9] Earthquake-Resistant Design with Rubber / Kelly, J., 1996, 2nd. edition, Springer-Verlag, London.

357

Ljubomir Taškov111, Lidija Krstevska212

SEIZMI KA BAZNA IZOLACIJA REZERVOARA I ZGRADA SA PRIMENOM SISTEMA ALSC

Rezime:

ALSC sistem pretstavlja specifi no tehni ko rešenje kada konstrukcija je do-vedena u stanju lebdenja primenom konstantnog potiska (uzgona) u kontaktu, ime je omogu eno lako klizanje. Efikasnost ALSC sitema je demonstrirana na modelu rezervoara u razmeri 1/3 sa masom od 10 t, kao i na modelu crkve u razmeri 1/3.5 sa masom od 30 t. ispitivanjem na seizmi koj platformi u IZIIS-u. Isti pozitivni efekat seizmi ke izolacije je dobiven u oba slu aja, obezbe uju i kompletnu zaštitu od rušenja ak i kod najja ih potresa.

Klju ne re i: Sistem za baznu izolaciju ALSC, efekat klizanja, potisak

SEISMIC BASE ISOLATION ON RESERVOIRS AND BUILDINGS BY APPLICATION OF THE ALSC SYSTEM

Summary: The ALSC system represents a specific technical solution when the main structure is split from the supporting structure, thus allowing the main structure to slide. The structure is kept in floating position by application of constant hydraulic up-lifting pressure at the contact, allowing easy sliding. The effectiveness of the ALSC system is demonstrated on a model of reservoir in scale 1/3 with mass of 10 tons, as well as on a model of church in scale 1/3.5 with mass of 30 tons, tested on IZIIS shake table. The same positive effect of seismic isolation was obtained in both cases, providing complete protection against collapse even for the strongest earthquakes.

Key words: ALSC base-isolation system, sliding effect, up-lifting pressure.

1 Prof. PhD, Institute of Earthquake Engineering and Engineering Seismology, University “Ss. Cyril and Methodius”, Skopje, Republic of Macedonia, [email protected] 2 Assoc. Prof. PhD, Institute of Earthquake Engineering and Engineering Seismology, University “Ss. Cyril and Methodius”, Skopje, Republic of Macedonia, [email protected]

358

1. DESCRIPTION OF ALSC SEISMIC BASE-ISOLATION SLIDING SYSTEM

“Almost Lifted Structure Concept” represents a specific solution belonging to the group of “sliding concepts”. The upper structure slides over a smooth surface under low pressure. Under horizontal load effect, the structure slides along the foundation base. By controlling the pressure of the liquid, which is located in the empty space under the sliding base, the friction force and the sliding of the structure are controlled as well. The friction force at the contact surface is defined by the following equation:

Fr= N (1)

a) Ordinary sliding: N=G, L=0

(2) b) ALSC sliding: N= G-L, G L N

N 0 L=pS ; (3) (for L=G ; N becomes 0 and Fr=0 i.e. floating state)

Fr = friction force;

= friction coefficient at the contact surface;

N = active compressive force;

L = uplifting force produced by the liquid;

G = weight of the principal structure;

p=liquid pressure;

S= contact surface between the sliding and the fixed plate

The contact surface must be smooth to reduce the friction between the plates and prevent leakage of the liquid, which is under pressure. Sufficient smoothness can be achieved by coating both contact surfaces with epoxy resin.The sliding and fixed foundation plates are designed to sustain the pressure of the liquid. The zone of the empty space between the two plates is filled with liquid on which pressure is exerted.The horizontal displacement of the structure as well the designed frequency is controlled by horizontal springs. The centering springs keep the structure in centered position always when sliding starts. The springs are designed as much flexible as possible, but stiff enough to resist the wind effects. ( Figs. 1 and 2).The low horizontal rigidity of the lateral springs and the very low friction force at the contact surface, contribute to a significant reduction of transmission of the vibration from the ground to the upper superstructure. The dimensions of the centering springs as well as the amount of the lifting pressure under the sliding foundation plate depend on the main dynamic characteristics of the structure, its weight, rigidity, designed vibration period, available space for lateral relative displacement between the sliding base and the fixed base, etc.

359

2. Description of the system for automatic control of the liquid pressure

The automatic system for activation of the working pressure under the sliding plate and keeping it constant as long as needed plays an important role in the effectiveness of the ALSC base-isolation system. During the tests, the system for pressure control was active, correcting and keeping the pressure at the required level. Having this system, the structure doesn't need to be subjected to a permanent pressure for a long time waiting for the earthquake. It will be activated just a few seconds before the shear seismic waves attack the structure. In the case of the shake-table test, the system for pressure control was activated a few seconds before the shake-table started to move and remained active throughout the testing time. Basically, the automatic pressure control system consists of a steel reservoir filled with liquid under pressure, kept constant by means of a compressor and a servo-valve. The servo-valve is closed until the pressure of the liquid under the sliding plate is decreased for more than 10%. After that, the valve opens and the liquid from the steel reservoir comes to the sliding plate increasing the pressure up to the required level. Some details of the automatic system for control of liquid pressure are shown in Figs. 3 and 4.

Figure 3 - Supply of the liquid under Figure. 4 - System for automatic control the foundation of the liquid pressure

Figure 1 - Seismic base isolation by “ALSC” system, model of reservoir, 1/3

Figure 2 - Seismic base isolation by “ALSC” system, model of church, 1/3.5

360

3. TEST RESULTS

3.1 TEST RESULTS from the model of reservoir in scale 1/3

The shake-table test on the model of reservoir was performed in the Laboratory of the Institute for Earthquake Engineering and Engineering Seismology in Skopje in the year 2003 and 2004. The objective of the test was to verify the efficiency of a base isolation system "ALSC". First step of the program was to investigate the sliding-friction force at the contact between RC plates under static and dynamic conditions, considering different springs and liquid pressure conditions. Three series of tests have been conducted for different type of dynamic excitations produced by shaking table. First of all, the resonant frequency of the model has been defined by random excitation test. In the next step, a series of harmonic excitations has been applied within the frequency range of 0.5-15.0 Hz, producing an acceleration of the shaking table within the range of 0.1-1.25g in horizontal and 0.1-0.5g in vertical direction. A series of earthquake excitations has been applied in the last phase of testing program in horizontal as well as by-axial, h-v direction. The same testing program has been applied for fixed base model. The test results are presented in table and graphs given bellow.The general expression about model behavior is that the base-isolated model behaves very stable under different earthquake and harmonic excitation both in horizontal and vertical direction. The amplification of the base motion on the top was les than one (0.2-0.32), which means that the input motion was attenuated 3-5 times. On the other hand, the fixed base model behaves very unfavorable, by damaging the fixing anchors at the base and producing large amplification on the top of the model (2.0-5.0).

Table 1 Response parameters of the model of reservoir under earthquake excitations N Record frequency

(Hz) inp.acc(g)

acc. base (g)

acc top (g)

rel. displ. (mm)

1 IZMIT -H 1.71 0.22 0.17 0.24 2.74 2 IZMIT -H 1.71 0.09 0.13 0.20 6.64 3 IZMIT -H 1.71 0.14 0.18 0.27 16.3 4 IZMIT -H 1.71 0.18 0.23 0.32 35.3 5 KOBE-H 0.83 0.10 0.14 0.23 4.6 6 KOBE-H 0.83 0.23 0.21 0.27 39 7 ELCENTRO-H 1.46 0.20 0.23 0.35 23.9 8 ELCENTRO-H 1.46 0.28 0.25 0.37 31.5 9 PETROVAC-H 1.46 0.38 0.48 0.79 76.5 10 NORTHRIDGE 1.07 0.29 0.20 0.37 64.2 11 MEXICO-H 0.48 0.19 0.11 0.17 4.5 12 ALMA ATA-H 2.39 0.14 0.17 0.32 16.4 13 ALMA ATA-H 2.39 0.37 0.24 0.44 53.4 14 ALMA ATA-H 2.39 0.44 0.26 0.43 67.6

361

Figure 5 - Comparative presentation of performance of the model of reservoir isolated by ALSC system and fixed base model under earthquake excitation

3.2 Test results from the model of church in scale 1/3.5

The shake-table test of the church model was performed in the Laboratory of the Institute for Earthquake Engineering and Engineering Seismology in Skopje in the year 2008. The model of the church was designed to the length scale of 1:3.5 according to the "gravity force neglected" modeling principles. The main objective of the testing was to investigate experimentally the effectiveness of the proposed reversible technology: base isolation of historical monuments by ALSC floating-sliding system. According to that, the seismic shake-table testing was performed in two main phases: phase 1-testing of the base-isolated model with the ALSC floating-sliding system and phase 2-testing of the original fixed-base model. The base-isolated model was tested by 5 different levels of input acceleration: 0.45g, 0,8g, 1.2g and 1.45 g. Uniform sliding of all parts of the structure was recorded (sliding plate, walls and tambour).The maximum response acceleration of the sliding plate in all the cases was about 0.2-0.3g (Table 2 and Fig.6). This level of acceleration didn’t produce any cracks in the model, except an increased relative sliding displacement between the basin and the sliding plate.

Table 2 Response parameters of a church model under harmonic and earthquake excitations

Acceleration(g) Rel. displ.(mm) Test

Excitation Freq. ( Hz) base sliding

plate top-tambour

tambour-sl.plate

base-sl.plate

1 Harmonic 7Hz 0.80 0.15 0.12 0.20 7.0 2 Harmonic 5Hz 0.60 0.15 0.16 0.22 3.0 3 Petrovac 7.7 0.20 0.20 0.18 0.15 3.5 4 Petrovac 7.7 0.45 0.25 0.18 0.20 6.0 5 Petrovac 7.7 0.80 0.25 0.25 0.30 15.0 6 Petrovac 7.7 1.20 0.32 0.35 0.38 20.0 7 Petrovac 7.7 1.45 0.25 0.22 0.28 35.0

362

On the other hand, the behavior of the fixed base church model was very unfavorable, producing the damage collapse of the dome and heavy damage of the walls on the intensity of earthquake 0.45g-0.70g. The amplification factor of the response of the model was 2-3.

Figure 6 - Response of the church model base-isolated by ALSC floating-sliding system

acceleration respons of the model-fixed base, Petrovac scaled 3.5 times

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 50 100 150 200

span of the input

acce

lera

tion

(g)

inpaccfixraccwallraccdome

Figure 7 - Response of the church model fixed to the base

acceleration response of sliding plate-petrovac scaled 3.5 times

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 100 200 300 400 500

span of the input acceleration

acce

lera

tion(

g)

input accALSC acc

363

4. ConclusionsThe comparative test between the base-isolated model by ALSC system and the classical fixed base model clearly shows the superior behavior of the ALSC floating-sliding base-isolation system. The same conclusion is valuable both for model of reservoir and church model. The excitation of 1.5g (maximum capacity of the shaking table and maximum peak acceleration of the Montenegro earthquake-scaled by a factor of 3.5) was not enough to produce any damage to the reservoir as well as the church. The tests show that the ALSC floating-sliding base–isolation system can protect the structure in any frequency and/or amplitude range and against the strongest earthquake, because of the specific property: limited transmissibility (0.2-0.3g for the scaled model i.e. 0.07-0.08g for the real structure), which can not produce damage. Unlike this model, the fixed base models are very vulnerable. The automatic system for control of the liquid pressure under the sliding plate was successfully designed. It effectively controls the pressure and keeps it constant all the time during the performed test. The role of the springs, controlling the lateral motion was also effective, allowing the structure to slide in the desired and controlled range. Comparative results between sliding model of reservoir and sliding model and church model show the same isolation effect, which means that ALSC system can be successfully applied in all type of structures which height to length ratio is less than 2.0

5. Acknowledgement The authors of the paper express their deep gratitude to the Ministry of Education

and Science of Macedonia and TUBITAC, for financing the test of the model of reservoir as well as to the European Commission for financing of the project PROHITECH within the sixth frame program FP6, in which the testing of the church model was realized.

REFERENCES

[1] "Seismic Base Isolation Based on Almost Lifted Structure Concept" Lj. Tashkov, A.Antimovski, M. Kokalevski, 9-th International Symposium of MASE, Ohrid, Republic of Macedonia, 2001

[2] "Comparison Between Fixed Base and Seismic Base Isolated Liquid Storage Tank by ALSC System" Lj. Tashkov, L. Krstevska, EE-21C-International Conference on Earthquake Engineering, 27.08-01.09. 2005, Skopje-Ohrid, CD Proceedings

[3] “Shake-Table Test of Model of St. Nicholas Church in Reduced Scale 1/3.5” Lj. Tashkov, L.Krstevska, K. Gramatikov, F. Mazzolani, Proceedings, Protection of Historical Buildings, Conference Prohitech 09, Rome, Italy, 2009, Taylor & Francis Group, London, ISBN978-0-415-55803-7, Vol 2, pp 1691-1697

365

1

Rezime

.,

, . „ “. : 1) -

2) . / -

.

Keywords: vulnerability models; spatial and temporal translation; Eurocode 8

CODE BASED CALIBRATION OF VULNERABILITY MODELS

Summary A method for spatial and temporal translation of vulnerability models is recently proposed. It is based on the aseismic design codes in the region of interest since it is assumed that if they are strictly followed in the construction practice, the seismic performance and level of damageability of building are "controlled by them". Two basic factors are considered: 1) the design base shear; and, 2) the ultimate deformation capacity of buildings. The referent vulnerability model is translated / calibrated to target vulnerability model through combination of a shift and rotation quantified by the differences in the building codes and construction practices.

Keywords: vulnerability models; spatial and temporal translation; Eurocode 8

_____________________________________________ , „ “ , [email protected]

366

1

j-

. ( ). -

. -

:- ( )

( ), ,

- ( ) .

.

.

:

1.

2.

3. : / .

.

.

--

,-

. ,

.

, .

367

1.1.

.,

,. , ,

. ,.

.

( ) ( )

( ) .

1.2.

.

.,

.

„ “ .

:

(1) (2) .

:, ,

. .

.,

(MDR). ,

368

.

2.

,,

. ( I),

I– I- ( I- MDR) .,

(SA) (PGA). (SA) (PGA) I

:

MMIbaPGALogorSALog *)(__)( (2-1)

:

bVVKI RT )/log(

1 (2-2)

I ; R ,

. PGA-MMI Murphy & O Brien (1977) : (Log(PGA)=0.25+0.25MMI).

, MMI 6-7 VT and VR .(2-2).

.I a I MDR

,I 6-7.

,I MMI (fig. 1).

, MMI 7 .

. MDR MMI MMI ,

369

. MMI ., MDRoriginal ,

I :

)()()( 11 IMDRIIMDR original (2-3)

(2-4)

Iinelastic

. Iinelastic 7-8

. .

(2-3) "intermediate" - (fig.1.).

Fig. 1.

3.

,

("intermediate" )

)(1

)5()5

5(

)5(0

)( 21

inelastic

inelasticinelastic

II

III

II

I

370

, I inelastic

MMI , I severe,:

(2-5)

R „ “ „ “

. 2 MDR .,

., 2 1.

MMI 10. MMI 10

. MMI Iinelastic ,

. .

'' '' Iinelastic

:2 11 SEVERE SEVERE

SEVERE INELASTIC

MDR I MDR II I (2-6)

MDR

:

22 2 1

1

1

0

11 1

10 1

INELASTIC

INELASTIC

INELASTIC

I I

MDR IMDR I MDR I MDR I I I and

MDR I

MDR II I and

MDR I (2-7)

MDR „ “ :

1 2T RMDR I MDR I MDR I MDR I (2-8)

, ,. ,

T

R

severeR

severeT KIMDRIMDR

22 )()(

371

. MMI

, ( ,) , .

5.EC-8

.

„ “ ( ) , „ “ ().

.

4.1.

RSA-83. - YUS-81.

..

Vt Vr :

Vt/Vr=0.05/0.08=0.893 ( 8. ) I= log(0.625)/0.25=-0.816

MMI : I(inelastic)=7+ I=6.184 I(severe) =10+ I=9.184

'' '' (RSA-83) '' '' (YUS-81) :

- r=1.5 (reference ultimate drift) , - t=h/600 x dukt.(8)=ili 1.33 (target ultimate drift)

372

0.00.10.20.30.40.50.60.70.80.91.0

5 7 9 11MMI

MDR

REFERENCE OMV

TARGET-JUS

, .,

. 1. :

2. (OMV) – (YUS-81) .

,.

( ) .

Fig. 2. (Victor Mendes)

:

(Victor Mendes )

( 4-1) . Iv,c=20 ( 2.).

'' '' :

(4-1) 0,5 0, 45 arctan 0,55 10, 2 0,05 Vd I I

373

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

5 6 7 8 9 10 11 12MMI

MD

R

reference-OMV

target-JUS

reference-JUS

Fig. 3.

4.2.

