Energy efficiency and saving on lighting systems in existing buildings

Monographs of the Doctoral School in Environmental Engineering

13

Energy efficiency and saving

on lighting systems in existing buildings:

intervention strategies

Michela Chiogna

2008

Based on the Doctoral Thesis in Environmental Engineering (XX cycle) defended in

February 2008 at the Faculty of Engineering of the University of Trento

Supervisor: prof. Antonio Frattari

Cover photograph: Paolo Bottura Copyright: Michela Chiogna (text and images when not differently specified) Direttore della collana: Alberto Bellin Segreteria di redazione: Laura Martuscelli Università degli Studi di Trento, Italia December 2008 ISBN: 978-88-8443-248-3

Contents

Contents I List of Figures III Summary XI

Chapter 1 – Literature overview 1

1.1 Sustainability and voluntary protocols 1 1.2 Innovation in field of energy saving 5 1.3 Home and building automation 7 1.4 Daylighting simulation tools 10 1.5 Comfort, illumination and vision 16

Chapter 2 – Innovative design tools for lighting systems 35

2.1 Methodology and evaluation tools to design an energy efficient lighting system 35

2.1.1 Design methodology 36 2.1.2 Software tools output: evaluation methodology 38 2.1.3 Data analysis method 45

2.2 Multiple criteria decision analysis: application for technological system choice 52

2.2.1. Problem definition and classification of the referred MCDM approach 52 2.2.2. Classification and definition of the MADM elements 55 2.2.3 Standard analysis sheets 63

2.3 A model sheet to monitor and analyze visual comfort 65

2.3.1 Objects and effects 65 2.3.2 Definition of fields sheet 65 2.3.3 Comparison sheets between existing analysis 76

II

Chapter 3 – Case study 81

3.1 Analytical and programmatic phase 81 3.1.1 Existing situation 81 3.1.2 Photometric parameters survey and software modelling of the lecture rooms 82 3.1.3 Design goals definition 84 3.1.4 Lighting energy use estimation in compliance with the prEN 15193, 2006 86 3.1.5 Lighting energy use estimation in compliance with DIN V 18599-4, 2006 94

3.2 Design synthesis phase 102

3.2.1 Light system design 102 3.2.2 Bus devices installed in the case study system: technical detail 107 3.2.3 System installation and configuration 120 3.2.4 Supervision system configuration 124

3.3 Evaluation phase 130

3.3.1 Comparison of the software tools output 130 3.3.2 Data recording evaluation 157 3.3.3 Visual comfort test evaluation 183

Chapter 4 – Conclusions 201 Reference 207 Acknowlegements 215

ANNEXES (in a separate volume)

Annex 1- Energy saving and Sustainability evaluation: monthly evaluation 1 Annex 2 - Daily data analysis 11 Annex 3 - Typical day data analysis 29

III

List of Figures

Fig.1.1_ 1: The proportion of the energy consumption in EU..............................................1

Fig.1.1_ 2: Structure of energy consumption in residential sector (left)...............................2

Fig.1.1_ 3: Expected energy savings technical potentials in building stock.........................4

Fig.1.2_ 1: Lighting energy demand in Italy .......................................................................5

Fig.1.4_ 1: Radiosity Method (Daylight in Building – IEA Solar Heating and Cooling Programme Task 21, section 6 Design Tolls) .............................................................12

Fig.1.4_ 2: Ray-Tracing Method (Daylight in Building – IEA Solar Heating and Cooling Programme Task 21, section 6 Design Tolls)................................................13

Fig.1.5_ 1: Diagram of the Eye (National Eye Institute – U.S. National Institutes of Health, available in http://www.nei.nih.gov/health/eyediagram/eyeimages3.asp).......16

Fig.1.5_ 2 : Scotpic, Mesopic and photopic range (IESNA, 2003, Advance Lighting Guideline, 2-Lighting and Human Performance, available in http://www.newbuildings.org/ALG.htm)....................................................................18

Fig.1.5_ 3: Spectral Louminous Efficiency Values, V’(l) – Unity at Wavelength of Maximum Luminous Efficacy...................................................................................18

Fig.1.5_ 4: Eye-to-screen distance and vertical location ...................................................28

Fig.1.5_ 5: Overview of ADELINE 3.0 Program System .................................................30

Fig. 2.1.2_ 1: Weather data available for Italian cities, in EnergyPlus weather format.......40

Fig. 2.1.2_ 2: Statistics for ITA_Bolzano_IGDG, Location: Bolzano - ITA .......................41

Fig. 2.1.2_ 3: Insulation hourly data [s] for the weather station in S. Michele all (TN)-Italy .................................................................................................................42

Fig. 2.1.2_ 4: SSP calculation sheet ..................................................................................42

Fig. 2.1.2_ 5: Global radiation calculation sheet ...............................................................43

Fig. 2.1.2_ 6: Calculation sheet for solar position and solar radiation: ..............................44

Fig. 2.2.2_ 1: Results of the expected value method calculation........................................55

Fig. 2.2.2_ 2: Value tree for the specific decision problem................................................57

Fig.2.2.3_ 1: Weights calculation of the decision problem ................................................64 Fig.2.3.2_ 1: Test section: boundary conditions definitions 66

Fig.2.3.2_ 2: Critical detail definition...............................................................................68

Fig.2.3.2_ 3: Optometric board definition.........................................................................69

Fig.2.3.2_ 4: Luxometer for the local parameter measurement:.........................................69

Fig.2.3.2_ 5: Test section: Visual acuity determination .....................................................70

IV

Fig.2.3.2_ 6: Disability glare evaluation........................................................................... 72

Fig.2.3.2_ 7: Polarity factor versus the target size for different luminance level ............... 74

Fig.2.3.2_ 8: Left: Letter identification accuracy vs. contrast. Right: Letter identification latency vs. contrast [Ahumada A., Scharff L., 2003]............................ 75

Fig.2.3.2_ 9: Test of different contrast condition by the Ophtalmology ward of S.Chiara.................................................................................................................... 75

Fig.2.3.2_ 10: Visibility calculation.................................................................................. 78

Fig.2.3.2_ 11: Visibility evaluation................................................................................... 79 Fig.3.1.1_ 1: Faculty plan, second floor 81 Fig.3.1.1_ 2: Light distribution and switching operation system 82

Fig3.1.2_ 1: Luminance Isolines for the natural light. Simulation date: 21/06/2006 10.30 o’clock............................................................................................................ 83

Fig3.1.2_ 2: Logitudinal and cross section of the luminance distribution. Simulation date: 21/06/2006 10.30 o’clock................................................................................. 83

Fig3.1.2_ 3: Measured illuminance level on 16/02/2006................................................... 84

Fig.3.1.3_ 1: Lighting automation system definition for each classroom .......................... 86

Fig.3.1.4_ 1: Flow chart illustrating alternative routes to determine energy use ................ 88

Fig.3.1.4_ 2: Large façade opening with moderate room depth (from EN15193-2006) ..... 89

Fig.3.1.4_ 3: Referred classroom geometry and window location..................................... 90

Fig.3.1.4_ 4: Window geometry and material definition................................................... 90

Fig.3.1.4_ 5: Classrooms location on the main façade ...................................................... 91

Fig.3.1.4_ 6: FD,C as a function of daylight penetration (EN15193-2006) ......................... 92

Fig.3.1.4_ 7: FD,N monthly calculation (EN15193-2006).................................................. 92

Fig.3.1.4_ 8: FOC values for different automation control systems (EN15193-2006) ......... 92

Fig.3.1.4_ 9: Calculation results of FO and FC for the case studied analyzed .................... 93

Fig.3.1.4_ 10: Energy use estimation using the quick method........................................... 93

Fig.3.1.4_ 11: Energy use estimation using the comprehensive method ............................ 94

Fig.3.1.5_ 1: Operating time estimation comparison......................................................... 95

Fig.3.1.5_ 2: Kunstlicht parameter calculation for the referred case study ........................ 95

Fig.3.1.5_ 3: Daylight supply factor comparison .............................................................. 96

Fig.3.1.5_ 4: Artificial light control factor comparison..................................................... 96

Fig.3.1.5_ 5: FTL monthly calculation............................................................................... 96

Fig.3.1.5_ 6: Comparison between FTL and FD calculation ................................................ 97

Fig.3.1.5_ 7: Occupancy dependency factor comparison .................................................. 97

V

Fig.3.1.5_ 8: Yearly energy consumption estimation in compliance with the DIN standard ....................................................................................................................97

Fig.3.1.5_ 9: Monthly energy consumption estimation in compliance with the DIN standard ....................................................................................................................98

Fig.3.1.5_ 10: Energy consumption estimation comparison ..............................................98

Fig.3.1.5_ 11: Energy consumption estimation comparison for the same operating time.................................................................................................................................99

Fig.3.1.5_ 12: Difference between the energy consumption estimated by the German and the European standard ........................................................................................99

Fig.3.1.5_ 13: Energy saving estimation comparison......................................................100

Fig.3.1.5_ 14: Monthly energy saving percentage comparison........................................100

Fig.3.1.5_ 15: Energy consumption estimation for automatic shading system.................100 Fig.3.1.5_ 16: Energy saving comparison using different shading systems. 101

Fig.3.2.2_ 1: Bus device components .............................................................................107

Fig.3.2.2_ 2: Universal interface.....................................................................................108

Fig.3.2.2_ 3: Universal interface circuit diagram [ABB on line product catalogue].........108

Fig.3.2.2_ 4: Communication objects when used as a pulse counter (4 byte)...................109

Fig.3.2.2_ 5: Communication objects when used as a switch/dimming sensor ................109

Fig.3.2.2_ 6: Light sensor...............................................................................................109

Fig.3.2.2_ 7: Directive Diagram of the acrylic glass rod .................................................110

Fig.3.2.2_ 8: Occupancy sensor......................................................................................110

Fig.3.2.2_ 9: Occupancy sensor circuit diagram and detection area [Merten on line catalogue] ............................................................................................................... 111

Fig.3.2.2_ 10: Communication objects of presence sensor.............................................. 111

Fig.3.2.2_ 11: Dimming actuator ....................................................................................112

Fig.3.2.2_ 12: Dimming actuator circuit diagram [Hager on line product catalogue].......112

Fig.3.2.2_ 13: Dimming actuator functioning .................................................................113

Fig.3.2.2_ 14: Functioning scheme of the dimming actu ator..........................................113

Fig.3.2.2_ 15: Communication objects for dimming mode .............................................114

Fig.3.2.2_ 16: Weather station ........................................................................................115

Fig.3.2.2_ 17: Weather station wiring [Theben on line product catalogue] ......................116

Fig.3.2.2_ 18: Weather station implemented communication objects ..............................116

Fig.3.2.2_ 19: Timer switch............................................................................................117

Fig.3.2.2_ 20: Timer switch circuit diagram [ABB on line product catalogue] ................118

Fig.3.2.2_ 21: Timer switch communication objects .......................................................118

VI

Fig.3.2.2_ 22: Functional diagram of the static watt-hour meter [product technical documentation] ....................................................................................................... 119

Fig.3.2.2_ 23: Static watt-hour meter wiring diagram [product technical documentation] ....................................................................................................... 119

Fig.3.2.3_ 1: Minimum KNX installation example ......................................................... 120

Fig.3.2.3_ 2: ETS3 Topology of the case study program................................................. 121

Fig.3.2.3_ 3: ETS3 physical address list of the case study installation ............................ 121

Fig.3.2.3_ 4: Group addresses view of the case study installation................................... 122

Fig.3.2.3_ 5: Device configuration by parameters setting ............................................... 123

Fig.3.2.3_ 6: Device view of the case study program...................................................... 124

Fig.3.2.4_ 1: EIB OPC server configuration................................................................... 125

Fig.3.2.4_ 2: Graphpic 7.1 Architecture.......................................................................... 125

Fig.3.2.4_ 3: Visualization interfaces of the case study. .................................................. 126

Fig.3.2.4_ 4: Integrated database of the case study measuring module............................ 127

Fig.3.2.4_ 5: Graphs displayed by the case study supervision. ........................................ 128

Fig.3.2.4_ 6: Example of the first elaboration data program. .......................................... 128

Fig.3.2.4_ 7: Example of the second elaboration data program....................................... 129 Fig. 3.3.1_ 1: Input model (generated as simple input) of the classroom monitored 131

Fig. 3.3.1_ 2: Photorealistic rendering output generated by Adeline ............................... 132

Fig. 3.3.1_ 3: Diffuse daylight output result for overcast sky conditions......................... 134

Fig. 3.3.1_ 4: Diffuse daylight output result for clear sky with sun conditions. ............... 135

Fig. 3.3.1_ 5: Diffuse and direct daylight output result for overcast sky condition .......... 135

Fig. 3.3.1_ 6: 3D daylight factor calculation................................................................... 136

Fig. 3.3.1_ 7: 3D diffuse and direct daylight output result for sky with sun condition..... 136

Fig. 3.3.1_ 8: Typical day inside illuminance calculation in ML (October) and discrete output data point of the Adeline calculation. ........................................................... 137

Fig. 3.3.1_ 9: Calculation of solar position based on NOAA's functions and solar radiation based on Bird and Hulstrom's model: input data....................................... 139

Fig. 3.3.1_ 10: Calculation of solar position based on NOAA's functions and solar radiation based on Bird and Hulstrom's model: output results. ................................ 139

Fig. 3.3.1_ 11: Global radiation calculation obtained by Hulstron model and typical year model; typical day in October.......................................................................... 140

Fig. 3.3.1_ 12: Percentage difference between global radiation calculation obtained by Hulstron model and typical year model; winter months........................................... 141

Fig. 3.3.1_ 13: Percentage difference between global radiation calculation obtained by Hulstron model and typical year model; summer months ........................................ 141

VII

Fig. 3.3.1_ 14: Calculation results of the hourly sun seconds for each day of the test reference year, the correspondent sunshine probability hourly calculation and the calculation of the mean year....................................................................................142

Fig. 3.3.1_ 15: Sunshine probability calculation for the January reference year month....142

Fig. 3.3.1_ 16: Comparison between SSP and TDP calculation for the day with the minimum variance in January .................................................................................143

Fig. 3.3.1_ 17: Comparison between SSP and TDP calculation for the first working week in June and percentage difference comparison................................................144

Fig. 3.3.1_ 18: Comparison between SSP and TDP calculation for winter and summer months ....................................................................................................................144

Fig. 3.3.1_ 19: Energy consumption (kWh/moth) for the refer situation and for the data monitored in ML.....................................................................................................145

Fig. 3.3.1_ 20: Percentage difference between refer situation and ML energy consumption ...........................................................................................................146

Fig. 3.3.1_ 21: Adeline output results for lightswitch on/off control system....................147

Fig. 3.3.1_ 22: Energy saving calculated by Adeline and recorded for Scenario1............148

Fig. 3.3.1_ 23: energy saving percentage difference between lightswitch on/off Adeline calculation and recorded data. .................................................................................148

Fig. 3.3.1_ 24: Adeline output results for continuous dimming control system................149

Fig. 3.3.1_ 25: Adeline output for reference illuminance level equal to 300 and 2 luminairs types:.......................................................................................................150

Fig. 3.3.1_ 26: Adeline output for reference illuminance level equal to 500 and 2 luminairs types:.......................................................................................................150

Fig. 3.3.1_ 27: Energy saving calculated by Adeline and recorded for Scenario 2...........151

Fig. 3.3.1_ 28: Energy saving percentage difference between Continous Dimming for Reference Point Adeline calculation and recorded data ...........................................151

Fig. 3.3.1_ 29: Adeline output for reference illuminance level equal to 500 and lightswitch on/off control system, setting the daylight-optimized shading system and the fixed one with a shading coefficient equal to 0.7.........................................152

Fig. 3.3.1_ 30: Adeline output for reference illuminance level equal to 500 and reference point continuous dimming control system, setting the daylight-optimized shading system and the fixed one with a shading coefficient equal to 0.7 .............................................................................................................152

Fig. 3.3.1_ 31: Energy saving percentage difference for Continous Dimming for Reference Point and Lightswitch Adeline calculation: results for 0.3 shading factor setting ...........................................................................................................153

VIII

Fig. 3.3.1_ 32: Probability of somebody switching on the light on minimum daylight illuminace level in working area [SUPERLINK/RADLINK Technical Manual, IEA SHC Task 21 / ECBCS Annex 29 ADELINE 3.0 Documentation] ............................ 154

Fig. 3.3.1_ 33: Adeline output results for manual on/off lighting control system ............ 154

Fig. 3.3.1_ 34: Energy saving calculated by Adeline using a manual light system and recorded for ML. .................................................................................................... 155

Fig. 3.3.1_ 35: Percentage difference for energy saving calculated by Adeline using a manual light system and recorded for ML............................................................... 155

Fig. 3.3.1_ 36: Occupancy level calculation versus inside illuminance for the typical day ......................................................................................................................... 156

Fig. 3.3.1_ 37: Operating time probability as function of the inside illuminance............. 156 Fig.3.3.2_ 1: Monthly value of the Presence Factor for the 3 scenario in winter semester159

Fig.3.3.2_ 2: Modulus of the maximum percentage variation in the energy consumption normalization by PF value ................................................................. 159

Fig.3.3.2_ 3: Energy consumption in November: scenario 2........................................... 160

Fig.3.3.2_ 4: Occupancy period in November in the 3 automated classrooms and in maximum level....................................................................................................... 160

Fig.3.3.2_ 5: Presence level in percentage for traditional classrooms.............................. 161

Fig.3.3.2_ 6: Mean presence value in the traditional and automated classrooms. ............ 161

Fig.3.3.2_ 7: Monthly value of the outside Illuminance Factor for the 3 scenario in winter semester....................................................................................................... 162

Fig.3.3.2_ 8: Mean outside illuminance level during occupancy period in October......... 162

Fig.3.3.2_ 9: Energy consumption in December scenario 2 ............................................ 163

Fig.3.3.2_ 10: Outdoor illuminance levels during the occupancy period in the traditional classrooms ............................................................................................. 163

Fig.3.3.2_ 11: Outdoor illuminance levels for occupancy hours: ML, scenarios 1,2 and 3 in October............................................................................................................ 164

Fig.3.3.2_ 12: Visual comfort index distribution in November ....................................... 164

Fig.3.3.2_ 13: Discomfort Index value in November ...................................................... 165

Fig.3.3.2_ 14: Indoor illuminance levels during the occupancy period in the traditional classrooms .............................................................................................................. 166

Fig.3.3.2_ 15: Indoor illuminance levels during occupancy period in the traditional classrooms in October............................................................................................. 166

Fig.3.3.2_ 16: Energy consumption in traditional classroom November ......................... 167

Fig.3.3.2_ 17: Daily energy consumption in ML per working day for each the winter semester months ..................................................................................................... 168

IX

Fig.3.3.2_ 18: Daily energy consumption in ML in the winter semester months..............168

Fig.3.3.2_ 19: Mean lighting energy use in the traditional classrooms ............................169

Fig.3.3.2_ 20: Energy use during the winter semester .....................................................169

Fig.3.3.2_ 21: Absolute energy consumption in November .............................................170

Fig.3.3.2_ 22: Daily energy consumption in November ..................................................170

Fig.3.3.2_ 23: Lighting energy use for scenarios 1 for three different months. ................171

Fig.3.3.2_ 24: Lighting energy use for scenarios 3 for three different months. ................171

Fig.3.3.2_ 25: Occupancy-normalized energy use for ML, scenarios 1, 2, and 3. ............172

Fig.3.3.2_ 26: Energy use for scenario 2 normalized for occupancy level and for occupancy and indoor illuminance levels. ...............................................................172

Fig.3.3.2_ 27: Comparison between ML energy use and that one for the three scenarios................................................................................................................................173

Fig.3.3.2_ 28: Energy saving of the three scenarios in the winter semester .....................173

Fig.3.3.2_ 29: Energy consumption in October, scenario 3 .............................................174

Fig.3.3.2_ 30: ML presence levels in percentage in summer semester months and mean presence value in ML and in each scenario..............................................................175

Fig.3.3.2_ 31: Correlation between outside illuminace level and energy demand, for traditional classrooms in winter semester ................................................................176

Fig.3.3.2_ 32: Outside illuminace level in each detected month. .....................................176

Fig.3.3.2_ 33: ML energy consumption in summer months and mean winter semester value .......................................................................................................................177

Fig.3.3.2_ 34: Occupancy-normalized energy use for ML in summer months .................177

Fig.3.3.2_ 35: Indoor illuminance level during the occupancy period in the traditional classrooms in summer semester ..............................................................................178

Fig.3.3.2_ 36: Discomfort Index value in April...............................................................178

Fig.3.3.2_ 37: Energy saving of the three scenarios in the summer semester ...................179

Fig.3.3.2_ 38: Comparison between ML energy use and the one for the three scenarios...............................................................................................................................180

Fig.3.3.2_ 39: Occupancy-normalized energy use for ML, scenario 1, 2 and 3................180

Fig.3.3.2_ 40: Occupancy-normalized energy use for scenario 3 in winter semester months ....................................................................................................................181

Fig. 3.3.3_ 1: Time table definition for the visual comfort evaluation .............................183

Fig. 3.3.3_ 2: Focusing problems....................................................................................184

Fig. 3.3.3_ 3: Student age...............................................................................................184

X

Fig. 3.3.3_ 4: Student position: distance from the blackboard......................................... 185

Fig. 3.3.3_ 5: Student position: distance from the windows ............................................ 185

Fig. 3.3.3_ 6: Critical detail definition for white background.......................................... 186

Fig. 3.3.3_ 7: Critical detail definition for black background.......................................... 187

Fig. 3.3.3_ 8: Visual discomfort evaluation .................................................................... 188

Fig. 3.3.3_ 9: Illuminance from each glare source at the eye in lux................................. 188

Fig. 3.3.3_ 10: Inside illuminance calculation by Relux: overcast sky acc. CIE; date: 02.05.2007; time: 8:40; location: Trento (47.50°, 7.60°) ......................................... 190

Fig. 3.3.3_ 11: Inside illuminance calculation by Relux: overcast sky acc. CIE; date: 02.05.2007; time: 18:30; location: Trento (47.50°, 7.60°)........................................ 190



Fig. 3.3.3_ 14: Inside illuminance calculation by Relux: clear sky acc. CIE; date: 02.05.2007; time: 10:30; location: Trento (47.50°, 7.60°)........................................ 192

Fig. 3.3.3_ 15: Detail definition for black background.................................................... 193

Fig. 3.3.3_ 16: Detail definition for white background.................................................... 194

Fig. 3.3.3_ 17: Visibility calculation for E=200.............................................................. 196



Fig. 3.3.3_ 20: Visibility calculation for E=700 .............................................................. 199

Fig. 3.3.3_ 21: Positive Visibility ................................................................................... 200

Fig. 3.3.3_ 22: Negative Visibility.................................................................................. 200

XI

Summary

Given new and emerging standards and requirements in Europe regarding energy performance of buildings (see, for example, EPBD 2002/91), it has become increasingly important to improve the energy effectiveness of building operation. Toward this end, efficient daylight-responsive systems for illumination of buildings (including installation of automatic lighting control systems) can provide a significant contribution (EN 15193). This research aims to define a new methodology for the design of lighting systems in the lecture halls. The experimental phase, focused on a specific case study, is based on the comparison between the energy consumption for the lighting control in the actual situation without control system and one designed with automation systems. A completed analysis of the state of the art about sustainability and voluntary protocols, innovation in field of energy saving, lighting automation systems, visual comfort, illumination and vision has been developed [Chapter 1]. Innovative design tools for lighting systems have been elaborated: specifically, the methodological approach for the energy efficiency evaluation (design methodology, software tool outputs evaluation methodology, data analysis method); the application for technological system choice of multiple criteria decision analysis; the development of specific model sheets to monitor and analyze visual comfort conditions [Chapter 2]. These design tools have been implemented in a specific case study: the Faculty of Engineering of Trento. The cross layout of this building is characterized by two parallel wings, both with the same south exposure. The referred lecture halls (manually and automatically operated) are symmetrical and have the same shape. This configuration allows their simultaneous comparison. The method of the case study evaluation has been divided in four phases [Chapter 3]: 1- Analytical phase: monitoring and analysis of existing situation This phase deals with the verification of the photometric parameters and the lighting standards in compliance with the regulation in force (UNI EN 12464-2/2004) and with the estimation of the lighting energy required (EN 15193/2007). 2- Programmatic phase: definition of project objectives and system requirements This phase focuses on the configuration of Scenarios to regulate the lighting system in

XII

relation with the lecture hall use, the natural light contribution, the maintenance state and the efficiency of the lights. Moreover, the requirements definition of the supervision system for the monitoring activity and for the data bank implementation (with photometric data and energy consumption calculation) has been carried out. 3- Propositional phase: system solution In this step, drawing boards and technical reports have been developed in order to document the adopted solutions: on the one hand both the activation typology and the activation modality of the automatic control system, on the other hand both the position and the technical characteristics of the devices installed in the automation system. In order to pursue the reduction of energy demand and the visual comfort improvement, the following factors have been considered:

1. control of the students presence in the lecture room, as a necessary condition to turn on the light;

2. regulation of the constant light, in relation to the inside illuminance level; 3. partial switching off of the lecture room, if only half space is occupied; 4. introduction of a new kind of high efficiency light.

For the evaluation of these aspects the monitoring of the following data has been carried out: - the energy consumption in each classrooms: the manually operated and the one

automated (with the implementation of three different scenarios) - the users presence in the space, in order to make a scientific analysis of the energy

consumption, normalized through the real occupation time - the verification of the user behaviour using the automation system, even if its

regulation has been forced by a manual regulation, in order to evaluate the visual comfort for the users in a building automation system

- the inside illuminance with and without intelligent devices, in order to evaluate the natural light use for the visual comfort.

- the dimming percentage for each light line in the room. In this way the evaluation of the function modality for the system and for the partial energy consumption will be more efficient.

4- Evaluation phase: evaluation and verification The main goal of the evaluation phase has been to individuate and to quantify the differences between the calculated model and the real behaviour of a monitored building. The developed methodology focuses on the following issues:

XIII

1. simulation, calculation and discussion of the Adeline software tool outputs, in collaboration with the Fraunhofer Insitute of Stuttgard, and comparison with the monitored data;

2. elaboration and statistical analysis of the users behaviour, calculated with the Technical University of Wien, in order to introduce improvements in the modelling process;

3. evaluation of the visual comfort maintained and perceived by the users, in relation with the illumination levels, in order to verify if the proposed model allows the visual comfort users requirements, in collaboration with the Ophthalmology ward of S.Chiara hospital in Trento.

1. The comparison between Adeline outputs and data monitored focuses on the following issues: to calculate the difference between the inside illuminance conditions for the simulation model and for the real class rooms; to calculate the outside conditions for the model (in particular the sunshine probability) and to verify the variance of statistical sky conditions; to understand and to quantify how much the recorded users behaviour differs in lecture halls, compared with the one used in the simulation tool; to analyze how the automated systems work in a real situation and how much energy saving has been expected looking for the Superlink output. 2. The energy saving potential for each scenario has been calculated related to a reference energy demand level, named ML (maximum level), that corresponds to the traditional classrooms. In order to understand and model users' behaviour in ML and to quantify the energy saving potentials of automation systems for lighting control, a typical day data analysis has been carried out. In this way it is possible to perform an hourly data analysis, depicting typical patterns of presence, actions, and energy use over time instead of mere daily overview. Monitoring results are stored in a databank and structured by the supervision software. The lighting energy use in the lecture rooms for the implemented scenarios has been compared using standardisation techniques. In particular the following normalisation factors have been considered: occupancy level, outside illuminance, indoor illuminance factor. 3. In order to analyze the visual comfort, a significant student sample has been tested. The goals of the subjective analysis are: to check, if present, the uncorrected refractive errors of the students, for different contrast conditions; to define the critical detail for the students during a typical lesson, for defined boundary conditions; to compare the veiling luminance level perceived and the discomfort value calculated; to evaluate the visual discomfort indexes of the students; to analyse the visibility, as calculated and perceived value, for defined boundary conditions.

Chapter 1

Literature overview

1.1 Sustainability and voluntary protocols

Climate changes are one of the main challenges that our society will face in the incoming years. An early definition of sustainable development was formulated in 1987 by the World Commission on Environment and Development (WCED). After twenty years, the importance of environmental issues has become a key aspect of political and scientific debates around the world. Given concerns regarding global warming, the urgency of reducing CO2 emissions is growing. Thereby the built environment plays an important role: the residential and commercial sectors account for more than 40% of end energy consumption in the European Community and are thus responsible for an important part of carbon dioxide emissions systems (Fig.1.1_1) [UNEP, 2007].

41%

31%

28%

RESIDENTIAL/COMMERCIALTRANSPORTINDUSTRY

Fig.1.1_ 1: The proportion of the energy consumption in EU

The fulfilment of the Kyoto commitment for the reduction of Green House Gas emissions cannot be obtained without the efforts of all Europeans, in particular, public and private organisations in all fields of our economy [ECCP, 2001]. Thanks to both the initiatives of the individual Governments and the European Union research programs, the recent research activities in Europe have lead to the definition and

CHAPTER 1

2

development of energy and environmental assessments tools. These tools can analyse the building performance from an environmental sustainability point of view, and respond thus to the applicable energy policies. Lighting has a substantial impact on the environment (Fig.1.1_2): recent studies carried out for the European Commission have shown that between 30% and 50% of electricity used for lighting could be saved investing in energy-efficient lighting systems. In most cases, such investments are not only profitable but they also maintain or improve lighting quality. [Bertoldi P., 2002].

25%

7%

11%

57%

WATER HEATING COOKING

ELECTRICAL APPL. SPACE HEATING

9%

25%

14%

52%

WATER HEATING LIGHTING

OTHER ELECTRICAL APPL. SPACE HEATING

Fig.1.1_ 2: Structure of energy consumption in residential sector (left)

and in non residential sector (right)

The main legislative instruments affecting the buildings sector is the Directive 2002/91/EC of the European Parliament and of the Council of 16 December 2002 on the energy performance of buildings [Directive 2002/91/EC]. The mean goal of this directive is to decrease the energy consumption implementating cost-effective savings potential for around 22% of present consumption in buildings within 2010. It represents 55 Mtoe energy consumption, about 20 % of the Kyoto Protocol target. Directive establishes requirements mainly as regards: - general framework for a methodology of calculation of the integrated energy performance of buildings; - application of minimum requirements on the energy performance of new buildings; - application of minimum requirements on the energy performance of large existing buildings that are subject to major renovation; - energy certification of buildings; - regular inspection of boilers and of air-conditioning systems in buildings and in addition an assessment of the heating installation in which the boilers are more than 15 years old. The adoption of a calculation methodology for an integrated energy performance of buildings requires every government to apply an own process which calculates the energy performance

Literature overview

3

of buildings. These calculations must be based on a general framework incorporating the

following items:

(a) thermal characteristics of the building;

(b) heating installation and hot water supply, including their insulation characteristics;

(c) air-conditioning installation;

(d) ventilation;

(e) built-in lighting installation ,mainly the non residential sector (prEN 15193-section 1.2);

(f) position and orientation of buildings, including outdoor climate;

(g) passive solar systems and solar protection;

(h) natural ventilation;

(i) indoor climatic conditions, including the designed indoor climate.

The world programmatic actions activated to prevent environmental catastrophes and the

declarations to apply solutions for a sustainable development have been presented in

numerous recent international conferences [e.g. International Conference Sustainable

Building 2006-2007 and 5° meeting “EPBD Concerted Actions“, Budapest, 8-9 May 2006].

In a previous research activity, the author analyzed with scientific rigor the methodologies

and the voluntary protocols applied in order to solve the emergent environmental problems

for different levels: worldwide, European, Italian, alpine and regional (regions of north Italy,

specifically Trentino)1.

This analysis has been carried out studying the original texts of national and international

standards and regulations. A structured synthesis has been developed, elaborating original

synoptic-comparative sheets. Based on a specific and accurate method, this tool permits to

detect and identify the main aspects of the articulated and wide field of study with a unitary

and global approach. The mentioned study begins from the world situation and programmatic

action analysis and continues deepening systematically more localized areas, in order to

focus on the state of the art about the sustainable approach in the alpine region.

Moreover the Directive 2006/32/EC of the European Parliament and of the Council of 5

April 2006 on energy end-use efficiency and energy services and repealing Council Directive

93/76/EEC establishes to improve energy efficiency delivered by energy services to citizens

and businesses [Directive 2006/32/EC].

It includes contribution to security of energy supply, by managing overall demand by

dissemination of energy efficient technologies and techniques and innovation boosts.

1 The Trento Common funds the research project entitled “Eco compatible planning: analysis of the technical

data used in the Italian alpine region, building’s performance and materials requirements for energy saving” for

the School of Doctoral Studies in Environmental Engineering, XX cycle. This research has been developed by

the author during the first year of the Phd study.

CHAPTER 1

4

In order to achieve these scopes the following strategies have been established:

- setting national indicative energy savings targets;

- removal institutional, financial and legal barriers to energy efficiency;

- set up conditions to develop and promote a market for energy services, together with

delivering other energy efficiency improvement measures to help final customers.

The energy saving target is 9% by 2017.

The SmartHouse Code of Practice CWA 50487:2005 is another importat document that

explains the possible impact of the smart technology application in order to create possibility

for efficient energy management and energy savings.

It is important to emphasize that: “SmartHouse includes the digital home, intelligent home,

connected home, and networked home. SmartHouse includes any “smart” activity, service or

application in the SmartHouse including any form of “office” or working environment in the

SmartHouse (but the smart office in commercial premises is excluded). SmartHouse covers

any residential premises where people live (e.g. house or apartment) but excludes

commercial and institutional premises”. [CENELEC, 2005]

As referred by Code of Practise, the applied technology itself and its usage could actually

cause significant increases in power consumption by facilitating and encouraging the use of

multiple appliances and systems. Smart house should be designed to allow intelligent

management of heating and lighting systems and should regulate energy consumption

depending on the occupancy of the house.

According to [Ashford P., 1999] the expected energy savings technical potential in building

stock has been calculated as reported in Fig.1.1_3 [Zalesak M., 2006].

Fig.1.1_ 3: Expected energy savings technical potentials in building stock

Literature overview

5

1.2 Innovation in field of energy saving

Global lighting electricity use

A first global estimate of lighting energy use, costs, and associated greenhouse-gas emissions

has been presented in recent research activities carried out at the IEA (International Energy

Agency) [Mills,2002].

Based on a compilation of estimates for selected countries representing approximately 63%

of the world’s population, a model for predicting lighting electricity use for other countries

where data are lacking has been developed. The corresponding lighting-related electricity

production for the year 1997 is 2016 TWh (21103 Petajoules), equal to the output of about

1000 electric power plants (assuming an average plant size of 400 megawatts, a 60% capacity

factor, and 10% transmission and distribution losses) and valued at about $185 billion per

year. Global lighting electricity use is distributed approximately 28% to the residential

sector, 48% to the service sector, 16% to the industrial sector, and 8% to street and other

lighting. The corresponding carbon dioxide emissions are 1775 million metric tonnes per

year.

An examination of global fuel-based household lighting suggests that it represents an amount

of primary energy of 3600 PJ ($48 billion). Although one third of the people obtain light with

kerosene and other fuels, representing about 20% of global lighting costs, they receive 0.2%

of the resulting lighting energy services.

While collecting end-use energy data is arguably not a high national priority in most

countries, this lack of attention is particularly problematic in the instance that lighting is a

preferred target for energy savings campaigns and policies.

Lighting electricity demand in the 23 IEA countries 2 represents approximately half of the

world’s total lighting use. Approximately 50% of IEA lighting energy (531 TWh) is used

within the IEA service sector.

Service sector lighting electricity average value is equal to 6% for the IEA countries

evaluated. It ranges from 39% (Germany and Japan) to 61% (Netherlands) of total service

sector electricity use in the IEA member countries.

country Total light

(TWh) Residential

(TWh) Services (TWh)

Industrial (TWh)

Ligting as fraction of total elec.

Italy 18.82 7.22 8.40 3.20 7%

Fig.1.2_ 1: Lighting energy demand in Italy [Mills E., 2002]

1 Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Japan,

Luxembourg, Netherlands, New Zealand, Norway, Portugal, Spain, Sweden, Switzerland, United Kingdom, USA.

CHAPTER 1

6

Lighting actions in the residential sector

Total domestic lighting consumes about 86 TWh in the Union and it is predicted to raise to

102 TWh by 2020.

The European Commission in cooperation with several national energy agencies and public

and private organisations is promoting end-use energy efficiency and conservation as a key

component of the EU energy policy and the common goal of reducing climate changes.

Three of the most important actions in the residential sector are:

- the mandatory EU energy label for lamps

- the voluntary ecolabel for lamps

- the European Quality Charter for Compact Fluorescent Lamps

As indicated in DELight study [Environmental Change Unit, 1998]: “Electric lighting is used

in practically all households throughout Europe and represents a key component of peak

electricity demand in many countries. There is already a well developed energy-efficient

technology available on the market, in the form of compact fluorescent light bulbs (CFLs),

that could deliver substantial savings. Such savings could be accessed quickly due to the

rapid turnover of light bulbs in the stock - the challenge is to get the more efficient

technology installed and guarantee the savings.”

The EU CFL QC started in 1998 on the initiative of the European Commission and

Eurelectric to support the European Wide Initiative for the Promotion of Efficient Lighting in

the Residential Sector. The original aim of the European CFL Quality Charter is to offer a

high quality standard to be used by utilities and other bodies in their promotion and

procurement campaigns. During the year 2002, the first revision of the European CFL

Quality Charter took place; other revisions were included in 2003 and 2005 (under way). The

European CFL Quality Charter is a voluntary set of criteria established by the European

Commission in collaboration with a number of private and public organisations.

Lighting actions in non-residential sector

The main European actions in field of lighting energy saving in non-residential sector are:

• Directive on energy efficiency requirements for ballasts for fluorescent lighting

• Energy-efficiency classification of ballast-lamp circuits (in collaboration with CELMA)

• The GreenLight Programme

Specifically the GreenLight Programme is a voluntary pollution prevention initiative

encouraging non-residential electricity consumers (public and private), referred to as

Partners, to commit towards the European Commission to install energy-efficient lighting

technologies in their facilities when it is profitable, and lighting quality is maintained or

improved. GreenLight was launched on 7 February 2000 by the European Commission

Directorate General Energy & Transport.

Literature overview

7

The objective of the GreenLight programme is to reduce the energy consumption from

indoor and outdoor lighting throughout Europe, thus reducing polluting emissions and

limiting the global warming. The objective is also to improve the quality of visual conditions

while saving money. [Bertoldi P., 2002].

European standard about energy requirements for lighting: prEN 15193-2006

This European standard was introduced to establish conventions and procedures for the

estimation of energy requirements of lighting in buildings, and to give a methodology for a

numeric indicator of energy performance of buildings. It also provides guidance on the

establishment of notional limits for lighting energy derived from reference schemes.

Having the correct lighting standard in buildings is especially importance and the convention

and procedures assume that the designed and installed lighting scheme conforms to good

lighting practices. For new installations, the design will be to EN 12464-1, Light and

Lighting – Lighting of work places – Part 1: Indoor work places.

The standard also gives advice on techniques for separate metering of the energy used for

lighting that will give regular feedback on the effectiveness of the lighting controls.

The methodology of energy estimation does not provide only values for the numeric

indicator but will also provide input for the heating and cooling load impacts on the

combined total energy performance of building indicator.

The methodology and format of the presentation results would satisfy the requirements of the

EC Directive on Energy Performance of Buildings 2002/91/EC.

In particular, this standard specifies the calculation methodology for the evaluation of the

amount of energy used for indoor lighting inside the building and provides a numeric

indicator for lighting energy requirements used for certification purposes. This standard can

be used for existing buildings and for the design of new or renovated buildings. It also

provides reference schemes to base the targets for energy allocated for lighting usage. This

standard also provides a methodology for the calculation of instantaneous lighting energy use

for the estimation of the total energy performance of the building. In this standard, the

buildings are classified in the following categories: Offices, Education buildings, Hospitals,

Hotels, Restaurants, Sports facilities, Wholesale and retail services and Manufacturing

factories.

1.3 Home and building automation

Building automation is a programmed, computerized, intelligent network of electronic

devices that monitor and control the mechanical and lighting systems in a building. The

intent is to create an intelligent building and reduce energy and maintenance costs [Bellantani

S., 2004].

Home automation (also called domotics) is a field within building automation, specializing in

CHAPTER 1

8

the specific automation requirements of private homes and in the application of automation

techniques for the comfort and security of its residents. Although many techniques used in

building automation (such as light and climate control, control of doors and window shutters,

security and surveillance systems, etc.) are also used in home automation, additional

functions in home automation include the control of multi-media home entertainment

systems [Capolla M., 2004].

Specific domotic standards include INSTEON, X10, LonWorks, KNX (standard), System

Box, Crestron, C-Bus, Universal power line bus (UPB), UPnP, ZigBee and Z-Wave that will

allow for control of most applications.

Some standards use control wiring, some embedded signals in the power line, some use radio

frequency (RF) signals, and some use a combination of several methods. Control wiring is

hardest to retrofit into an existing house. Some appliances include USB that is used to control

it and connect it to a domotics network. Specific gateways translate information from one

standard to another.

X10 is an international and open industry standard for communication among electronic

devices. It primarily uses power line wiring for signalling and control, where the signals

involve brief radio frequency bursts representing digital information. A wireless radio based

protocol transport is also defined. X10 was developed in 1975 by Pico Electronics of

Glenrothes, Scotland, in order to allow remote control of home devices and appliances. It was

the first domotic technology and remains the most widely available [Quaranta G., Mongiovì

P., 2004].

LonWorks is a networking platform specifically created to address the unique performance,

reliability, installation, and maintenance needs of control applications. The platform is built

on a protocol created by Echelon Corporation for networking devices over media such as

twisted pair, powerlines, fiber optics, and RF. It is popular for the automation of various

functions within buildings such as lighting and HVAC.

In 1999 the communications protocol (then known as LonTalk) was submitted to ANSI and

accepted as a standard for control networking (ANSI/CEA-709.1-B). Echelon's power line

and twisted pair signalling technology was also submitted and accepted to ANSI for

standardization. Since then, ANSI/CEA-709.1 has been accepted as the basis for IEEE

1473-L (in-train controls), AAR Electro-pneumatic braking systems for freight trains, IFSF

(European petrol station control), SEMI (semiconductor equipment manufacturing), and in

2005 as EN 14908 (European building automation standard). The protocol is also one of data

link/physical layers of the BACnet ASHRAE/ANSI standard for building automation [Sauter

T., Dietrich D., Kastner W., 2001].

Literature overview

9

C-Bus is a home and building automation protocol that is used in Australia, New Zealand,

Asia, the Middle East, Russia, USA, South Africa, the UK and other parts of Europe

including Greece and Romania. C-Bus was created by Clipsal's Clipsal Integrated Systems

division for use with its brand of home automation and building lighting control system.

C-Bus has recently become available in the USA under the 'SquareD Clipsal' brand name.

C-Bus is used in the control of domotics, or home automation systems, as well as commercial

building lighting control systems. C-Bus uses a dedicated low-voltage cable or two-way

wireless network to carry command and control signals. This improves the reliability of

command transmission and makes C-Bus far suitable for large, commercial applications.

The C-Bus System can be used to control lighting and other electrical systems and products

via remote control and can also be interfaced to a home security system, AV products or other

electrical items. The C-Bus system is available in a wired version and a wireless version, with

a gateway available to allow messages to be sent between wired and wireless networks.

[Sauter T., Dietrich D., Kastner W., 2001]

KNX is a standardised (EN 50090,ISO/IEC 14543), OSI-based network communications

protocol for intelligent buildings. KNX is the successor to, and convergence of, three

previous standards: the European Home Systems Protocol (EHS), BatiBUS, and the

European Installation Bus (EIB).

EIB-Instabus, was a decentralized open system to manage and control electrical devices

within a facility. Berker, Gira, Jung, Merten and Siemens AG developed it. There are about

120 companies of electrical supplies using this communication protocol. The EIB (European

Installation Bus) allows all electrical components to be interconnected through an electrical

bus. Every component is able to send commands to other components, no matter where they

are. [EIB Konnex Association, 1999]. The commissioning principle of this standardized

protocol has been analyzed in section 3.2.3.

From the point of view of the location of the intelligence of the domotic system resides, there

are three different architectures:

• Centralized Architecture: a centralized controller receives information of multiple

sensors and, once processed, generates the opportune orders for the actuators.

• Distributed Architecture: all the intelligence of the system is distributed by all the

modules that are sensors or actuators. Usually it is typical of the systems of wiring in

bus.

• Mixed Architecture: systems with decentralized architecture as far as they have

several small devices able to acquire and to process the information of multiple

sensors and to transmit them to the rest of devices distributed by the house.

CHAPTER 1

10

1.4 Daylighting simulation tools

Introduction

Daylighting design is a creative process. It aims to generate appropriate architectural and/or

technical solutions to achieve an enjoyable and productive built environment while

simultaneously reducing the energy consumption of buildings through the substitution of

daylight for electric light.

Design tools are intended to help designers with the qualitative and quantitative elements of

daylighting design through features that commonly include:

• visualisation of the luminous environment of a given daylighting design;

• prediction of daylight factors in a space lit by diffuse daylight;

• identification of potential glare sources and evaluation of visual comfort index;

• prediction of potential energy savings achievable through daylighting;

• control of the penetration of the sun’s rays and visualisation of the dynamic behaviour of

sunlight.

Architects have traditionally used scale models as tools for predicting the available

illumination in a building and for visualizing the overall lighting quality of the environment.

Although scale modeling techniques have some advantages relative to calculation methods,

they also have several technical shortcomings in actual application. Scale model photometry

requires considerable time in building physical models and accurately calibrating costly

photometric instruments. The test results from outdoor measurements may not always be

reproducible. Once set up, models cannot always be easily modified in response to changes

in design features.

Realizing these limitations, lighting researchers have tried to develop daylighting calculation

tools for use in the design process. These mathematical tools are based on theories of lighting

physics under hypothetical indoor and outdoor conditions.

Daylighting calculation methods can be categorized into two types depending on their form:

graphic methods and numerical calculations. Based on their theoretical approaches, the

numerical calculation techniques can be further divided into two groups [IEA, 2000]:

• The daylight factor method, was originally developed for uniform sky conditions.

This method consists of two principal calculation procedures: integrating the sky

luminance that is visible through openings for the sky component calculation, and

approximating the interreflected light in a space by the split-flux method. Using

manual calculation techniques, only spaces with simple windows could be analyzed

for uniform sky conditions. In addition, the effect of the transmittance variation with

the angle of incidence could not be accounted for in the manual calculation methods.

This method was later expanded to include overcast and clear sky conditions.

Literature overview

11

• the lumen method, was derived from the principle that indoor illuminance levels are

proportional to the light flux entering through openings due to different daylight

sources. This method was initially developed for estimating light levels from electric

light sources. In daylighting applications, this flux-averaging method does not

account for the spatial distribution of source luminance, nor does it separate the

indoor illuminance into sky and reflected components. A series of indoor illuminance

measurements were made for various space proportions and the results were

tabulated in look-up tables. The lumen method is designed to enable quick estimation

of illumination levels for a given space configuration by interpolating between values

in the look-up table. This engineering approach is useful in estimating the indoor

daylight level before the room and window geometry are precisely determined.

The application of computers to daylighting calculations has provided two major benefits.

For sky component calculations, the complexity introduced by the presence of internal or

external obstructions can be handled easily by using geometric algorithms that identify

patches of sky visible from indoors through openings. These geometric algorithms also make

it possible to accurately calculate radiation flux exchange between internal surfaces within a

space having a complex configuration. Secondly, the luminance integration for

non-rectangular windows under any kind of sky luminance distribution can be calculated to a

higher degree of accuracy by numerical integration techniques. In fact, if the luminance

distribution of actual sky conditions can be measured, then indoor illumination levels can be

calculated for actual sky conditions, including partly cloudy cases.

The greatest contribution of computers to daylighting calculation, however, is in the

calculation of reflected components. Previously, these components had been estimated by the

split flux method, which gives approximated results only for roughly spherical rooms, under

the assumption that reflected light is uniformly distributed over the entire work plane.

Despite its limitations, the split flux method has been widely used for several practical

reasons. First, for typical interior surface textures and reflectances (where these values are

not too high and their spatial variation is not too drastic) the contribution of reflected

components is often only a small portion of the total illuminance. Second, when only

diffusive outdoor sources are considered, the split-flux method yields reasonable results.

However, when directional light such as sunlight is considered, the calculated results deviate

from actual behaviour. Finally, the alternate method of calculating reflected light, requires

rigorous computation that can be done only by computers.

Global illumination models

Two main categories of computer-based tools can be distinguished based on the calculation

methods they use: the radiosity technique and the ray-tracing technique.

CHAPTER 1

12

Fig.1.4_ 1: Radiosity Method (Daylight in Building – IEA Solar Heating and Cooling

Programme Task 21, section 6 Design Tolls)

The radiosity method is probably one of the first lighting calculation techniques applicable to

the evaluation of the interchange of light among all the surfaces defining an architectural

space. This method has a significant advantage over former analytical techniques because it

allows for light inter-reflections between surface walls.

Originally developed for energy calculations, the radiosity method was used to determine the

energy balance of a set of surfaces exchanging radiant energy (Figure. 1.4_1). Some of its

basic hypotheses and limitations are that:

• wall surfaces must be subdivided into small finite elements characterised by homogeneous

photometric properties (e.g., reflection coefficient);

• all elements must be perfect diffusers (Lambert’s law);

• similar hypotheses must be applied to all of the external obstructions situated in front of

windows and openings.

The radiosity method is used to determine the illuminance and luminance of a set of points

located at the centres of different surface elements.

The ray-tracing technique determines the visibility of surfaces by tracing imaginary rays of

light from a viewer’s eye to the objects of a rendered scene. A centre of projection (the

viewer’s eye) and an arbitrary view plane are selected to render the scene on a picture plane.

Thanks to the power of novel computer algorithms and processors, millions of light rays can

be traced to achieve a high-resolution rendered picture.

Originally developed for imaging purposes, some ray-tracing programmes were adapted and

optimised for calculation of daylighting within building spaces. In this case, light rays are

Literature overview

13

traced until they reach the main daylight source, which is usually the sun position (clear and

intermediate skies) or the sky vault (cloudy skies). The Fig. 1.4_2 illustrates the principle of

ray tracing, showing the viewpoint (P) and view direction of the observer as well as the main

light source, represented by the sun.

Fig.1.4_ 2: Ray-Tracing Method (Daylight in Building – IEA Solar Heating and Cooling

Programme Task 21, section 6 Design Tolls)

Evaluation of available daylighting software programs

By the mid-1980’s, a number of software packages were under development to predict

daylighting performance in buildings, in particular illumination levels in daylighted spaces.

The evaluation of these first tools demonstrated that none of this software then available was

capable of predicting the simplest of real daylighting designs [Ubbelohde et al, 1988]. In the

last years computer capabilities have evolved rapidly. The current software packages are far

more powerful and nuanced in their ability to predict daylight than previously. Some can

accurately predict quantitative daylight performance under varying sky conditions and

produce handsome and accurate visualizations of the space. The programs differ

significantly, however, in their ease of use, modeling basis and the emphasis between

quantitative predictions and visualization in the output.

Daylighting prediction software is often reviewed as a subset or feature of lighting design or

energy simulation software, although attention has also been paid to these programs as

visualization software, already from the ninetieths years. [Novitski 1992, 1993]. Additionally

after that, articles on the performance and features of individual software packages appear

regularly in the lighting press, computer graphics publications and architectural journals

[Mahoney 1994; Dubiel et al. 1995; Sullivan 1996].

The Illuminating Engineering Society of North America (IESNA) published an annual

software survey in Lighting Design + Application, an extensive, broad matrix which

categorizes the features of each package, including price, addresses of vendors and computer

hardware requirements.

CHAPTER 1

14

Such technical papers are regularly published in conference proceedings from SIGGRAPH

and the Journal of the IES. A glimpse into the range of technical conference papers is offered

by Ian Ashdown’s radiosity bibliography. [Ashdown 1996].

The IEA Task 21 validation program investigates the accuracy of Radiance, Genelux and

SuperLite. Simulations using these three software packages were compared to data sets

developed with extremely simplified physical models (a side lighted room and a top lighted

courtyard) measured in artificial skies. Within these highly controlled comparisons, the three

software packages delivered close and accurate predictions of illumination levels

[IEA,1998].

A comparison among four of the main daylighting software programs (Lumen Micro from

Lighting Technologies Inc. www.lighting-echnologies.com; Superlite and Radiance from the

Environmental Energy Technology Division, Building Technologies Program at Lawrence

Berkeley National Laboratory etd.lbl.govandradsite.lbl.gov/radiance/HOME.html;

Lightscape Visualization System from Lightscape Technologies, Inc. www.lightscape.com)

has been developed by [Ubbelohde, 1999].