-- ( ,

- ) YUS-81-a EC8. . ( -

Iv,c.

:

1. RC1 (Iv=20) 2. RC4 (Iv=10) a)b)3. - RC2 (Iv=0) a)

( ):

374

0.00.1

0.20.3

0.40.5

0.60.7

0.80.9

1.0

5 6 7 8 9 10 11 12

I

MDR

r

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

5 6 7 8 9 10 11 12

MD

Rr

RC2

RC4

RC1

Fig. 4.

-

„ “, RC1

( ) RC2 ( - ) .,

.

Fig. 5. „ “

(Iv):

1. .

„ “

Iv,c=20 RC 1 Iv,c=10 RC 4 Iv,c=0 RC 2

DUAL SYSTEMWALL SYSTEM

FRAME SYSTEM

375

RC 1 Iv=20 Iv,c=6.316 RC 2 Iv=0 Iv,c= -29.627RC 4 Iv=10 Iv,c= -7.52

-YUS-81 .

1.

:

1.

2. ( )

3. („ “ )

4. I

5. : Iinelastic Isevere

6. r t

7. - „ “ 1MDR I

8. -„ “ 2 MDR I

9. „ “ TMDR I

376

[1] Bertero, V.V. (1997). Performance-based seismic engineering: A critical review of proposed guidelines. Proceedings of the International Workshop on Seismic Design Methodologies for the Next Generation of Codes, P.Fajfar and H. Krawinkler (editors), Bled Slovenia, June 1997, A.A. Balkema, Roterdam, 1997.

[2] Chopra, A.K. and R.K. Goel (2000). Capacity-demand diagram methods based on inelastic design spectrum, Proceedings of the 12WCEE, Auckland, New Zealand,2000

[3] EUROCODE 8: Design provisions for earthquake resistance of structures. Part 1.1 General rules – Seismic actions and general requirements for structures. DD ENV 1998-1-1: 1996.

[4] Freeman, S.A. (1998). The capacity spectrum method as a tool for seismic design. Proceedings of the 11ECEE, 1998 Balkema, Rottedam, ISBN 90 54 10 982 3

[5] Giovinazzi, S. and S. Lagomarsino (2001). A methodology for the vulnerability analysis of built-up areas, Department of structural and Geotechnical Engineering, University of Genoa, 2001.

[6] HAZUS-Earthquake loss estimation methodology. Prepared by National Institute of Building Sciences for federal Emergency Management Agency, 1997.

[7] HAZUS 99 SR2 - Earthquake loss estimation methodology (1999). Prepared by National Institute of Building Sciences for Federal Emergency Management Agency, Washington DC, 1999.

[8] Holms, W.T. (1996). Seismic evaluation of existing buildings-State of the practice. Proceedings of the 11WCEE, Paper No. 2008 1996 Elsevier Science Ltd, ISBN 0 80 Tokyo- Kyoto, Japan (Vol. II).

[9] Huo, J.R., R.S. Lawson and N.G. Kishi (1998). Translation of earthquake vulnerability functions. Proceedings of 11th ECEE, 1998 Balkema Rotterdam, ISBN 90 5410 982 3

[10] Kerstin Lang-Seismic vulnerability of existing building –Zurich February 2002 [11] King, S. and C. Rojahn (1996). A comparison of earthquake damage and loss

methodologies. Proceedings of the 11WCEE, Paper No. 1482 1996 Elsevier Science Ltd, ISBN 0 80 042822 3.

[12] Lungu D., A. Aldea, C. Arion, R. Vacarenau, F. Petrescu, and T. Cornea (2001). European distinctive features, inventory database and typology, RISK-UE WP1 Report, December 2001 – typology of objects

[13] Milutinovic, Z., T. Olumceva and G. Trendafiloski (1998). Earthquake related emergency capability of large hospital campus in Republic of Macedonia, Institute of Earthquake Engineering and Engineering Seismology, Report IZIIS 98-43/1, World Health Organization Regional Office for Europe, Copenhagen, Denmark, September 1998.

[14] Milutinovic, Z.V. and G.S. Trendafiloski (2003). WP4: Vulnerability of current buildings, RISK-UE Handbook, September 2003.

[15] Murphy, J. R. & O’Brien,L.J.1997.The correlation of peak ground acceleration amplitude with seismic intensity and other physical parameters, Bulletin of Seismological Society of America,67(3):877-915

377

[16] Newmark, N.M., and E. Rosenblueth (1971). Fundamentals of Earthquake Engineering, Prentice-Hall, Inc., Englewood Cliffs, N.J., 1971.

[17] Nozar G. Kishi, Jun-Rong Huo, R. Scott Lawson-Regional translation of earthquake vulnerability functions. Proceedings of 6th National Conference on Earthquake Engineering.

[18] Petrovic Bosko – Selected chapters from the earthquake engineering – Gradjevinska knjiga

[19] Priestley, M.J.N. (2000). Performance based seismic design, Proceedings of the 12WCEE, Auckland, New Zealand, 2000.

[20] Regulations for seismic design – A world list 2004. International Association for Earthquake Engineering (IAEE), 2004.

[21] Singhal, A. and A.S. Kiremidijan (1996). Method for probabilistic evaluation of seismic structural damage, Journal of structural Engineering ASCE, Vol.122, No. 12, December, 1996, pp. 1459-1467

[22] Trendafiloski, G.S. (2003). GIS-oriented method for elaboration of probabilistic earthquake scenarios, Doctoral Dissertation, Institute of Earthquake Engineering and Engineering Seismology (IZIIS-Skopje), Skopje, January 2003 (in Macedonian).

[23] V c reanu, R. (1998). Seismic risk analysis for high rise buildings, Proceedings of the 11ECEE, Paris, 1998.

[24] Williams, M.S. and R.G. Sexsmith (1995). Seismic damage indices for concrete structures: A state- of-the-review. Earthquake Spectra, Vol. 11 No. 2, May 1995.

[25] Vladimir Sigmund, Mirjana Bosnjak, Ivica Guljas, Andreas Stanic – Comparison of the implementation of the Croatian regulations and Eurocode 8 – Gradjevinar 52(2000)

[26] YUS-81, Code of Technical Regulations for the Design and Construction of Building in Seismic Regions (Official Gazette of S.F.R. Yugoslavia, No.31/81)

379

Violeta Mir evska1 Vladimir Bi kovski2 Mihail Garevski3

BENCHMARK TEST softvera PROC3DN-IZIIS

Rezime

Cilj ove analize je benchmark test geotehni kog softwera PROC3DN-ver.1.2 razvijen u IZIIS-u sa ciljem demonstriranje objektivnosti i primenljivosti izlaznih rezultata. Za uspore uju i softwer odabran je komercijalni geotehni ki kompjuterski program PLAXIS-ver.8.1 Univ. of Delft-Netherland. Spore enirezultati su dobijeni od oba programa kao efektivna naprezanja, deformacije, manifestovane zone plasti nosti kao i vrednosti koeficijenta sigurnosti sistema brana-duboki iskop, budu i naponsko-deformaciono stanje je tipa plošne deformacije i kako PLAXIS podržava 2D analize a PROC3DN 3D podržava 3D analize, komparativni model definiran je grani nim uslovima koji u celosti simuliraju površinsko stanje deforamcija.. Kljucne reci: koeficenat sigurnosti, benchmark test.

A BENCHMARK TEST OF THE SOFTWARE PROC3DN-IZIIS

Summary

The objectiv of this work is a benchmark run of the geotechnical software PROC3DN-ver.1.2 developed at IZIIS in oder to assess it’s objective performance as well as correctness and applicability of the software output results. For benchmarking software choosen is the comercial geotechnical computer programs PLAXIS ver.8.1 Univ.of Delft – Netherland. Compared and presented are the obtained results from both computer programs, such as the effective stresses, the deformations, the manifested plastic zones as well as the value of short term coefficient of safety of the dam-excavation system. The stress-strain state is of plain strain type. PLAXIS supports 2D analysis while PROC3DN supports 3D analyses wherefore the comparative model has been defined by a boundary conditions that thoroughly simulate the plain strain state of deformations. Keywords : temporary safety coefficient, benchmark test

_______________________________________________ 1.Assoc. Prof. Dr., (IZIIS),Skopje, Republic of Macedonia 2. Prof.Emeritus, (IZIIS), Skopje, Republic of Macedonia 3. Prof. Dr., (IZIIS) Skopje, Republic of Macedonia

380

1. INTRODUCTION The existing analyzed dam belongs to the type of low hydrotechnical structures

with a height of 12 m from the lake bottom to the dam crest. The width of the dam crest is 10.50 m. The slopes on the upstream and the downstream side of the dam are 1:3. The lithological composition of the terrain is assumed to consist of 8 layers,. The layers are mainly described as subclay brown color media, dust subclay beige to brown color, subsand, dusty sand beige to yellow color, airy sandy layer, gray to beige color, airy alevrolit gray to beige color, airy clay, gray to beige color with blue stains, etc. The layers are compacted sufficiently enough. The physical-mechanical characteristics of the materials within the dam body and the terrain are presented in the text below. The objective of the performed analysis is the proving of the short term stability of the dam “Arti-ficial Lake’s Dyke” in the vicinity of Tirana immediatelly after the excavation of the construction pit for the purpose of construction of the “Building for Living and Services” , at level + 8 above ground and level 2 below ground. For that reason, it is anticipated to cut the downstream slope of the dam at a distance of 10 m from the dam crest, parallel to the direction of the crest within length of 40 m. With this cut with a depth of 7 m, the consolidated stability of the dam-soil system is disturbed. It is necessary to stabilize the dam-excavation system. The supporting structure adopted in the analyses represents an RC wall with a thickness of

1,0 m along the entire length of the excavation with embedment of 2m. Since PLAXIS ver. 8 supports a 2D analysis, the stress-strain state of the system has been defined as a plain state of deformations. The PROC3DN program supports 3D analyses wherefore the corresponding plain strain conditions has been defined by use of 1 m thickness and boundary conditions that thoroughly simulate the plain state of deformations. The discrete model evaluated in using the PLAXIS version 8 program is composed of 2D triangular finite elements with 15 nodal points. The corresponding discrete model evaluated in using the PROC3DN program is composed of 3D finite elements with 8 nodal points. In both analyses, the Mohr-Coulomb’s criterion for nonlinear behaviour of soil media has been treated. Both applied models for analysis define the stress and strain state in the media as well as the expected coefficient of safety of the system.

2. BENCHMATK RESLUTS FROM THE STATIC ANALYSIS The static analysis of the dam-deep excavation system has been carried out for the

effect of gravity forces (self weight of the system), the hydrostatic pressure of the water in the reservoir and the active pore pressure resulting from the steady seepage state in the model under the effect of the active filtration forces. It has been defined the short term coefficient of safety against failure of the dam-wall support -deep excavation system. Futher on presented are the output results of the analysis by use both computer programs.

381

Figure 1: Total displacement increments indicating the most possible failure mechanism

382

Figure 2: Coefficent of safetty for the most critical sliding surface

Figure 3: Deformed mesh of the analysed system mU x 011.0max,

Figure 4: Deformed mesh of the analysed system mU x 01.0max,

383

Figure 5: Plastic points and “tension cut-off”’points in the model

Figure 6: Plastic points in the model

Figure 7: “tension cut-off”’ points in the model

384

Figure 8: Shear stresses 2/28.70 mkNxy

Figure 9: Shear stresses2/35.70 mkNxy

385

Figure 10: Vertical effectiv stresses 2' /77.263 mkNyy

Figure 11: Vertical effectiv stresses 2' /94.253 mkNyy

386

Figure 12: Horizontal total stresses2/22.318 mkNxx

Figure 13: Horizontal total stresses 2/18.320 mkNxx

387

MohrCoulomb 1 2 3 4

unsat [kN/m³] 22.00 19.00 Undrain B=1 Undrain B=1 sat [kN/m³] Drain B=0 Drain B=0 19.500 19.30

Eref [kN/m²] 150000.000 100000.00 80000.000 30000.00 [-] 0.260 0.250 0.350 0.300

Gref [kN/m²] 59523.810 40000.000 29629.630 11538.46 Eoed [kN/m²] 183531.746 120000.00 128395.062 40384.61 cref [kN/m²] 1.00 1.00 100.00 25.00

[°] 40.00 33.00 19.00 21.00 MohrCoulomb 5 6 7 8

unsat [kN/m³] Undrain B=1 Undrain B=1 Undrain B=1 Undrain B=1 sat [kN/m³] 19.60 19.60 19.80 19.90

Eref [kN/m²] 30000.000 43200.000 60000.000 84000.00 [-] 0.280 0.280 0.300 0.350

Gref [kN/m²] 11718.750 16875.000 23076.923 31111.111 Eoed [kN/m²] 38352.273 55227.273 80769.231 134814.81 cref [kN/m²] 20.00 45.00 75.00 80.00

[°] 24.00 21.00 20.00 22.00

Figure 14: Geotechnical properties of the layers

4. CONCLUSION

Comparing the corresponding analyses carried out by use of the PLAXIS and PROC3DN programs it can be concluded that there is a certain identity of compared values. While analyzing the strain states, the difference is negligible. Comparing the corresponding stress states, it can be concluded that the shape of the zones of concentration of stresses is identical whereat the difference in intensity is less than 3-5% . The shape and the propagation of the manifested plastic zones and tension cut of zones coincide. The difference between the computed minimal safety coefficient obtained by application of Plaxis and the corresponding coefficient obtained by using the PROC3DN program results from the differences in the method for definition of the safety coefficient within each program. According Plaxis factor of safety is defined as a ratio of the true strength to the computed minimum strength required for equilibrium calculated by use of the step by step iterative Phi-C- reduction procedure. The procedure is special kind of plastic analysis based

388

on reduction of the strength parameters in the same proportion that equals the value of the total multiplier. The value of this multiplier at the last step of the iterative procedure that maintains equilibrium is the value of the coefficient of safety. So Plaxisi is giving us coefficient as a ratio of available extreme strength and minimum strength needed for equilibrium of the system. According PROC3DN the factor of safety is a ratio of the available extreme shear strength producing passive shear forces and the manifested working shear stresses producing active shear forces. If the element is in elastic behaviour then the active working shear stress is always less than the those needed for maintaining the limit equilibrium at the beginning of the plastic flow.

REFERENCES [1] "Study on the Stress-Deformation State for Preparation of Own Software for 3D

Static and Dynamic Nonlinear Analysis of Earth-Fill Dams PROC3DN ver 1.2, IZIIS,Skopje, V.,Mircevska, V., Bickovski,

[2] Finite Element Code for Soil and Rock Analyses, " PLAXIS version. 8.1, Technical University of DELFT- Netherland

[3] A 3D Nonlinear dynamic analysis of a rock-fill dam based on IZIIS software ,V. Mircevska, V., Bickovski, M., Garevski, ACTA GEOTECHICA SLOVENICA” vol.4, 2007/2,pp.17-32

389

Violeta Mir evska1 Vladimir Bi kovski2 Mihail Garevski3

BENCHMARK TEST softwera FILT3D-IZIIS

Rezime

Cilj ove analize je benchmark test geotehni kog softwera FILT3D-ver.1.2 razvijen u IZIIS-u sa ciljem demonstriranje objektivnosti i primenljivosti izlaznih rezultata. Za uspore uju i softwer izabran je komercijalni geotehni ki kompjuterski program FLAXIS ver.8.1 Univ. Of Delft-Nitherland. Uspore eni su rezultati dobijeni od oba programa kao porni pritisci, potencijali i brzine filtracija, za sistem brana-duboki iskop. Budu i da je filtraciono stanje plošnog tipa a kako PLAXIS podržava 2D analizu a FIL3D podržava 3D analizu, komparativni model je definisan grani nim uslovima koji u potpunosti simuliraju površinsko stanje filtracije.