The cited study is neither a validation exercise nor research into the calculation algorithms of

these software packages, but rather focuses on a single case of predictions and monitored

data, with all the potential difficulties both can offer in terms of accuracy. The results should

be viewed with regard to these limitations. In the quantitative results, this study indicates that

the more restrictive an input model is, the more likely a real design will not be modelled

accurately in the input state. This means it is less likely that the output will be an accurate

prediction of the daylight in the real building.

Both Lumen Micro and SuperLite have been shown to be accurate if the space being

modelled matches the limitations of the input model requirements. These two programs offer

daylighting predictions with objectives other than visualization and rendering. For Lumen

Micro, the daylighting analysis is essentially a supplement to a powerful industry-standard

electrical lighting software and is appropriate in generic daylighting applications such as

sidelighted offices. SuperLite can be used in Adeline to generate inputs for sophisticated

thermal analysis programs. However, when Radlink connects a Radiance rendering to these

thermal analysis programs smoothly, SuperLite may cease to have a clear function in the

Adeline software package. SuperLite is also used as a validation tool and a base case for

comparison in situations where the model geometry can be completely trusted.

Radiance has proven in this study to be much accurate in predicting illumination levels.

However, Radiance does not yet have a reasonable user interface and requires a great deal of

time and training to use well.

In the last years, multiple studies have been carried out internationally to collect data on

building users’ interactions with building control systems and devices, in order to develop a

Literature overview

15

stochastic model for predicting lighting energy consumption. Some examples of how

researchers modelled manual lighting control are summarized in the followings.

Hunt (1980) suggested a prediction method for manual lighting control that is based on field

study data. Based on a switch-on probability function for electric lighting and annual

frequency distributions of indoor illuminance, he generated mean switch-on probabilities for

certain times of a weekday. He assumed that lighting is switched on at the start of a period of

occupation, left on throughout the day and switched off at the end [Hunt, 1979].

The first paper that followed a stochastic approach to manual lighting control was presented

in [Newsham, Mahdavi and Beausoleil-Morrison,1995]. The authors developed a model

called Lightswitch that simulated user occupancy at the work place based on measured field

data in an office building in Ottawa, Canada. The resulting user occupancy profiles were then

used to estimate the energy benefit of an occupancy sensor. The lighting was constantly

activated throughout the day for the manually operated reference lighting system. For the

occupancy sensor controlled system the lighting was switched on upon occupant arrival at

the work place and switched off whenever the user left the workplace for a time step longer

than the delay time of the occupancy sensor.

Reinhart proposes a new manual lighting and blind control algorithm that is dynamic and

stochastic: Lightswitch-2002. Dynamic indicates that instead of looking at an average day in

a year or month, user occupancy, indoor illuminance and the resulting status of the electric

lighting and blinds are considered in 5min time steps throughout the year. Stochastic means

that whenever a user is confronted with a control decision i.e. to switch on the lighting or not,

a stochastic process is initiated that determines the outcome of the decision [Reinhart 2002].

The integration of the Lightswitch2002 behavioural algorithms in the online design support

tool Lightswitch Wizard [www.buildwiz.com] and the expert daylighting analysis software

DAYSIM [www.daysim.com], allows for a more realistic estimate of lighting use under

dynamic conditions. The current downside of these approaches is that the whole building

energy impact of manual changes in blind settings and lighting use is not considered

[Dubrous, 2004].

In [Bourgeois D., Reinhart C., Macdonald I., 2005], a sub-hourly occupancy-based control

(SHOCC) model is presented, which allows advanced behavioural models to be integrated in

whole building energy simulation programs. The enhanced functionality is demonstrated

through annual energy simulations in a private office.

The [Mohammadi A., Kabir E., Mahdavi A., Pröglhöf C., 2007] describes an effort to

observe control-oriented occupant behaviour in 29 offices of a large high-rise office complex

over a period of one year.

Such data can bring about a better understanding of the nature, type and frequency of

control-oriented user behavior in buildings and thus support the development of

CHAPTER 1

16

corresponding behavioral models for integration in building performance simulation

applications. Moreover, such data could support the effective (and proactive) operation of

building service systems for indoor environmental control.

1.5 Comfort, illumination e vision

Light and vision

Eyes anatomy

Vision depends on light. Eye is a complex sensory organ which maintains the spatial and

temporal relationships of objects in visual space and converts the light energy received into

electrical signals processed by the brain.

The eye can be divided into:

optical components

- cornea

- crystalline lens

- pupil

- intraocular humors

neurological components

- retina

- optic nerve

Fig.1.5_ 1: Diagram of the Eye (National Eye Institute – U.S. National Institutes of

Health, available in http://www.nei.nih.gov/health/eyediagram/eyeimages3.asp)

The cornea is the transparent front part of the eye that covers the iris, pupil, and anterior

chamber, providing most of an eye's optical power. Together with the lens, the cornea refracts

Literature overview

17

light, and as a result helps the eye to focus, accounting for approximately 80% of its

production to 20% of the lens focusing power.

The lens or crystalline lens is a transparent, biconvex structure in the eye that, along with the

cornea, helps to refract light to focus on the retina. By changing the curvature of the lens, one

can focus the eye on objects at different distances from it. This process is called

accommodation. In humans, the refractive power of the lens in its natural environment is

approximately 15 dioptres, roughly one-fourth of the eye's total power.

The aqueous humor and vitreous humor help maintain the shape of the globe and provide

nutrients to non vascular structures within the eye.

The retina is a thin layer of neural cells that lines the back of the eyeball. It is comparable to

the film in a camera: the retina and the optic nerve originate as outgrowths of the developing

brain. Hence, the retina is part of the central nervous system (CNS). It is the only part of the

CNS that can be imaged directly. The vertebrate retina contains photoreceptor cells (rods and

cones) that respond to light; the resulting neural signals then undergo complex processing by

other neurons of the retina [Marsigliante, 2006].

Eyes physiology

Rods and cones

Radiation that reaches the posterior eye must be converted into an electrical signal that can be

analyzed by the brain. The incident radiation is absorbed by photopigments located I the

outer segments of retinal receptors: roads and cones.

The greatest concentration of cones occurs at the fovea. The number of cone drops

precipitously away from the fovea, although they can be found even in the far periphery of

the retina. Rods, on the other hand, are scarce near the fovea and increase in number toward

the periphery.

For this reason has been defined a “peripheral vision” and a “foveal vision”.

At low light levels the road system determinate the sensitivity and there is a scotopic vision: it

occurs at illumination levels under which the cones cease to function, at sustantially less than

1 footcandels, such as those illuminances experienced on a starlit night. With scopic vision,

there is no perception of color, and central or foveal vision is impaired.

At higer light levels, the cone sistems determine the relative spectral sensitivity and the

photopic vision is activated. It is defined as vision at relatively high light levels where the

cones are fully activated. It occurs at illumination levels above 3 footcandles (30lx). Almost

all research on visual acuity and visual preferences had occurred in the illumination range

from 10-200 footcandles (100-2000lx), to represent indoor work environments. Illumination

meters are typically adjusted to the ranges of sensitivity of the eye in this range [IESNA,

2000, 2003].

CHAPTER 1

18

Fig.1.5_ 2 : Scotpic, Mesopic and photopic range (IESNA, 2003, Advance Lighting

Guideline, 2-Lighting and Human Performance, available in

http://www.newbuildings.org/ALG.htm)

Scotopic and Photopic Spectral Luminous Efficiency

0,0

0,2

0,4

0,6

0,8

1,0

380

400

420

440

460

480

500

520

540

560

580

600

620

640

660

680

700

720

740

760

Wavelenght of maximum louminous efficacy (nm)

%

Scotopic Vision

Photopic Vision

Fig.1.5_ 3: Spectral Louminous Efficiency Values, V’(l) – Unity at Wavelength of

Maximum Luminous Efficacy

Adaptation

The human eye can process information over range of illuminance from about 1 lx to over

100.000lx. The eye change its sensitivity to light: this process is called adaptation.

It involves three main processes:

- change in pupil size, the pupil can constrict in response an increase in light level

about 5 time faster than it can dilate in response to a drop in light level;

- neural adaptation, a change in neural sensitivity that allows us to instantly adapt to a

range of illumination levels typical of most interior level (about 3 orders of

magnitude, but in working places from 100 to 1000lx)

Literature overview

19

- photochemical adaptation, involving the bleaching and regeneration of the pigments

in the rods and cones under more extreme range of illumination. The cone system can

regenerate in 10-12minutes, while the rod system may require up to 60 minutes for

full generation.

Eye movements

It is the function of the oculmotor system of the eye to position the lines of sight of the two

eyes so that they are pointed at the object of regard. If the image of the target will be reduced.

This is important in certain tasks, such as driving at night, because an object detected in the

periphery of vision often causes the driver’s visual attention to be momentarily diverted from

road to the object. Changes in the illumination of stationary object, such as flash or flicker,

also draw our attention.

Accommodation

As the distance between the viewer and the object is decreased, the refracting power of the

eyes must be increased to maintain a clear image on the retina, such as change in the shape

and position of the crystalline lens within the eye.

Accommodative function decreases rapidly with the age such that, by the mid-forties, most

individuals can no longer see clearly at normal near-working distances and may need optical

assistance.

Color vision

Our eyes can interpret color across of the visible spectrum. However, we are most sensitive to

light in the green-yellow (550 nm) portion of the spectrum. Daylight and sunlight provide

illumination across the entire color spectrum, but change in content over the course of the day.

Electric light sources vary widely in their spectral content, and should be carefully selected

for their color characteristic.

The perception of colors is a function not only of the eye’s sensitivity and the intrinsic color

of object, but also of the brain’s adaptation and the spectral content of the light.

The color perceived depends on the light source and on the colors nearby that may be reflect

the light.

Lighting and human performance

Introduction

The right artificial lighting design for a working/living/learning environment can not be

determined by illuminance levels alone. Humans are complex physiological beings who

respond to their environment. Without the optimum lighting environment we can not operate

CHAPTER 1

20

as efficiently as possible. Lighting conditions can affect our well-being and productivity.

While lighting designers and engineers are constantly urged to reduce energy and limit CO2

emissions, this would be counter-productive if it had a detrimental affect on the ergonomics

of the end user’s environment. [Brennan J., 2007]

It is important to realize that daylighting is not ‘only’ an energy-efficiency technology but

also an architectural discipline, and a major factor in occupants’ perception and acceptance of

workspaces in buildings.

The importance of daylighting for building design has fluctuated throughout history. It

started out as the primary light source and a significant architectural form giver before the

first part of the 20th century, and was then largely ignored in the post-war era as fluorescent

light, air conditioning, and cheap energy drove building design. Daylighting attracted

renewed interest as a result of the oil crises of the 1970s, and suffered again from declining

interest in the 1980s and 1990s as energy concerns lessened. Nowadays, the quest to light

buildings with daylight and sunlight is enjoying increasing interest from building owners and

architects alike. The source of this interest often lies beyond the energy-efficiency concerns

of the past decades. Instead, a ‘new’, emerging school of daylighting design has become

more occupant-centred, concerning itself with questions such as how can one design a

building that satisfies occupant needs for comfort and health, and, in a commercial setting,

positively influences the productivity of the organization hosted? Within that school of

thought energy savings remain important but the real challenge is to find design solutions

that simultaneously serve both goals.

Three major developments are contributing to this recent surge in interest:

- recent discoveries of the impact of light on human health;

- growing influence of ‘green building rating schemes’;

- progress on lower-cost, reliable, integrated control technologies to provide the

responsiveness needed for comfort and energy savings [Reinhart C. Selkowitz S., 2006].

Occupant preferences and satisfaction

There is a need for a comprehensive understanding of the occupants’ needs and preferences

in daylit spaces, even because recent developments in automated control systems and novel

materials and technologies require new investigative directions. These researches should be

based upon the foundation of notable work that exists in the scientific literature.

A review of research studies on daylighting covering the period from 1965 and 2004 reveals

the limitations of current knowledge about how people respond to daylight, and particularly

how they respond to automated photocontrolled lighting and shading controls [Galasiu A. D.,

Veitch J. A., 2006; Veitch J.A., Gifford R., 1996; Veitch J.A., Hine D.W., Gifford R., 1993].

Literature overview

21

Current knowledge may be succinctly expressed as follows.

Examinations about the preferred physical and luminous conditions in daylit office

environments [Galasius A., Veitch J., 2006]:

- there is a strong preference for daylight in workplaces, associated particularly with

the belief that daylight supports better health;

- when both daylight and electric light are used, people overestimate the contribution

of daylight to the overall illumination, and the degree of overestimation increases with

the distance from the windows;

- preferred window size probably varies for different settings, but in general larger

windows are preferred. Optimal window size for offices appears to be in the range

1.8–2.4 m in height and somewhat wider than taller, to provide a wide lateral view;

- when manually operated shading devices are available, people tend to set them and

then rarely to change them;

- preferred illuminance levels in offices with daylight are very variable from one

person to another. In addition, desired quantities of additional electric light vary with

the type of task and the distance from the window;

- discomfort glare from windows is less problematic than daylighting glare index

models would predict, although it is very variable from one person to another. The

degree of discomfort reported depends in part on the quality of the view outside the

window, as well as on the distance from the window and on the task.

Investigations about occupant satisfaction and acceptance in relation to the electric lighting

and window shading in daylit offices:

- photocontrolled lighting systems have best acceptance when there is individual

override control provided to users. Fully automated systems have low occupant

acceptance, and are sometimes too complex for facility managers to maintain;

- photocontrolled shading devices also need overriding occupant controls if they are to

be accepted;

- integrated controls for both lighting and shading can be acceptable, but are most

accepted when a degree of manual control is provided;

- control systems are more acceptable to both occupants and facility managers when

they are simple and easy to use.

In particular, concerning the lighting control systems, different research questionnaires

reveal that people did not use their light dimmers to save energy (despite working for an

organization concerned with environmental issues), but rather to accommodate for the tasks

CHAPTER 1

22

being performed.

In the building with photocell control, the majority of the subjects had not noticed any change

in light levels during the day. They did not even know that an automatic lighting system was

in place too.

Where controls were difficult to use, occupants chose lighting levels that reduced the need

for using the controls and increased the energy consumption. When the occupants perceived

that a particular set of environmental conditions was imposed upon them, the control systems

were deactivated.

Measurements carried out from Carter research [Carter D., Slater A., Moore T., 1999]

showed a high preference for daylighting and electric light levels below the current standards.

While the British Office Lighting Guide recommends 500 lux for general office lighting,

between 300–500 lux for computer workstations, and 750 lux for deep-and open-plan offices,

in over half of the shallow- plan offices the occupants chose average levels of task

illuminance below 300 lux, while the average illuminance in the deep-plan buildings was

below 750 lux.

The installations that had no user control accomplished better workplane illuminance and

luminance ratios according to current recommendations. However, users viewed the

installations that they could control more positively, even when the lighting conditions did

not meet current lighting practice guidance. This suggests that occupants preferred to have

the capability to choose their own lighting environment rather than having to accept lighting

levels chosen for them, even when these lighting levels were ‘‘better’’ according to

recommendations.

The blinds-closed/lights-on scenario was very common. Especially in the buildings with tall

windows, this was done intentionally by the occupants in order to keep the photocontrolled

lights on, because of complaints from people seated further from the windows and not in

direct control of the blinds, who often complained about glare from the upper sky.

Improving the energy-efficiency of commercial building lighting should include better use of

daylight, but it will requires the development of control systems that result in luminous

conditions that are suitable to occupants.

Such the summary above reveals, we do not yet know what control system features would be

most acceptable, nor what range of luminous conditions the system should permit.

Specific research gaps that should be investigated include the following:

- Study the relationship between discomfort glare reports, use of the window view,

Literature overview

23

satisfaction with overall luminous conditions, and chosen luminous conditions in relation to

outdoor conditions, to assess the trade-off between access to view, glare control, and lighting

control for energy efficiency. Automated controls based on maintained illuminance alone are

unlikely to achieve a balance between these considerations.

- Compare automated systems using behaviourally-derived algorithms, to semi-automated

systems (i.e., add manual over-ride controls) to determine the incremental benefit of allowing

individuals to modify the conditions. This work should begin in the laboratory, but should be

replicated in the field.

- Analyse the energy use resulting from the choices made by individuals using manual

controls, in contrast to automated or no controls. Both laboratory and field data may be used

for this purpose.

- Study the potential for conflict in open-plan spaces having manual or semi-automatic

lighting and shading controls, to identify system designs that minimize conflict while

maintaining energy savings and widely acceptable lighting conditions. Field studies, possibly

including interventions, are likely to be most effective in addressing this question.

Lighting mediating processes: activation and meaning

Psychologists know now that the effects of physical conditions on human behaviour are

mostly indirect, via complex cognitive processes.

The physical environment was irrelevant to their performance, rather that it influenced the

participants’ expectations and beliefs, which in turn changed their behaviour. Psychologists

call this type of chained effect mediation.

Today, we are only at the beginning to understand how the lit environment influences mental

states and processes that, in turn, determine work performance, satisfaction, and other

important outcomes. The motivation for much of this research, for lighting and for other

indoor conditions, is economic. Buildings cost less than employees, so any environmental

condition that decreases individual performance (either in quantity or quality), increases

absenteeism, or contributes to turnover, is more expensive to organisations than the capital

and operating costs of better indoor environments [Veitch J.A., 2006].

Productivity, the organisational outcome most sought after, is a complex concept defined in

different ways in various disciplines, but which is not synonymous with individual

performance. One common definition of productivity is efficiency; in this definition,

productivity is an index of output relative to inputs. Using the efficiency definition, a poor

indoor environment decreases organisational productivity both by reducing the value of

outputs and by increasing input costs [Pritchard, R.D., 1992].

The research literature concerning micro-level environmental conditions and productivity

CHAPTER 1

24

does not lend itself to direct calculations of the consequences for organisations [Rubin, A.,

1987]. However, by understanding the psychological and organisational processes that

influence employees in their working environments, we can develop advanced design

recommendations that support employee workspace needs.

The effects of lighting conditions on people should to be described via three mediating

processes: vision (see section 1.5.2), activation and meaning.

Light and activation: the circadian system

Circadian rhythms occur naturally in the human body over a 24-hour period and regulate

body temperature, alertness, secretion of hormones like melatonin (which regulates sleep

patterns) and cortisol (which regulates stress levels). Our circadian clock (biological clock) is

located in the suprachiasmatic nucleus (SCN) in the brain and is synchronised by light

transmitted through the eye to the circadian system.

Like the visual system the circadian system’s point of influence is through the eye. In 2003,

Peter Boyce wrote ‘The photo-receptor, or photo-receptors, used to influence the human

circadian system have not yet been identified until now [Boyce P., 2003]. Recent research has

shown that about 3% of the ganglion cells, mentioned earlier, contain a photo-pigment called

melaopsin [Berson D.M., Dunn, F.A. Motoharu Takao, 2002]. These nerve cells pass light

messages directly to the hypothalamus and from there to the pineal gland, inhibiting the

secretion of melatonin.

Light enters the eye, it is absorbed by the photo-receptors in the ganglion cells and converted

into electrical discharges which are transmitted through the retinahypothalamic tract (RHT)

to the SCN and then, by way of the paraventricular nucleus (PVN) and the superior cervical

ganglion, to the pineal gland. This route is called the Retinahypothalamic-pineal axis (RHP

axis).

The pineal gland is where melatonin is secreted during periods when insufficient light

reaches the eye – principally at night. Melatonin can be thought of as a messenger hormone.

Melatonin detectors have been found throughout the body and melatonin itself has the

purpose of transmitting messages from the SCN (the master clock) to these parts of the body

to synchronise their physiological functions to start at their appropriate time in the 24-hour

period. It is important that these hormones hit their targets on time to create healthy circadian

rhythms.

Humans need to receive the radiation during the day to suppress melatonin. However, for

humans working in artificial environments, such as offices, where access to direct daylight

may be limited, melatonin levels may not be suppressed sufficiently. This will affect the

rhythms of the circadian system.

Literature overview

25

The light influences for health and well-being, but it appears that workers might benefit from

having access to higher illuminances for at least some of the workday. This exposure might

be effectively provided by judicious use of daylighting within buildings, or by spending part

of the day outside. The current evidence does not suggest the need for an increase in general

illumination levels. Using daylighting to achieve the higher daily light exposure indoors

would be an energy-efficient solution (when coupled with appropriate electric lighting

controls), and in addition would probably be biologically effective because daylight is rich in

the spectral regions to which the novel photoreceptors are most sensitive (450–470 nm).

Light and meaning: positive affect

An important reason for understanding existing beliefs and preferences about lighting is to

create luminous conditions that match what the occupants want and expect. Obtaining these

lighting conditions is believed to lead to a pleasant motional state that psychologists call

positive affect. Positive affect theory states that environmental conditions that reate positive

affect lead to better performance, greater effort, less conflict, and greater willingness to help

others. Experiments in which positive affect was induced using fragrance have supported this

theory [Baron R.A., 1994].

Application of positive affect theory to lighting requires, first, that we know which luminous

conditions people prefer.

Individual lighting controls can address the problem of individual differences in lighting

preferences. When one does not know which conditions will create positive affect, individual

controls allow people to self-select their preferred conditions.

Regulations and visual ergonomics

Visual ergonomics

As it is defined by the International Ergonomics Associatin (IEA)

[http://www.humanics-es.com/def-erg.htm] "Ergonomics (or human factors) is the scientific

discipline concerned with the understanding of interactions among humans and other

elements of a system, and the profession that applies theory, principles, data and methods to

design in order to optimize human well-being and overall system performance."

Domains of specialization within the discipline of ergonomics are broadly the following

[Booth J., 1998]:

- physical ergonomics is concerned with human anatomical, anthropometric,

physiological and biomechanical characteristics as they relate to physical activity;

CHAPTER 1

26

- cognitive ergonomics is concerned with mental processes, such as perception,

memory, reasoning, and motor response, as they affect interactions among humans

and other elements of a system;

- organizational ergonomics is concerned with the optimization of sociotechnical

systems, including their organizational structures, policies and processes.

In particular “visual discomfort” and “visual ergonomics” have been related actually with the

use of video display terminals (VDT) for extended periods.

After a detailed bibliography research, realized in collaboration with the Ophthalmology

ward of S.Chiara hospital in Trento (Prof. De Concini), in the following sections a literature

overview of the ergonomics improvement in work station design has been presented.

Typical symptoms of vision challenges include the factors discussed above [De Concini M.,

Betta A., 1997].

- Eye strain refers to ocular fatigue, eye discomfort and headaches associated from intensive

use of the eyes. Common causes include:

glare on the computer screen

poor visual correction (out of date eyeglass prescription)

reading small character sizes on the screen

poor contrast between text and background on the monitor

noticeable screen flicker

dry eyes

- Blurred vision can be caused by normal physiological changes in the eye (i.e. aging or

disease). It can also be caused by constant focusing on objects within 12" of the eyes, which

often occurs when reading in low light.

- Dry and irritated eyes occur when there is insufficient fluid in the eyes to keep them moist.

Eyes are kept moist and refreshed by a normal blink reflex which is present from birth. Blink

rates vary with different activities and can become slower when concentrating. Eyes can

become red and itchy. Common causes include:

reduced blinking when using the computer

air movement that is noticeable in the face area.

International Ergonomic studies, conducted especially in Switzerland and Sweden, conclude

that frequency of eye discomforts are related to the amount of working time at the computer

display (significant increase after four hours).

Displays glare is a major contributor to visual discomfort. Reflections from the display

overhead lights, windows and even clothing can cause glare and wash-out effects which, "fog

Literature overview

27

over" the screen. This decreased character contrast so that users cannot read displayed

information clearly. The result is a marked drop in accuracy and loss of operation efficiency.

Ergonomic Directives: recommendations only for video display terminals

Several ergonomics directives have recently become law which may have potential impact

upon CBT because of their encompassing nature. The two most salient to this discussion will

be briefly summarized here.

EU 90/270/ECC

On December 31, 1996 the European guidelines for computer-monitor use (EU 90/270/ECC)

became law. As described by Zwingmann (1996) the directive contains new directions for

work safety that, among other aspects, place minimum safety standards for all employees –

private industry as well as government and public/civil service. In addition to ergonomic

requirements respective to HW, general office equipment and work environments, the

directive also addresses man-machine interface aspects – the key aspect of which orients

upon user-friendliness of SW.

EN ISO 9241-10

This directive (Ergonomic requirements for office work with visual display terminals –

Dialogue principles) replaces in 1998 the former directive DIN 66234-8 which, essentially,

covered the same major ergonomic issues, but has added two new general principles to them.

In general, the directive covers SW-ergonomic aspects such as task suitability,

self-descriptiveness, controllability, conformity with user expectations and error tolerance.

The newly added principles involve individualisability of SW systems and learning

suitability.

The Italian regulation D. Lgs. 19/09/1994 n. 626, “Attuazione delle direttive 89/391/CEE,

89/654/CEE, 89/655/CEE, 89/656/CEE, 90/269 /CEE, 90/270/CEE, 90/394/CEE,

90/679/CEE, 93/88/CEE, 97/42/CEE e 1999/38/CE riguardanti il miglioramento della

sicurezza e della salute dei lavoratori durante il lavoro”, includes specifics requirements for

working places at VDT.

For instance the art.51 define the VDT, the working place and the worker 1.

The art. 52 lists the employer’s requirement in order to guarantee an ergonomics working

place2.

The art. 55 indicates the sanitary check up requirements that the employer must guarantee,

especially for what concerns the vision ability3.

The annex VII contains specific requirements for the indoor environmental, especially for the

illumination condition and the discomfort glare protection 4.

CHAPTER 1

28

Nevertheless the indications included in the Italian regulation can result too much general

and with any numeric threshold has been indicated.

Fig.1.5_ 4: Eye-to-screen distance and vertical location

Some more detailed recommendations for monitor placement and lighting are based on the

latest scientific research [Jaschinski-Kruza W., 1990].

- Eye-to-screen distance: at least 25", preferably more.

- Vertical location: viewing area of the monitor between 15° and 50° below horizontal eye

level.

- Monitor tilt: top of the monitor slightly farther from the eyes than the bottom of the

monitor.

- Screen colours: dark letters on a light background.

One of the main reasons for computer-related eyestrain is the closeness of the monitor. It

seems easy to understand that, if having the monitor too close contributes to the problem, one

of the solutions is to place it farther away. When viewing close objects the eyes must both

accommodate and converge. The farther away the object of view, the less strain there is on

both accommodation and convergence. Reducing those stresses will reduce the likelihood of

eyestrain.

Early recommendations said that the monitor and document had to be at the same distance.

But to do that often means moving the monitor closer. Research by Jaschinski-Kruza (1990)

found that eyestrain was not increased when the monitor and document distance differed. In

fact, users preferred that the monitor be farther away.