Kljucne reci: stacionarna filtracija, benchmark test

A BENCHMARK TEST OF THE SOFTWARE FILT3D-IZIIS

Summary

The objectiv of this work is a benchmark run of the geotechnical software FILT3D–ver.1.2 developed at IZIIS in oder to assess it’s objective performance as well as it’s correctness and applicability. For benchmarking software choosen is the comercial geotechnical computer programs PLAXISver.8.1,univ.Delft – Netherland. Analyzed and compared are the distribution of potentials, pore pressures and filtration velocities through the porous media. PLAXIS program supports 2D analysis while FILT3D program supports 3D analyses wherefore the comparative model has been defined by boundary conditions that thoroughly simulate the plain state of filtration

Keywords : stationary filtration. benchmark test

___________________________________________ 1.Assoc. Prof. Dr., (IZIIS),Skopje, Republic of Macedonia 2. Prof.Emeritus, (IZIIS), Skopje, Republic of Macedonia 3. Prof. Dr., (IZIIS) Skopje, Republic of Macedonia

390

1. INTRODUCTION

The existing analyzed dam belongs to the type of low hydrotechnical structures with a height of 12 m from the lake bottom to the dam crest. The width of the dam crest is 10. m. The slopes on the upstream and the downstream side of the dam are 1:3. This dyke forms an artificial lake, with depth of 3 m. On the downstream side of the dam, for the needs of construction of the Building for Living and Services, a cut in the dam body, at 10 m from the crest, parallel to the direction of the crest within length of 40 m, is to be made. With this cut with a depth of 7 m, the consolidated stability of the dam-soil system is disturbed. It is necessary to stabilize the system dam-excavation. The supporting structure adopted in the analyses represents an RC wall with a thickness of 1,0 m along the entire length of the excavation with embedment of 2m. The objective of the analysis has been the proving of the short term stability of the dam “Artificial Lake’s Dyke” in the vicinity of Tirana after the excavation. The filtration state of the dam-excavation system is of a plain type since the geometry of the filtration domain and the filtration conditions existing at the boundaries of the filtration domain are invariable in longitudinal direction of the dam, i.e., the dam crest. The PLAXIS ver. 8.1 program supporting 2D analysis has been used. The FILT3D ver. 1.2 program supports 3D analyses wherefore the corresponding comparative model has been defined by a thickness of 1 m and boundary conditions that thoroughly simulate the plain state of deformations.

2. RESULTS FROM THE STEADY SEEPAGE ANALYSIS

The analysis of stationary filtration through the dam and the soil has been carried out for a reservoir water level at altitude of 117.25. The filtration coefficients for the whole

porous media have been adopted as follows sec/10 7 mkkk zzyyxx . The filtration

conditions on the external geometry boundaries of the mathematical model are of two types. The prescribed specific discharge normal to the bottom boundary and both lateral boundaries are of type .0/ nW , which means that the specific discharge is equal to zero. The prescribed active groundwater head (potential) on the face of the upstream side of the dam and the bottom of the reservoir has a value equal to 21.73m, which is, in fact, the water level in the reservoir in respect to the referent coordinate system selected for analysis of the stationary filtration. In reality, this corresponds to the altitude of 117.25. At all free surfaces, the value of the active ground water head equals the geodetic height in respect to the referent coordinate system, fig (1). Fig. (2) shows the active ground water head, whose

extreme value amounts to mW 73.21max , whereas its minimal value is mW .13min

obtained by use of Plaxis, whereat the corresponding results obtained by use of FILT3D are presented on fig.(6). Figures (3) and (4) show the distribution of the active pore pressure obtained by use of Plaxis. The extreme value of the pore pressure is

2max /217 mkNU and minimal value is

2min /0.0 mkNU . The pore pressure in the

zone below the left supporting wall ranges within 2/7550 mkNU

. The corresponding results obtained by use of FILT3D are presented on fig.(7). Fig. (5) shows the velocity of

391

the stationary filtration of the water fluid through the porous media, obtained by use of Plaxis. Extreme value of filtration velocity is detected in the immediate vicinity below the

left supporting wall amounting to sec/10*06.75 9max mV . The corresponding results

obtained by use of FILT3D are presented on fig.(8).

Figure 1: Filtration conditions on the boundaries of the analyzed domain

Figure 2: Active groundwater head with extreme head Wmax=21.73m

392

Figure 3: Active pore pressures with extreme pore pressure Umax=217.19kN/m2

Figure 4: Active pore pressure Umax=217.19kN/m2

393

Figure 5: Velocity of the flow with extreme value Vmax=75.06 10-9 m/sec

Figure 6: Active groundwater head with extreme head Wmax=21.725m

Figure 7: Active pore pressure with extreme value Umax=220.5kN/m2

394

Figure 8: Velocity of the flow with extreme value Vmax=78.10 10-9 m/sec

4. Conclusion

Generating completely the same mathematical model for analysis of stationary filtration and assigning the same boundary filtration conditions, the steady seepage state has been solved by using both software FILT3D ver.1.2 – IZIIS and Plaxis ver. 8.1 - Technical University of Delft – Netherland.. The obtained results are almost identical.

References

[1] "Study on Stationary Filtration Processes in porous media” Software FILT3D ver 1.2, IZIIS,Skopje, V.,Mircevska, V., Bickovski, [2]Finite Element Code for Soil and Rock Analyses, " PLAXIS version. 8.1, Technical University of DELFT- Netherland

[3] A 3D Nonlinear dynamic analysis of a rock-fill dam based on IZIIS software ,V. Mircevska, V., Bickovski, M., Garevski, ACTA GEOTECHICA SLOVENICA” vol.4, 2007/2,pp.17-32

395

Dragan Luki 113, Elefterija Alatanovi 214

MERODAVNI INIOCI SEIZMI KE ANALIZE DVA PARALELNA BLIKA TUNELSKA OBJEKTA

Rezime: Danas je u svetu sve eš a praksa izgradnja dva paralelna tunelska objekta ije su prednosti smanjenje pre nika tunelskog profila, a samim tim i veli ine pomeranja tla izazvanog konstrukcijom tunela. Do današnjih dana, naime, veoma mali broj nau nostraživa kih radova se bavio odgovorom dva pralelna tunelska iskopa na veoma bilskom me usovnom rastojanju, njihova interakcija ne može biti zanemarena. U prvom delu rada analizira se ponašanje ove vrste objekata na osnovu iskustva nekih od najzna ajnijih zabeleženih zemljotresa. U drugom delu posebno su istaknuti merodavni parametri seizmi ke analize dva bliska tunelska objekata. Klju ne re i: zemljotres, odgovor tunela na seizmi ke uticaje, dva bliska tunelska objekata, merodavni inioci

RULING PARAMETRES FOR SEISMIC ANALYSIS OF TWIN'TUNNELSSummary:

Two smaller one-way tunnels are preferred to construct nowadays, which presents major advantages, such as the reduction of both the tunnel diameter and the soi8l movement resulting from the tunnel construction. Up to now, however, rather limited work has been focused on the dynamic response of two proximate paralllel tunnel structures, particularly from the aspect of their minimum safe distance, which is important especially during earthquakes. If twin tunnnels are very close, their interaction is not negligible. The firs part of the paper discusses an analysis of behavoir of this type of facilities given according to some of the most significatnt earthyuakes experienced so far. In the second part, a special attention was given to the problem of ruling parameters for seismic analysis of two tunnels driven in clos proximity.Keywords: earthyuake, seismic response of a tunnel, twin-tunnels, ruling parameters

1 Associated professor, grad.civ.eng., Faculty of Civil Engineering, University of subotica, Serbia2 Assistant, grad.civ.eng., Faculty of Civil Engineering and Architecture, University of nisš, Serbia

396

1. INTRODUCION

Modern flows and everyday people's lives point out the fact that the need to make use of space under the ground today is the greatest than ever. Ever increasing population of large cities, density of transportation and need for storage capacity have led, invetably, to an increased use of undergound facilities, that are an tegral part of the intreastructure of the modern society and are used for a wide range of applications, including subways and railways, higways, material storage, and swage and water transport. For the reasosns of the overpppulation and the lake of a space, tunnels have a significant role in the development of urban areas. Some of these areas are prone to frequakes. Even fifty percent of the world's population live in urban areas, and seventy percent of that population live in earthquakeprone areas. The city of Niš, Serbia, and Balkan region as well, are in the seismic active area.

Initially, tunnels were designed with no regard to seismic effects, while, being confined by the surrounding rock or soil, these structures habe long been assumed to have good antiseismic ability, unless they are located within active faults, or within liquefied soil zones. Thus in a very big time, the damage of tunnels did not take enough atention as free-standing structures. A s the tunnel number and its seismic damage incereased, particularly during several recent earethyuakes, the assumption is shown to be incorrect. There is an increasing awareness of the vulnerability of these structures to seismic activity and this problem attracts the attention of experts of all countries, reviving the interest in the associated analysis and desing methods.

Quite often tunnels are located under densely populated urban areas, and reyuire very high standards in a sense of their safety. Yet, even in the most developed industial countries this engineering tiled is not fully researched and well known so far, and there is a huge discrepancy between current regulations for underground facilities, particularly in terms of earthyuake acitvity, and the requirements for desing and construction of cost'efficient and safe underground structures.

2. ANALYSIS OF BEHAVOIR OF TUNNELS ACCORDING TO SOME OF THE MOST SIGNFICANT EARTHQUAKES

Historically, underground facilities have experienced a lower rate of damage than surface structures. Nevertheless, some underground structures have experienced significant damage in recent large earthquakes, including the 1995 Kobe, Japan earthquake, the 19999 Chi-Chi, Taiwan earthquake, and the 1999 kocaeli and the 2001 Duzce, Turkey earthguakes.

In 1923, Kanto earthyuake in japan, when over 80% of the more than hunderd tunnels in the disaster area damaged, in one of the structures movement of the arc wall reached 25 cm, the tunnel bottom knobbed reach to 1 m, the shrinkage of cross section reached 50 cm.

In 1971, St. Fernando earthquake of USA – it was the first time that earthquake engineers linked the structural damage caused by an earthquake to the impulisve character of the nearfault grund motion. One tunnel close to Sylmar fault reached vertical displecement up to 2.29 m.

397

Hyogoken Nanbu (Osaka-Kabe) earthquake in 1995 is the first case of severe earthquakeinduced damage to modern underground facilities. The earthquake caused the failure of parts of undergrund Daikai Station that was located away from an active fault, and where the ground did not experience soil liquefaction. The earthquake stimulated the sharp rise in demond for rational desing procedures for urban facilities.

In Chi-Chi earthquake of Taiwan in 1999, tunnel damage have a lot to do with the construction age – 44% of damaged tunnels were those built before 1980, 14% built after 1980.

In 2001, Duzce earthquake in Turkey, 400 m long stretch under construcition of the Bolu tunnel collapsed.

Some ilustrative examples are presented at the folloxing figures:

Image 1 – Behavoir of tunnels according to some most significant earthquakes: a) Kern County earthquake; b) Mid North Iwate earthquake, c) Off Iyu Oshima earthquake

1952 Kern County earthquake – the rails are bent, showing that the 46-cm-thick tunnel wall was lifted up enough for the rail to silde underneath.

1998 Mid Nort Iwate earthquake (Kakkonda hydroelectric power station) – outlet tunnel was crack up into sevaral large block, some thick blocks were pushed in the tunnel reducing the tunnel's cross-section, one of them completely fell down allowing the soil with big boulderes (30-50 cm) to fall into the tunnel.

1970 Off Izu Oshima earthquake – the example given herein shows that a tunnel can keep its cross section almost intact even after experiencing some large fault dislocations, reaching about 1m .The cross sections were only pushed slightly out of shape.

Dowing and Rozen (1978) studied the response of 71 tunnels in rock to earthquake motions. The dmage ranged from cracking to slosure in 42 cases. Sharma and Judd )1991) compiled a database on the response od 192 tunnels during 85 earthquakes throughout the world; 94 of the tunnels suffered from small to heavy damage. More than half the damage reporeted was caused by events that exsceeded magnitude 7 od the Richter scale, and nearly 75% of the damage reported occurred within 50 km of the earthquake epicenter. There was no damage in tunnels where the horizontal peak ground accelartation was up to 0.2 g. In most cases where damage was reported, the peak ground accelartions were larger than 0.4 g. The data show that shalow tunnels are at greater risk during earthquakes than deeper tunnels; roughly 60% of the total cases had overburden depths less than 50 m and suffered some damage. Ground type is also important; 79% of the opening sxcavated in soil were reported to have suffered some damage [1]

398

3. RESPONSE OF TUNNEL STRUCTURES TO SEISMIC WAVES

Ground shaking refers to the deformation of the fround procuced by seismic waves propagating throungh the earth's crust. The major factors influencing shaking damage inculude the shape, dimaensions and depth of the structure, the properties of the surrounding soil or rock, the preoperties of the structure and the severtiy of the ground shaking.

Underground structures have features that make their seimic behavior distinct from most surface structures, most notably, their complete enclosure in soil or rock, and their significant length. The design of underground facilities to withstand seismic loading thus has aspects that are very different from the seismic desing of surface structures. In general, seismic desing load for underground structures are characteriyed in terms of the deformation and stains imposed on the structure by the surroundin ground (image 2, left above), often due to the interaction betweren these two. In contrest, surface structures are desingned for the inertial forces caused by ground yccelartions. Earthyuake damage to tunnel structures has also proven to be better coralated with paticle velocity and displacement than accelaration.

Seismic desing of tunnel structures is unique in several ways. For most underground structures, the inertia of the surrounding soil is large realtive to the inertia of the structure. Up to now experience has shown that the response of a tunnel is dominated by the surrounding ground response and not the inertial properies of the tunnel structure itself. The focus of underground seismic desing is on the free-field deformation of the fround and its interaction with the structure. The emphasis on displacement is in stark contrest to the design of surface structures, which focuses on inertal effects of the structure itself.

Tunel structure – fround intereaction under seismic impact is to a great extent more complex agains that one considering surface structures, in which case only foundations are exposed to soil-structure interaction, and vibrations of soil particles imposed to fundations are being transmitted to a structure above the ground. On the contraty, as to tunnel structures, soilstructure interaction is induced along an overalll controur of the structure, and a shape of interaction depends mainly on a type of a construction prcedure, i.e. on a methodology of excavation and instalin of a tunnel support system.

For long structures such as tunnels, different ground motions may be encountered by different parts of the structure (the motion could vary significantly in amplitude and phase along tunnel's length), and traveling wave effects must be considered. This spatial incoherence may have a significant imapct on the response of the structure, since it tend to increase the strains and stresses in the longituidanal direction. There are forum major factors that may cause spatial incoherence: wave-passage effects, extended source effects, ray-path effects saused by inhomogenties along the travel path and local soil site effects. The generation of ground motion time histories with appropriate spatial incoherence is a critical task if the designer is to compute differential strains and force buildup along a tunnel length. The designer will have to work closely bith and engineering seismologist to identify the relevant factors contributing to ground motion incohernce at a spesific site and to generate appropriate ground motion time histories.

399

The deformation can be quite complex due to the interaction of seismic waves with surficial sof deposits and the generation of surface waves. For engineering desing purposes, these complex deformation modes are simplified into their primary modes. Tunnel structures can be assumed to undero three primary modes of deformation during seismic shaking [2

3.1. AXIAL DEFORMATIONS

Axial deformations (imate 2, right above) are the simpliest mode to consider, and they are generated by the cmponents of seismic waves that procude motion parallel to the axis of the subsurface exavation and couse alternating longitudinal compession and tension. The casse of a tunnel structure subjected to an axially propagating wave is slightly more complex since there will be some interaction between the structure and the ground. The interaction becomes more important if the ground is soft and shear stress transfer between the ground and the structure is limited by the interaface shear strength.

3.2. CURVATURE (BENDING) DEFORMATIONS

Bending deformations (image 2, lift below) are caused by the components of seismic waves that propagate in the drection along tunnel axis, producing partical motions perpendicular to the longitudinal axis. Diagonally propagating waves subject different parts of the structure to out-of-phase displacements (image 2, middle below), resulting in a longitudinal compression-rarefaction wave traveling along the structure. In general, larger displacement amplitudes are associated with longer wavelengths, while maximum curvatures are produced by shorter wavelengths with relatively small displacement amplitudes (Kuesel, 1969).

3.3. SHEAR DEFORMATIONS Ovaling defromations of a tunnel section (image 2, right below) develop when

shear waves propagate normal or nearly normat to the tunnel axis, resulting in a distrorsion of the cross-sectional shape of the tunnel lining.

Image 2 – Propagation of seismic waves and three types of deformations as the response of tunnel structures to seismic motions (Owen and Scholll, 1981).

400

4. TWIN TUNNEL STRUCTURES

The latest research studies in the field of tunneling and earthquake engineering has underlined developing udnerground traffic as very imperative. For a single tunnel, a large cross section area would be necessary, in which case the volume of excavated material, the influence range and required techniques for contruction would be much greater than the construction of twin tunnnels. Thus two smaller one-way tunnels are proffered to construct nowadays (image 3, a), which present major advantages, such as the reduction of the both the tunnel diameter and the soil movement resulting from the tunnel construction. In some cities, the geotechnical and underground conditions impose the construction of new tunnels close to existing one.

In the recent time, construction of the third tunnel between two main structures became a practice, for safety tunnels in case of bursting out the fire (image 3, b), or as a ramp for the need to connenct the main tunnels located deep underground to ground level (image 3, c) [6 ]

Image 3 – Examples of tunnel structures as multiple closely spaced cavities

One of the most interesting cesses in a contemporary world practice is the Wushaoling Tunnnel, famous for the length of22.05 km (the longest tunnel in China at present, it has been completed), along which are encountered severeal regional fault yonse. The Tunnel comprised of two parallel separate tunnels. Especially, within the Fault Zone No. 7 that has a width of about 785 m, and the most complex geological conditions such as squeezing strata, due to excessive deformations, in order to speed up the tunnel construction, another two advancing drifts were made to go throught the collapsed regions. There exist four closely spaced tunnels within the fault zone, and the distances between every two tunnels are 40 m (centre-to-centre).