Jaschinski-Kruza compared work performance with subjects working at viewing distances of

20" and 40". The task was to find mistakes in a database and he found better performance at

the 40" distance.

The old guidelines that recommended that the monitor be placed at eye level were based in

part on the belief that the resting position of the eyes (considered to be the most comfortable

>60 cm

-15°

Literature overview

29

gaze angle) is 15° below the horizontal. New evidence (and some that has been around for a

while) shows that, while the eyes might be most comfortable with a 15° gaze angle when

looking at distant objects, for close objects they prefer a much more downward gaze angle.

Figure shows the optimum position for the most important visual display, according to the

International Standards Organization (ISO 1998).

In an office of any size, the best solution for glare and reflections on the screen, as well as for

overall visual performance, is ceiling suspended, indirect lighting. This is sometimes referred

to as "uplighting." The underside of the lamps should be the same colour as the ceiling. Wall

mounted sconces may also be appropriate in certain instances. Because some tasks and

workers require more light than others, it is best to keep the overall light level low and allow

workers to supplement it with individually controlled task lights.

Aside from absolute brightness, a big problem with direct ceiling lights is that they provide a

high contrast with the rest of the ceiling. That contrast can reflect onto the screen. Many

guidelines mistakenly specify only a luminance (brightness) value for ceilings and walls.

While absolute intensity is important (a bright light reflecting off the screen will always

cause problems), reducing the contrast is much more critical. Interrupting the ceiling with

patches of bright light almost guarantees competing reflections on the screen.

Conclusions

Based on the literature overview, it is becoming increasingly important to establish a realistic

baseline of the actual lighting energy consumption in buildings for the different scenarios

used nowadays (both manually and automatically operated), which incorporates occupant

behaviour. At the same time it is strategic to explore and to quantify the benefits of typical

energy saving design measured (automated systems) compared with a traditional operation

system (manual system).

ADELINE has been used for the physically based computer simulation/visualization

techniques developed for the specific case study analyzed in this research project (Chapter 3).

The previous results about accuracy of validated illuminance calculation results for the

analyzed software tool have been assumed, as consolidated start point for our research.

For what concerns the stochastic and dynamic approach, we discussed the ADELINE

calculation and an effort to observe control-oriented behaviour in an educational building,

specifically the classrooms in the University of Engineering in Trento (Italy) has been

developed.

The efficiency of the lighting system as been evaluated not only in terms of energy saving

CHAPTER 1

30

but also in terms of visual comfort maintained and perceived by the users.

ADELINE (Advanced Daylighting & Electric Lighting Integrated New Environment)

Simulation based design aid tools that address lighting issues require a lighting analysis tool

capable of determining interior lighting levels and some measure of lighting comfort. These

building performance characteristics must be calculated, using actual weather data as input,

so that both spatial distribution within a building interior and temporal variations during an

individual day and throughout an entire year, can be assessed. The software tool used for this

calculation should be fast enough to allow rapid experimentation with various lighting design

parameters and should be also accurate enough to account variations in design parameters

such as room and building site configuration, interior and exterior surface reflectance,

complex fenestration system types, and commercial electric lighting system types.

Among the other software tools, in this research ADELINE has been used to create the model

of the classrooms analysed, because of the output program and the usability of this tool.

In the section 3.3.1 Adeline has been compared with other software more user friendly and

faster to use but that can include less input parameter, specifically Relux 2006 and Dialux

4.1.

Fig.1.5_ 5: Overview of ADELINE 3.0 Program System

Two main paths are offered by ADELINE; one provides 3D accurate and detailed displays of

various lighting scenarios in colour with complex graphic representations to analyse the

luminance distribution, glare sources and visual comfort using RADIANCE; the other path

calculates illumination levels and daylight factors on any plane using SUPERLITE. Both the

input and outputs of SUPERLITE can be analysed with corresponding plot programs

included in ADELINE. For both main paths, the lighting energy consumption can be

calculated using SUPERLINK or RADLINK, respectively.

SUPERLITE, developed in 1985, is really the classic daylighting program, using the flux

Literature overview

31

transfer method.

The advantage of SUPERLITE over other daylighting programs is that it can simulate several

daylighting strategies. However, due to the underlying radiosity method, all reflecting

surfaces are treated as diffusing. SUPERLITE calculates hourly values of illuminance or

daylight factors. SUPERLITE treats an uniform sky, a standard CIE overcast sky or a CIE

clear sky with or without sun.

Due to the radiosity calculation method used, the orientation of surfaces is very important.

Light can only be exchanged between surfaces which at least partly face each other. All

surfaces are subdivided into so-called "patches" or "nodes". During simulation, it is assumed

that there is a uniform brightness on the whole patch area. This makes it difficult for example

to calculate sharp shadows correctly. All surfaces are treated as perfectly diffuse, which

makes it impossible for example to correctly model daylighting systems with specular

components.

The main results of SUPERLITE are illuminance and daylight factor values calculated for all

nodes of the work surfaces defined. Work surfaces are imaginary surfaces representing the

areas of interest, e.g. the desktops.

RADIANCE calculates the luminance inside a space of any shape which can be furnished. A

visualization of the given space is presented. The surfaces can be specular, semispecular,

diffuse, refracting or translucent. General bidirectional reflection or transmission distribution

functions may be applied. The number of surfaces in a model is only restricted by the amount

of memory available.

RADIANCE uses the ray-tracing technique and creates a realistic 3-D image of a space.

The user then obtains a colour visual representation of the space and furniture with shading

and most important with calculated luminosity values.

RADIANCE uses the so-called "backward ray-tracing" method to simulate lighting

situations. In contrast to the radiosity method, subdivision of surfaces into patches is not

necessary.

RADIANCE does not require additional topological information like in SUPERLITE. Due to

the ray-tracing method it is possible to treat specular surfaces correctly. In contrast to many

other raytracing programs, RADIANCE can handle diffuse reflections on surfaces in an

adequate way.

SUPERLINK and RADLINK were developed to link daylighting programs, such as

SUPERLITE or RADIANCE, to an energy program so that the influence on the total energy

balance of replacing electric lighting by daylighting can be studied.

The main point of the link is the mixing of hourly standard sky conditions, calculated by

SUPERLITE, to simulate real weather conditions. SUPERLITE and RADIANCE are run for

CIE overcast skies, clear skies with and without sun within SUPERLINK and RADLINK,

respectively.

CHAPTER 1

32

These programs then use the hourly illuminance distribution on a work surface for every hour

of one day per month for the three standard conditions. The calculated hourly daylight

illuminance is then compared to the required design illuminance and the necessary artificial

lighting obtained for the CIE skies. The hourly sunshine probability, which is derived from

actual weather data, is used for weighting the standard conditions and obtaining the electric

lighting for real sky conditions.

Literature overview

33

Note: 626/94 citations

1 ART. 51 - Definizioni

1. Ai fini del presente titolo si intende per:

a) videoterminale: uno schermo alfanumerico o grafico a prescindere dal tipo di

procedimento di visualizzazione utilizzato;

b) posto di lavoro: l'insieme che comprende le attrezzature munite di videoterminale,

eventualmente con tastiera ovvero altro sistema di immissione dati, ovvero software per

l'interfaccia uomo-macchina, gli accessori opzionali, le apparecchiature connesse,

comprendenti l'unità a dischi, il telefono, il modem, la stampante, il supporto per i

documenti, la sedia, il piano di lavoro, nonché l'ambiente di lavoro immediatamente

circostante;

c) lavoratore: il lavoratore che utilizza un'attrezzatura munita di videoterminale in modo

sistematico e abituale, per almeno quattro ore consecutive giornaliere, dedotte le pause

di cui all'art. 54, per tutta la settimana lavorativa.

2ART. 52 - Obblighi del datore di lavoro

1. Il datore di lavoro, all'atto della valutazione del rischio di cui all'art. 4, comma 1, analizza i

posti di lavoro con particolare riguardo:

a) ai rischi per la vista e per gli occhi;

b) ai problemi legati alla postura e all'affaticamento fisico o mentale;

c) alle condizioni ergonomiche e di igiene ambientale

3ART. 55 - Sorveglianza sanitaria

4. Il lavoratore è sottoposto a controllo oftalmologico a sua richiesta, ogni qualvolta sospetti

di una sopravvenuta alterazione della funzione visiva, confermata dal medico competente.

5. La spesa relativa alla dotazione di dispositivi speciali di correzione in funzione dell'attività

svolta è a carico del datore di lavoro.

3 ALLEGATO VII

2. Ambiente

b) Illuminazione

L'illuminazione generale ovvero l'illuminazione specifica (lampade di lavoro) devono

garantire un'illuminazione sufficiente ed un contrasto appropriato tra lo schermo e l'ambiente,

tenuto conto delle caratteristiche del lavoro e delle esigenze visive dell'utilizzatore.

Fastidiosi abbagliamenti e riflessi sullo schermo o su altre attrezzature devono essere evitati

strutturando l'arredamento del locale e del posto di lavoro in funzione dell'ubicazione delle

CHAPTER 1

34

fonti di luce artificiale e delle loro caratteristiche tecniche

c) Riflessi e abbagliamenti

I posti di lavoro devono essere sistemati in modo che le fonti luminose quali le finestre e le

altre aperture, le pareti trasparenti o traslucide, nonché le attrezzature e le pareti di colore

chiaro non producano riflessi sullo schermo.

Le finestre devono essere munite di un opportuno dispositivo di copertura regolabile per

attenuare la luce diurna che illumina il posto di lavoro.

Chapter 2

Innovative design tools for lighting systems

2.1 Methodology and evaluation tools to design an energy efficient lighting system

Introduction

Nowadays two items have a very great importance in the realisation of a correct project, both

concerning the building in a general acceptation and in the choice of the specific typology for

the work and study areas.

The first one is related to the environmental sustainability and takes into account energy

demand control issues and the use of renewable resources in relation with the Kyoto protocol.

The second is the way in which the project solutions meet the user’s needs, in terms of

comfort and system usability.

A sustainable building process concerns the plan, the construction and the management of a

building through criteria that assure environmental balance, the sustainable management of

energy and material fluxes, the meditated use of natural resources, that should be renewable

and not noxious.

Comfort is the particular psychophysical condition that occurs when a person is satisfied with

the microclimate. Specifically we can speak about tactic, acoustic, visual comfort, thermal

comfort in the winter or summer season, respiratory and olfactory comfort and air quality. In

particular by visual comfort we mean the optimal level of luminosity that can assure the best

conditions for the human eye in order to improve the visual perception in relation with the on

going activities. The desirable luminosity level is therefore strictly connected to the different

activities in terms of intensity, quality and distribution.

The control of the inside luminance depends on a pondered regulation of the natural

luminance, on the artificial illumination and on the shading systems in order to improve the

visual perception avoiding glare phenomena. The use of automation systems can be a

solution to improve the performance of the lighting system both energy saving and user

comfort.

There are several simulation tools currently available for the lighting design with technical

CHAPTER 2

36

solutions. They aim to simulate and estimate the potential energy saving, but until now they

can not include correctly the use of automation systems. The research goal is to individuate

and to quantify the differences between the calculated model and the real behaviour of a

monitored building with automated lighting system in order to improve the actual available

design tools to introduce smart devices for lighting.

The developed methodology focuses on the following issues:

1- definition, programming and monitoring of different scenarios for the lighting control

system (section 2.1.1);

2- simulation, calculation and discussion about outputs of different software tools

-Adeline, Relux and Dialux- and comparison with data monitored for a specific case

study (section 2.1.2);

3- elaboration and statistical analysis method of the users behaviour, in order to

introduce improvements in the modelling process (section 2.1.3);

4- evaluation of the visual comfort maintained and perceived by the users, in relation

with the illumination levels standard, in order to verify if the proposed model allows

the visual comfort users requirements (section 3.3)

In the following section the factors related to the method to develop the first 3 aspects

will be examined in an eco-compatible perspective. The methodology has been

implemented for a specific case study and the comparison between data calculated and

real measurements has been developed for the specific condition of educational building,

as presented in Chapter 3.

Section 2.3 deals with the last point concerning visual comfort.

2.1.1 Design methodology

This research defines a new methodology for the design of lighting systems in lecture halls,

finalized to minimize the energetic consumption of the building.

In order to carry out this analysis it is necessary to compare environments homogeneous or

normalized by specific factor that simulate the same boundary conditions.

The method developed is divided into four phases:

1- Analytical phase: monitoring and analysis of existing situation

In order to analyze exhaustively the existing situation the criteria described in the following

section have been adopted.


37

- The verification of the photometric parameters (average luminance, luminance

uniformity, glare verification, chromatic variation), as defined in the UNI 10380/1994

in comparison with the new standard required in the UNI 12464/2004. Light and

illumination – Illumination of workplaces – Part 1: inside work places. In particular

this regulation, that substituted the previous one, redefines the average luminance (Em)

as the limit-value for the luminance on a work space at the maintenance time.

According to this new regulations, a direct calculation and check of the glare is not

necessary, but the value of UGR (Unified Glare Rating) calculated by a software is

considered sufficient. This value has been defined by CIE in 1995.

- The calculation of the Lighting Energy Numeric Indicator (LENI), in compliance

with the PrEN 15193/2006: Energy performance of buildings — Energy requirements

for lighting, in order to evaluate the total annual lighting energy required in the space

analysed and provides a numeric indicator for lighting energy requirements using the

existing lighting system.

- The realisation of a model for the analyzed space using specific software tools. The

photometric data calculated with the simulation tool will be compared with those

obtained by local measured in order to verify the environmental characteristic model.

2- Programmatic phase and performance control: definition of project objectives and

system requirements

The main goal of this research step is to define Scenarios configuration in order to control

and reduce the energy demand in the analysed space and to guarantee at the some time the

visual comfort improvement.

In order to do so, the following factors have been considered:

- control of the users presence in the space , as a necessary condition to turn on the light

- regulation of the artificial light, in relation to the natural light level

- partial switching off of the light system, if only a part of the space is occupied

- improvement of the luminaires efficiency

For the evaluation of these aspects, the monitoring of the following data is necessary:

- the dimming percentage for each light line in the room. In this way the evaluation of

the function modality for the system and for the partial energy consumption will be

more efficient.

- the users presence in the space, in order to make a scientific analysis of the energy

consumption, normalized through the real occupation time.

CHAPTER 2

38

- the verification of the user behaviour using the automation system, even if its

regulation has been forced by a manual regulation, in order to evaluate the visual

comfort for the users in a building automation system

- the inside illuminance with and without intelligent devices, in order to evaluate the

natural light use for the visual comfort.

3- Propositional phase: system solution

In this phase the automated system architecture, the devices to be installed and their position

and the supervision software program implementation has to be defined.

This entails the elaboration of an architectural/domotic design, that includes drawing boards

and technical reports for the adopted solutions, including both the activation typology

(function of each single actuator: on/off, dimming, and so on) and the activation modality

(definition of the interaction among the devices system) of the automatic control system, and

the definition of position and technical characteristics of the devices installed and integrated

in the automation system.

4- Evaluation phase: evaluation and verification

This phase described in detail in the sections 3.3.1 and 3.3.2, includes the verification of the

photometric parameter standards for the designed system using model software, in

compliance with the regulations in force. The model output data will be compared with the

values monitored in real life conditions. The evaluation of the user satisfaction for the new

control system will be examined also, as well the energy demand comparison between

systems with or without the intelligent devices.

2.1.2 Software tools output: evaluation methodology

There are numerous simulation tools currently available for lighting analysis (see section 1.4).

They aim to generate appropriate architectural and/or technical solutions to achieve an

enjoyable and productive built environment while simultaneously reducing the energy

consumption of buildings through the substitution of daylight for electric light. Simulation

based design aid tools that address lighting issues require a lighting analysis tool capable of

determining interior lighting levels and some measure of lighting comfort.

Among these software tools ADELINE has been used in this research activity, in order to

realize the model of the rooms analysed.

The goals of the data comparison presented focus on the following issues:

1- to calculate the difference between the inside illuminance conditions for the simulation


39

model and for the real class rooms;

2- to calculate the outside conditions for the model (in particular the sunshine probability)

and to verify the variance of statistical sky conditions;

3- to understand how much the recorded users behaviour differs in lecture halls, compared

with that one used in Adeline simulation tool

4- to analyze how the automated systems work in a real situation and how much energy

saving has been expected looking for the Adeline output.

Inside illuminance

The inside illuminance, calculated with Adeline (Superlite) for different time and sky

conditions, has been compared both with the real condition measured in the room by a

luxmeter placed in reference points on the working places and the inside illuminance

calculated for a typical day of the month considered using the data measured by the light

sensors installed into the rooms monitored.

To realize the model, Adeline uses the tool Superlite, which typically involves two steps: the

former creates an input file, the latter defines selected sky and location conditions.

The first step consists of creating an input file that contains a geometric description of the

subject building space, as well as the solar and luminaire data to be used for the simulation.

The input file can be created in different formats.

Once an input file has been created, the program can be run. It is possible to type the sky

definition, selecting among standard CIE sky condition (overcast sky, clear sky with sun,

uniform sky, clear sky without sun).

The solar and weather data input for the program can be supplied as geographical and

atmospheric data, that means that it is necessary to specify the latitude and longitude of the

building location, time and date of the simulation under given sky conditions. In this way the

simulation program provides for a series of simulations for given times of day and year. The

output can be displayed as contour plots of illuminance.

Sunshine probability

The design and performance of a daylight system depends strictly on the duration and

frequency of sunshine over the year at the location of the building. Adeline tool uses a

combination of detailed daylighting calculation programs for interior spaces and dynamic

energetic computing routines based on hourly weather data, in which the sunshine

probability (SSP) within a given time step is the central parameter. To obtain the SSP input

file it is possible to implement an Adeline calculation based on the data contained in the test

reference year file (TRY) for the specific location.

CHAPTER 2

40

Different data sources have been consulted to get this data format for Italian locations.

The IWEC (International Weather for Energy Calculations) are the result of ASHRAE

Research Project 1015 by Numerical Logics and Bodycote Materials Testing Canada for

ASHRAE Technical Committee 4.2 Weather Information. The IWEC data files are typical

weather files suitable for use with building energy simulation programs for 227 locations

outside the USA and Canada. All 227 locations in the IWEC data set are available for

download in EnergyPlus weather format. The files are derived from up to 18 years of

DATSAV3 hourly weather data originally archived at the U. S. National Climatic Data Center.

The weather data is supplemented by solar radiation estimated on an hourly basis from

earth-sun geometry and hourly weather elements, particularly cloud amount information.

The IWEC CD-ROM is available from ASHRAE. Only for some big cities IWEC data file

are available [ASHRAE, 2001] but these cities could not be representative of the whole

national territory.

Alghero (IGDG) Ancona-Falconara (IGDG) Ancona (IGDG) Aviano (IGDG) Bari-Palese Macchie (IGDG) Bergamo-Orio al Serio (IGDG) Bologna-Borgo Panigale (IGDG) Bolzano (IGDG) Bonifati (IGDG) Brescia-Ghedi (IGDG) Brindisi (IGDG) Brindisi (IWEC) Cagliari-Elmas (IGDG) Campobasso (IGDG) Capo Bellavista (IGDG) Capo Palinuro (IGDG) Catania-Fontanarossa (IGDG) Catania-Sigonella (IGDG) Cozzo Spadaro (IGDG) Crotone (IGDG) Gela (IGDG) Genova-Sestri (IGDG) Genova (IWEC) Gioia del Colle (IGDG) Grosseto (IGDG) Lecce (IGDG)

Enna (IGDG) Firenze-Peretola (IGDG) Foggia (IGDG) Marina di Ravenna (IGDG) Messina (IGDG) Messina (IWEC) Milan (IWEC) Milano-Linate (IGDG) Milano-Malpensa (IGDG) Monte Cimone (IGDG) Monte Terminillo (IGDG) Naples (IWEC) Napoli-Capodichino (IGDG) Novara-Cameri (IGDG) Olbia-Costa Smeralda (IGDG) Paganella (IGDG) Palermo-Boccadifalco (IGDG) Palermo-Punta Raisi (IGDG) Palermo (IWEC) Parma (IGDG) Patelleria (IGDG) Perugia (IGDG) Pescara (IGDG) Piacenza (IGDG) Pianosa (IGDG)

Pisa-S Giusto (IGDG) Rome (IWEC) S Maria di Leuca (IGDG) San Remo (IGDG) Pisa (IWEC) Ponza (IGDG) Potenza (IGDG) Pratica di Mare (IGDG) Rimini (IGDG) Roma-Ciampino (IGDG) Roma-Fiumicino (IGDG) Taranto (IGDG) Tarvisio (IGDG) Torino-Caselle (IGDG) Torino (IWEC) Trapani-Birgi (IGDG) Treviso-Istrana (IGDG) Treviso-S Angelo (IGDG) Trieste (IGDG) Udine-Campoformido (IGDG) Ustica (IGDG) Venezia-Tessera (IGDG) Venice (IWEC) Verona-Villafranca (IGDG) Vicenza (IGDG)

Fig. 2.1.2_ 1: Weather data available for Italian cities, in EnergyPlus weather format

The weather data contained in the IGDG data source include some more locations however it

is not possible to use these data because they are structured as hourly data for only one typical

day for each month [U.S. Department of Energy-Weather Data], instead 365 days of hourly

data have been required for the Adeline simulation.


41

Fig. 2.1.2_ 2: Statistics for ITA_Bolzano_IGDG, Location: Bolzano - ITA

Data Source: Custom-160200 WMO Station 160200 –

Using Design Conditions from "Climate Design Data 2005 ASHRAE Handbook"

Furthermore it is not possible to get a complete TRY data format for each monitoring site,

because only some of the parameter required could be recorded for a significant number of

years. In particular only the global radiation is usually available by the weather station

monitoring of the mainly locations, whereas diffuse and direct radiation have been not

recorded or the monitoring of these parameters start only from the last years.

Nevertheless, if the hourly sunshine seconds have been monitored day by day, it will be

possible to calculate the hourly SSP for the considered location, according to the method to

compute a test reference year exposed in IGDG Italian Climatic data Collection [Servizio

Metereologico del Trentino-Rete Agrometeorologica IASMA], implementing the method for

an hourly calculations to obtain 365 test reference days. The result is a vector that contains

the amount of the hourly sunshine seconds measured for each day, selecting from the total

CHAPTER 2

42

amount of the years sample the real days for which has been calculated the minimum

variance. In this way a virtual year could be calculated that contains the hourly sunshine

duration of real days, selected from different years of the sample.

Fig. 2.1.2_ 3: Insulation hourly data [s] for the weather station in S. Michele all

(TN)-Italy

Fig. 2.1.2_ 4: SSP calculation sheet

To understand the possible deviation between the weather data calculated for a probabilistic

year using measured data and the ones computed with a theoretic method, the two data series

have been statistically compared. The method used to get the theoretical radiation is the

Hulstro Model [Richard E. Bird, Roland L.Hulstro, 1981].


43

Fig. 2.1.2_ 5: Global radiation calculation sheet

In particular the calculation of solar position based on NOAA's functions and solar radiation

based on Bird and Hulstrom's model is founded on the following input data:

• latitude in decimal degrees (positive in northern hemisphere)

• longitude in decimal degrees (negative for western hemisphere)

• time zone in hours relative to GMT/UTC (PST= -8, MST= -7, CST= -6, EST= -5)

• daylight savings time (no= 0, yes= 1)

• start date to calculate solar position and radiation

• start time

• time step (hours)

• number of days to calculate solar position and radiation

• barometric pressure (mb, sea level = 1013)

• ozone thickness of atmosphere (cm, typical 0.05 to 0.4 cm)

• water vapor thickness of atmosphere (cm, typical 0.01 to 6.5 cm)

• aerosol optical depth at 500 nm (typical 0.02 to 0.5)

• aerosol optical depth at 380 nm (typical 0.1 to 0.5)

• forward scattering of incoming radiation (typical 0.85)

• surface albedo (typical 0.2 for land, 0.25 for vegetation, 0.9 for snow)

Implementing an excel calculation sheet with these input it is possible to calculate the

following data:

• date+time

• solar azimuth (deg)

• solar elevation (deg)

• Extra-terrestrial radiation normal to the beam, W/m2, corrected for earth-sun distance

variations and Julian day

CHAPTER 2

44

• Bird model direct radiation normal to the beam at the earth surface (W/m2)

• Bird model direct radiation incident upon a horizontal surface (W/m2)

• Bird model global radiation incident upon a horizontal surface (W/m2)

• Bird model diffuse radiation incident upon a horizontal surface (W/m2)

This calculation has be carried out with the supervision of the technical staff of Prof. Zardi.

Inputlatitude in decimal degrees (positive in northern hemisphere)longitude in decimal degrees (negative for western hemisphere)time zone in hours relative to GMT/UTC (PST= -8, MST= -7, CST= -6, EST= -5)daylight savings time (no= 0, yes= 1)start date to calculate solar position and radiationstart timetime step (hours)number of days to calculate solar position and radiationbarometric pressure (mb, sea level = 1013)ozone thickness of atmosphere (cm, typical 0.05 to 0.4 cm)water vapor thickness of atmosphere (cm, typical 0.01 to 6.5 cm)aerosol optical depth at 500 nm (typical 0.02 to 0.5)aerosol optical depth at 380 nm (typical 0.1 to 0.5)forward scattering of incoming radiation (typical 0.85)surface albedo (typical 0.2 for land, 0.25 for vegetation, 0.9 for snow)

Calculation of solar position based on NOAA's funct ions and solar radiation based on Bird and Hulstrom's model.

run

date+time

solar azimuth (deg)

solar elevation (deg)

Extra-terrestrial radiation normal to the beam, W/m2, corrected for earth-sun distance variations and Julian day

Bird model direct radiation normal to the beam at the earth surface (W/m2)

direct radiation incident upon a horizontal surface (W/m2)

Bird model global radiation incident upon a horizontal surface (W/m2)

Bird model diffuse radiation incident upon a horizontal surface (W/m2)

1/1/01 1.00 AM1/1/01 2.00 AM1/1/01 3.00 AM1/1/01 4.00 AM1/1/01 5.00 AM1/1/01 6.00 AM1/1/01 7.00 AM1/1/01 8.00 AM1/1/01 9.00 AM1/1/01 10.00 AM1/1/01 11.00 AM1/1/01 12.00 PM1/1/01 1.00 PM1/1/01 2.00 PM1/1/01 3.00 PM1/1/01 4.00 PM

Fig. 2.1.2_ 6: Calculation sheet for solar position and solar radiation:

input data sheet and calculations output.


45

Probability to turn on the light as function of the inside illuminance

The probability to turn on manually the light considered by Adeline tool to calculate the

energy saving for a manual use of the artificial light has been compared with the one

calculated for a typical day in the monitored classrooms.