Image 4 – Four tunnels driven in squeeying fault yone of Wushaoling Tunnel [3]

401

Up to now, however, rather limited work has been focused on the dynamic response of twvo proximate parallel tunnel structures. Considering all the facts above, a main goal of this work is to highlight a great importance of finding an optimal solution for the construction of twin-tunnels, particularly from the aspect of their minimum safe distance, which is important especially during earthyuakes. If twin parallel tunnels are very close, their interaction (i.e. overlapping of indluence zones of individual tunnel) is not negligible. The interaction is an important factor to affect the ground settlement.

6. IMPORTANT ASPECTS AND RULING PARAMETERS OF SEISMIC ANALYSIS OF TWIN TUNNELS

5.1. Analytical solutions on tunneling-induced movements are useful; however, they cannot accommodate all important factors such as complex soil stress-strain behavior, construction details and geological conditions. A potential course of the research is numerical study to be done. The complex nature of the seismic soil-structure interaction problem for tunnel structures may require the use of numerical methods. The method of dynamic analysis or the “soil-structure interaction” is based on the fact that the presence of an underground structure modifies the fre field ground deformation. In this approach, a dynamic soil-structure interaction is conducted using numerical analysis tools such as finite element or finite difference methods. Dynamic analysis approach is mostly applicable in a case of structures laid in saturated soil of low bearing capacity [4], which will contribute to developing a new or to improving the existing desing models.

Besides, for the sake of the comprehensive understanding of the effects induced by construction of tunnels or by an earthquake action, it would be appropriate to combine both the experimental (image 5) and numerical analysis of the seismis response of tunnels. A verification of the proposed mathematical model would be based on a comprison of computed values with values measured at an experimental model of a tunnel. Thus, the obtained results of a numerical analysis would be verified, justifiability of application of the initial hypotheses would be checked, and by that a detailed and comprehensive seismic analysis of twin-tunnels could be accomplished. Realization of experimental investigations is often not certain, because of the too complex requirements, not only for the financial reasons, but also from the aspect of place and time needed for organizing the experiment.

Image 5 – Seismic response analysis of twin-tunnels in shaking table test [5]

When the word about the type of the analysis is (2D or 3D), it should be emphasiyed that there could be inconsistencies between twvo analyses. Namely, 2D and 3D

402

analyses do not agree in the soft soil, and 2D underestimates displacements occur ar different time steps. One of the possibilities is to approach the problem from the aspect of a plane-strain state, but conditions such as the model dimensions, boundary conditions and analysis constants are to be set identically to those in the 3D analysis (image 6).

Image 6 – Comparision between 2D and Image 7 – Horizontal, vertical and 3D analyses [6] inclined alignment of twin-tunnels [7]

5.2. A research sould be done referring to three configurations of the twin tunnels: aligned horizontally, vertically and inclined. The highest soil settlement is obtained for vertical aligned tunnels, while horizontal aligend tunnels cause the lowest setlements (image 7).

5.3. When it comes to surrounding medium, there are plenty of possibilities: - tunneling-induced soil movement in soft ground is increasingly common

geotechnical activity for construction of transportation in populated urban; this is because ground setlement and subsurface movement usually cause potential damge to existing structures (image 8).

Image 8 – Twin tunnels' induced Image 9 – Fluid saturated settlement in soft soils [8] porous elastic formation [9]

- the analytical or numerical solutions involving multiple cavities in a fluidsaturated porous elastic formation (multi-phase medium) seem to be nonexsistent in the literature (image 9); presence of water in the soil with low bearing capacity (soft soil) further aggravates soil response to seismic influence [15], [16]; namely, when subjected to action of seismic waves, such deposits are prone toliquefaction (soil flow), which results in floating of the tunnel structure on the water saturated subsoil;

- layered formations with different material proparties, because of usually encountering in urban areas (image 10);

403

Image 10 – Multi-layered soil medium – two and three layers [10]

- fractured medium, considering that fractures not only reflect seismic energy, but also serve as host for interface waves that can propagate for long distances with less attenuation than direct waves.

5.4. In oreder to include soil-structure interaction, the analysis should empoloy lined circular tunnels where the lining could be sonsidered as integrated or jointed. Esitmation of the dynamic internal forces in the lining structure is a key procedure in the seismic design of tunnels, allowing a better understanding of fynamic interaction between the lining and its surrounding media. In the course of the past years, appliance of TBM gave rise to the jointed linings where could be distinguished straight-jointed and staggered-jointed rings. Presence of joints could result in up to 50% reduction in the developed moment (image 11).

Image 11 – Straight-jointed and Image 12 – Interaction of twvin straggered-jointed lining structures [11] tunnels – depth relationship [8]

5.5. Tunnel structure deformation decreases with the increase of the thickness of layer above the tunnel and the deep-laid tunnels are safer and less sensitive to earthyuakes, as opposed to the tunnels with shallow embedement depth. Taking into consideration the conclusion that when burial depth is large, interaction of the twin-tunnels is smaller, for the sake of the safe construction in urban areas, it would be favorable to study shallow-buried tunnel structures, as the interaction is an important factor to affect the ground settlement (image 12).

5.6 When it comes to the loading which should be employed in the analysis, three different stress fields could be pointed out:

404

a) Initial stress field (in situ stresses prior to tunnel excavation) induced by weight of soil, earth pressure, and eventually water pressure. Impact of initial stresses on the over-stressed zones for tunnels for the first time was considered recently. As depicted in the following figures, firmer soil influence over-stressed zone decreasing (image 13, left), and when K (horizontal pressure – vertical pressure ratio) closes to 1, overstressed zones and induced displacements around each tunnel cecome more symmetry (image 13, right);

Image 13 – Impact of initial stresses [10]

b) Induced stresses should be taken into consideration in the form of disturbed or degraded zone around bored tunnels (image 14) due to the effects of construction and disturbance on the in situ stress field, which results in increasing moment and decreasing thrust [15], [16]. It would be of great importance to try take into consideration the influence of TBM loading (image 15), and to examine what would be the seismic response of a tunnel when it is under construction. Thus there could be two comparable possibilities – simultaneous excavation of twin-tunnels vs. excavation of a new tunnel close to existing one [17].

Image 14 – Degraded yone Image 15 – Analysis of the around a bored tunnel [12] influence of TBM loading [13]

c) Incident (special) loading by seismic waves and faults. Dynamic response of twin-tunnels could be analyzed drugin the passage of

compressive and shear waves (separately of combined) [9]. In case of shallow-buried tunnels surface Rayleigh waves should be considered too, because of the fact that they induce motions and stresses in the soil and structure which are substantially different. Despite the significant progress in understanding the seismic behavior of underground structures, vey little is known about their response, when the soil is excited from the passage of Rayleigh waves. The deformation of the strcture is usually greater under these

405

waves, and the bending deformation is the greatest [14]. Earthyuake effects on tunnel structures take the form of deformations that cannot be changed significantly bu strengthening the structure. The structure should instead be designed with sufficient ductility to absorb the imposed deformations without losing the capacity to carry static loads.

A careful review of the damages saused by earthyuakes to underground structures shows that most damaged tunnels were located in the vicinity of the causative fault. Uner such conditions ground motion in affected by near-fault effects and the induced strain filed is quite complex, characterized by a strong intensity and a relatively short duration (pulse-like motion). Thus, instead of investigation of on wave scattering of seismic waves, alternative in raserach work could be twin-tunnels affected by the focal mechanism, the direction of rupture propagation relative to the site (directivity effects) and by the permanent ground displacement resulting from the fault slip (fling step), where one tunnel structure is at a footwall, and the other one at the hanging wall of the fault [17]. The general philosophy is to desing the structure to accommodate expacted fauld displacements and allow repair of damaged lining afterwards.

6. CONCLUSION

Analysis of seismic behavior of a tunnel is a complex task since it involves the interaction with several disciplines including soil, rock and structural dynamics, structural geology, seismotectonics and engineering seismology. In addition, underground structures are crucial elements in transporation network and the occurrence of a seismic event not only determine a potential loss of human lives, but also damages to the infrastructures that could severaly affect the economy of a region due to the time required to restore the functionality of the network.

Taking all into account, finding an optimal solution for the construction of twin-tunnels is assumed to be a serious task, but it would be of great importance, particularly considering that desing rules for tunnels are not introduced in Eurocode 8 (EN 1998-5, 2003).

REFERENCES

[1] Geotechnical Earthquake Engineering – Simplified analyses with case studies and examples / Mulutin Srbulov // Springer , 2008, 244 pages

[2] Seismic design and analysis of underground structures / Hasash Y. et al. // Tunneling and Underground Space Technology 16, 2001, pp. 247-293

[3] Interaction of four runnels driven in squeeying fault zone of WushaolingTunnel / Yang J.S., Yan L., Deng S.J, Li G.L. // Tunneling and Underground Space Technology 21, 2006

[4] Determination of the minimum seismically safe distance between two parallel tunnels / Fotieva NN. // Osnovania, Fundamenti I Mehanika Gruntov, 1980

[5] Numerical simulation of dynamic soil-structure interaction in shaking table testing / Pitilakis D., Dietz M., Wood D.M., Clouteau D., Modaressi A. // Soil Dynamics and Earthquake Engineering 28, 2008, pp. 453-467

406

[6] Full 3D seismic response analysis of underground ramp tunnel using large-scale numerical computation / Dobaschi H. et al. // World Conference on Earthquake Engineering, Beijing, China, 2008

[7] Numerical analysis of the interaction between twin tunnels: Influence of the relative position and construction procedure / Hage Chehade, Shahrour // Tunneling and Underground Space Technology 23, 2008, pp. 210-214

[8] Twin tunnels-induced ground settlement in soft soils / Wang et al. // Sino-Japanese Symposium on Geotechnical Engineering, Beijing, China, 2003

[9] Harmonic wave diffraction by two circular cavities in o poroelestic formation / Seyyed Hasheminejad M., Avazmohammadi R. // Soil Dynamics and Earthquake Engineering 27, 2007, pp. 29-41

[10] Mechanical behavior of a twin-tunnel in multi-layered formations / Bin-Li Chu et al. // Tunneling and Underground Space Technology 22, 2007, pp. 351-362

[11] Dynamic behavior of shield tunnels in the transverse direction considering the effects of secondary lining / Mizuno, Koiyumi // European Conference on Earthquake Engineering, Geneva, Swityerland, 2006

[12] Simplified analysis of seismic in-plane stresses in composite and jointed tunnel lining / Hany El Naggar et al. // Soil Dynamics and Earthquake Engineering 28, 2008, pp. 1063-1077

[13] On the influence of face pressure, grouting pressure and TBM design in soft ground tunneling / Kasper T., Meschke G. // Tunneling and Underground Space Technology 21, 2006, pp. 160-171

[14] Dynamic response of lutility tunnel during the passage of Raylegh waves / Yue Q.X., Li J. // World Conference on Earthyuake Engineering, Deijing, China, 2008

[15] Seismic loading on shallow-laid underground structures / Zlatanovi E., Luki D. // International Conference on Civil Engineering Desing and Donstruction, Subotica, Serbia, 2007, pp. 377-382

[16] Influence of earthquakes on the stress and strain state of the shallow-buried tunnel structures in saturated soil of low bearing capacity / Zlatanovi E. // Scientific journal “Facta Universitatis”, Serise: Architecture and Civil Engineering Vol. 6, No 2, Niš, Serbia, 2008, pp. 221-227

[17] Twin-tunnel structures from the aspect of seismic influences / Zlatanovi E. // 1 st SCEES – German – Sout East Eruopean Conference on Civil and Environmental Engineering Sciences, Bochum, Germany, Decembar 7-10, 2008

407

Ratko Salati 115

DISIPACIJA ENERGIJE U POLUKRUTIM VEZAMA

Rezime:Energija zemljotresa predata konstrukciji apsorbuje se i disipira na razli ite na ine. U eli nim ramovima najve i deo energije se disipira u vornim vezama kroz mehanizam histerezisnog ponašanja. Polukrute veze smanjuju ukupnu krutost ramova, ali istovremeno zna ajno pove avaju kapacitet disipacije enrgije rama. Ova injenica se može iskoristiti za projektovanje seizmi ki otpornih objekata. Formiran je numeri ki model grede kojim se može obuhvatiti i histerezisno i viskozno prigušenje u vezama. Fleksibilna veza je modelirana oprugom sa nelinearnom vezom moment-rotacija. Numeri ki rezultati analize desetospratnog rama dati su na ilustrativnim dijagramima. Klju ne rije i: polukrute veze, viskozno i histerezisno prigušenje

ENERGY DISSIPATION IN SEMI-RIGID CONNECTIONS

Summary:In structures subjected to earthquake, the input energy is absorbed and dissipated through different mechanisms. The main source of energy dissipation in steel frames can be hysteretic behavior of connections. Semi-rigid connections reduce the overall frame stiffness, but also significantly increase the energy dissipation capacity of steel frames. This conclusion may be used to achieve economical earthquake-resistant building structures. A numerical element model of beam that can involve both hysteretic and viscous damping is developed. A flexible connection is modeled by rotational spring with nonlinear moment rotation relationship. The results of ten-story frame calculations have been presented with illustrative diagrams and figures. Key words: Semi-rigid connections, Viscous and hysteretic damping

1 , , , e73

408

1 INRODUCTION The analysis of steel frames is traditionally based on the assumption that a

connection between beam and column is either infinitely rigid or perfectly pinned. However, numerous experimental investigations show that the actual response of joints is between these two extreme cases, i.e. all connections transmit some moments and exhibit certain degree of flexibility. This problem is of particular importance in frames subjected to earthquake loading as a substantial amount of energy is dissipated in connections due to friction, local buckling, plastic yielding and distortions. According to this concept the whole energy dissipation is lumped in joints.

Viscous damping due to relative rotational velocity in connection is introduced. The flexural complex stiffness matrix for a beam element with semi-rigid connections and viscous rotational dashpots is evaluated.

Flexible connections through energy dissipation greatly influence the dynamic behaviour of steel frames. Under cyclic loads, the connection hysteretic loop increases the energy absorption capacity and hysteretic damping may significantly reduce dynamic response of real structures. Therefore, modeling of the nodal connection is important for the design and accuracy in the dynamic frame structure analysis.

Two types of nonlinearities are considered: geometric nonlinearity of the structure and material (constitutive) nonlinearity of the connections. Numerical experiments are conducted on a simple frame subjected to seismic excitation in order to show the influence of various parameters on dynamic response due to hysteretic and viscous damping in connections.

2 BEAM ELEMENT WITH SEMI-RIGID CONNECTIONS In our research, semi-rigid connections between beams and columns are modelled

by rotational springs with nonlinear moment-rotation relationships. The proposed element has the same number of degrees of freedom as the corresponding model in the conventional analysis used for the frames with fully rigid connections. In order to study viscous damping phenomena in beams with flexible connections, rotational springs and rotational viscous dashpots are attached at beam ends (Fig.1). Note that a finite size of nodal assemblages is considered and beam ends are assumed to be at distances e1 and e2 from the nodal points.

Figure 1- A beam element with flexible, eccentric and viscous damping connections

409

1.1 STIFFNESS MATRIX, MASS MATRIX, DAMPING MATRIX For a uniform beam with rotational springs and dashpots attached at its ends the

complex dynamic stiffness matrix is obtained. The stiffness matrix has been obtained based on analytical solutions of governing differential equations second order analysis, so that each beam represents one element. Nodal displacements and rotations are chosen as the primary unknowns, while displacements and rotations of the element ends are eliminated. Thus, the number of degrees of freedom is the same as for the system with rigid connections. Besides, the consistent mass matrix and damping matrix of element are derived. These matrices are based on the physical properties of the member and given in an explicit form, Sekulovic [1].

Expanding the elements of the dynamic stiffness matrix in series with respect to the circular frequency and neglecting higher terms than the third order, the following expansion is obtained in the decomposed form:

...2* mckk j ,

(1) where k is the static stiffness matrix, c the damping matrix and m the mass matrix for the uniform beam with flexible springs and dashpots at its ends.

1.2 SEMI-RIGID CONNECTION MODELING Numerous experimental results show that connection moment - rotation

relationships are non-linear in entire range of loading. In this study three-parameter power model, Richard [2] is used. The generalized form of this model is:

nn

kM 1

0

0

1

(2) where Mu is ultimate moment capacity, n shape parameter, k0 initial connection stiffness and 0=Mu/k0 reference plastic rotation. The shapes of the above equation for two types of connections (Double Web Angle - DWA, Top and Seat Double Web Angle - TSDWA) are shown in Fig. 2 (left). The details of these connections can be seen in Chen [3]. The proposed moment-rotation relationship has been verified through a number of full-scale experiments conducted by various investigators, Kishi [4]. However, the experimental data for the connection behaviour under cyclic loading are rather poor.