Adeline tool uses the Manual On/Off Probability defined by Hunt that predicts the

probability for use of artificial lighting in a manually operated on/off-switching control

system [International Energy Agency, 2000]. The method is based on patterns of switching

behavior observed in field studies in England. Hunt found that the probability of someone

switching on the artificial lights in a space is correlated with the minimum daylight

illuminance on the working plane. From the data set of the field study, an empirical algorithm

was defined.

This algorithm has been compared with the probability to have the light on in the classrooms

monitored as a function of the inside illuminance measured. For this aim a typical day

calculation has been used, derived from the yearly data analysis.

Energy saving comparison

The energy saving calculated by Superlink for 2 light systems (Lightswith on/off and

Continuous Dimming) has been compared with the data monitored in the rooms where

automated light control systems have been installed.

2.1.3 Data analysis method

The goals of the complete data analysis, concerning a whole year of data monitoring, are:

- to calculate the different percentage of the energy saving potential for various applications

(scenarios) of building automation systems for lighting control;

- to understand the users behaviour in lecture halls with the traditional lighting system and

to quantify the amount of energy loss for an improper use of the artificial light;

- to deeply analyze how the automated systems work, in order to improve the scenarios

efficiency.

The comparison objects in this research are the performances, in terms of energy efficiency

and visual comfort maintained of:

- lighting system operating manually

- lighting system operating by automation system described by Scenarios.

The manual light system is assumed in this research as operated by switching on/off; the

CHAPTER 2

46

Scenarios are defined as follow:

Scenario 1 (SL1):

- occupancy detection and minimum illuminance level control, as primary condition to

switch on the lighting system (if Em> Ethreshold then switch off );

Scenario 2 (SL2):

- occupancy detection and minimum illuminance level control as primary condition to switch

on the lighting system (if Em> Ethreshold then switch off );

- dimming regulation of the artificial light as function of the natural light level detected in

discrete points, as second automation level;

Scenario 3 (SL3):

- occupancy detection by zones (the space has to be divided into discrete parts, each one

detected by a presence sensor) and minimum illuminance level control as primary condition

to switch on the lighting system (if Em> Ethreshold then switch off );

- dimming regulation of the artificial light in more rows of dimming channels in order to

approximate an ideal continuous dimming in the space, as second automation level;

- use of different and more efficient lighting sources.

Energy saving percentage

The energy saving percentage for each scenario described above has been calculated in

comparison with a reference level of energy demand considering the manual control of the

lighting system, hereafter named ML (maximum level).

As ML, it could be assumed that during a normal working day, the light could be turned on

for the whole working day (maximum 10 hours considering a working time from 8.00 until

18.00), hypothesizing in this way a continuous use of the artificial light (including lunch and

coffee break). This way to define the maximum energy demand for a standard user attitude

could differ from the real human behaviour.

The energy consumption of both automated systems and traditional one has been recorded, in

order to use these last data to calculate the ML energy consumption level to be compared with

the ones of different scenarios (SL1, SL2, SL3 levels).

The geometrical parameters checked in order to guarantee the same conditions in terms of

built space are:

• geometric base case (orientation, dimension, geographical position, etc.)

• parameters for the base case selected

• materials for walls, ceiling(s) and floor(s)

• openings (windows, roof lights) and their position

• outside obstructions for the windows defined


47

• luminairs

The energy consumption of the i-nth room has been recorded in Wh/m2. This data has been

divided by a defined operating time interval, so that it has been calculated in W/m2 (2.1.3_1).

In order to have a continuous profile of energy demand, a system with a significant detection

interval has to be used. For this issue an impulses counter has been interfaced to a static

active watt-hours meter.

A detection time interval has to be fixed. The discrete number of Wh cumulated in this time

interval has been recorded and displayed by the supervision system.

(2.1.3_1)

where:

eci,j = energy consumption of the i-nth room during the j-nth time interval, for the observed

operating period, in W/m2

j = j-nth time interval

si = the surface in m2 of the i-nth room

N = number of intervals in a day

The energy consumption for the ML during the j-nth interval, expressed in Wm-2, has been

calculated as follow:

(2.1.3_2)

where:

eci,j = energy consumption of the i-nth room operated by manual system during the j-nth time

interval, in W/m2

j = j-nth time interval

si = the surface in m2 of the i-nth room

N = number of the room operated by manual system considered

Nevertheless the comparison of the energy consumption amount is not sufficient in order to

make a correct calculation of the energy saving using intelligent devices, if it has not been

linked to three factors: the observed occupancy time, the outside illuminance level and the

visual comfort values.

∑ == N

i jiecNML1 ,

1

i

N

jj

ji

i s

t

ecN

ec

=∑ =1

,1

CHAPTER 2

48

Observed occupancy time

The occupancy time of the users in each room has to be recorded and monitored in order to

normalize the energy consumption by the necessity and the effective use of the lighting

system in a space.

The presence is recorded by a numerical variable defined as integer that assumes the value 1

if occupancy is detected or the value 0 if it is not (absence). Each data recorded is correlated

with the detection time. The time included between two value equal to 1 of the presence

variable is a presence interval (2.1.3_3), the time included between two value equal to 0 of

the presence variable is an absence interval (2.1.3_4)

The space is detected by the occupancy sensors after a defined time interval, implemented in

relation with the accuracy of the measure and the precision of the device.

These data have to be collected and registered in a databank by the supervision software,

calculating a minimum significant absence period Dta,min, defined in dependence on the space

size and the activity typology. This time interval could be smaller than the one implemented

in the presence sensor detection program.

The normalization factor for presence PFi (2.1.3_5) has to be calculated for each automated

room as the ratio between the arithmetic mean of observed presence (in sec) in each

automated room PA,i (2.1.3_6) and the arithmetic mean of observed presence in the

traditional rooms considered PT (in sec) (2.1.3_7).

(2.1.3_3)

if the presence variable value is 1

where

i = room detected index

j = time interval index

(2.1.3_4)

if the presence variable value is 0 and

while jiji pa ,, =

where

i = room detected index

j = time interval index

(2.1.3_5)

T

iAi P

PPF ,=

jjji ttp −= +1,

jjji tta −= +1,

min,aa tt ∆>∆


49

(2.1.3_6)

where:

pi,j = presence seconds in the i-nth automated room during the j-nth presence time interval

N = numbers of time intervals in which a day has been divided

(2.1.3_7)

where:

pi,j = presence seconds in the i-nth traditional room during the j-nth presence time interval


(2.1.3_8)

where:

N = numbers of room operated manually monitored

Outside illuminance level

Each room monitored has a different occupancy time related with the time table of the

workday. In order to compare the same outside conditions during the occupancy period it is

necessary to check and compare the outside illumination conditions.

For this reason a normalisation factor IFi (2.1.3_9) has been introduced. It is calculated as the

ratio between the arithmetic mean of outside illuminance during the work period in each

automated rooms IAi (2.1.3_10) and the arithmetic mean calculated for the traditional ones IT

(2.1.3_11).

The outside illumination value in lx is monitored and recorded by a numerical variable

defined as integer. Each data recorded is correlated with the detection time. The detection

interval is implemented in the weather device program in dependence on the technical

characteristic of the instrument and on the climatic conditions of the monitored site.

(2.1.3_9)

∑ =⋅= N

j jiiA pN

P1 ,,

1

∑ =⋅= N

j jiiT pN

P1 ,,

1

∑ =⋅= N

i iTT PN

P1 ,

1

T

iAi I

IIF ,=

CHAPTER 2

50

(2.1.3_10)

where:

i i,j = outside illuminance in the i-nth automated room during the j-nth time interval


(2.1.3_11)

where:

N = numbers of room operated manually monitored

Visual comfort values

The intelligent light system is regulated and dimmed by means of light detectors and

dimming actuators in order to guarantee a minimum illuminance level (in compliance with

the regulation UNI EN 12464-2/2004). Nevertheless in traditional rooms build before 2004,

the minimum illuminance level could be lower, in compliance with the preview regulation

UNI 10530/97.

In order to evaluate different levels of lighting system performances, concerning visual

comfort conditions as function of the illuminance level maintained on the task area, the

discomfort factor DFi has been introduced (2.1.3_12). This factor is calculated as weighted

mean between the illuminance level detected Ei,j and the minimum illuminance level Emin.

(2.1.3_12)

where:

Ei,j = inside illuminance in the i-nth traditional room during the j-nth time interval


Calculation of the energy saving for each scenario

Using the normalisation factors explained above the performance of each scenario is defined

in conformity with the boundary conditions of the other compared rooms (2.1.3_13), so that

it is possible to establish the energy saving percentage for different scenarios in comparison

with the ML.

−⋅−= ∑ =

N

j

jii E

EE

NDF

1min

,min11

∑ =⋅= N

j jiiA iN

I1 ,,

1

∑ =⋅= N

i iTT IN

I1 ,

1

iii

N

j jii DFIFPFecN

SL ⋅⋅⋅

⋅= ∑ =1 ,

1


51

(2.1.3_13)

With the pair comparison between the operation systems of each scenario, it is possible to

define the energy saving percentage for each automation system solution.

Typical day data analysis

In order to understand and model users' behaviour in traditional classrooms (ML) and to

quantify the energy saving potentialities of automation systems for lighting control, a typical

day data analysis has been carried out. In this way it is possible to perform an hourly data

analysis, instead of a daily overview of the system functions and user interactions.

To perform this statistical analysis, it is necessary to consider the conditions operating in the

monitored environment during a defined time interval over the course of the day.

The data monitored by the input devices of the automation system are to the bus not regularly,

but in an event-based manner (recorded by the supervision system). For this reason, the data

analyzed for each day analysed have been averaged considering a discrete temporal interval,

using a Lab view calculation programmed for the research data analysis. In this way the data

collection results are synchronized and it is possible to relate different factors occurring at the

same time in a specific room.

CHAPTER 2

52

2.2 Multiple criteria decision analysis: application for technological system choice

2.2.1. Problem definition and classification of the referred MCDM approach

Multiple Criteria Decision Making (MCDM) is a structured framework for evaluating

decision alternatives against multiple, and often conflicting, criteria. Its ability to handle

complex trade-offs in a variety of quantitative and qualitative units gives it potential in order

to choice automation system technology. Studies and graduation thesis on these topics have

been carried out at the CUnEdI (University Center for Intelligent Building - Centro

Universitario Edifici Intelligenti), as answer to specific and contingent requests of public

administration and private institute that are involved in lighting automation system design.

Many different criteria should be considered, in order to design a lighting automation system

in compliance with environmental sustainability principles (efficient use of electrical light

and control of the user’s visual comfort) and at the same time with feasibility principles (use

simplicity, economic convenience in installation and maintenance phase, esthetical

characteristics, ect.): technical characteristics, product effectiveness, installation techniques,

maintenance requirements, esthetical and economic factors.

For this purpose a specific design tool has been developed, based on multiple criteria

decision analysis. The decision maker (DM) for the developed case study is the lighting

designer/system integrator.

MCMD problems can be classified on the basis of three dichotomies [Sharifi A., Herwijnen

M., Toorn W., 2004]:

1. multi-attribute decision-making (MADM) versus multi-objective decision-making

(MODM);

2. individual decision maker problems versus group decision-maker problems;

3. decisions under certainty versus decisions under uncertainty.

Fig.2.2.1_ 1: Classification of multicriteria decision problems [Sharifi A., Herwijnen

M., Toorn W., 2004]


53

The specific approach developed in this research can be described as MADM, because it

includes a unique object: the definition of the best automation devices for the implementation

of an energy saving oriented lighting system and for the development of the system

functionality measurement, monitoring and analysis. The system components have been

defined in relation to a finite number of alternative solutions that are described by several

attributes. The specific decision problem is defined as individual decision making problem,

because it refers to a problem that has a single goal-preference structure.

Each attribute introduced (see section 2.2.2) is a characteristic of the object, which can be

evaluated objectively by the DM (lighting automation system designer) according to a

measurement scale. On the basis of these attributes, also called criteria, and the priorities of

each one, mostly referred as weights, the alternative options are evaluated and a complete

ordering of the alternatives (ranking of alternatives from best to worst) is generated.

The starting points for any MADM techniques are the generation of a discreet set of

alternatives, the formulation of a set of criteria, and the evaluation of the impact of each

alternative on every criterion. The estimated impact of each alternative on every criterion is

the criterion score. These scores are organized in effect/impact table. Most of the methods

build an aggregation of the evaluations associated with each attribute, so that a unique

preference can be derived on the whole set. According to the aggregation method and to the

way of elucidation of the preferences, three approaches are dominant: compensatory method,

outranking method and non-compensatory approach.

In the referred application a compensatory method, based on the aggregation of criteria to

form unique meta-criteria, has been applied.

This research is concerned with the stage of weighting the criteria, which is the major

judgmental component of MADM. The primary purpose of weighting the criteria is to

develop a set of cardinal or ordinal values which indicate the relative importance of each

criterion. These values will then be used by a ranking algorithm to determine the relative

value of each alternative.

Five weighting methods were analyzed. They are considered representative of the many

techniques that are available: fixed point scoring, graphical weighting, paired comparison,

rating ordinal raking [Hajkowicz S., McDonald G., Smith P., 2000].

Fixed Point Scoring

In this technique the decision maker has to distribute a fixed number of points amongst the

criteria. The higher is the point score that indicates the more important criterion. Often

percentages are used as this is a measure with which many decision makers are familiar. The

CHAPTER 2

54

key advantage of fixed point scoring is that it forces decision makers to make trade-offs in a

decision problem. Through fixed point scoring it is only possible to ascribe higher

importance to one criterion by lowering the importance of another. This presents a difficult

task to the decision maker which requires careful consideration of the relative importance of

each criterion.

Rating

The rating technique obtains a score from a decision maker to represent the importance of

each criterion. Often the numbers 1–5, 1–7 or 1–10 are used to indicate importance [Nijkamp

et al., 1990]. It is possible to alter the importance of one criterion without adjusting the weight

of another. This represents an important difference between the two approaches.

Graphical Weighting

There are many variations on the graphical weighting of criteria. The approach enabled

decision makers to express preferences in a purely visual manner.

Paired Comparisons

Paired comparisons involve the comparison of each criterion against every other criterion in

pairs. It can be effective because it forces the decision maker to give thorough consideration

to all elements of a decision problem. A popular form of paired comparison is the analytic

hierarchy process by Saaty [Saaty R.W., 1987]. It requires the decision maker to rate the

importance of each attribute in its pair on a nine-point scale, ranging from equal importance

(1) to absolutely more important (9). Once all the paired comparisons have been made, Saaty

proposes to calculate eigenvalues to represent weights. This method, in order to result

consistent, should be applied for a limited number of criteria.

Ordinal Ranking

Ordinal ranking requires the decision maker to rank the criteria in order of importance. This

method requires minimal information from the decision maker and is probably the easiest to

handle conceptually. A drawback associated with ordinal ranking is that it will significantly

limit the number of ranking methods that can be applied. For example, weighted summation,

one of the most commonly applied ranking algorithms, cannot be used when only ordinal

weights information is available. In order to use ordinal weights with cardinal ranking

methods it is necessary to estimate cardinal weights from the ordinal information.

The expected value method works in such a way that differences in quantitative weights for

objectives at the top of the ordinal ranking (i.e. those that are more important) are greater than

differences between those at the bottom of the ordinal ranking (i.e. those that are less


55

important). A detailed description of how the expected value method can be used to derive

cardinal weights is provided in [Nijkamp P., Rietveld P. & Voogd H., 1990].

number of criteria expected value of criterion weights J E(g1) E(g2) E(g3) E(g4) E(g5) E(g6) 2 0,25 0,75 / / / / 3 0,11 0,28 0,61 / / / 4 0,06 0,15 0,27 0,52 / / 5 0,04 0,09 0,16 0,26 0,46 / 6 0,03 0,06 0,10 0,16 0,26 0,41

Fig. 2.2.2_ 1: Results of the expected value method calculation

.

The ordinal ranking method has been applied in the specific case study analysis, because it

has been estimated to be the easiest to use and the most useful in terms of how much it helps

to clarify the decision problem.

A number of commercially available computer software that supports these methods are

available. The one used in this analysis is DEFINITE 2.01.

2.2.2. Classification and definition of the MADM elements

As reported in the previous section, a multi-criteria decision problem can be described by the

set of alternative courses of the action and by the objective that the problem solution should

achieve. Moreover the objective can be measured by a set of criteria.

The objective of the specific decision problem is the definition of the best automation

devices for the implementation of an energy saving oriented lighting system and for the

development of the system functionality measurement, monitoring and analysis.

The objective can be considered reached, if the technology individuated achieves also

feasibility criteria. Indeed in a realistic prospective it is necessary to evaluate not only the

performance of the system for the energy consumption, but also the efficiency of the system

in building and maintenance condition.

An objective must be seen as the direction to aim for, in striving to do better. How well an

objective is achieved can be measured with one or more criteria. A goal differs from an

1 DEFINITE has been developed at the Institute for Environmental Studies of the Vrije Universiteit of

Amsterdam. The program was commissioned and developed in co-operation with the Ministry of Finance of the

Netherlands. Without their support this program would not have been made.

The Evalue procedure was implemented in close co-operation with prof. dr. T. Stewart (University of Cape

Town, South Africa). Prof. dr. P. Rietveld (Vrije Universiteit, Amsterdam) made contributions to the procedures

for sensitivity analysis. Arcadis Heidemij Advies is kindly acknowledged for providing the material of the case

study on the Highway 73-south. The software is written by drs. D.A. Schweizer (iec ProGAMMA, Groningen)

CHAPTER 2

56

objective in that a goal is a target, or desired level of achievement for each attribute. In this

decision problem any minimum level has been defined.

The automation lighting systems typologies considered as homogenous categories are four:

manual regulation, light switching on/off by occupancy sensor, light switching on/off by

occupancy sensor and minimum illuminance level maintained, continuous dimming and

presence control. In the specific case study application, the lighting system typology selected

requires the dimming regulation of the artificial light, so that the alternatives for the specific

system selected are the following: light sensor, presence sensor, dimming actuator, and

switch. All the required devices have been analyzed in reference to each of the following

manufacturer’s products:

• Hager lumen: www.hager.de

• ABB: www.abb.it

• Siemens: www.siemens.it

• Merten: www.merten.de

• Theben: www.theben.de

• Zumtobel (Luxmate basic, emotion, professional): www.zumtobel.com

• Osram (Dim Pico-Mico, Daly Easy-Basic-Advanced-Multi): www.osram.com

• IGuzzini (Scene-Light-Colour Equalizer): www.iguzzini.com

The alternatives represent the possible solutions to the problem. It is important to notice that

for the specific case study each alternative is represented by the combination of the devices

components in order to design a complete automation lighting system. The effectiveness of

each single component has been calculated intrinsically, but the focus of the decision

problem is based on the evaluation of a global lighting system.

The definition of formal value structure has been developed, formally in form of tree of

objective, criteria and attributes. Such structures are referred in literature as criteria tree,

criteria structure, value tree or in case of Saaty analytical hierarchies. In criteria tree, the

higher-level objective may have some lower level objective assessed by one or more

sub-criteria. Finally indicators or attributes measure the criteria. In the following section the

value tree elements of the specific MADM have been presented.

The meta-criteria of the analyzed decision problem are the following:

1. Technical analysis

2. Installation analysis

3. Maintenance analysis

4. Effectiveness analysis

5. Formal analysis

6. Economical analysis

Each meta-criterion should fulfill the following requirements: completeness, homogeneity,


57

avoid double counting. In other terms they should be SMART, i.e. specific, measurable,

attributable, realistic, time bound. In order to specify and qualify each meta-criterion, the

correspondent criteria reported in Fig.2.2.2_2 have been definite and ordered in an

importance growing scale.

1 TECHNICAL ANALYSIS 1.1 functioning principle

1.2 visualization and operating outputs 1.3 wiring/transmission system 1.4 power supply 1.5 installation

2 DESIGN ANALYSIS 2.1 design

2.2 commissioning 2.3 supervision 2.4 diagnosis 2.5 technical support

3 MAINTENANCE ANALYSIS 3.1 glitch/electrical failure signal

3.2 minimum operation level guarantee 3.3 technical maintenance

4 EFFECTIVENES ANALYSIS 4.1 flexibility

4.2 interface possibility 4.3 installation modality

5 FORMAL ANALYSIS 5.1 color

5.2 shape/dimension

6 ECONOMICAL ANALYSIS 6.1 price

Fig. 2.2.2_ 2: Value tree for the specific decision problem

1. The technical analysis gives the indication about how well the alternatives achieve the

own specific function in the automation system (measure of a physical parameter for sensors

or operating actions for actuators), in relation with the technical characteristics of the

hardware/covering device components.

The meta-criterion n°1 has been assessed by the following criteria:

1.1 Functioning principle: with this criterion the measurement system (physical

apparatus for direct or indirect measure) and the accuracy of the measure (range of

possible measure and sensibility of the tool) are analyzed, considering the performances

of devices commercially available. In particular the measure could be defined:

1.1.1 very accurate, if it corresponds to the best solution available among the

alternatives analyzed

1.1.2 accurate, if it corresponds to a mean performance level among the

CHAPTER 2

58


1.1.3 not accurate, if it corresponds to a low performance level among the


The attributes and the critical points of this criterion have been specifically defined for

each devices typology.

Example: for lighting sensor three different measure ranges have been analyzed, as

reported in ascending order of accuracy:

1.1.1 range: 0-5000 lx

1.1.2 range: 0-1000 lx

1.1.3 range: 100 – 1000 lx

1.2 Visualization output: with this criterion the visualization availability /readability of

the data measured are analyzed. In particular the event/parameter output is defined (in

ascending order):

1.2.1 complete, if it is possible to read the measured value

1.2.2 partial, if it is possible to know only the range in which the measured

value is included

1.2.3 not available, if the measured value is detected by the device and used

for the programmed functions, but it is not possible to have the measured

value available.

Example: for lighting sensor the three different options have been distinguished as

follow:

1.1.1 measure of the illuminance level detected, in lux

1.1.2 definition of the illuminance level as function of a threshold level (set

from the system integrator) or as function of pre-set range

1.1.3 measure of the illuminance level not readable from the bus.

1.3 Wiring/transmission system: with this criterion the output transmission system is

evaluated. In particular the following possibilities are considered and classified in

importance order:

1.3.1 direct, if there is a direct connection bus-device (integrated device,

binary coupling unit and application module indivisible)

1.3.2 indirect, if there is a connection to the bus by the binary coupling unity

1.3.3 central unit, i.e. each telegram is directly managed by the c.u.

1.4 Device power supply system: with this criterion the effectiveness of the power

supply system is analyzed, in order to evaluate the device effectiveness.

1.4.1 high, if the power supply is the bus (29V) or the electrical system


59

(220V)

1.4.2 low, if the power supply derives from batteries or specific supply unit

1.5 Installation: with this criterion the possibility to install correctly the device has

been evaluated. In particular the installation has been defined:

1.5.1 standardized, if specific requirements (position, wiring, connection, etc.)

referred to standards should be satisfied in order to install the device. Example: DIN

rail mounted unit.

1.5.2 partially standardized, if specific requirements referred to the

dimensions and the location of the devices are recommended by the manufacturer but

not by a specific standard. Example: bus devices for flush mounting

1.5.3 not standardized, if specific auxiliary attachments should be used for the

devices. Example: bus devices for flush mounting

2. The design analysis gives the indication about how well the alternatives achieve the own

specific function in the automation system, in relation with the application programs and the

number of the communication objects available. In other terms the flexibility of the program

and commissioning phases has been evaluated for the specific issue.

The meta-criterion n° 2 has been assessed by the following criteria:

2.1 Project design: with this criterion the necessity to define the logical structure of

the lighting system and the interaction between the devices has been evaluated. In

particular the project and planning phase of the system could be defined:

2.1.1 indispensable, if the automation system allows the interaction of all the

devices connected to the bus and requires a specific planning of the functionality of

each element by mean of an open communication protocol (open standard). This

configuration permits a flexible use of the system potentialities and allows to choice

devices of different manufacturers, with the same communication protocol (for

example konnex standard). Devices of different systems (heating-cooling-lighting

system, etc.) can be logically connected. Moreover it is possible to use different

application programs for the specific design solution.

2.1.2 necessary, if the automation system allows the interaction of the devices

but in compliance with a proprietary protocol. It means that it the interaction of the

devices of different systems is not guarantee, if different protocols are used. The

flexibility and the complexity of the system are lower than in the previous case.

2.1.3 not necessary, if each device allows only one application program and

the interaction with a prefixed number and typology of devices.

CHAPTER 2

60

2.2 Commissioning: with this criterion the commissioning flexibility of the lighting

system has been evaluated. In particular the logical configuration of the system could

be defined:

2.2.1 indispensable, if a software tool is required for commissioning the bus

devices, for example the ETS (eib tool software)

2.2.2 necessary, if the initialization and the parameterization of the devices is

required, but not the application program configuration. It means that the

devices are already pre-programmed. A simplified software tool is used in

these systems, in order to get the commissioning of each single device but not

of the global system.

2.2.3 not necessary, if the device does not required the commissioning,

because it have only one operating modality available. A central unit might

regulate the interaction among the system elements or it might be the case of

the so called “stand alone solutions”.

2.3 Supervision system: the necessity to have a supervision system in order to control

the system process or in order to monitor the data transmitted and detected has been

evaluated with this criterion. In particular the presence of a supervision system could

be defined:

2.3.1 necessary, if it operates the data monitoring, but not the system process.

2.3.2 indispensable, if it operates both the system process and the data

monitoring; it is the case of the centralized systems with a specific central

unit.

2.3.3 not available, if the telegrams on the bus are not available/readable. It is

not possible to record and monitored the system functionalities.

2.4 Diagnostics: with this criterion the possibility to locate and define any foul has

been evaluated. If an installation does not function correctly, the errors should be

located as quickly as possible and rectified. To do so, it is important to describe any

problems that arise precisely, so that detailed and current system documentation is

essential to be able to detect these types of errors. In particular the diagnosis

functionality could be defined:

2.4.1 complete, if it is possible to operate the diagnosis checking the fault in the

system (location of the device and of the error typology) and moreover

operating directly in order to restore the correct functionality.

2.4.2 partial, if it possible to locate the error but it is not possible to operate

directly in order to re-establish the correct functionality

2.4.3 not available, if it is not possible to operate any diagnosis.


61

2.5 Technical documentation: with this criterion the content of the technical

documentation available has been evaluated. In particular it could be defined:

2.5.1 complete, if installation modality and commissioning configuration

available are offered.

2.5.2 partial, if only the installation requirements are available, but the

commissioning characteristics (especially number of communication objects,

content of the application program) are not expressed.