410

Figure 2 - Three parameter power model and independent hardening model

To describe the nonlinear behaviour of the connection under cyclic loading, the independent hardening model is used, Fig. 2-right. The effect of hysteretic damping on dynamic behaviour of the structure is directly included through the connection constitutive relation. The skeleton curve used in this model is obtained from three-parameter model.

3 ENERGY DISSIPATION Energy dissipation exists in frame structures under dynamic loads. The primary

sources of energy dissipation may be hysteretic behaviour of connections and the friction between elements forming the beam-column assemblage. In addition, different types of energy dissipation devices can be installed into connection in order to increase the structural energy absorption capacity. For this reason, in the present model, the total energy dissipation is confided to the joint connections. Two types of energy dissipation are assumed. They are: hysteretic damping due to nonlinear behaviour of connections and viscous damping at the connections. In general, the effects of these dampings are coupled. Also, they can be considered separately using either linear constitutive relation for the connections or zero value for the viscous damping coefficients at the connections. As it is assumed that all structural elements, except the connections, remain elastic through the whole loading range, the energy dissipation at plastic hinges cannot be observed. Other types of energy dissipation that may exist in real frame structures can be included in the present model in a usual way, by mass and stiffness proportional damping matrix.

4 NUMERICAL PROCEDURES AND NUMERICAL EXAMPLES The equations of motion of a frame are formed using well-known procedures. The

diagonal system mass matrix consisting of terms corresponding to translational degrees of freedom is used. The system damping matrix and the system stiffness matrix are assembled from element matrices. The equations of motions are integrated using Newmark average acceleration method. Within each time step, an iterative algorithm based on evaluating secant stiffness matrix and damping matrix has been performed to resolve unbalanced forces to zero.

411

Figure 3 - Layout and properties of ten-story frame

Based on the above theoretical considerations, a computer program has been developed and dynamic analysis of ten - story single bay steel frame has been carried out. The geometrical and material properties of the frame investigated are shown in Fig. 3.

One type of semi-rigid connection (TSDWA- Top and Seat angle Double Web Angle) with both linear and nonlinear moment-rotation relation is considered. In order to compare obtained results, the same frame with fully rigid connections is analyzed. Linear (first order) and geometrically nonlinear (second order) analyses were carried out.

The natural frequencies and the corresponding periods for the first three modes are determined for the cases of fully rigid and linear semi-rigid connection and shown in Table 1.

Table 1 - Natural frequencies of the frame investigated Type of

connectionNatural frequencies

(rad/s) Periods

(s)mode First Second Third First Second Third Rigid 6.328 17.523 31.116 0.993 0.359 0.202

TSDWA 5.727 16.088 28.611 1.097 0.391 0.220

The frame is assumed to be subjected to the first four seconds of Petrovac, (NS component) earthquake motion. The displacement response at the top of the frame with both linear and nonlinear connection TSDWA and rigid jointed frame according to the linear and second order analyses are plotted in Fig. 4 and Fig. 5.

There is a considerable difference between the responses for rigid jointed frames and frames with nonlinear connections, due to hysteretic damping, which exists only in nonlinear connections. The linear moment-rotation TSDWA connection type exhibits similar behaviour as the fully rigid connection type and can be a good model only for low-level loads. Fig.4 and Fig.5 show that in nonlinear types of connection there are permanent deflection drift (due to large connection rotation) in both first order and second order analysis.

412

The lateral displacement and shear force envelopes of the frame with the previous connection types obtained according to the linear and second order analyses are shown in Fig.6 and Fig.7. The frames with flexible nonlinear connection under applied earthquake motion have smaller both lateral displacements and shear forces when compared with the rigid jointed frame. That can be explained by taking into account the fact that any earthquake is an excitation with wide range of frequencies. The predominant frequencies of the applied earthquake are within the range from 2 to 10 Hz (periods 0.1 to 0.5 second). The lowest natural frequencies of the investigated frames (Rigid, TSDWA) are much higher than the predominant earthquake frequencies, while the second and the third natural frequencies are within the range of predominant frequencies of the applied earthquake. It obviously has a great influence on displacement response of these frames.

Time history acceleration responses of the frame with rigid and both linear and nonlinear (TSDWA) connections according to linear and second order analyses are shown in Fig. 8. There is a great hysteretic damping effect on the acceleration response of the frame with nonlinear connection. On the contrary, in the case of rigid jointed frame, the acceleration response is not dampened, so the large amplification of the acceleration response exists. The hysteretic M- loop at joint C of the frame with TSDWA type of connection is shown in Fig.9. It can be seen that the connections undergo strong rotational deformations during the applied earthquake motion.

1 2 3 4 5 6 7 8 9 10 Time, t(sec)

Dis

plac

emen

t, u A

(cm

)

-20

-10

10

20TSDWA linear TSDWA nonlinear

Rigid

Figure 4 - Time-history displacement at the top node of the frame, linear analysis

1 2 3 4 5 6 7 8 9 10 Time, t(sec)

Dis

plac

emen

t, u A

(cm

)

-20

-10

10

20TSDWA linear

TSDWA nonlinear

Rigid

Figure 5 - Time-history displacement at the top node of the frame, second-order analysis

413

1

2

3

4

5

6

7

8

9

10

Floo

r lev

el

TSDWA linear

TSDWA nonlinear

Rigid

10 20 Displacement, u(cm) Shear force, T (kN)

Floo

r lev

el

1

2

3

4

5

6

7

8

9

10

200 300 400 500 600 100

TSDWA linear

TSDWA nonlinear

Rigid

b)a)

Figure 6 - Lateral displacement and shear force envelopes according to the linear analysis

Shear force , T (kN )

Floo

r lev

el

1

2

3

4

5

6

7

8

9

10

200 300 400 500 600 100 b )

1

2

3

4

5

6

7

8

9

10

Floo

r lev

el

10 20 D isp lacem ent, u(cm )

TSD WA linear

TSD WA non linear

R ig id

a)

Figure 7 - Lateral displacement and shear force envelopes, second order analysis

1 2 3 4 5 6 7 8 9 10

Time, t(sec)

1 2 3 4 5 6 7 8 9 10Time, t(sec)

Acce

lera

tion,

(m/se

c2 )

-20

-10

10

20

Acce

lera

tion,

(m/s

ec2 )

-20

-10

10

20

30

-30

TSDWA linear TSDWA nonlinear

Rigid

TSDWA linear TSDWA nonlinear

Rigid

a)

b)

Figure 8 - Time-history at the frame’s top node, first order (a), second-order (b) analysis

414

Rotation, (rad) 10-4

5

100

-10 -5-20 -15-25

Bend

ing

mom

ent,

M (k

Nm

) 200

300

400

500

-500

-400

-300

-100

-200

10

Figure 9: Hysteretic M- loop at joint C with TSDWA connections

The foregoing results and conclusions based on time history responses and displacement and force envelopes can be presented through energy balance approach. In the case of fully rigid or flexible connections with linear moment-rotation relations, the input seismic energy is transformed into potential and kinetic energy of the frame investigated, Fig. 10a and Fig. 10b. On contrary, in the case of frame with flexible nonlinear connections, the input energy is dissipated by hysteretic behaviour of joint connections, which significantly decreases both the kinetic and recoverable (elastic) strain energy of the system, which diminish at the end of ground shaking, Fig. 10c. Also, it can be noticed that these three connection types do not absorb the same amount of input energy, which decreases as the flexibility of frame connections increase. Time variation of energy dissipated by both viscous and hysteretic damping for nonlinear TSDWA connection type is shown in Fig.11. Viscous damping additionally decreases the kinetic energy of the frame and dissipates less energy than the hysteretic damping.

5 CONCLUSION An efficient method to perform the dynamic analysis of steel frame structures with

flexible connections has been presented in this paper. A numerical model that includes both nonlinear connection behavior and geometric nonlinearity of the structure has been developed. The stiffness matrix was derived according to the analytical solutions of the second order analysis equations, so that each beam is one finite element. Based of the above theoretical considerations and the results of the applied numerical analysis, it is evident that the flexible joint connections greatly influence the dynamic behaviour of steel frames.

An increase in the connection flexibility reduces the frame stiffness, and thus the eigenfrequencies, particularly the lower, which may have a primary influence on dynamic response of the structure.

415

1 2 3 4 5 6 7 8 9 10 Time, t(sec)

1 2 3 4 5 6 7 8 9 10 Time, t(sec)

1 2 3 4 5 6 7 8 9 10 Time, t(sec)

10

20

30

40

50

60

70

80

Ener

gy, E

(0kJ

)

10

20

30

40

50

60

70

80

Ener

gy, E

(0kJ

)

10

20

30

40

50

60

Ener

gy, E

(0kJ

)

Input energy Kinetic energy Elastic strain energy Hysteretic energy

Input energy Kinetic energy Strain energy

Input energy Kinetic energy Strain energy

Rigid connections

TSDWA linear

TSDWA nonlinear

a)

b)

c)

Figure 10 - Time variation of energy for various types of frame connections

416

10

20

30

40

50

60

Ener

gy, E

(0kJ

)

1 2 3 4 5 6 7 8 9 10 Time, t(sec)

Input energy Kinetic energy Elastic strain energy Hysteretic energy

TSDWA nonlinear

Viscous damping energy

c=10000 kNmrad/sec

Figure 11 - Time variation of energy dissipated by both viscous and hysteretic damping

The results of numerical examples show considerable difference in structural responses of frames with flexible nonlinear connections and frames with conventional connection types (fully rigid or linearly flexible). They also show that the effect of hysteretic damping on structural response is significant. Therefore, the nonlinear constitutive model for connections should be used in design and response analysis of real frame structures. It can also be concluded that the viscous damping at connections may considerably reduce the displacement response and internal forces of the frame. Moreover, the conclusions based on the dynamic evaluation of displacements and forces completely correspond to the results based on energy balance approach.

REFERENCES 1. Sekulovic M, Salatic R, Mandic R. Seismic analysis of frames with semi-rigid eccentric

connections. Twelfth Conference on Earthquake Engineering. New Zealand 2000 2. Richard RM, Abbott BJ. Versatile elastic-plastic stress-strain formula. Journal of

Engineering Mechanics Division, ASCE 1975; 101(EM4): 511-515.3. Chen WF, Goto Y, Liew JYR. Stability design of semi-rigid frames. New York: John

Willey & Sons; 1996. 4. Kishi N., Chen WF, Steel Connection Data Bank Program, CE-STR-86-18, West Lafaete,

In: School of Civil Engineering, Purdue University 5. Chen WF, Cishi N. Semirigid steel beam-to-column connections: Data base and modeling.

Journal of Structural Engineering, ASCE; 120 (6): 1703-1717. 6. Chan SL, Shui PPT. Nonlinear static and cyclic analysis of steel frames with semi-rigid

connections. Elsevier; 2000. 7. Sekulovic M, Salatic R. Nonlinear analysis of frames with flexible connections. Computers

and Structures 2001; 79(11): 1097-1107. 8. Salati R. Analysis and Response Control of Steel Frames Subjected to earthquakes, Ph.D.

thesis (in Serbian), University of Belgrade, 2000. 9. Sekulovic M., Salatic R., Mandic R. and Nefovska M., Energy dissipation in steel frames

with semi-rigid connections, 12-th ECEE, London 2002.

417

Matjaž Godec116, Renato Vidrih217, Peter Sin i 318

OSMATRANJE SEIZMI NOSTI NA PODRU JU VELIKIH BRANA U SLOVENIJI

Rezime:Svake godine na našoj planeti nastane više mo nih zemljotresa sa opsežnim posledicama tako e na ve im objektima. Tu svrstavamo tako e velike brane od kojih su mnoge izgra ene na potresno aktivnim podru jima. Za prosu ivanje projektnih optere enja i ponašanje brana, u Sloveniji smo se odlu ili, da je za obezbje enje sigurnosti brana, potrebno nastaviti sa posmatranjem njihove seizmi nosti. U službenem listu RS godine 1999 je bio objavljen Pravilnik o osmatranju seizmi nosti na podru ju velikih brana, koji propisuje na ineosmatranja seizmi nosti, tehni ke normative seizmoloških instrumenata te uslove koje mora ispunjevati izvo a osmatranja uticaja seizmi nosti na velike brane. Key words: velike brane, osmatranje seizmi nosti

SEISMOLOGICAL MONITORING OF LARGE DAMS IN SLOVENIA

Summary: There are many strong earthquakes on our planet each year. They also cause extensial consequences on large structures, such as large dams, which are often build on active seismic regions. To estimate seismic actions and behaviour of the dams, we decided it is necessary to continue seismic monitoring of the dams in order to be sure of their safety. The Regulation of seismic monitoring of large dams was issued in the National Gazette in 1999. It regulates methods of seismic monitoring, technical standards of the seismological equipment, conditions which must be fulfilled by the performer of the seismic monitoring and also defines the conception of large dams. Key words: large dam, seismological monitoring

1 Agencija Republike Slovenije za okolje, Urad za seizmologijo in geologijo 2 dr., Agencija Republike Slovenije za okolje, Urad za seizmologijo in geologijo 3 Agencija Republike Slovenije za okolje, Urad za seizmologijo in geologijo

418

1 UVOD Godišnje je na zemlji zabilježenih preko 3 miliona potresa. Ve ina od njih je tako

slaba da ih ljudi ne osjetimo. Me utim bar 900 potresa godišnje je mo niji od magnitude 5. Takvi potresi mogu prouzrokovati tako e opsežne posledice. Od godine 1900 je v zemljotresima izgubilo život preko 1,6 miliona ljudi, posledice potresa u pojedinim državama zna e pravu privrednu katastrofu. Mnoge brane u svijetu su izgra ene na potresno aktivnim podru jima i po znanim podacima je 74 brana pretrpjelo ošte enje zbog potresa, od toga 27 brana teška ili vrlo teška ošte enja. Zbog potresa su tako pretrpjele ošte enja tako e nama blizu brane u Makedoniji, Rumuniji i Velikoj Britaniji (Huber, 1995).

Slika 1. Razorena brana prilikom zemljotresa 20. septembra 1999 na Tajvanu. Brana Ši – Gang je bila postavljena na tektonskem rasjedu, gdje je došlo pri potresu do površinskog

pretrganja. Na jednoj strani rasjeda je došlo do dizanja 3 – 4 m, na drugoj pa do tonjenja (spuš anja) 1 – 2 m. (foto R. Vidrih).

2 OSMATRANJE SEIZMI NOSTISastavni dio ocjene potresne opasnosti podru ja i lokaliteta brana objekata je

osmatranje seizmi nosti branskih objektata. Osim toga je to osmatranje bistveno važno za ocjene potresne odtpornosti objekata. Tako dobijeni podaci omogo avaju smisaonost projektne odluke kod novogradnji, kod ve postoje ih brana pa omogu avaju što realnije odluke kod popravaka ili oja avanja poslije mogu nih ošte enja zbog potresa. Bilješke

419

stvarnih potresa nam zna e jedine prave rezultate. Tako dobijeni rezultati nam služe za prosu ivanje projektnih optere enja, ponašanje i cjelokupnu ocjenu sigurnosti brana.

Na osnovu Pravilnika o tehni kih normativih za seizmi ko osmatranje visokih brana (Sl. l. SFRJ 6/88) i Zakona o zaštiti okoline (Sl. l. RS 32/93) smo se zato u Sloveniji odlu ili, da je za obezbje ivanje sigurnosti brana potrebno nastaviti sa njihovim potresnim osmatranjem. Kod pripremanja pravilnika bila su upotrebljena kod osiguravanja potresnog osmatranja tako e iskustva drugih država. Iskustva sa potresnim osmatranjem imaju izme u ostalih u Australiji, Austriji, Kanadi, Italiji, Japanu, Švajcarskoj i SAD (Fajfar, 1996).

Pravilnik o osmatranju seizmi nosti na podru ju velikih brana objavljen u Službenem listu RS br. 92 godine 1999 propisuje:

- na in osmatranja inducirane seizmi nosti; koju pri injava voda u sabiralniku, ogra enom sa velikom branom,

- na in osmatranja dinami nog ponašanja tjela i temelja velikih brana, sabiralnika (kolektora) vode, odnosno prostora za njima, te slobodne površine u njihovoj direktnoj blizini prilikom djelovanja potresa,

- tehni ke normative seizmoloških instrumenata i normative za njihovo održavanje i - uslove, koje mora ispunjevati izvo a osmatranja uticaja seizmi nosti na velike

brane.

2.1 VELIKA BRANA Velika brana (dalje u tekstu – brana) po tom pravilniku je: - svaka brana, koja je viša od 15 metara ili - svaka brana izme u 10 i 15 metara visine, koja ispunjava bar jedan od slijede ih

uslova: - dužina krune manja od 500 metara, - sadržaj sabiralnika, kojega napravi brana, ni manja od jednog miliona kubnih

metara, - maksimalna visoka voda, koja uti e na branu, ni manja od 2000 kubnih metara na

sekundu, - brana je imala teške uslove kod postavljanju temelja i - brana je neuobi ajene konstrukcije.

Visina brane je opredeljena u skladu uzetog iz terminologije velikih brana, i visina brane meri se od najniže ta ke temelja do vrha brane (Zadnik, 1997).

2.2 CILJ PRAVILNIKA Namjera pravilnika je osigurati osmatranje inducirane seizmi nosti i osmatranje

dinami noga ponašanja brane. Kod toga zna i: - osmatranje inducirane seizmi nosti obilježavanje i zapisivanje promjena potresne

aktivnosti, koje nastaje zbog vode u sabiralniku, obuzdane u prostoru za velikom branom,

- osmatranje dinami noga ponašanja brane je ozna avanje in bilježenje odziva tjela i temelja brane te slobodne površine oko nje na potres. Obveznik za obezbje ivanje osmatranja inducirane seizmi nosti i osmatranje

dinami noga ponašanja brana po pravilniku je vlastnik brane.

420

Obveznik za postavljanje instrumenta odnosno instrumenata za osmatranje inducirane seizmi nosti (seizmograf) i instrumenata za osmatranje dinami noga ponašanja brane (akcelerograf) je vlastnik brane.

Obveznik za obezbje enje izvo enja radnoga monitoringa je onaj koji upravlja branom.

3 INDUCIRANA (TRIGERIRANA) SEIZMI NOSTPojavljanje potresa, koji su povezani sa djelatnosti ovjeka nazivamo inducirana ili

pravilnije trigerirana seizmi nost. Samo da se ta manifestuje u širokom prostorsko-vremenskom i energetskom rasponu od mikropotresa u neposrednoj blizini izvora promjena do rušila nih zemljotresa sa žariš em u dubini ve oj od deset kilometara.

Poznamo više uzroka za induciranu (trigeriranu) seizmi nost. Naj eš i su: - vodna zajaženja, - injektiranje teku ina i gasova u zemljinu unutrašnjost, - crpanje nafte i gasa, - rudarski radovi u kamenolomima, - crpanje geotermalne energije, - podzemne nuklearne probe,

Potresi trigerirani sa zaježenjima vode spadaju me u mo nije. Do sada je nesumnjivo utvr ena i dokumentirana pojava promjene potresne aktivnosti na bar 120 vodni zaježenja. Najmo niji su bili:

- Koyna, India, 10.12.1967, M = 6,5 - Kremasta, Gr ija 05.02.1966, M = 6,3 - Kariba, Rodezija – Zambija, 23.09.1963, M = 5,8

U svim tim primjerima je bila dubina vode ve a od 80 m. Jednostavna statistika kaže, da je svako pedeseto vodno zaježenje sa dubinom vode ve om od 80 metara aktiviralo potres magnitude 5,7 ali još više.

Najbliži primeri Sloveniji su se desili u Piave de Cadore (Italija), gdje je 13.01.1960 nastao potres i Vajont u Italiji sa potresima magnitude oko 3.

Iako je istraženost ti pojava i znanje o njihovim uzrocima za sada prili no nepotpuno dosta je vjerovatno, da te potrese ne pri injava ovjekova aktivnost. U podru ju, gdje taki potresi nastaju moraju postojati tektonsko ugodni uslovi za postanak potresa i promjene seizmi nosti.

Za jedan put su utvr ena dva mogu a mehanizma, koji mogu aktivirati promjenu prirodne seizmi nosti. U oba primjera ide za smetnje u prirodnom napetosnem stanju. Teža vode u zaježenju izvodi dodatni pritisak u vertikalnom smjeru. Potresi, aktivirani sa tim mehanizmom su po pravilu slabiji (jer je masa vode tako e kod najve ih zajaženja relativno mala u uspre ivanju sa masom stijena ispod zajaženja) i pojavljuje se uskoro (nekoliko dana ili mjeseci) po za etku punjenja zajaženja na manjim dubina (do 3 km) i u neposrednoj blizini. Vrlo esto je broj potresa povezan sa visinom vode ili sa brzinom punjenja i pražnjenja zajaženja. U odre enim uslovima je promjena taka, da se prirodna seizmi nost smanji. Tako je u uslovima, gdje je najve a napetost vertikalna ili ima strm pad, o ekivati pojavu novi potresa (podru ja, u kojima preovla uju normalni i strižni rasjedi) jer težina vode pove ava prirodne napetosti. U podru jima, u kojima je najve anapetost horizontalna, je esta pojava smanjenje prirodne seizmi nosti jer težina vode djeluje suprotno od prirodnih uslova.

421

Drugi mehanizam je pove anje pornega pritiska podzemnih voda koji smanjuje odpor stijene uz rasjed proti strižnim napetostima. Potresi, aktivirani na taj na in nastaju u ve im dubinama (tako e ve im od 10 km) mogu tako e na ve oj udaljenosti (nekoliko puta deset kilometara) od vodnog zajaženja. Pošto je vodi potrebno prili no vremena da prodre do dubina (odvisno od permeabilnosti stijena) taki potresi se prvi put pojavljuju tako e više godina po punjenju (do 20 godina). Potresi ve ih magnituda nastaju uz ve e rasjede i njihovo postajanje je preduslov tako e uz pojavu potresa. Vodna zajaženja su po pravilu u rije nim dolinama, koje su postale sa eroziom u podru jima sa aktivnom tektonikom, koja ve inom presjecaju brojni razsjedi.

Dinamika pojava trigerirane seizmi nosti je lako razli ita. Za sada su upoznali nekoliko »tipi nih« ponašanja:

- potresi odmah po punjenju (naj eš i) su povezani sa promjenom nivoa vode i ponekad se prestanu ponavljati poslije nekoliko godina,

- stalni potresi (re a pojava) govore na stalnu promjenu seizmi nosti,- seizmi nost u podru jima krasa, koji je ponekad tako e u vezi sa vrlo malim

zajaženjima, - aseizmi na zajaženja, kod kojih dolazi do smanjenja potresne aktivnosti i - mješana, u kojima sa vremenom dolazi do prelaza iz jednog tipa u drugoga.

Pojavljanje idducirane seizmi nosti predstavlja opasnost kako za sam nasip tako tako er za okolinu. U slu aju mo nijeg potresa može do i do ošte enja brane i oticanja vode, ili do klizišta u vodno zajažanje, koja brzo dignu vodni nivo i aktiviraju vodeni talas. Rizikovanju su posebno izpostavljena podru ja sa niskom prirodnom seizmi nostji jer brane nisu projektovane za mo nije potrese, kao što je to po pravilu u podru jima sa visokom seizmi nosti i jakim potresima u prošlosti. (slika 1a in 1b).

Poznavanje problema inducirane seizmi nosti u vodnim zajaženjima je nedovoljno i nije mogu e sa sigurnoš u tvrditi, da bi bilo kakvo vodno zajaženje bilo gdje na svijetu sigurno pred mogu nostima induciranih potresa.

Kontinuirano pra enje potresne aktivnosti u blizini vodnoga zajaživanja pomože boljemu razumjevanju pojava trigerirane seizmi nosti i mehanizama, koju prouzrokuju, kao tako e tektonskih uslova, u kojima je ta pojava više manje vjerovatnija. Za osmatranje slabijih potresa je potrebna upotreba osjetljivih instrumenata u neposrednoj blizini, da možemo utvrditi nastanak potresa i probamo ga usporediti sa drugim parametrima (visina vode, brzina promjene nivoa vode ...) Za sve do sada utvr ene mo ne trigerirane potrese je ustanovljeno, da su im predhodili brojni manji potresi. Osmatranje trigeriranih potresa je specifi no podru je seizmologije, u kojem se stru njaci slažu, da je prognoziranje potresa mogu e. Neki autori su ak mišljenja, da je trigeriranu seizmi nost mogu e kontrolirati sa režimom djelovanja vodne brane.

Osmatranje inducirane seizmi osti se mora izvesti na branama, ija visina je ve aod 40 metara i to:

- brana, ija visina je ve a od 100 m, mora imati na slobodnoj površini u udaljenosti do pet kilometara od obale sabiralnika vode (kolektora) lokalnu mrežu najmanje od tri seizmografa,

- brana, ija je visina izme u 40 i 100 metara mora imati na slobodnoj površini u udaljenosti do pet kilometara od obale sabiralnika najmanje jedan seizmograf. Za garantovanje optimalnog ozna avanja i zapisivanja te odre ivanja parametara

lokalnih potresa je potrebno izraditi posebni projekat, u kojem se odrede lokaliteti seizmografov, vrsta opreme i na in njenog postavljanja. Posebni projekat odobri

422

ministarstvo, pristojno za zaštitu okoline. Ta posebni projekat je sastavni dio projekta za pridobijanje dozvole za posezanje u prostor za gradnju brane.

Vlasnik brane mora osigurati po etak osmatranja inducirane seizmi nostinajmanje tri godine prije po etka punjenja sabiralnika, odnosno prostora za branom te obezbjediti to posmatranje deset godina poslije (prvog) zapunjavanja do kote punjenja, odre ene sa projektom za pridobijanje dozvole za posezanje u prostor za gradnju brane.

Ministarstvo može poslije isteka predvi ene dobe na prijedlog obveznika odlu iti da osmatranje inducirane seizmi nosti prestane ako iz analize s posmatranjem dobijenih podataka proizlazi da se seizmi nost podru ja, na kojem stoji brana, zbog brane nije promjenila.

U pravilniku je odre ena oprema seizmografa.

4 OSMATRANJE DINAMI NOGA PONAŠANJA BRANE Zahtjevi pravilnika:

- za brane s visinom 60 ili više metara mora biti postavljena mreža od najmanje etiri akcelerografa, od koji je jedan u temelju, dva u tjelu brane, jedan na

slobodnoj površini, - za brane visine od 30 do 60 metra mora biti postavljena mreža od najmanje tri

akcelerografa, od koji je jedan u temelju, jedan u tjelu brane i jedan na slobodnoj površini,

- za druge brane moraju biti postavljena najmanje dva akcelerografa, od koji je jedan u temelju brane, drugi na slobodnoj površini.

Svi akcelerografi, postavljeni na pojedinoj brani, moraju biti povezani u jedinstven sistem mjerenja i skupljanja podataka o osmatranju dinami noga ponašanja brane.

Za osiguranje optimalnoga ozna avanja i zapisivanja dinami noga osmatranja ponašanja brane je potrebno upravo tako izraditi posebni projekat osmatranja, u kojem se odredi broj i lokalitete akcelerografa, vrsta opreme i na in njenog postavljanja. Posebni projekat odobri Ministarstvo i sastavni je dio projekta pridobivanje dozvole za posezanje u prostor za gradnju brane.

Pravilnik propisuje opremu akcelerografa sastavne djelove odnosno osobine.

5 POSTAVLJANJE INSTRUMENATA

5.1 INSTRUMENTI NA SLOBODNOJ POVRŠINI Zapisuju potresno njihanje tla, na koje ne uti e postojanje brane i vodene

zajažitve. Pokazuju njihanje, koje bi bilo na podro ju brane, ako te nebi bilo. Instrumenti moraju biti postavljeni što bliže brani. Me utim dovoljno daleko, da

možemo zanemariti uticaj objekta na bilješke (zapise). Ta udaljenost bi morala biti jednaka duploj visini brane.

5.2 INSTRUMENTI U TEMELJU BRANE Te instrumente instaliramo na karakteristi na temeljna tla. Lokacijsko se

postavljaju u elije na betonskim temeljima, koje omogu uju povezanost sa osnovnom stijenom, na kojoj stoji brana.

423

5.3 INSTRUMENTI U TIJELU BRANE Sa tim instrumentima se mjeri odziv konstrukcije. Glavno mjesto je na najve oj

visini brane, gdje o ekujemo najve e pomjerenje objekta.

6 OBEZBJE IVANJE OSMATRANJA Osmatranje inducirane seizmi nosti i dinami noga ponašanja brane može za

obveznika iz pravilnika da izvodi pravna ili fizi ka osoba, koja ima punomo Ministarstva. Ovlaš enje Ministarstva može dobiti pravna ili fizi na osoba, koja ima registrovanu geofizikalnu ili drugu odgovaraju u djelatnost osmatranja, mjerenja i kartiranja (seizmološki osmatra ).

Ministrarstvo izdaje ovlaš enje uz ispunjavanje uslova iz pravilnika u obimku za kojega obveznik odnosno osoba zamoli, s obzirom na vrstu i obseg izvo enja osmatranja.

Seizmološki osmatra mora za pridobivanje ovlaš enja ispunjavati slijede euslove:

- da je privredna organizacija, zavod ili samostalni preduzima ,- da ima sjedište u Republici Sloveniji.

Seizmološki osmatra dobije ovlaš enje na osnovi molbe na Ministarstvu. Molba mora sadržavati podatke o moliocu te opis vrste i obsega izvo enja seizmološkog osmatranja, za kojega molioc želi punomo .

Seizmološkom osmatra u se može izdati punomo za najviše šest godina. Punomo se može obnoviti na osnovi nove molbe, ako ispunjava uslove odre ene u 26. lanu toga pravilnika.

O osmatranju inducirane seizmi nosti te osmatranju dinami noga ponašanja brane za vrijeme potresa, seizmološki osmatra je dužan svake godine izraditi godišnji izvještaj u propisanom obliku, najkasnije do 31. marta za prošlu godinu.

Seizmološki osmatra je dužan prilikom svakog potresa, pri kojem ubrzanje na vrhu na prostom površju prelazi vrednost 5 procenata zemljinog ubrzanja, pripraviti poseban izvještaj, koji mora sadržavati sve originalne registracije potresnoga njihanja i odgovarjaju u obradu. Taj izvještaj je potrebno dostaviti Ministarstvu u roku 30 dana po doga aju. Izvještaj mora obveznik uvati deset godina.

Nadzor nad izvo enjem toga pravilnika vrše inspektori, pristojni za o uvanje okoline.

Bez obzira na odredbe pravilnika mogao je seizmološki osmatra do 31.12.2004 dobiti punomo tako e, ako ispunjava slijede e uslove.

- da je privredna organizacija, zavod ili samostalni preduzima ,- da ima sjedište u Republiki Sloveniji, - da ima odgovaraju i broj tehni no i stru no osposobljenih radnika sa

odgovaraju om stru nom spremom i iskustvom za vršenje seizmološkoga osmatranja po tom pravilniku,

- da razpolaže sa odgovaraju im instrumentima te probno i merilno opremo za izvo enje seizmološkoga osmatranja te za održavanje opreme u skladu sa tim pravilnikom.

Ispunjavanje uslova utvr uje Ministarstvo u saradnji sa upravnim organom, pristojnim za standardizaciju i mjere.

424

Obveznici za izvo enje seizmološkoga osmatranja moraju za postoje e brane osigurati osmatranje u skladu sa tim pravilnikom najkasnije u jednoj godini po njegovoj važnosti. U istom roku moraju obveznici usaglasiti sa odredbama pravilnika tako eobstoje e sisteme seizmološkoga osmatranja, koji su bili postavljeni po Pravilniku o tehni ki normativi za seizmi no osmatranje visokih vodnih brana (Službeni list SFRJ broj 6/88).

Bez obzira na odredbe 8. lana toga pravilnika moraju obveznici najkasnije u jednoj godini po važnosti toga pravilnika obezbjediti tako e osmatranje inducirane seizmi nosti za brane iz 5. lana pravilnika. To osmatranje mora trajati najmanje tri godine, po isteku toga vremena pa može prestati uz propisane uslove i po propisanom postupku.