2.5.3 not available, if any technical content is available.

3. The maintenance analysis gives the indication about how well the alternatives achieve

the own specific function in the automation system, in relation with the maintenance of the

system. Possible causes for non-responsive bus devices within a line can include: wire

breakage on the bus line, polarity reversal of the bus line at a device, preprogrammed devices

installed on the wrong line, bus devices incorrectly configured and/or parameterized,

incorrectly programmed device, faulty device.


3.1 Failure signal: with this criterion the visibility of the failure signal has been

evaluated. In particular it could be defined:

3.1.1 adequate, if the failure can be visualized both by a local signal (led alarm)

and by supervision system/installation software tool

3.1.2 partially adequate, if a supervision system is required in order to visualize

the failure/installation software tool

3.1.3 not available

3.2 Minimum operation level: with this criterion the device capability to guarantee the

operation function in case of faulty device has been evaluated, considering the

accessibility of the system both for final users and for not specialized operators. In

particular the operating level could be defined:

3.2.1 total, if the a reset procedure is available in order to restart the device as

previously, operating manually on the device

3.2.2 partial, if the reset procedure is available but operating by software

(user integrator)

3.2.3 not available, if any reset procedure is available.

3.3 Technical maintenance: with this criterion the requirement of a technical

maintenance operated by technical-specialized staff (in compliance with specific

standard) has been evaluated. In this case only two options have been considered:

CHAPTER 2

62

3.3.1 Required

3.3.2 Not required

4. The effectiveness analysis gives the indication about how well the alternatives achieve

the conversion, modification and renovation of the system during the operation time.


4.1 Flexibility: with this criterion the flexibility of the system has been evaluated,

relating to the functional parts (commissioning and control functions) of the devices

that could be updated; for example new application programs, new communication

objects, etc. In particular the flexibility level could be defined:

4.1.1 good, if it is possible to update the device software, by downloading the

data sheet of the products.

4.1.2 partial, if it is possible to update the device functionalities only by mean

of new hardware elements.

4.1.3 not available, if it is not possible to update the device software (especially

in centralized systems).

4.2 System extension: with this criterion the possibility to extend the system with

additional devices has been evaluated. In this case, the flexibility of the whole system,

not only of the single devices, has been considered. In particular, the extension level

could be defined:

4.2.1 good: if it might be possible to add new devices to the bus system with

extra power supply.

4.2.2 partial: if the number of the devices installed in the system is fixed, i.e. the

maximum number of the system elements can not be raised unlimited

4.2.3 not available: if it is not possible to add new devices to the system after

the commissioning of the system.

4.3 Usury level: with this criterion the durability of the structural part of the system

have been evaluated, in relation with the usury of the physical components. In

particular the usury of the system could be defined:

4.3.1 low, if the device cannot be reached and damaged by the users and it

should be not operated manually (for example smoke detectors) and if it is not

submitted to environmental factors (for example DIN rail mounted devices).

4.3.2 partial, if the device can be reached and damage by the users and it should

not be operated manually. It is the case of device with a removable application

module, for example flush mounted room controller or multi-functions switch

controller.

4.3.3 high, if the device is submitted to indoor or outdoor environmental factors

(such example, respectively, smoke or gas detectors and weather station).


63

5. The formal analysis gives the indication about how well the alternatives achieve the

esthetical value of the device as function of the options available.


5.1 color: with this criterion the range of different color available has been evaluated.

In particular the durability of the system could be classified as follow, considering only

binary options:

5.1.1 different colors

5.1.2 only black/white

5.2 shape and dimension

5.2.1 different shape/design available

5.2.2 only one shape

6. The economical analysis gives the indication about how well the alternatives achieve a

sustainable cost in relation with the mean price of the device typology.

The meta-criterion n° 6 has been assessed by only one criterion: the percentage difference

between the price of the devices examined and the calculated lower price (reference price).

The following attributes define the economic convenience:

6.1 percentage difference in the price

6.1.1 low: if the price of the considered device is between 0% and 10% higher

than the reference price

6.1.2 middle: if the price of the considered device is between 10% and 25%

higher than the reference price

6.1.3 high: if the price of the considered device is more than 25% higher

2.2.3 Standard analysis sheet

Using the ordinal ranking weighting system analyzed in section 2.2.1 and the criteria tree

described in section 2.2.2, the evaluation scores reported in Fig. 2.2.3_1 has been defined.

A graphical sheet has been developed, that can be filled simply by the system integrator and

that calculates directly the score in tenth of the device tested.

The calculation output includes an evaluation table as well, that is not only useful to get an

overview of the problem but is also used to recover how the data are measured.

CHAPTER 2

64

Fig.2.2.3_ 1: Weights calculation of the decision problem


65

2.3 A model sheet to monitor and analyze visual comfort

2.3.1 Objects and effects

The goals of the subjective test experimented and tested are:

- to check, if present, the uncorrected refractive errors of the students, for different contrast

conditions;

- to define the critical detail for the student in a typical lesson condition, for defined boundary

conditions;

- to compare the veiling luminance level perceived and the literature discomfort value;

- to evaluate the visual discomfort indexes of the students;

- to analyse the visibility, as calculated and perceived value, for defined boundary conditions.

2.3.2 Definition of fields sheet

The sheet is divided into the following 5 SECTIONS.

1. Boundary condition definition

As first the boundary conditions of the evaluated situation have been checked and monitored,

as primary factors affecting the visual test results.

For this aim, it was necessary to use a luxmeter to measure the inside illumination level, a

weather station to evaluate the outside illuminance conditions, and a questionnaire to

describe the individual visual deficit.

In particular it has been defined:

- the optical systems deficit of the student sample, concerning the myopia, hyperopia

and astigmatism level (in dioptres);

- the characteristic environmental factors in the class room monitored, in particular

the lighting operating system (automated or manually controlled); the shading

position and the lighting operation state during the last lesson taken;

- the position of the student in the class room, to establish the task distance (position of

the work desk occupied during the lesson as consequential number) and the glare

source distance (position of the work desk from the windows as consequential

number);

CHAPTER 2

66

- the lesson condition, that means the lesson typology (if a projection has been done as

teaching tool or if just the blackboard has been used); the number of the lesson

already taken, to consider the tiredness of the students; the time of the lesson

(measuring the illuminance and luminance level, considering the weather conditions

too).

1 BOUNDARY CONDITIONS DEFINITION 1,1 UNCORRECTED REFRACTIVE ERRORS

Do you use glasses or contact lens?

Myopia 0 dioptres <6 dioptres >6 dioptres Hyperopia 0 dioptres <3 dioptres >3 dioptres Astigmatism 0 dioptres <3 dioptres >3 dioptres

Which is the age of the users? < 23 >23 years 24 25 26 or more 1,2 LECTURE ROOM: ENVIRONMENTAL CHARACTERISTICS

Which is the lesson room? automated class room 2M 2N 2P traditional class room 2A 2B 2C

Which is the lighting and shading use during the lesson?

shading position closed partially open openi lighting operation state off partially on on

Which is your distance from the blackboard?

work desk number from the blackboard 1 2 3 4 5 6 7 8 9

work desk number from the window 1 2 3 4 5 6 7 8 9

1,3 LESSON CONDITIONS

Which was the lesson condition?

lesson hour I II III hour IV-V-VI hour VIII-IX-X hour

number of the lesson already taken 1 2 3 4 5 6 7 8 9

Slide show time the whole lesson

a part of the lesson time no slide show

blackboard use time the whole lesson

a part of the lesson time

no blackboard use

Fig.2.3.2_ 1: Test section: boundary conditions definitions


67

2. Critical detail definition

The word acuity is often used to describe the visibility of fine details, involved in various

kinds of displays. Two different kinds of acuity are recognized: resolution and recognition

acuity.

In particular the recognition acuity represents the ability to correctly identify a visual target,

such as differentiating between a “G” and a “C”. Usually it is measured in terms of the

smallest size target, in angular measure, that can be discriminated. Visual acuity testing

performed using letters, such as done clinically, is a form of recognition acuity. Visual acuity,

like detection, varies with exposure duration and illuminance, but in the case of a class room

target definition on a blackboard there is not a critical value of exposure duration (that is

always more than one second).

In this treatment the definition of the critical detail is dependent from the ratio between the

surface of the target and the surface of the background, according with the following relation:

(2.3.2_1)

where:

d = size of critical detail

So = occupied surface, by target and space

Sr = real surface, only the target

Note that the critical ratio So/Sr represents the minimum value between So’/Sr’ and So’’/Sr’’,

as expressed in the Figure 2.3.2_2 The detail dimension is an important factor that influences

visual comfort, the attention and the learning performance during a lesson when slide show is

used, where it is possible to have a very small dimension of the data and text projected.

dH

L1

s

So'

Sr'

So''

Sr''

5.0

arg

⋅=

r

oett S

SdD

CHAPTER 2

68

Letter contour External measure

3/10 L 1 32 8,00 24,00 L 2 32 8,00 24,00 L 3 32 8,00 24,00 H 48 12,00 s 8 d 12,00

So' 1536,00 mmq

Sr' 960,00 mmq

So'' 288,00 mmq Detail 14,7 mm

Sr'' 192,00 mmq

Fig.2.3.2_ 2: Critical detail definition

It is necessary to define the critical detail in the classrooms monitored, considering the two

room sites: the 2M is 12 m long; the 2N and 2P are 6 m long.

For this reason, two different optometric boards have been prepared:

- one with the standard dimension to detect the visual acuity from a distance of 5m

- one with a double dimension to detect the visual acuity from a distance of 10m

These two measures represent the critical distance, from the last work desk of the classrooms

to the black board. It is important to note that the visual acuity has been related for the whole

student sample considering the student position in the classroom (see above “Boundary

condition definition”).

In order to relate the optometric measure in tenth with the detail size in mm, we have adapted

a standard optometric board including the some selected letter “E” in each detail category

detected.

In particular the following detail detections, with the correspondent value in mm, have been

included in the test (detail measures of the optometric board for a 10m observation distance):

- 3/10 : 14.7 mm

- 5/10 : 9.2 mm

- 7/10 : 7.3 mm

- 9/10 : 5.5 mm

The luminance contrast is defined as follow:

b

bd

L

LLC

−=


69

Where

Ld = luminance of the detail

Lb = luminance of the background

Two different contrast conditions have been examined:

- positive contrast where Ld = 98% (greater luminance) and Lb = 6% (lesser luminance)

- negative contrast where Lb = 98% (greater luminance) and Ld = 6% (lesser luminance)

V CA T H ZL U D E R

B R T N O P

E T L Y U S C

EN H M L I Q G

B Z A V U R E P C

G O D B E T N S A

R A H E P R T F U

OD T L H B P E Y S

T L E P S C Q G U V Y

TAVOLA OTTOMETRICA DECIMALEDISTANZA 5 METRI

1/10

2/10

3/10

4/10

6/10

5/10

7/10

8/10

9/10

10/10

11/10

V CA T H ZL U D E R

B R T N O P

E T L Y U S C

EN H M L I Q G

B Z A V U R E P C

G O D B E T N S A

R A H E P R T F U

OD T L H B P E Y S



1/10

2/10

3/10

4/10

6/10

5/10

7/10

8/10

9/10

10/10

11/10

V CA T H ZL U D E R

B R T N O P

E T L Y U S C

EN H M L I Q G

B Z A V U R E P C

G O D B E T N S A

R A H E P R T F U

OD T L H B P E Y S



1/10

2/10

3/10

4/10

6/10

5/10

7/10

8/10

9/10

10/10

11/10

V CA T H ZL U D E R

B R T N O P

E T L Y U S C

EN H M L I Q G

B Z A V U R E P C

G O D B E T N S A

R A H E P R T F U

OD T L H B P E Y S



1/10

2/10

3/10

4/10

6/10

5/10

7/10

8/10

9/10

10/10

11/10

Fig.2.3.2_ 3: Optometric board definition

Fig.2.3.2_ 4: Luxometer for the local parameter measurement:

CHAPTER 2

70

2 VISUAL ACUITY DETERMINATION: critical detail definition

2,1 READING PROVE positive contrast = 0,3

write the correspondent line

1° visual acuity detection (detail 3/10)

it is possible to read easily

it is possible to read with difficulty

it is not possible to read








it is possible to read easy








2,2 READING PROVE WITH DIFFERENT CONTRAST CONDITIONS negative contrast = 0,3





















Fig.2.3.2_ 5: Test section: Visual acuity determination


71

3-4. Visual discomfort evaluation: veiling luminance and eyes sickness indicators

Glare can occurs in 2 ways: when there is too much light or when the range of luminance in a

visual environment is too large.

Glare of this sort can have 2 effects: a reduction of visual performance and a feeling of

discomfort. Glare that reduces visual performance is called disability glare and is due to light

scattered in eye, reducing the luminance contrast of the retinal image. The effect of scattered

light on the luminance contrast of the target can be mimicked by adding a uniform “veil” of

luminance to the target. The magnitude of display glare can be estimated by calculating this

equivalent veiling luminance [IESNA, 2000].

(2.3.2_2)

where

Lv = equivalent veiling luminance in cd/m2

Ei = illuminance from the ith glare source at the eye in lux

Ji = angle between the target and the ith glare source in degree

For different illuminance conditions the veiling luminance has been calculated, using the

equation above. The illuminance level has been measured in different significant points of

the classrooms, describing the mean student’s condition and as well the more critical

position.

With this data a model of the room has been adapted to the measured conditions and the

contribution of each single glare source for different position has been calculated.

In the experimental test the glare perception has been evaluated. In particular the

environmental conditions and the consequent glare feeling changing the shatter and blend

position has been considered.

The effect of disability glare on the luminance contrast of the perceived target can be defined

by adding the equivalent veiling luminance in the formula (2.3.2_2).

The other form of glare is called discomfort glare, defined as a sensation of annoyance or

pain caused by high luminance in the field of view (see the next section “ Visibility

evaluation”).

With the supervision of Professor De Concini, top clinician of the Ophthalmology ward of

S.Chiara hospital in Trento, the visual discomfort perception has been considered in the

questionnaire, including some significant symptoms of eyestrain discomfort.

( )∑ +=

n

i ii

ive

EL

5.12.9

ϑϑ

CHAPTER 2

72

3 DISABILITY GLARE EVALUATION

3,1 DISABILITY GLARE PERCEIVED

Do you perceive disability glare reading on the blackboard? no partially yes

Do you read difficulty in some part of the blackboard?

on the right side centrally on the links site

Do you perceive disability glare during the projection? no partially yes

3,2 DISABILITY GLARE CALCULATED

windows luminance (by Relux model, for a specific daily calculation) Lv<0,25 Lv>0,25

windows luminance (by Relux model, for local parameter measurement) Lv<0,25 Lv>0,25

3,3 BOUNDARY CONDITION CHANGE

shatter completely open

discomfort glare perception

Partially discomfort glare perception

no discomfort glare perception

shatter open for 3/4




shatter half open




4 VISUAL DISCOMFORT EVALUATION

Do you feel visual discomfort?

1 burning eyes no partially yes 2 periocular pain no partially yes 3 headache no partially yes

4 eyes itch no partially yes 5 lachrymation no partially yes

6

blurred vision (close vision) no partially yes

7 blurred vision (far vision) no partially yes

Fig.2.3.2_ 6: Disability glare evaluation

( )∑ +=

n

i ii

ive

EL

5.12.9

ϑϑ


73

5. Visibility evaluation

Visual acuity and threshold contrast are two different aspects to define the visibility of a

target. Visual acuity sets the minimum size for a target to be seen and threshold contrast sets

the minimum luminance contrast that is required for a target of a given size to be seen. The

contrast sensitivity function combines these two measures by showing the minimum contrast

required for a target of different size to be seen. Specifically, the contrast sensitivity function

is a plot of contrast sensitivity against spatial frequency [IESNA, 2000].

The visibility levels for different illuminance conditions have been calculated by the Adrian

Model [Adrian W., 1989], elaborating a calculation table (Fig. 2.3.2_7). This method has been developed to compute the luminance difference threshold thresholdL∆ of visual target for

diverse detail size and for positive and negative contrast (2.3.2_3). Another parameter

evaluated is the observation time in practical viewing conditions. The effect of age on the

threshold contrast and that of disability glare has been incorporated. The following equations

(2.3.2_4-2.3.2_12) contain the numerical description to determinate factors influencing the

visibility level VL of object in the visual field (2.3.2_13). These factors are functions of the

background luminance (Lb) and of the target size (α ).

( ) ( ) AFLaLFLL bbCPthreshold ⋅⋅⋅

+Φ=∆ ,,6.2

2

2/12/1

ααα (2.3.2_3)

Where:

• 2

2/12/1

6.2

+Φ

Lα

(2.3.2_4)

is the threshold luminance difference for positive contrast, observed average age 23 years

and a 2 sec or unlimited observation time;

• for Lb > 0.6 cd/m2

(2.3.2_5)

(2.3.2_6)

• for 0.0048 cd/m2 <Lb < 0.6 cd/m2

(2.3.2_7)

(2.3.2_8)

(2.3.2_9)

• is the contrast polarity factor (2.3.2_10)

( ) 5867,01556,02/1 1684,01925,4log bb LL ⋅+⋅=φ

( )22/1 log0866,0log3372,0072,0log bb LL +⋅+−=φ

bLL ⋅+−= 319,0256,1log 2/1

466,02/1 05946,0 bLL ⋅=

( ))2(4.2

1,=

−

∆−=

tposbcp L

mLF

βαα

466,02/1 05946,0 bLL ⋅=

CHAPTER 2

74

• is is the exposure time influence (2.3.2_11)

• ( )

99.02160

19 2

+−= ageAF is the influence of the age (2.3.2_12)

In particular to obtain the difference between DL for positive and negative contrast, the factor

FCP was derived from Aulhorn’s data [7]. The luminance difference threshold DLneg for a

target in negative contrast can be computed with the term:

(2.3.2_13)

The contrast polarity factor depends on target sizea and background luminance. The Fig.

2.3.2_7 displays the relationship between positive and negative target contrast. In the negative

contrast the threshold of a target of a defined size is always lower than in positive contrast at

the same background luminance. This explains why darker target appear to be better

perceived than brighter target at the same luminance difference. The curves indicate that Fcp

is always<1, which yields smaller DL thresholds for negative contrast reflecting the

phenomenon that a negative contrast can be seen better than the positive for the same DL.

Prameter:Background luminance [cd/m2]

0,20

0,30

0,40

0,50

0,60

0,70

0,80

0,90

1,00

1,10

1,0 1,6 2,5 4,0 6,3 10,0 16,0 25,0 40,0 63,0 100,0

Target size [min]

Po

lari

ty f

acto

r

0,0318 0,3180 3,1800 31,8000

Fig.2.3.2_ 7: Polarity factor versus the target size for different luminance level

( ) ( ) ( )[ ]1.2

,2/122

FF

LaaLa

+= αα

cpposneg FLL ∆=∆


75

Moreover it is important to note that, as the study of Ahumada demonstrates [Ahumada A.,

Scharff L., 2003], latency and accuracy are better for negative contrast conditions at 20% and

40% contrast, but were not different at 10%.

Fig.2.3.2_ 8: Left: Letter identification accuracy vs. contrast. Right: Letter identification

latency vs. contrast [Ahumada A., Scharff L., 2003].

Fig.2.3.2_ 9: Test of different contrast condition by the Ophtalmology ward of S.Chiara

For contrast levels lower than the 10% a positive contrast could be better, as proved using the

photometric control system available in the Ophtalmology ward of S.Chiara. This is the

reason for the positive contrast adopted for VDT designed for persons with visual disability

[Ahumada A., Scharff L., 2003].

The visibility levels calculated for different illuminance conditions have been compared with

the detail definition perceived by the observers, as it is reported in the last section of the test.

The calculation sheet presented in Fig. 2.3.2_10 has been realized in collaboration with Eng.

I. Zancarli, of the STAIN Engineering s.r.l, based on [Adrian, 1993].

CHAPTER 2

76

It is important to note that in the yellow column the DL considering the effect of the disability

glare has been calculated, in the grey column just the background illuminance has been

considered.

DL indicates at least a value at which a target of defined size becomes perceptible with near

100% probability under the observation conditions as used in the laboratory experiments.

Under practical observation conditions, however, a multiple of DL is needed depending on

the visual task demand. In most cases the luminance difference has to reach a level that

allows for form perception or to render conspicuity to the target.

For this reason a descriptive term has been introduced, the visibility level VL:

(2.3.2_14)

where:

DLeff = C*Lb

2.3.3 Comparison between existing analysis sheets

There have been a number of studies that have examined the acceptability of specific aspect

of lighting conditions, especially for what concerns:

1. the acceptability of different illuminance levels

2. the vision effects changing the spatial distribution of the light

3. the calculation of the VCP (visual comfort probability) for artificial light sources

4. the effect on discomfort and performance of VDT user

Points 1 and 2 have been analysed above all for one specific context: the working place in the

office. The results examination [section 3.3.3] suggests that the illuminance ranges are wide

and that there is no sharp preference for a specific illuminance.

Any specific study has been developed to describe the lighting preference in educational

building with the some specific aim.

For what concerns point 3, the calculation of the VCP is used in North America as empirical

prediction system for artificial lighting sources use acceptability. It is based on assessment of

discomfort glare for different sizes, luminance and numbers of glare sources, their location in

field of view and the back ground luminance against which they are seen, for conditions

likely to occur in interior lighting [Wolska A., 2003].

The VCP system evaluates lighting systems in terms of the percentage of the observer

population that will accept the light system and its environment as being comfortable, using

threshold

eff

L

LVL

∆∆

=


77

the perception of glare due direct light from luminaires to the observer as a criterion.

Point 4 includes only analysis related to the evaluation of visual discomfort and ergonomic

aspects for VDT users, as diffusely examined in section 1.5 [Ahumada A., Scharff L., 2003].

To structure the test for the visual comfort condition analysis in the classrooms monitored, all

the aspects mentioned above have been considered.

CHAPTER 2

78

Visibility calculation

Input Data Calculated Data

Description Symbol Value m.u. Symbol Value m.u. Symbol Value m.u.

Illuminance E 200,0 lux 200,0 lux 200,0 luxReflection factor ρρρρ Nero 3 % Bianco 94 %

Luminance L 1,91 cd/m2 59,84 cd/m2Veiling luminance L ve 0,8 cd/m2 0,0067 cd/m2 0,2094 cd/m2

Detail d 24,0 mmDistance D 10,0 m

Detail αααα 8,25 minEsposure time t (esp.) 2,0 sec 2

k factor k 2,6age age 23 years 23 years

∆∆∆∆L eff/E0 0,91C=∆∆∆∆L eff/L b 0,30

k 9,20A f1 0,997Af 1,000

Visual luminance L v 2,73 cd/m2Log(Lb+6) 6,4357

a(Lb) 0,1667Log(αααα+0.523) 1,4395

a(αααα) 0,1981(a(αααα, Lb)+t)/t 1,0000 1,0616

φφφφ1/21/21/21/2 0,9124 0,9936L1/2 0,0804 0,0949

∆∆∆∆L pos (t>=2) 0,0948 0,1206cd/m2

∆∆∆∆L pos (t,Af) 0,0948 0,1206cd/m2VL pos (t,Af) 6,11 4,81

Ccr 0,0496 0,0631TI 21,4 %m 0,2574 0,3005ββββ 0,5449 0,5168

Fcp 0,6418 0,6511

∆∆∆∆L neg (t) 0,0609 0,0785cd/m2

∆∆∆∆L neg (t,AF) 0,0609 0,0785cd/m2VL neg (t,Af) 2,58 9,52 7,38

Ccr 0,0319 0,0411TI 22,5 %

Negative Contrast

Age

Esposure time

Positive Contrast

Scene Positive NegativeContrast

Background TargetScene

CALCULATION

( )fcp

bthreshold AF

t

tLaLkL

+

+=∆

,2

2/12/1 α

αϕ

threshold

eff

L

LVL

∆∆

=d

bd

L

LLC

−=

Fig.2.3.2_ 10: Visibility calculation

.


79

5 VISIBILITY EVALUATION: detail dimension for defined contrast

Adrian W. Visibility of targets: model for calculation. Lighting Res. Technol. 1989; 21: 181-188.

visibility calculation for the luminance condition during the lesson DL<0,7 DL>0,7

visibility calculation for the luminance condition simulated with Relux DL<0,7 DL>0,7

5,1 BOUNDARY CONDITIONS CHANGE (WITH POSITIVE CONTRAST)

Detail definition with illuminance < 200 lx good weak bad

DL calculated

Detail definition with illuminance included between 200 lx and 500 lx good weak bad

DL calculated

Detail definition with illuminance > 500 lx good weak bad

DL calculated 5,2 BOUNDARY CONDITIONS CHANGE (WITH NEGATIVE CONTRAST)

Detail definition with illuminance < 200 lx good weak bad

DL calculated

Detail definition with illuminance included between 200 lx and 500 lx good weak bad

DL calculated

Detail definition with illuminance > 500 lx good weak bad

DL calculated

Fig.2.3.2_ 11: Visibility evaluation

Chapter 3

Case study

3.1 Analytical and programmatic phase

3.1.1 Existing situation: description of the built environment and of the lighting

system

The cross layout of the Faculty of Engineering of Trento is characterized by two parallel

wings, both with the same south exposure [Fig.3.1.1_1]. The lecture halls are symmetrical

and have the same shape. This configuration allows the simultaneous comparison between

the existing situation in the east wing (with the traditional electric system) and the new one

with automation system projected and realized in the west wing.

Fig.3.1.1_ 1: Faculty plan, second floor

CHAPTER 3

82

In particular the following lecture halls will be examined (in light colour the new designed

lecture rooms, in dark colour the lecture rooms that will be compared).

- 2P: 32 seats, length 7m, width 6.3m, 3 windows each 1.4m length – compared with

the symmetrical 2C

- 2N: 32 seats, length 6.5m, width 6.3m, 3 windows each 1.4m length - compared with

the symmetrical 2B

- 2M: 77 seats, length 12.8m, width 6.3m, 3 windows each 1.4m length - compared

with the symmetrical 2C

The actual distribution of the switching on system is in alternate line, both for the lecture

rooms and for the corridor [Fig. 3.1.2._2]. The switch is on/off.

Fig.3.1.1_ 2: Light distribution and switching operation system

3.1.2 Photometric parameters survey and software modelling of the lecture rooms

After a first measure campaign, carried out following the regulations described in Chapter

1.2, two conclusion can be drawn:

- it is necessary to evaluate the natural light contribution in relation to the lecture room

width (more than 6m);

- it is important to verify the real performance of the existing lights subjected to the

ordinary maintenance; that consists in the substitution of the fluorescent tubes

without a control of the lights that are opened and so subjected to get dusty.

Case study

83

These data, compared with the ones obtained with a simulation model using the software

Relux 2006 show that:

- the natural light contribution is reduced to a fourth from the study desks close to the

windows to the most interior one [Fig. 3.1.2_1];

- the actual light efficiency is the same of a low maintenance light system with a very

low maintenance factor (the lights have to be changed)

Fig3.1.2_ 1: Luminance Isolines for the natural light. Simulation date: 21/06/2006 10.30

o’clock

Fig3.1.2_ 2: Logitudinal and cross section of the luminance distribution. Simulation date:

21/06/2006 10.30 o’clock

Through qualitative verifications in situ we have discovered that students are used to occupy

the lecture rooms for their individual study. Therefore usually small groups occupy a large

room turning on the complete lighting.

Moreover lights are often forgotten switched on, both in the lecture halls and in the corridor.

CHAPTER 3

84

Fig3.1.2_ 3: Measured illuminance level on 16/02/2006

3.1.3 Design goals definition

The goal of the research, applied in the specific case study analyzed, is the control and the

reduction of energy demand in the lecture halls lighting and the visual comfort improvement.

In order to do that, the following factors have been defined:

1. control of the students presence in the lecture room, as a necessary condition to turn

on the light

2. regulation of the constant light, in relation to the window distance

3. a partial switching off of the lecture room, if only half space is occupied

4. introduction of a new kind of high efficiency light

The Scenarios used and monitored in the research case study are defined as follow:

- Scenario 1: use of occupancy sensors and minimum illuminance level control (if Em> 500 lx

then switch off). That means that if nobody is in the class room, the light will be turned off

automatically after 15 minutes.

Case study

85

This scenario has been used between September and December in the class 2P.

- Scenario 2: scenario 1 and dimming regulation of the artificial light in 2 rows of dimming

channels. That means that the light system is regulated and dimmed through 2 light detectors

and 2 dimming channel in order to have a minimum illuminance (500 lx, in compliance with

the regulation UNI EN 12464-2/2004).

This scenario has been applied in the new classroom 2N for the whole year.

- Scenario 3: scenario 1, dimming regulation of the artificial light in 3 rows of dimming

channels, new fluorescent tubes (T5 instead of T8), switching on divided by zones (the

lecture room has been divided into two parts, each one detected by a presence sensor).

The light system is regulated and dimmed through 3 light detectors and 3 dimming channel in

order to have a minimum illuminance of 500 lx.

This scenario has been applied in the new classroom 2M during the whole year.

- Scenario 1 bis: scenario 1 and dimming regulation of the artificial light in 3 rows of

dimming channels. That means that the light system is regulated and dimmed through 3 light

detectors and 3 dimming channel in order to have a minimum illuminance (500 lx, in

compliance with the regulation UNI EN 12464-2/2004).

This scenario has been applied in the new classroom 2P between March and June.

It is necessary to specify that in the traditional class rooms the minimum illuminance level

should be only 300 lx, in compliance with the old regulation UNI 10530/97.

In the second semester (Spring semester from March to June) the illuminance level required

has been changed in 300 lx instead of the 500 lx of the first semester. It is possible in this way

to verify the effective energy saving percentage for the same performance of the light system.

The Fig.3.1.3_1 reports the main characteristics of the electric system for each room. The

dark grey colour of the cell indicates an additional automation level that characterizes a

specific classroom. In order to separate the energy saving amount of each specific

automation level the following pair comparisons have been done:

2P’ + ML => presence sensor (scenario 1)

ML + 2P => dimming 1+1+1 (scenario 2)

2N + 2P => dimming 1+2 (scenario 3)

2P + 2M => room division (scenario 4)

CHAPTER 3

86

Fig.3.1.3_ 1: Lighting automation system definition for each classroom

The supervision system has been programmed for the monitoring of the following data:

1. counting the dimming percentage for each light line in the three rooms. In this way the

evaluation of the function modality for the system and for the partial energy consumption

will be more efficient

2. monitoring the presence of student in the lecture rooms, both the ones with intelligent

devices and the traditional ones, in order to make a scientific analysis of the energy

consumption, normalized through the real occupation time of the two different wings

3. verify when and how many time the automatic regulation is forced by the manual

regulation, in order to evaluate the visual comfort for the users in a building automation

system

4. measuring and registering the lux located in the lecture rooms without intelligent devices,

in order to evaluate the natural light use for the visual comfort. This evaluation is not

necessary for the lecture rooms with the constant light automation system that is

intrinsically designed to maintain the programmed luminance.

3.1.4 Lighting energy use estimation in compliance with the EN 15193, 2007

EN 15193:Energy performance of buildings — Energy requirements for lighting

The analytical and programmatic phase of the design of the referred case study takes into

account the new European standard EN15193, in order to evaluate the impact of automation

system in lighting design.

This European standard, approved on April 2006, establishes conventions and procedures for

the estimation of energy requirements of lighting in buildings, and gives a methodology for a

numeric indicator of energy performance of buildings.

LECTURE ROOM (application period) presence sensor Dimming

channels = 1+1+1 Dimming

channels = 2+1 New luminaires and 2 pr. sens.

2B/2C (winter and summer)

no no no no

2A (winter and summer)

no no no no

2P (winter)

yes no no no

2P’ (summer)

yes yes no no

2N (winter and summer)

yes no yes no

2M (winter and summer)

yes yes no yes

Case study

87

It is assumed that the designed and installed lighting scheme conforms to good lighting

practices. For new installations the design will be developed in compliance with EN 12464-1,

Light and Lighting – Lighting of work places – Part 1: Indoor work places.

The standard also gives advice on techniques for separate metering of the energy used for

lighting that will give regular feedback on the effectiveness of the lighting controls. The

methodology of energy estimation does not only provide values for the numeric indicator but

also provides input for the heating and cooling load impacts on the combined total energy

performance of building indicator.

In order to estimate the total energy use for lighting the following terms have been defined:

5. built-in luminaires

6. control glare

7. power: for luminaires, parasitis and emergency

8. energy: for lighting, parasitic and emergency

9. time: operating, daylight and non daylight usage

10. useful area

11. dependency factors: daylight dependency factor (FD),occupancy dependency

factor (FO), absence factor (FA), constant illuminance factor (FC),maintenance

Factor (MF)

In the following section the calculation of these parameters for the referred case study will be

presented, in order to calculate the total estimated energy required for the analyzed period in

the monitored rooms.

For this purpose the equations 3.1.3_1 has been applied:

Wt = WL,t + WP,t [kWh] (3.1.4_1)

An estimate of the lighting energy required to fulfil the illumination function in the building

(WL,t) shall be established using the following equation (3.1.3_2):

WL,t = _{ (Pn x Fc) x [(tD x Fo x FD) + (tN x Fo)]}/1000 [kWh] (3.1.4_2)

An estimate of the parasitic energy (Wp,t) required to provide charging energy for emergency

lighting and for standby energy for lighting controls in the building shall be established using

the equation (3.1.3_3):

WP,t = _{{ Ppc x [ty – (tD + tN)]} + (Pem x tem)}/1000 [kWh] (3.1.4_3)

Both the calculations obtained using the Quick method and the Comprehensive method,

defined respectively in 6.2.1 and 6.2.2 of EN 15193, will be presented.

CHAPTER 3

88

Fig.3.1.4_ 1: Flow chart illustrating alternative routes to determine energy use

The lighting consumption of the monitored classrooms has been separately measured using

one of the methods reported in the section 5 of EN15193: a lighting management system that

can calculate the local consumed energy and makes this information available to a building

management system (BMS). This data have been recorded in an exportable format (a spread

sheet format) by the supervision system used for the specific case study.

At least the LENI (Lighting Energy Numeric Indicator) calculation for the referred

classrooms will be presented. The LENI is a numeric indicator of the total annual lighting

energy required in the building.

LENI = W/A [kWh/(m2 × year)] (3.1.4_4)

Where:

W is the total annual energy used for lighting [kWh/year]

A is the total useful floor area of the building [m2]

Determination of the daylight dependency factor FD,n

The FD,n for the nth room is the factor relating the usage of the total installed lighting power

to daylight availability in the referred area. It is determined as a function of the daylight

supply factor FD,S,n and of the daylight dependent electric lighting control factor FD,C,n by the

equation (3.1.4_5).

FD,n = 1- (FD,S,n X FD,C,n) (3.1.4_5)

where

FD,S,n is the daylight supply factor that takes into account the general daylight supply in the

zone n. It represents, for the considered time interval, the contribution of daylight to the total

required illuminance in the considered zone n.

Case study

89

FD,C,n is the daylight control factor that accounts for the daylight depending electric lighting

control system’s ability to exploit the daylight supply in the considered zone n.

As reported in the ANNEX C of EN15193, the procedure to calculate FD,n incorporates the

following 5 steps:

1) segmentation of the building into zones with and without daylight access;

2) determination of the impact of room parameters, facade geometry, and outside obstruction

on the daylight penetration into the interior space using the concept of the daylight factor;

3) prediction of the energy saving potential described by the daylight supply factor FD,S,n as a

function of local climate, maintained illuminance and daylight factor;

4) determination of the exploitation of the available daylight by the type of lighting control

by the daylight control factor FD,C,n;

5) conversion of annual value FD,n to monthly values.

The calculation of each step for the specific case study will be presented in the next

paragraphs.

1) The classroom are has to be sub-divided into a daylight zone AD,j and a zone AND,j not

receiving any daylight (Fig3.1.4_2).

Fig.3.1.4_ 2: Large façade opening with moderate room depth (from EN15193-2006)

Considering the geometry of the room (Fig.3.1.4_3) and the position of the windows, the

maximum possible depth of the daylight area has been calculated as follow:

aD,max = 2,5 ( 2,00m-1,00m) = 5 m

CHAPTER 3

90

Because the classroom depth is 6,2 m, that is less than 1,25 times the calculated maximum

depth, the real depth of the space of calculation can be used as aD . Thus the AD has been

calculated as follow [C.3 - EN15193-2006]:

AD,J= 6,2mx6,5m= 40,3 m2

Fig.3.1.4_ 3: Referred classroom geometry and window location

2) Daylight supply of a zone benefiting from daylight depends on the geometric boundary

conditions described by the transparency index IT, the depth index IDe, and the obstruction

index IO.

Fig.3.1.4_ 4: Window geometry and material definition

Considering for the referred classroom the area of the façade opening (AC = 8,7 m2) and the

total area of horizontal work plane benefiting from natural light calculated above, it has been

calculated IT=0,2 and IDE= 3,1.

Case study

91

Fig.3.1.4_ 5: Location of the classrooms on the main façade

The obstruction index accounts for following effects reducing light incident onto the façade:

- other buildings and natural obstacles such as trees and mountains, that is not the case of the

university referred building, because of the open horizon in front of the exposure façade

(IO,OB=1);

- simple courtyard and atrium design of the building itself , that does not interest the

classrooms analyzed, as Fig.3.1.4_5 shows;

- horizontal and vertical overhangs attached to the façade, that are not present in the building

analyzed (IO,OV=1 and IO,VF=1)

From these considerations it results IO equal to 1.

From the geometric indices, DC = 4.0 (daylight factor) has been calculated. It expresses the

access of the room to daylight for the carcass facade opening (i.e. without fenestration and

sun-protection system). The impact of the fenestration and shading system on the indoor

lighting levels should be determined by using facade type dependent correlations of DC with

the expected energy demand, i.e. methods deriving the daylight supply factor FDs as function

of the façade system (see DIN V 18599-4-2007).The simplified estimation reported in

EN15193-Annex C has been adopted in the calculation presented here for the classrooms

analyzed. It results D = 1.79. The daylight penetration as function of the daylight factors has

been classified Weak for the referred case study.

3) The daylight supply factor FD,S can be approximated as a function of latitude γSite for

latitudes ranging from 38° to 60° north. Because the latitude of the university building is 46°,

it results FD,S=0.51.

4) The Daylight Dependent artificial lighting Control (FD,C ) describes the efficiency of how a

control system or control strategy exploits the given saving potential, i.e. the daylight supply

CHAPTER 3

92

in the considered space. It does not consider the power consumption of the control gear itself.

The factor FD,C provides the correction of the daylight supply as a function of daylight

penetration.

Fig.3.1.4_ 6: FD,C as a function of daylight penetration (EN15193-2006)

The results reported in Fig.3.1.1_12 have been obtained for the classroom analyzed using the

monthly values of the daylight dependency factor:

Fig.3.1.4_ 7: FD,N monthly calculation (EN15193-2006)

The determination of occupancy dependency factor FO has been calculated, using FA value

for the following building classification: as Building type Educational Building, and as

Room by room calculation a mean between Classroom and Lecture hall, considering the

dimension and the use of the classrooms monitored.

For the energy consumption estimation, the lighting control systems evidenced in the

Fig.3.1.4_8 have been considered.

Fig.3.1.4_ 8: FOC values for different automation control systems (EN15193-2006)

Case study

93

The “Auto On/Dimmed” is defined as the control system that switches the luminaires

automatically on whenever there is presence in the illuminated area, and automatically

switches them to a state with reduced light output (of no more than 20 % of the normal 'on

state') no later than 5 minutes after the last presence in the illuminated area. In addition, no

later than 5 minutes after the last presence in the room as a whole is detected, the luminaires

are automatically and fully switched off [D.2-EN15193, 2006].

It is important to observe that the dimming regulation of the artificial light as function of the

natural light contribution is not evaluated but the lighting control system implemented in the

case study is.

For the case study analyzed the FO and FC values reported in Fig. 3.1.4_9 have been

calculated.

FO FC

Manual On/Off Switch 0,9 0,9

auto on/dimm 0,85 0,9

auto on/off 0,8 0,9

Fig.3.1.4_ 9: Calculation results of FO and FC for the case studied analyzed

Operating the total estimated energy use using the quick method the annual results in Fig.

3.1.4_10 have been obtained.

Energy consumption estimation [kWh/year]

Energy consumption estimation

[kWh/mq year]

LENI [kWh/mq year]

Manual On/Off Switch 1225,919 30,65 36,65

auto on/dimm 1089,705 27,24 33,24

auto on/off 717,339 17,93 23,93

Fig.3.1.4_ 10: Energy use estimation using the quick method

By the comprehensive method application, for a more accurate determination of monthly

energy estimation value, the results reported in Fig.3.1.4_11 have been calculated.

CHAPTER 3

94

ENERGY CONSUPTION prEN

0,0010,0020,0030,0040,0050,0060,0070,0080,0090,00

100,00110,00120,00130,00140,00

Jan Feb Mrz Apr Mai Jun Jul Aug Sep Okt Nov Dez

kWh

/mo

nth

manual on/off lightswitch on/off dimming

Fig.3.1.4_ 11: Energy use estimation using the comprehensive method

3.1.5 Lighting energy use estimation in compliance with DIN V 18599-4, 2007

DIN V 18599-4: Energy efficiency of buildings – calculation of the net, final and

primary energy demand for heating, cooling, ventilation, domestic hot water and

lighting Part 4: net and final energy demand for lighting.

These results calculated in compliance with EN15193-2006 have been compared with the

one resulting from DIN V 18599-4. This standard contains some differences in the

coefficients definition. In the following paragraphs these diversities will be emphasized. In

particular the following calculation will be analyzed:

1) annual operating hours

2) electrical power

3) daylight supply factor

4) daylight dependency factor

5) occupancy dependency factor

At last the DIN calculation results will be presented in comparison with the one resulting

from EN15193.

Case study

95

1) The annual operating hours are defined more detailed in DIN V 18599-4 that in EN15193

[see DIN V 18599-4: (4) and (5)]. Indeed using the German standard tTag and tNight are not

related only to the building type (moreover defined much more accurately by DIN V

18599-10 than by EN15193in Tab.G.1) but also to the daylight dependency factor (in DIN

FTL ) and to the occupancy dependency factor (in DIN FPr), so that the factors are introduced:

12. teffTag,TL,j

13. teffTag,,KTL,j

14. teffNacht,,KTL,j

DIN18599-4 EN1593

tDay 2543 1800

tNight 207 200

tO 2750 2000

Fig.3.1.5_ 1: Operating time estimation comparison

2) The electrical power associated with the luminaires installed is calculated by DIN using

coefficients related with the illuminance task level, the building type and the luminairs

typology [5.4 Kunstlich, DIN V 18599-4]. This calculation does not deal with the European

standard. By the parameter definition reported in Fig.3.1.5_2, pj has been calculated equal to

22,19 kWm-2 for T8 luminaire, equal to 17,9 kWm-2 for T5 luminaire. By the EN15193

calculation the electrical power is lower: equal to 16 kWm-2 for T8 luminaire, equal to 12,7

kWm-2 for T8 luminaire.

Fig.3.1.5_ 2: Kunstlicht parameter calculation for the referred case study

3) The daylight calculation is the same in the two standards, indeed the geometrical factors

calculation is the same in both cases. Instead, the daylight classification follows two different

CHAPTER 3

96

approaches: indeed the DC European factor corresponds to τEff,SNA [(27) DIN V 18599-4] but

in the German regulation the FTL [(14) DIN V 18599-4] calculation is related to the opening

exposition and to the shading typologies. The EN15193 does not deal with these two aspects.

The difference results of the daylight supply factor calculations are reported in Fig.3.1.5_3. It

is important to note that the result with the EN15193 is similar to the DIN result considering

room without shading system.

CTL,Vers

DIN18599-4

FD,S

EN15193

With shading system 0.572 -

Without shading system 0.753 -

One sigle result - 0.704

Fig.3.1.5_ 3: Daylight supply factor comparison

The artificial light control factor is defined for 3 different automation systems in DIN instead

of the two in EN. As the calculation results for the two regulations reported in Fig.3.1.5_4

show, the manual operating system coefficient is higher in the German regulation instead of

in the European one.

CTL,kon

DIN18599-4

FD,C

EN15193

manual 0.52 0.3

dimmed no off 0.73 0.77

dimmed auto off 0.78 0.77

Fig.3.1.5_ 4: Artificial light control factor comparison

4) Using the monthly values of the daylight dependency factor of DIN, the results reported in

Fig. 3.1.5_5 have been obtained for the classrooms analyzed.

Fig.3.1.5_ 5: FTL monthly calculation

Case study

97

Comparing the mean annual results for the two standards, reported in Fig.3.1.5_6, it is

possible to note that the DIN values are higher than the EN ones. That means that a higher

energy demand has been estimated by DIN for both the lighting operating systems.

FTL

DIN18599-4

FD

EN15193

manual 0.863 0.789

dimmed no off 0.583 0.458

dimmed auto off 0.555 0.458

Fig.3.1.5_ 6: Comparison between FTL and FD calculation

5) The occupancy dependency factor calculated according to DIN is related to the absence

period estimated for the specific building type, in accordance with the CAj definition (DIN V

18599-10). The DIN parameters are lower than the EN ones, as reported in Fig. 3.1.5_7. That

means that considering the German standard the occupancy factors reduce more significantly

the energy consumption estimation.

FPr

DIN18599-4

FO

EN15193

manual 0.75 0.9

Auto on/dimmed 0.525 0.85

auto on/off 0.525 0.8

Fig.3.1.5_ 7: Occupancy dependency factor comparison

Operating the total estimated energy use using the simplified method for the annual

calculation, the results reported in Fig. 3.1.1.5_9 have been obtained.

Energy consumption estimation [kWh/year]

Energy consumption estimation

[kWh/mq year]

LENI [kWh/mq year]

manual on/off 1328,86 33,22 36,65

lightswitch on/off 930,20 23,26 33,24

dimming 754,31 18,86 23,93

Fig.3.1.5_ 8: Yearly energy consumption estimation in compliance with the DIN

standard

CHAPTER 3

98

By the monthly method application, for a more accurate determination of energy estimation

values, the results reported in Fig.3.1.5_9 have been calculated.

ENERGY CONSUPTION DIN

0102030405060708090

100110120130140


kWh

/mo

nth

manual on/off lightsw itch on/off dimming

Fig.3.1.5_ 9: Monthly energy consumption estimation in compliance with the DIN

standard

Comparing the annual results, the differences reported in Fig. 3.1.5_10 have been calculated:

the European standard estimates a lower energy consumption level for every automation

system installed.

Energy consumption estimation difference

[kWh/year]


[kWh/m 2 year]

Percentage difference

manual on/off 431,50 10,79 32,47%

lightswitch on/off 132,54 3,31 14,24%

dimming 218,57 5,46 28,97%

Fig.3.1.5_ 10: Energy consumption estimation comparison

In order to quantify the difference in the calculation method in relation with the dependency

factors definition, the same operating time used for the DIN has been implemented for the

Case study

99

EN calculation too. By this daily and non-daily time usage parameterization, the differences

reported in Fig. 31.5_11 have been calculated. These results express higher energy use

estimation for the manual and lightswitch calculation using EN15193 than using DIN18599,

a lower one for the dimming calculation.


[kWh/year]


[kWh/m 2 year]

Percentage difference

manual on/off 102,20 2,55 7,69%

lightswitch on/off -160,16 -4,00 -17,21%

dimming 35,16 0,88 4,66%

Fig.3.1.5_ 11: Energy consumption estimation comparison for the same operating time

Using the monthly data comparison reported in Fig. 3.1.5_12 it is possible to observe that the

energy use for manual on/off lighting operating system is always estimated higher by DIN

(from 6.12% to 8.14% in the winter months included between October and March; from

7.12% to 8.22% in the winter months between April and September), whereas the lightswitch

on/off is always estimated lower (from -9.93% to 16.43% in winter months; from14.78%

to15.64% in summer months). Instead for dimming lighting system there is a distinct

seasonal trend: higher values in summer (between -1.95% to -14.54%), lower values in

winter (between 8.93% to 43.51%).

-20,00

-15,00

-10,00

-5,00

0,00

5,00

10,00

15,00

20,00

Jan Feb Mrz Apr Mai Jun Jul Aug Sep Okt Nov Dez media

kWh

/mo

nth manual on/off lightswitch on/off dimming

Fig.3.1.5_ 12: Difference between the energy consumption estimated by the German and

the European standard

The annual energy saving estimate using lighting automation systems has been calculated as

reported in Fig. 3.1.5_13 in compliance with the two regulations.

CHAPTER 3

100

Energy saving estimation

DIN18599-4

Energy saving estimation EN15193

lightswitch on/off 30.0% 11.11%

dimming 43.24% 41.37%

Fig.3.1.5_ 13: Energy saving estimation comparison

The monthly energy saving percentage estimated using the dimming automation systems has

been calculated as reported in Fig.3.1.5_14 in compliance with the two regulations. The

energy saving percentage for the lightswitch on/off automation system is independent from

the monthly evaluation.

0

10

20

30

40

50

60

70

Jan

Feb Mrz

Apr

Mai

Jun

Jul

Aug

Sep Okt

Nov

Dez

LIG

HT

SW

ITC

HO

N/O

FF

%

DIN18599 EN15193

Fig.3.1.5_ 14: Monthly energy saving percentage comparison

Considering an automatic shading system the CTL,Vers results equal to 0,6546. Changing this

factor the energy consumption estimated has been calculated as reported in Fig. 3.1.5_15.


manual on/off

120,43 109,86 101,70 95,93 92,57 91,13 92,09 94,97 100,74 108,42 118,03 130,52

lightswitch on/off

84,30 76,90 71,19 67,15 64,80 63,79 64,46 66,48 70,52 75,90 82,62 91,36

dimming 77,30 67,60 60,10 54,81 51,73 50,40 51,29 53,93 59,22 66,28 75,09 86,55

Fig.3.1.5_ 15: Energy consumption estimation for automatic shading system

The energy saving difference between these last results (defined for building with Automatisc

Betriben Sonnen-und/oder Blend Schutzsystemen) and the one obtained before (defined for

building with Bledschutz) is reported in Fig.3.1.5_16. As the Fig.3.1.5_16 shows, the energy

Case study

101

saving percentage grows up in relation with the outside illuminance level. The sunny months

correspond to a lower absolute energy consumption amount, so that in terms of kWh saved

the contribution of the shadow system estimated is low.

0

2

4

6

8

10

12

14

16

18


% manual on/off lightswitch on/off dimming

Fig.3.1.5_ 16: Energy saving comparison using different shading systems.

CHAPTER 3

102

3.2 Design synthesis phase

3.2.1 Light system design

The technology used to realize the lecture halls is a comprehensive, integrated system for

home and building automation for the implementation of upwardly compatible and flexible

systems: the Konnex solution. The KNX standard complies with the regulations CEI 50090

and allows a remarkable flexibility for the electric system, not only regarding the bus wiring

but also the quantity of devices available on the market that is possible to interface using this

protocol. Due to its functional versatility, its use is not confined to simple and limited

installations but also enables solutions for the complete building sector. KNX fulfils the

requirements of project design, installation, commissioning and operation of the bus system.

For these operations a software tool is needed: ETS3.

The light system requirements of the classrooms analyzed in the case study are synthesized in

the Figures 3.2.1_1, in which ideograms have been used in order to describe the selected

technological solution. This design step is named “architectonical-automation design step”.

Any technical characteristic and logical connection of the smart devices is not jet specified in

this phase.

The 2M lecture room will be divided into 6 lines: 3 for each half room. The light used is a

ceiling light 4x14W T5. The constant light regulation entails a longitudinal division of the

light ignition (in Figure 3.2.1_1 different colour gradations). 3 light sensors have been

installed, one for each longitudinal line division, and 2 presence sensors to scan without

interference the 2 half room. The switches have been integrated in order to have the priority

of the manual switching compared with the automated one. The 2N lecture room has been

divided into 2 lines. The light used is a ceiling light 4x18W T8. The constant light regulation

entails a longitudinal division of the light ignition, so it was just necessary the installation of

2 light sensors and 1 presence sensor to detect the whole room. The switches have been

integrated in order to have the priority of the manual ignition compared with the automation.

The 2P lecture room has been divided into 3 lines. The light used is a ceiling light 4x18W T8.

The constant light regulation entails a longitudinal subdivision for the light ignition (in Fig.

3.2.1_1 three different colour gradations). 3 light sensors and 1 presence sensor have been

installed to scan the whole room. The switches have been integrated to have the priority of

the manual ignition compared with the automation.

The energy consumption has been calculated separately for each classroom.

In each traditional classroom a presence/light sensor has been installed, only with monitoring

function.

Case study

103

Fig.3.2.1_ 1: Architectonical-Automation design

CHAPTER 3

104

In order to choose technological equipment for the light system installation, the multiple

criteria analysis sheet has been applied for input - output devices (see section 2.2).