Pregledna tablica 1 - Spisak velikih brana u Sloveniji

BR

AN

A

ZAJA

ŽIV

AN

JE

GO

DIN

A

IZG

RA

DN

JE

KO

NST

RU

KTI

VN

A

VIS

INA

(m)

HID

RA

VLI

NA

V

ISIN

A (m

) D

UŽI

NA

K

RU

NE

(m)

ZAPR

EMIN

A

ZAJA

ŽITV

E (1

000m

3 )

DU

ŽIN

A

ZAJA

ŽAV

E (k

m)

BR

OJ

AK

CEL

ERO

GR

AFA

B

RO

JSE

IZM

OG

RA

FA

1 Dravograd 1942 23,0 8,9 180 7000 10,0 2 2 Vuzenica 1952 34,0 13,8 191 14200 12,0 3 3 Vuhred 1956 33,0 17,4 167 19300 13,0 3 4 Ožbalt 1960 33,0 17,4 167 12880 13,0 3 5 Fala 1928 34,0 14,6 248 4095 8,0 36 Mariborski 1943 33,0 14,2 184 18700 16,0 3 7 Melje 1977 17,0 8,2 160 4600 6,0 2 8 Zlatoli je 1968 54,0 24,8 50 17,0 3 1 9 Markovci Ptujsko jez. 1968 19,0 11,5 120 23000 6,0 2 10 Formin 1977 49,0 29,0 49 8,0 3 1 11 Moste 1952 59,6 48,0 52 6240 5,0 3 1 12 Završnica 1914 15,0 32 135 1,0 2 13 Mav i e 1986 38,0 17,5 118 10700 7,0 3 14 Medvode Zbiljsko jez. 1953 30,0 21,2 134 7000 6,0 3 15 Vrhovo 1993 24,0 8,1 140 8650 10,0 2 16 Boštanj 2006 27,0 8,0 170 8000 10,0 2 17 Podsela Doblarsko j. 1939 55,0 56 5800 8,0 3 1 18 Ajba 1940 39,0 72 1600 5,0 3 19 Solkan 1984 35,0 22,0 138 7600 10,0 3 20 Lo e Šmartinsko j. 1970 16,0 11,0 205 6500 2,0 2

21 Tratna Slivniško j. 1975 17,0 13,0 81 4000 2,5 2

425

BR

AN

A

ZAJA

ŽIV

AN

JE

GO

DIN

A

IZG

RA

DN

JE

KO

NST

RU

KTI

VN

A

VIS

INA

(m)

HID

RA

VLI

NA

V

ISIN

A (m

) D

UŽI

NA

K

RU

NE

(m)

ZAPR

EMIN

A

ZAJA

ŽITV

E (1

000m

3 )

DU

ŽIN

A

ZAJA

ŽAV

E (k

m)

BR

OJ

AK

CEL

ERO

GR

AFA

B

RO

JSE

IZM

OG

RA

FA

22 Ra igaj Braslovško j. cca 20 nepr

2

23 Trnava Žovneško j. 1978 13,5 7,5 333 1720 1,5 2

24 Vodnarje Sotelsko jez. 1980 19,0 13,3 120 12400 6,5 2 25 Prigorica 9,8 7,3 960 8800 2 26 Vogrš ek 1988 37,0 31,0 200 8500 2,7 3 27 Klivnik 1987 28,0 252 4300 3,0 2 28 Mola 1979 23,5 90 4300 3,7 2 29 Vanganel 1964 19,0 17,3 130 244 0,2 2 30 Bukovžlak 41,0 520 3 1

31 ZaTravnikom 49,0 630 3 1

32 Drtijš ica 2002 18,2 265 5900 0,5 2

UKUPNO 80 6

7 ZAKLJU AKPostavljanje instrumenata je jedno od najracionalniji oblika i metoda zaštite pred

potresom na branama. Podaci, dobijeni sa instrumentima za zapisivanje mo nih potresa, se mogu koristiti kao osnovni podaci za definisanje projektnih kriterija i parametara. Bez odgovarjaju ih bilježenja tako e nije mogu e upore ivati ponašanje brane u vrijeme djelovanja potresa sa projektnim seizmi nim parametrima. Bez tih podataka tako e nije mogu e odlu ivati da li je još brana sigurna ili je potrebno sanirati neposredno po potresu.

Iz navedenoga se preporu uje seizmi no osmatranje sa instalacijom instrumenata za zapisivanje potresa na svim hidrotehni kim objektima, a posebno na velikim branama, koje su izgra ene na potresno aktivnim podru jima.

Još u SFRJ je važio Pravilnik o tehni kim normama za posmatranje visokih brana (Sl.l. SFRJ 6/1988). Na njegovoj osnovi je bio pripremljen i u službenem listu RS godine 1999 objavljen Pravilnik o osmatranju seizmi nosti na podru ju velikih brana, koji propisuje na ine osmatranja seizmi nosti, tehni ke normative seizmoloških instrumenata te uslove koje mora ispunjevati izvo a osmatranja uticaja seizmi nosti na velike brane. Pravilnik izme u nabrojanih uslova, odlu uje tako e o pojmu velike brane, nadzor nad njegovim izvo enjem izvršavanju inspektori, pristojni za uvanje okoline.

426

Uvo enje svakog Pravilnika je povezano sa financijskim sredstvima, to su po obi aju tako e teško e. Zato se postavljanje instrumenata na branama ne odvija tako kao što je bilo predvi eno.

U Sloveniji je ukupno 32 brana, koje odgovaraju definiciji velike brane. Od toga je 10 brana na reki Dravi, 6 na reki Savi i 3 na reki So i te 11 vodoprivrednih objekta, te još 2 za spre avanje muljastih odpadaka. Na tim branama bi e potrebno postaviti ukupno 80 akcelerografa i 6 seizmometara (u Preglednoj tablici 1 je prikazan spisak brana u Sloveniji).

Za sada (2009) se osmatranje seizmi nosti na podru ju velikih brana u Sloveniji izvodi na 18 velikih brana. Problem prestavlja ukupno 14 velikih brana na kojima se još ne izvodi osmatranje seizmi nosti.

LITERATURA

1 Earthquake Induced Damage to Dams – Classification and Statistical Evaluation / B.Huber // University of Technology Vienna, 1995, Vienna.

2 Tehni ni slovar za pregrade / B.Zadnik // Slovenski nacionalni komite za velike pregrade, 1997, Ljubljana.

3 Strokovne podlage za pravilnik o seizmološkem monitoringu velikih pregrad / P.Fajfar, B.Zadnik //. Fakulteta za gradbeništvo in geodezijo, 1996, Ljubljana.

4 Pravilnik o opazovanju seizmi nosti na obmo ju velike pregrade / Uradni list Republike Slovenije št.92/99, 1999, Ljubljana.

427

Radenko Pejovi 119, Radivoje Mrdak220, Jelena Pejovi 321, Nina Serdar422,

SEIZMI KI ODGOVOR VISOKE LU NE BRANE “MRATINJE”

Rezime:U ovom radu je prikazan dio rezultata iz numeri ke analize stati ke i seizmi keotpornosti visoke lu ne brame “Mratinje”. Za potrebe analize konstruisan je prora unski model koji se temelji na unaprijed definisanim nelinearnim zonama i presecima. Te zone su modelirane elementima tipa nelinearnih opruga. U prora unski model za dinami ku analizu ove brane je uklju eno tijelo brane, stjenski masiv, voda u akumulaciji, stvarne mehani ke karakteristike materijala i realni seizmi ki parametari mikrolokacije. Analiza se sastoji od dva dijela. Prvi dio je linearna analiza sistema brana-fundament-rezervoar za dejstvo projektnog zemljotresa. Drugi dio je nelinearna analiza brane za dejstvo maksimalnog zemljotresa relevantnog za sigurnost brane. U radu su prezentirani karakteristi nirezultati analize. Key words: lu na brana, seizmi ka analiza,brana-fundament,rezervoarat the most

SEISMIC ANALYSIS OF HIGH ARC DAM “MRATINJE”

Summary:This paper shows part of results of static and seismic numeric analysis of high arc dam ‘’Mratinje’’. For purpose of analysis design model with ahead defined nonlinear zones and sections was created. Zones were modeled by nonlinear spring elements. The numerical model for the dynamic analysis of this dam included the dam body; rock mass, reservoir water, real material mechanical characteristics and real seismic parameters of micro location. Analysis was consisted of two parts. First part is linear analysis of the system dam-foundation-reservoir for DBE-Design Basic Earthquake. Second part is nonlinear analysis of dam for occurrence of Maximal Credible Earthquake (MCE), relevant for the dam safety.Key words: arch dam, seismic analysis, system dam-foundation-reservoir

1 Prof.dr,Univerzitet Crne Gore,Gra evinski fakultet, Podgorica 2 Mr,Univerzitet Crne Gore,Gra evinski fakultet, Podgorica 3 Mr,Univerzitet Crne Gore,Gra evinski fakultet, Podgorica 4 Mr,Univerzitet Crne Gore,Gra evinski fakultet, Podgorica

428

1. UVOD

U ovom radu prikazan je dio rezultata numeri ke analize stabilnosti betonske lu ne brane “Mratinje”. Analiza je ra ena za potrebe Elektroprivrede Crne Gore A.D. Nikši iz Nikši a, kao naru ioca. Ura ena je od strane Gra evinskog fakulteta iz Podgorice i Energoprojekta-hidroinženjeringa A.D. iz Beograda u skladu sa projektnim zadatkom naru ioca. Konsultanti pri izradi analize bili su Prof.dr Miodrag Sekulovi i Aleksandar Saša Božovi .

Betonska lu na brana ”Mratinje” (sada nosi ime “Piva”) spada u red izuzetno visokih brana. Gra evinske visine je 220 m što je svrstava me u 25 najviših brana na svijetu. Brana je završena 1976.godine, kada se po elo sa prvim probnim punjenima i pražnjenjima akumualcije, tj. nalazi se u eksploataciji preko 30 godina.

Provjere konstrukcije brane i njenih oslonaca vršene su koriš enjem tada raspoloživih ra unskih metoda. Uporedo sa ra unskim analizama vršena su obimna modelska istraživanja.

Provjere pogonskog stanja brane tokom njene eksploatacije vrši se od strane posebno formirane službe koja radi kontinuirano. Me utim, mjerni instrumenti, na in mjerenja i prikupljanje podataka, na in obrade i interpretacija izmjerenih veli ina prakti no se nijesu mijenjali od prvog probnog punjenja akumualcije 1976.godine do danas, iako je ve realizovano preko 30 jednogodišnjih ciklusa mjerenja. Pokazalo se da do sada koriš ene metode obrade i interpretacije podataka ne mogu dati adekvatan odgovor i objašnjenja za sve pogonske doga aje koje je brana pretrpjela u dosadašnjem periodu eksploatacije. S druge strane prikupljeni fond podataka samo je djelimi no interpretiran, jer nijesu napravljeni ra unski modeli koji obuhvataju interakciju tijela brane sa okolnom stijenom i vodom u akumulaciji (interakcija: brana – fundament – rezervoar). Brana “Piva” analizirana je na numeri kim modelima koji su bili na raspolaganju projektantu u momentu projektovanja po etkom 70-tih godina.

Shodno navedenom, pristupilo se izradi novog, savremenijeg, numeri kog modela na osnovu koga bi se, uz koriš enje raspoloživih podataka, mogla uraditi provjera projektnih veli ina brane (dimenzije, naponi, pomjeranja i dr.).

Koriš enjem osavremenjenog prora unskog modela ura ena je analiza stanja napona i deformacija i na osnovu nje date ocijene:

-Postoje eg stanje brane u odnosu na projektovano i izvedeno stanje, kao i ocijena njene sigurnost za stati ko i dinami ko optere enje;

-Koje veli ine izmjerenih deformacija brane, pri stati kom optere enju mogu ugroziti stabilnost objekta;

-Koje veli ine maksimalnih urbranja i pomjeranja tla ili brane dovode u pitanje seizmi ku stabilnost brane.

U radu su prezentirani karakteristi ni rezultati linearne i nelinearne analize brane “Mratinje”.

2. OSNOVNI PODACI O BRANI „MRATINJE” Brana HE “Piva”(“Mratinje”) je betonska, nesimetri na, lu na sa dvojnom

krivinom. Izgradnjom brane stvoreno je vješta ko akumulaciono jezero ije se vode koriste za proizvodnju vršne elektri ne energije u pribranskom postrojenju HE “Piva”. Profil pregradnog mjesta je geometrijski i geotehni ki nesimetri an. Lijevi bok kanjona je strmiji

429

od desnog. U vrhu desnog boka moduli deformacije stijenske mase imaju niske vrijednosti u odnosu na ostalu konturu. Zbog ovoga je brana konstruisana kao nesimetri na. U gornjoj etvrtini na desnom boku se oslanja preko sistema horizontalnih šipova na bolje partije

stijenske mase koje imaju približno iste geotehni ke karakteristike kao i na odgovaraju im nivoima u lijevom boku. U centralnom dijelu na kruni su tri prelivna polja širine po 13 m i visine 5 m sa kotom praga 670,00 mnm.

Brana se sastoji do 18 konzola i ima pet revizionih galerija na kotama: 642, 602, 562, 522 i 482 mnm.

Osnovni tehni ki podaci za branu su: - Konstruktivna visina 220 m - Hidrauli na visina 190 m - Dužina luka u kruni 268,56 m - Dužina luka u nivou korita 40,00 m - Debljine u tjemenim presjecima od 4,514 m u kruni do 29,917 m u dnu - Debljine u temeljima od 6,46 m u kruni do 45,00 m u dnu - Kota krune brane 678,00 mnm. - Kota maksimalnog uspora 677,50 mnm - Kota normalnog uspora 675,22 mnm - Kota najnižeg radnog nivoa 595,00 mnm - Projektovan amarka betona MB 30

Osnovne karakteristike betona i stijene koje su usvojene u prora unskom modelu utvr ene su na osnovu rezultata ispitivanja i iznose:

Beton 41000 MPa

Stijena : Lijevi Bok 10000 – 15000 MPa Desni bok 5000 (7000) MPa until 15000 (16000) MPa

Za vodu u akumulaciji usvojen je modul kompresije 2.07 *106 MPa i = 0.00.

Eksperimentalnim ispitivanjem brana “Mratinje“ prinudnim vibracijama koje je izvršio IZIIS Skoplje, 1980.godine izmjerene su dinami ke karakteristike. Mjerene su rezonantne frekvencije za vibracije u pravcu kanjona (radijalna komponenta) i popre no na kanjon (tangencijalna komponenta), sopstveni oblici i kapacitet prigušenja. Ispitivanja su vršena za nivo vode u akumulaciji od 11.00 m ispod kote krune brane koja je zate ena u vrijeme izvodjenja eksperimenta.

Table 1 – Izmjerene vrijednosti sopstvenih frekvencija

mode 1 mode 2 mode 3

S. frekfencija (Hz) (radijalna komponenta) 2.80 4.56 5.34 S. frekfencija (Hz)(tangencionalna komponenta) 2.18 4.07 4.46

430

U Tabeli 1 dat je prikaz izmjerenih vrijednosti sopstvenih frekvencija pri eksperimentalnom ispitivanju brane. Izmjerene vrijednosti koeficijenta prigušenja su u granicama od 1.030 do 2.430 %. Malo prigušenje je rezultat malog nivoa dinami ke pobude pri utvr ivanju dinami kih karakteristika brane.

3. OSMATRANJE BRANE

Osmatranje visokih brana definisano je kao obavezuju e za korisnika brane doma im i stranim propisima. Me unarodni komitet za visoke brane ICOLD (International Commission of Large Dams) definiše kriterijume i na in osmatranja.

Tehni ko osmatranje brane HE “Piva” sprovodi se u skladu sa Pravilnikom o tehni kom osmatranju visokih brana, Sl. list SFRJ br. 7/66. Na HE “Piva” postoji posebna služba koja je vršila osmatranje u fazi izgradnje i nastavila u fazi eksploatacije i ona i danas postoji. Osmatranje se vrši na oko 1200 mjernih mjesta.

Seizmi ko osmatranje brane vrši se u skladu sa Pravilnikom o tehni kim normativima za seizmi ko osmatranje visokih brana, Sl. list SFRJ br. 6/88.

Postoji i poseban Pravilnik o tehni kom osmatranju brane HE “Piva” koji je donešen 22.03.2988.godine. Ovim internim dokumentom izvršena je razrada i prakti na primjena važe ih zakonskih dokumenata i propisa u oblasti tehni kog osmatranja. Obra ene su vrste, na in i metodologija sprovo enja tehni kog osmatranja stru nog tima službe za osmatranje. Posebno poglavlje ovog Pravilnika odnosi se na postupanje u slu ajuopasnosti pri pojavi odre enih ošte enja i defekata same brane ili pribranskog tla.

4. PRINCIPI NA KOJIMA JE ZASNOVANA URA ENA ANALIZA BRANE

Analiza ponašanja i prora un seizmi kog odgovora visokih lu nih brana su veoma kompleksni problemi. Tome doprinose težina posljedica koje bi nastale njihovim rušenjem i složene geometrijske i geomehani ke karakteristike brane i okolnog terena kao i stohasti kapriroda zemljotresa. Savremeni zahtjevi u pogledu optimalnog projektovanja seizmi kiotpornih brana zasnivaju se na novim saznanjima o prirodi ponašanja materijala i konstrukcija pri dejstvu zemljotresa kao i na novim mogu nostima formulisanja i rješavanja složenih matemati kih modela. Klasi ni modeli koji se zasnivaju na predpostavkama o krutoj vezi objekta i stijene ne obezbje uju utvr ivanje realnih dinami kih karakteristika konstruktivnog sklopa brane i fundamenta u uslovima kada se voda nalazi u akumulaciji i djeluje na branu i tlo. Usled toga ne postoji mogu nost ni za kvalitativnu predstavu o ponašanju objekta tokom seizmi kog dejstva. Savremeni prora unski modeli za analizu ponašanja lu nih brana zasnivaju se na predpostavkama o interakciji tijela brane sa okolnom stijenom i vodom u akumulaciji (interakcija: brana-fundament-rezervoar). Oni omogu avaju obuhvatanje stvarne geometrije i karakteristika materijala kao i simulaciju neograni enih medija stijene i vode i njihove interakcije pri dejstvu jakih zemljotresa. Pri tom, modeli zasnovani na interakcijama dovode do redukcije uticaja u odnosu na klasi nemodele.

Pri dejstvu zemljotresa umjerene ja ine struktura brane ostaje u domenu linearne teorije elasti nosti. U drugoj fazi, pri dejstvu jakih zemljotresa brana postaje sistem deformabilnih blokova sa odre enim unutrašnjim stepenima slobode, sa izrazitim nelinearnim ponašanjem.

431

Analiza sigurnosti visoke lu ne brane podrazumijeva dvije vrste analize koje su posledica razli itih kriterijuma za projektovanje i sigurnost objekta. Prvi nivo se odnosi na optimalmo projektovanje (optimum earthquake-resistant design) za koji, kao kriterijum stabilnosti, nijesu dopuštene pojave prslina. Radi se za slu aj spoljašnjeg uticaja tipa zemljotresa mjerodavnog za projektovanje (DBE-Design Basic Earthquake) u kombinaciji sa drugim uticajima. Drugi nivo je procjena stabilnosti objekta (earthquake-resistant design) za koji su dozvoljena ošte enja izuzev onih koji ugrožavaju globalnu stabilnost objekta. Mjerodavni zemljotres za ovu analizu je maksimalno mogu i zemljotres (MCE-Maximal Credible Earthquake).

Za optimalno projektovanje i provjeru sigurnosti brana od presudne važnosti je iznalaženje ta nog seizmi kog odgovora prora unskog modela. Model treba da što vjernije aproksimira me usobnu interakciju složenog sistema brana-fundament-rezervoar. Na predpostavkama o simultanom radu objekta sa tlom i fluidom i njihovoj dinami kojinterakciji, razvijen je veliki broj razli itih prora unskih modela na bazi njihove materijalne diskretizacije. Mnogi problemi, a naro ito oni iz domena nelinearne analize, istraživa ki su aktuelni. Zbog veli ine sistema, raznorodnosti mehani kih karakteristika materijala, velikih i nepravilnih kontaktnih površina izme u dijelova sistema, naro ito su aktuelni problemi u pogledu traženja modela racionalnih za primjenu.

Uovom radu je izložen originalan pristup konstrukciji racionalnog prora unskog modela za nelinearnu analizu. Umjesto uobi ajene racionalizacije modela za nelinearnu seizmi ku analizu sa usvajanjem grublje podjele kontinuma na manji broj elemenata pošlo se od slede ih predpostavki koje dovode do uproš enja modela:

- Linearnom analizom brane može se odrediti zona u kojoj e se desiti nelinearno deformisanje pri dejstvu zemlotresa sa nivoom mjerodavnim za ocjenu stabilnosti objekta. Elementi za simulaciju nelinearnog ponašanja brane ugradili bi se samo u toj zoni. Na osnovu dobijenih rezultata nelinearne analize model bi se, po potrebi, korigovao i analiza ponovila na korigovanom modelu.

- Lu ne brane se grade od nearmiranog betona kao niz konzolnih blokova. Blokovi su me usobno vezani nearmiranim kontaktnim spojnicama koje ne primaju napone zatezanja pa predstavljaju potencijalno mjesto za pojavu i propagaciju prslina i pojavu plasti nih deformacija. Posle prestanka dejstva zemlotresa spojnice se , usled stati kihoptere enja, zatvaraju. Blokovi i spojevi brane se razli ito ponašaju. Pretežan dio deformacija brane ostvari se u spojevima pa se pri konstruisanju prora unskog modela mogu uvesti predpostavke da su dijelovi brane izme u spojeva linearno elasti ni i izotro-pni dok nelinearno ponašanje nastaje samo u spojevima. Iz ovoga proizilazi da je mogu esamo spojnice blokova modelirati elestoplasti nim vezama.

- Imaju i u vidu tip nelinearnog ponašanja mogu se primijeniti jednostavni kontaktni elementi sastavljen od nelinearnih opruga koji su ugra eni u vorovima mreže kona nih elemenata duž spojnica u zoni i u okolini zone u kojoj su se pojavila zatezanja. Za simulaciju nelinearnog ponašanja mogu se primijeniti kontaktni elementi tipa nelinearnih opruga kod kojih je odnos napon-deformacija nezavisan jedan od drugog.

Na navedenenim predpostavkama je konstruisan prora unski model i ura ena seizmi ka analiza visoke betonske lu ne brane “Mratinje” za koju su, eksperimentalnim istraživanjima, utvr ene dinami ke karakteristike, ime je omogu ena kontrola prora-unskog modela i njegova kalibracija. U prora unski model za dinami ku analizu ove brane

je uklju eno tijelo brane, stjenski masiv, voda u akumulaciji, stvarne mehani ke karak-teristike materijala i realni seizmi ki parametari mikrolokacije.

432

5. SEIZMI KI PARAMETRI Za seizmi ku analizu brane koriš eni su akcelerogrami zemljotresa koji se

dogodio u aprilu 1979.g na Crnogorskom primorju i to:- Petrovac, hotel "Oliva", componenta N-S,15.04.1979. - Petrovac, hotel "Oliva", componenta t N-S, 9 15.04.1979. - Kotor, 24.05..1979. - Budva, 24, 12.05.1979. - Titograd-2, Seizmološka stanica, komponenta N-S, 15.04.1979. Svi zemlotresi su skalirani na maksimalno ubrzanje tla za nivo od 0.25 g za

projektni zemlotres (za linearnu analizu) i 0.40 g za maksimalno mogu i zemljotres (za nelinearnu analizu).

6. NUMERI KI MODELI Model brana-fundament-rezervoar je složen. dinami ki sistem po geometriji,

granicama domena, promjeni fizi kih osobina i stohasti koj prirodu dinami ke pobude. Sastoji se od tijela brane, fundamenta kao beskona nog poluprostora i vode u akumulaciji. U uslovima dejstva zemljotresa sastavni elementi modela se nalaze u simultanoj dinami koj interakciji brane-stijene i vode u akumulaciji. Za linearnu seizmi ku analizu brane „Mratinje” konstruisan je model baziran na numeri koj analizi simultanog interaktivnog sistema brana-fundament-rezervoar koriš enjem 3-D kona nih elemenata. Dinami ka analiza je sprovedena uz predpostavku o fundamentu bez mase i kompresibilnom ponašanju vode u rezervoaru.

Slika 1- Prora unski model sistema brana-fundament-rezervoar brane “Mratinje”

Za diskretizaciju tijela brane i fundamenta upotrebljeni su 3-D elementi elasti nogkontinuma sa 24 stepena slobode. Za tijelo brane usvojena je debljina elemenata jednaka etvrtini debljini brane. U prora unskom modelu za fundament, angažovana je debljinja

stjenske mase koja je približno jednaka 1.5 H , gdje je H visina brane na mjestu centralne

433

konzole. Upotrebljeni su elementi prizmati nog oblika sa postepenim pove anjem dimenzija elementa sa zalaženjem u stijensku masu.

Debljina stjenske mase od 1.5 H je usvojena na osnovu rezultata istrazivanja koja su obavljena na brani "Andrijevo" na rijeci Mora i gdje je pokazano da se ne mijenjaju rezultati prora una za usvojene ve e debljine stjenske mase.Angažovana dužina fundamenta ispod rezervoara je: sa uzvodne strane brane 2H a sa nizvodne strane 1H.

Prigušenje u prora unskom modelu je usvojeno kao zamjenjuju e viskozno prigušenje sa koeficijentom prigušenja od 0.05 i kre e se u granicama koje su utvr ene za ovaj tip konstrukcije.

Model za nelinearnu analizu formiran je pod predpostavkom da e se, usled prekora enja vrsto e betona na zatezanje u spojnicama, od po etnog monolitnog stanja tijelo brane transformisati u niz blokova sa fleksibilnim vezama duž spojnica. Osnovni vid nelinearnog ponašanja je pojava i propagacija prslina i pojava relativnih pomjeranja duž spojnica (otvaranje, zatvaranje i smicanje).

Usled prekora enja vrsto e betona u spojnicama na zatezanje od po etnog monolitnog stanja tijelo brane e se transformisati u niz blokova sa fleksibilnim vezama duž spojnica.

Za nelinearnu analizu sistema upotrebljen je modifikovani model upotrebljen za linearnu analizu unošenju u model nelinearne elemenata. Oni su postavljeni u vertikalnim spojnicama izme u solid elemenata kojima je diskretizovano tijelo brane gdje su linearnom analizom ustanovljena najve a zatezanja. Polaze i od predpostavke da je osnovni vid nelinearnog ponašanja je pojava i propagacija prslina i pojava relativnih pomjeranja duž spojnica (otvaranje, zatvaranje i smicanje za simulaciju relativnih pomjeranja primijenjeni su kontaktni elementi tipa nelinearnih opruga koji su ugra eni u vorovima mreže kona nihelemenata duz spojnica u zoni u kojoj su se pojavila zatezanja . Svaki nelinearni elemenat sastavljen je od šest nelinearnih opruga, za svaku od šest unutrasnjih deformacija. Usvojeno je da je odnos napon-deformacija ovih opruga nezavisan jedan od drugih.

7. REZULTATI I DISKUSIJA U prikazu rezultata ura ene linearne analize isti e se veoma veliki uticaj

frekventnog sastava zemljotresa. Analiziraju i uticaj zemljotresa razli itog frekventnog sastava, a u cilju izbora

merodavnog akcelerograma, uo ena je bitna razlika u veli ini za pojedine vrste zemljotresa. Za ilustraciju, na slici 2a, prikazani su dijagrami maksimalnih pomjeranja vorova na centralnoj konzoli nastalih usled dejstva zemljotresa sa razli itim frekventnim

karakteristikama koji su skalirani za isti nivo maksimalnog ubrzanja. Na dijagramu se jasno uo ava da se uticaji, usled razli itih akcelerograma, kre u u veoma širokim granicama. Maksimalno pomjeranje vrha centralne konzole za akcelerogram „Petrovac” je za preko deset puta ve i od odgovaraju eg pomjeranja za akcelerogram „Budva”

Na slici 2b dat je prikaz maksimalnog pomjeranja vorova na centralnoj konzoli nastalih usled dejstva zemljotresa sa istim karakteristikama na modelima za linearnu i nelinearnu analizu.Isti e se da je sra unato maksimalno pomjeranje na modelu za nelinearnu analizu ve e za 2,38 puta.

434

Maximum displacements of joints during five characteristic earthquakes

8.3

7.7

6.5

5.9

4.9

3.9

3.2

2.3

2.0

1.8

1.6

1.5

1.4

1.3

1.2

1.1

1.1

1.0

1.0

0.9

0.9

8.5

7.9

7.3

6.7

6.1

5.1

3.0

2.6

2.4

2.1

1.9

1.8

1.7

1.6

1.5

1.4

1.3

1.2

1.2

1.1

1.1

1.0

3.2

3.0

2.7

2.4

2.1

1.6

1.2

1.0

0.8

0.8

0.7

0.6

0.6

0.5

0.5

0.5

0.4

0.4

0.4

0.4

0.3

0.3

0.3

0.3

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.4

0.3

0.3

0.3

0.0

13.3

12.2

11.1

10.0

9.0

7.0

5.3

4.0

1.5

0.5

450

460

470

480

490

500

510

520

530

540

550

560

570

580

590

600

610

620

630

640

650

660

670

680

0 5 10 15

Displacements U1 (cm)

Z co

ordi

nate

(m)

max displacements KO

max displacements PET 15.04

max displacements PG

maxdisplacements BD

max displacements PET 9.04

Slika 2- Max. pomjeranje vorova centralne konzle a/ usled zemljotresa sa razli itimkarakteristikamab/ za modele za linearnu i nelinearnu analizu

Slika 3- Slika napona u tangencijalnom pravcu Slika 4- Slika napona u tangencijalnom s22max(uzvodna strana),linearna analiza, pravcu s22max(uzvod. strana),ne linearna za slu aj kombinacija optere enja: analiza, za slu aj kombinacija optere enja: ST+HS1+T1+HD+Z2 ST+HS1+T1+HD+Z2

Maximum displacement for earthquake in Montenegro derived from linear and non linear analysis

20.3

18.0

14.3

13.3

11.5

9.9

8.4

13.3

14.4

14.7

14.1

13.3

11.9

10.2

8.4

6.5

4.5

1.7

1.3

1.5

1.5

8.5

7.9

7.3

6.7

6.1

5.1

3.0

2.6

2.4

2.1

1.9

1.8

1.7

1.6

1.5

1.4

1.3

1.2

1.2

1.1

1.1

1.0

450

460

470

480

490

500

510

520

530

540

550

560

570

580

590

600

610

620

630

640

650

660

670

680

0 5 10 15 20 25

Pomjeranja U1 (cm)

Kot

a (m

)

displacements from non-linearanalysis

displacements from linear analysis

b/a/

435

Na slikama 3 i 4 dat je prikaz maksimalnih tangencionalnih napona sa uzvodne strane brane za linearnu i nelinearnu analizu.

Rezultati pokazuje veliku redukciju uticaja nakon ulaska brane u nelinearno podru je deformisanja U odnosu na linearnu analizu naponi su i za 20% manji.

Na slici 5 dat je prikaz ubrzanja vrha centralne konzole usled pobude akce-lerogramom „Petrovac”. Faktor amplifikacije iznosi 15,6 i pokazuje veliku osjetljivost bra-ne na dejstvo zemljotresa sa ovakvim frekventnim karakteristikama,

-50

-40

-30

-20

-10

0

10

20

30

40

0 10 20 30 40 50 60 70

Time (s)

acce

lera

tion

(m/s

2)

Slika 5- Akceleracija vrha centralne konzle

Na osnovu upore enja rezultata izvršene seizmi ke analize brane i ekspe-rimentalnih rezultata, pri istom nivou vode u akumulaciji, može se zaklju iti da je odabrani prora unski model brane u koga je uklju ena interakcija brana-stijena-voda i stvarne mehani ke karakteristike materijala pogodan za iznalaženje dinami kih karakteristika i dinami kog odgovora sklopa. Dinami ki odgovor prora unskog modela, dobijen usled dejstva zemljotresa razli itog frekventnog sastava, se kre e u veoma širokim granicama i ukazuje na veliku važnost istraživanja seizmi kih parametara mikrolokacije.

Maksimalni glavni zatežu i naponi u brani usled sopsvene težine, hidrostati kog pritiska i dejstva zemljotresa (projektni zemljotres sa maksimalnom akceleracijom od 0.25 g) je manji od dozvolenih za beton koji je ugra en u tijelo brane. Usled dejstva zemljotresa sa maksimalnim ubrzanjem od 0.40 g (MKE) u kombinaciji sa sopsvenom težinom i hidrostati kim pritiskom, glavni naponi zatezanja prora;unati na nelinearnom modelu su u granicama dozvoljenih. Zona najve ih glavnih napona zatezanja je u gornjoj tre ini brane u zoni oko centralne konzole.

Upore enje rezultata nelinearne analize usled maksimalno mogu eg zemljotresa mjerodavnog za sigurnost brane (maksimalnim ubrzanjem od 0.40 g) sa rezultatima linearne analize, za zemljotres istog inteziteta, pokazuje velike razlike u pogledu pomjeranja, rasporeda i nivoa napona. Pokazano je da brana pri dejstvu zemjotresa MKE ulazi u nelinearno podru je deformisanja, trpi ošte enja, ali da pri tom nije ugrožena stabilnost brane.

REFERENCES

1 M.Sekulovic R. Pejovic, R..Mrdak , Earthquake response of high arch dams” 11 ECEE, Paris, 1998.

1 M.Sekulovic R. Pejovic, R..Mrdak , Analysis of seismic safety of high arch dam, 12 WCEE, Auckland, New Zeland,2000.

CIP - Каталогизација у публикацијиНародна и универзитетска библиотекаРепублике Српске, Бања Лука

711.4 : 624.131.55(082)624.042.7(082)550.34(082)

INTERNATIONAL conference on earthquakeengineering (2009 ; Banja Luka)

Планирање, пројектовање, изградња ирехабилитација зграда и других инжењерскихобјеката у сеизмички активним подручјима = Planning, design, construction and rehabilitationseismically active areas / [editors Mirko Aćić, Drago Trkulja]. - Banja Luka : Zavod za izgradnju = Institute for Construction, 2009 (Banja Luka : Nezavisne novine). - 423 str. : ilustr. ; 24 cm

Radovi na srp. i engl. jeziku. - Tiraž 500. -Bibliografija uz sve radove.

ISBN 978-99955-630-4-2

COBBISS.BH-ID 1255960

Documents

Planning, design, construction and rehabilitation