32 input device typologies (among switches, presence sensors, light sensors and weather

stations) of different factories and 15 output devices typologies (among shutter, dimming and

on-off actuators) have been analyzed. The main companies considered are (see section 2.2):

• Hager lumen: www.hager.de

• ABB: www.abb.it

• Siemens: www.siemens.it

• Merten: www.merten.de

• Theben: www.theben.de

• Zumtobel: Luxmate basic, emotion, professional

• Osram: Dim Pico-Mico, Daly Easy-Basic-Advanced-Multi

• IGuzzini: Scene-Light-Colour Equalizer

The technical characteristic of each device included in the case study installation is reported

in the next section 3.2.2.

In order to identify the role and the position of each device 2 more design plans have been

elaborated: the technological system plan (Fig.3.2.1_3) and the wiring system plan

(Fig.3.2.1_4).

In the first one an identification code has been assigned to each device with the following

specifications:

- Device typology

- Progressive number of the room

- Progressive number of the device

- Device specification

It is important to note that the devices located in the cabinet have a number of the room equal

to 20 and they do not appear in the classrooms plan (Fig. 3.2.1_3).

The wiring plan relates the devices codes to each actuators channel (Fig. 3.2.1_4).

Fig.3.2.1_ 2: Installation and wiring phase in the lecture rooms

Case study

105

L2101 o L2104 o L2107 o

L2102 o L2105 c L2108 o

L2103 o L2106 o L2109 o

S2101 p

S2101 l S2102 l S2103 l

I2101 s

I2101 s

L2201 o L2204 o L2207 o

L2202 o L2205 o L2208 o

L2203 o L2206 o L2209 o

I2201 p

I2201 l I2202 l

I2201 s

I2201 s

L2301 n L2304 n L2307 n

L2302 n L2305 n L2308 n

L2303 n L2306 n L2309 n

I2302 s

I2301 s

I2301 p I2302 p

I2301 l I2302 l C2303 l

L2310 s L2312 n L2314 n

L2311 n L2313 n L2315 n

I2304 s

I2303 s

Dev

ice

tip

olo

gy

Pro

gre

ssiv

e n

um

ber

of

the

dev

ice

Pro

gre

ssiv

e n

um

ber

of

the

roo

m

Dev

ice

spec

ific

atio

n

INPUT devicesswitch IXXYYs I XX YY spresence sensor IXXYYo I XX YY plight sensor IXXYYa I XX YY lweather station IXXYYw I XX YY w

electrical outputsnew luminaires T5 LXXYYn L XX YY nold luminaires T8 LXXYYo L XX YY o

OUTput devicesdimming actuator OXXYYd O XX YY d

classroom 2P:21

4x18W Fluorescent tubes T8 L2101 o L 21 01 o4x18W Fluorescent tubes T8 L2102 o L 21 02 o4x18W Fluorescent tubes T8 L2103 o L 21 03 o4x18W Fluorescent tubes T8 L2104 o L 21 04 o4x18W Fluorescent tubes T8 L2105 o L 21 05 o4x18W Fluorescent tubes T8 L2106 o L 21 06 o

I2101 o L 21 01 oI2101 l I 21 01 lI2102 l 21 02 lI2103 l 21 03 lI2101 s 21 01 sI2101 s 21 01 s

clssroom 2N

4x18W Fluorescent tubes T8 L2201 o L 22 01 o4x18W Fluorescent tubes T8 L2202 o L 22 02 o4x18W Fluorescent tubes T8 L2203 o L 22 03 o4x18W Fluorescent tubes T8 L2204 o L 22 04 o4x18W Fluorescent tubes T8 L2205 o L 22 05 o4x18W Fluorescent tubes T8 L2206 o L 22 06 o

I2201 p I 22 01 pI2201 l I 22 01 lI2202 l I 22 02 lI2201 s I 22 01 sI2201s I 22 01 s

clssroom 2M4x14W Fluorescent tubes T5 L2301 n L 23 01 n4x14W Fluorescent tubes T5 L2302 n L 23 02 n4x14W Fluorescent tubes T5 L2303 n L 23 03 n4x14W Fluorescent tubes T5 L2304 n L 23 04 n4x14W Fluorescent tubes T5 L2305 n L 23 05 n

time swith IXXYYt I XX YY t

occupancy sensorlight sensorlight sensorlight sensor

switch

light sensoroccupancy sensor

light sensor

switch

switchswitch

4x14W Fluorescent tubes T5 L2306 n L 23 01 n4x14W Fluorescent tubes T5 L2307 n L 23 02 n4x14W Fluorescent tubes T5 L2308 n L 23 03 n4x14W Fluorescent tubes T5 L2309 n L 23 04 n4x14W Fluorescent tubes T5 L2310 n L 23 05 n4x14W Fluorescent tubes T5 L2311 n L 23 01 n4x14W Fluorescent tubes T5 L2312 n L 23 02 n4x14W Fluorescent tubes T5 L2313 n L 23 03 n4x14W Fluorescent tubes T5 L2314 n L 23 04 n4x14W Fluorescent tubes T5 L2315 n L 23 05 n

I2301 p I 23 01 p

I2301 l 23 01 lI2302 l 23 02 lI2303 l 23 03 lI2301 s 23 01 sI2302 s 23 02 s

occupancy sensor

light sensorlight sensorlight sensor

switchswitch

I2302 p 23 01 poccupancy sensor

I2303 s 23 03 sI2304 s 23 04 sswitch

switch

o4x18W Fluorescent tubes T8 L2107 o L 21 07 o4x18W Fluorescent tubes T8 L2108 o L 21 08 o4x18W Fluorescent tubes T8 L2109 o L 21 09 o

IIII

dimming actuator O2001 d 20 01 dO

energy couter OXXYYe O XX YY e

energy counter O2001 d 20 01 eO

4x18W Fluorescent tubes T8 L2208 o L 22 07 o4x18W Fluorescent tubes T8 L2209 o L 22 08 o4x18W Fluorescent tubes T8 L2206 o L 22 09 o

dimming actuator O2002 d 20 02 dOenergy counter O2002 d 20 02 eO


energy counter O2003 d 20 03 eO

IXXYYz

OXXYYz

LXXYYz


IIIIIIII

Fig.3.2.1_ 3: Technological system plan

CHAPTER 3

106

L3L2

PEN

L1

BACK BONE

BCU

EIBKNX+

-

-

+ KNX

EIB

EIBKNX+

-

SUB LINE

- +- +

KNX EIB

MAIN LINE

LN

+ -

!

BUS

24Vcc 630mAEIB KNX

EIB

230V 50Hz 2.5A

L2303 n

!

BUS

EIB KNX

BUS

+ -

A B C

- +

- +

N

L2 NL1

L1

- +

- +

NL1

- +

- +

NL1

ENERGY COUNTER

-

+ KNX

EIB

EIBKNX+

-

prog. led prog. button

I2301l

LINE COUPLER

ELECTRICAL SUPPLY

L1N

PE

L2L3





I2302l

I2301s DIM1+I2301s DIM1-I2303s DIM2+I2303s DIM2-

I2303l

I2302s unlock presence detectionI2304s unlock presence detection

I2301pI2302p

!

BUS

EIB KNX

BUS

+ -

A B C

- +

- +

N

L2 NL1

L1

- +

- +

NL1

- +

- +

NL1

L2301 n L2304 n L2307 n

L2302 n L2305 n L2308 n

L2303 n L2306 n L2309 n

I2302 s

I2301 s

I2301 p I2302 p

I2301 l I2302 l C2303 l

L2310 s L2312 n L2314 n

L2311 n L2313 n L2315 n

I2304 s

I2303 s

L2301 nL2302 n

L2306 n

L2304 nL2305 n

L2309 n

L2307 nL2308 n L2310 n

L2311 nL2312 nL2313 n

L2314 nL2315 n

BCU

Fig.3.2.1_ 4: Wiring plan

Case study

107

3.2.2 Bus devices installed in the case study system: technical detail

A functioning bus device principally consists of three parts:

- bus coupling unit (BCU)

- application module (AM)

- application program (AP)

Bus coupling units and application modules can be either separated or integrated into one

device.

Installation bus Bus device

PEI

BCU AM

PEI = Physical ext. interface BCU = Bus coupling unit AM = Application module

Fig.3.2.2_ 1: Bus device components

In a KNX installation each bus device has its own intelligence owing to the integrated BCU:

this is the reason why KNX works as a decentralized system and does not need a central

supervising unit (e.g. a computer). Central functions (e.g. supervision) can however be run

by visualization and control software installed on PCs, if needed.

BCU are currently available for connection to two different media: Twisted Pair (Safe Extra

Low Voltage 32V) or Powerline (mains power). The system installed in the 3 classrooms

analyzed uses a TP connection.

Bus devices can be principally divided into three classes:

- input devices or sensors, in the case study design switches, light sensors, occupancy

sensors

- output devices or actuator, in the case study design on/off actuators and dimming

actuators

- controller

In the case of a sensor, the application module transfers information to the BCU. The BCU

codes this data and sends it on the bus. The BCU therefore checks the state of the application

module at appropriate intervals. In the case of an actuator, the BCU receives telegrams from

CHAPTER 3

108

the bus, decodes them and passes on this information to the application module. A controller

will influence the interaction between sensors and actuators (e.g. logical module).

In the following sections the technical data and the functions setting of the devices installed

in the case study installation have been reported. In particular, as input devices, universal

interface, light sensor, occupancy sensor have been examined; as output devices dimming

actuators; as system devices weather station, energy counter and timer.

Universal interface: US/U 2.2 by ABB

Fig.3.2.2_ 2: Universal interface

The device has four channels which can either be parameterised as inputs or outputs by

selecting the application in the ETS program. Using the colour-coded connecting cables, it is

possible to connect conventional push buttons, floating contacts or light-emitting diodes.

The bus connection is carried out via the bus connecting terminal supplied.

The parameter setting defines the operating mode of the input. The following functions can

be selected: switch sensor, switch/dimming sensor, shutter sensor, value-forced operation,

scene control, control of electronic relay (heating actuator), LED control, switching sequence

(“latching relay”), push button with multiple operation, pulse counter.

Fig.3.2.2_ 3: Universal interface circuit diagram [ABB on line product catalogue]

Case study

109

Application in the case study system

In the described installation the universal interface has been set as switch/dimming sensor

and as pulse counter. If the function of the input is set as a switch/dimming sensor, it is

possible to connect conventional push buttons to the inputs of the universal interface. The

ETS program makes at least one 1 bit communication object available for each switch sensor

input. If the setting “Distinction between long and short operation” is set to “yes”, the switch

sensor can distinguish between a short and a long input signal. This function has been use for

the classrooms switches in order to set the dimming percentage with a long operation and the

total switch on/off with a short operation. Indeed, when used as pulse counter, the universal

interface can count up input signals and send them on the EIB. Depending on the parameter

setting, up to four communication objects are displayed.

No. Type Object name Function 0 1bit Input A Disable 1 4byte Input A Telegr. counter value 3 1bit Input A Request counter values 7 1bit Input B Disable 8 4byte Input B Telegr. counter value

10 1bit Input B Request counter values

Fig.3.2.2_ 4: Communication objects when used as a pulse counter (4 byte)

No. Type Object name Function 0 1bit Input A Disable 1 1bit Input A short Telegr. switch 2 1bit Input A Telegr. dimming 7 1bit Input B Disable 8 1bit Input B short Telegr. switch 9 1bit Input B Telegr. dimming

Fig.3.2.2_ 5: Communication objects when used as a switch/dimming sensor

Light sensor: TK022 by HAGER

Fig.3.2.2_ 6: Light sensor

CHAPTER 3

110

With the light sensor, it is possible to determine the brightness level in closed rooms. The

light sensor is mounted in a standard installation box in the ceiling. The cover of the sensor

is stuck firmly onto the device. The complete unit is then screwed into a flush-type box.

When combined with the lighting controller, the light sensor is used for constant light

control. The electrical connection to the lighting controller is carried out with a twin core

MSR cable (SELV). The total length of this cable may not exceed 100m. This sensor is

connected with the dimming actuator described in the correspondent following section.

Fig.3.2.2_ 7: Directive Diagram of the acrylic glass rod


The light sensor is supplied with 2 different acrylic glass rods. In the referred classrooms the

flat type has been installed.

The Fig. 3.2.1_7 shows the distribution of the light sensibility in the room as regards the

selected acrylic glass rod. The percentage values refer to the maximum sensibility of the light

sensor.

Occupancy sensor: INSTABUS ARGUS by MERTEN

Fig.3.2.2_ 8: Occupancy sensor

Installed under the ceiling, the ARGUS presence detector can detect people up to 14.5 metres

away at an installation height of 3.00 metres for sitting person’s area, as in the case study

installation. Two setting Channels are available for this device. It measures at the same time

Case study

111

the intensity of the natural light: if brightness has fallen below a pre-selected light threshold,

the smallest movements in the room are sufficient to switch on the lighting via Channel 1

automatically. The brightness sensor can be set in a continuously variable manner between

10 - 1,000 lux.

However, if the ambient brightness is adequate, or if it does not detect any person in the

room, the presence detector switches off the lights again. The ON time (after the last

movement detection) can be set in a continuously variable manner between 10 s - 30 min.

With 5 sensors, 6 extensive detection levels and 544 switching segments, it can pick up even

small movements. The Channel 2, acting independently from the first channel, controls all

systems connected to it, such as heating or ventilation.

Fig.3.2.2_ 9: Occupancy sensor circuit diagram and detection area [Merten on line

catalogue]


Only the Channel 1, dependent on brightness, has been set in the case study installation,

because any other control system, in addition to the light system control, has been applied.

One occupancy sensor is enough in order to detect the automated classrooms 2P and 2N, and

the symmetrical traditional classrooms 2B and 2C. In the classroom 2M two sensors have

been installed in the middle of the ceiling, both of them partially covered with a dark layer in

order to detect only an oriented zone (one half of the classroom).

No. Type Object name Function 0 1bit Switching Ob. Presence block 1 1bit Locking Ob. Presence block 2 1bit Always dark Ob. Presence block 5 1bit Switching Ob. Movement block 6 1bit Locking Ob. Movement block

Fig.3.2.2_ 10: Communication objects of presence sensor

CHAPTER 3

112

Dimming - On/off actuator: Hager TK022

Fig.3.2.2_ 11: Dimming actuator

The light controller/switch dim actuator is a DIN rail mounted device for insertion in the

distribution board. It is connected to the EIB via a bus connecting terminal.

It is used to switch and dim luminaires with electronic ballast devices with 0 to 10 V control

inputs. It has two independent channels. Four application programs are available: dim-switch

control (8 communication objects); dim-switch control 2 (8 communication objects);

dim-switch control limit (8 communication objects); dim-switch slave (14 communication

objects).

Fig.3.2.2_ 12: Dimming actuator circuit diagram [Hager on line product catalogue]


The characteristic application function of the case study application is the light dimming by

manual or automatic control.

For the manual control the operation function is the following.

Case study

113

Start dimming long operation

Stop dimming release dimming rocker plate

Switch off: short operation of rocker plate

Start dimming long operation

Stop dimming: release rocker plate

100%

0%

Dimming actuator switches to last reached

Switch on: short operation of rocker plate

tlim

Fig.3.2.2_ 13: Dimming actuator functioning

The duration of the key operation determines whether the switching function or the dimming

function is activated. If the pressure time of the key is shorter than tlim a switch telegram is

transmitted. Longer periods of key operation after the period tlim cause the transmission of a

'start dimming' telegram. As soon as the key is released again, a 'stop dimming' telegram is

transmitted.

BCU PEI AU

15 V 24 V 5 V

SR

DAC

Dimming electronic choke 0-10 V

AC 230 V

BCU = Bus coupling unit SR = Shift register AU = Application unit DAC = Digital-analogue converter PEI = Physical external interface

Fig.3.2.2_ 14: Functioning scheme of the dimming actu ator

During the dimming period, the bus coupling unit increases or decreases the digital

brightness value according to the set regulating time. The brightness value is continuously

passed on to the shift register (SR) in the application unit. The 8 bit long data word allows the

generation of 28 = 256 brightness values. The data word is fed into the digital/analogue converter (DAC), which then generates the appropriate control voltage in the range of 0 to

CHAPTER 3

114

10V. The dimmer's electronic choke uses the voltage to control the light emission of a

fluorescent tube. The power circuit breaker in the application unit is used to connect/

disconnect the mains voltage (Fig.3.2.1_13).

For the automated control the operation function is the following.

The function “closed loop control and dimmer” is selected in the general parameters, so that

the actuator can be used as a light controller. The current brightness value is conveyed via the

input of the light sensor.

The closed loop control can be carried out in two ways:

1. the brightness set point is set once and may not be changed by the user.

2. the brightness set point may be changed temporarily by the user. This new brightness set

point is maintained until the next switching command is sent.

This second function has been implemented in order to give a command priority to the

manual control. This is a crucial point in order to manage the lesson scenario during slides

screening.

The light controller is set by carrying out the following steps:

– darken the room

– change the light intensity by dimming up or down until the room reaches the required

brightness

– send a telegram with the value “1” to the “Set point” object e.g. via a separate switch sensor

– the brightness value that is measured via the light sensor is accepted as the new set point

– the lighting becomes significantly darker to indicate that the setting process has been

successful and then slowly sets itself to the new brightness set point

Further dimmers can be controlled via the 1 byte object. Phase controlled or phase aligned

dimmers can therefore be included in the closed loop control.

The control response on bus voltage recovery can be set. The controller can be regulated to a

new brightness state or the closed loop control can be switched off.

No. Type Object name Function

0 1bit Channel A Switch 1 1bit Channel B Switch 2 4bit Channel A Relative dimming 3 4bit Channel B Relative dimming 4 1byte Channel A Brightness value 5 1byte Channel B Brightness value

Fig.3.2.2_ 15: Communication objects for dimming mode

Case study

115

Weather station: Teben

Fig.3.2.2_ 16: Weather station

The weather station is used primarily for detection of brightness (1,000...99,000 lux), rain

(upper sensor and lower surfaces are permanently heated), temperature (– 30 ... + 50 °C),

day/night (under 10 lux is night at more than 10 lux it is day -one minute and 15 seconds after

the brightness value has exceeded 10 lux) and wind speeds (0…24.0 m/s).

The weather station has 2 different channel types:

• universal channels

• the sun protection channels

The universal channels can be used for sub-tasks (e.g. a pure brightness threshold) or for a

free combination of measured variables.

A channel is made up of up to 4 logically linked weather conditions:

• If the brightness is above/below the threshold AND

• If the temperature is above/below the threshold AND

• If the wind speed is above/below the threshold AND

• If rain is present/not present

A non-relevant condition can be set to the value "any" and so it is ignored during logical

linking. As a result of the satisfaction or non-satisfaction of this AND link, a telegram is sent

to the associated channel object. Each universal channel has one lock object and one teach in

object.

If required, a universal channel can also be parameterized as a safety channel. Here, the

relevant variables – i.e. temperature, rain and wind – are linked with an OR operation.

The sun protection channel comprises:

• a dawn/dusk threshold

• 3 brightness thresholds

• 3 objects for actuating the drive (up/down – height % – slats %)

• 1 sun control object (morning/evening)

CHAPTER 3

116

• 1 teach in object

• 1 safety object

Fig.3.2.2_ 17: Weather station wiring [Theben on line product catalogue]


The weather station has over 41 communication objects. Some objects can assume various

functions and names depending on their configuration. Moreover in the case study

application only the measured values function has been used in order to send the current

actual values for wind speed, brightness, temperature and rain.

It is necessary to select a place for the installation on the building where the wind, rain and

sun can be measured uninhibited by the sensors. In particular, the light dome of the cover

may not be in a shadow of the building or other external obstructions, for example, trees. At

least 60 cm free space must be left underneath the weather sensor to enable correct wind

measurements and to prevent covering by snow in case of snowfall. In order to observe these

installation requirements, the weather station has been installed closed to the roof, on the

upper part of the south façade of the university building.


0 2 byte Brightness value Phisical value

1 2 byte Temperature value Phisical value

2 2 byte Wind speed Phisical value 3 1 bit Rain sensor Rain/ no rain …40

Fig.3.2.2_ 18: Weather station implemented communication objects

Case study

117

Τhe object 0 "Brightness value" sends the current brightness value either if there is a change

in brightness (it is possible to send the brightness value in the event of a change of

10%-20%-30%- 50 %, but at least 1lux) and/or cyclically. Only the value measured directly

by the weather station is sent. Received external actual values are not considered.

The object 1 "Temperature value" sends the current temperature value either if there is a

change (it is possible to send temperature value in the event of a change of

0.5°-1.0°-1.5°-2.0°-2.5°) and/or cyclically (every 2-3…-60min).

The object 2 "Wind speed" sends the current wind speed either if there is a change (it is

possible to send wind speed in the event of a change of 20%-30%- 40 %, but at least 1 m/s)

and/or cyclically (every 2-3…-60min). The units (m/s or km/h) can be chosen on the

“Measured values” parameter page.

The object 3 "Rain sensor" sends the current rain status – "1" for "rain" and "0" for "no rain".

Depending on how it is configured, it can be sent only when the status has changed, or after a

change, or cyclically.

Time Switch: ABB SW/S 4.5

Fig.3.2.2_ 19: Timer switch

The 4-fold week time switch with a day, week and year program is a DIN rail mounted device

for insertion in the distribution board. Connection to the EIB is carried out via the bus

connecting terminal at the front of the device.

There are 324 memory locations available with free weekday block formations. Using a

program for use during holidays, the execution of the programming can be interrupted for up

to 45 days. The time switch has a priority switching operation (single operation) for special

days and holidays databank. It is also possible to program switching impulses. The cover in

front of the keypad and the display can be sealed.

The time switch can send switching or value telegrams to EIB actuators at the specified

times.

The time switch has four channels each having the same parameters. ETS displays various

communication objects depending on the parameter selection.

CHAPTER 3

118

The communication objects of the four channels send telegrams in accordance with their

switching programs at the times that are programmed in the clock. It is also possible to select

separately for each channel whether its telegrams should be sent cyclically.

It is possible to define the switching times so that the clock functions “ON” and “OFF” are

not used alternately.

Fig.3.2.2_ 20: Timer switch circuit diagram [ABB on line product catalogue]


The application program used in the case study program is the Switch Value Cyclic, in order

to lock the presence sensors after 19.00 and 21.00. The unlock function is operated manually

by switching. This function has been applied in order to switch off the light during the

cleaning time, and to allow just the manual control. Indeed the presence sensors can detect

someone moving outside the room if the door is open.

For this scope the parameter “Function” of the channel is set to “Send telegr.switch”, so that

the channel has a 1 bit communication object. The value of the communication object (“0” or

“1”) is dependent on the programmed switching function of the clock application module.


0 1 bit Channel 1 - switch Telegr. Switch 1 1 bit Channel 2 - switch Telegr. Switch 2 1 bit Channel 3 - switch Telegr. Switch 3 1 bit Channel 4 - switch Telegr. Switch 4 3 byte Time Telegr.time 5 3 byte Date Telegr.date 6 1 bit Date/time Send request

Fig.3.2.2_ 21: Timer switch communication objects

Case study

119

Static Watt-Hour Meter

Fig.3.2.2_ 22: Functional diagram of the static watt-hour meter [product technical

documentation]

The static watt-hour meter used in the case study system has the following applications:

- single phase line

- active energy count

- measure and display of voltage, current, active power in true RMS value

- direct connection up 240V 63A

- pulse output for remote monitoring.

The contact range is 110Vdc/ac 50mA and the pulse output is 1imput/10Wh.

This device has been used in connection with the Switch - universal interface, programmed

for this specific application as “pulse counter”: when used with this setting, the universal

interface can count up input signals and send them on the EIB. Depending on the parameter

setting, up to four communication objects are displayed.

Fig.3.2.2_ 23: Static watt-hour meter wiring diagram [product technical documentation]

CHAPTER 3

120

3.2.3 System installation and configuration

In the KNX system used in the case study presented, the network nodes are connected via

twist pair lines. Other media are in use: coaxial cable, supply main and even infrared and

radio. In a TP version the bus line is used to transmit the supply for the electronics and for the

exchange of the information. KNX systems are characterized by a clear, hieratic structure:

zones and lines. Each line is separate from the others by line couplers. A minimum TP KNX

installation consists of the following components:

• a power supply unit (29V DC)

• a choke (can also be integrated in the power supply unit)

• sensors (a single switch sensor is represented in Fig.3.2.3_1)

• actuators (a single switch actuator is represented in Fig.3.2.3_1)

• bus cable (only two wires of the bus cable are required).

1

1

KNX/EIB

230V

P.A.: 1.1.2 GA 5/2/66

P.A.: 1.1.1 GA: 5/2/66

Fig.3.2.3_ 1: Minimum KNX installation example

After the installation, a KNX system is not ready for operation until sensors and actuators

have been loaded with application software using ETS3 program. The configuration steps

are the followings:

• assignment of physical addresses to the different devices (for the unique identification of

a device in the installation);

• selection and setting (parameterization) of the appropriate application software for each

device;

• assignment of group addresses (for linking the functions of sensors and actuators).

Physical address

The source address is the physical address that specifies the area and line to which the

sending device is assigned. The physical address is permanently assigned to the bus device

Case study

121

during the project design stage and is mainly used for commissioning and service functions.

It has the following format: Area [4 bit] - Line [4 bit] – Bus device [1 byte]. In the case study

system the topology is structured in 1 area and 2 lines, because the classrooms system has

been connected with another system already existing in the University, named CUnEdI.

Some devices allocated in line 1.1 have been connected in the line 1.2 functionality: the

weather station and the time controller.

Fig.3.2.3_ 2: ETS3 Topology of the case study program

Each bus device is normally prepared for the acceptance of its individual address by pressing

a programming button on the bus device. The programming LED is lit during this process.

The individual address is also used for the following purposes after the commissioning stage:

- diagnosis, error rectification, modification of the installation by reprogramming

- addressing of the interface objects using commissioning tools or other devices.

Fig.3.2.3_ 3: ETS3 physical address list of the case study installation

CHAPTER 3

122

Group address

Communication between devices in an installation is carried out via group addresses: indeed,

in KNX installations, devices that belong to the same functional group are “logically wired”

by means of group addresses.

When setting the group address via ETS, it can be selected as a “2-level” (main group/

subgroup) or “3-level” structure (main group/middle group/subgroup). In the case study

configuration system the second setting modality has been used.

The following pattern has been used in the project configuration:

a) Main group = functional domain (lighting, heating, metering, ect.)

b) Middle group = secondary functionality (for lighting switching, dimming, scenario, etc.)

c) Subgroup = sequential number device The Fig.3.2.3_4 shows the Group Addresses view of the ETS program. This view is required

together with the Building view to link the communication objects with the corresponding

group addresses.

Fig.3.2.3_ 4: Group addresses view of the case study installation

Communication objects represent a neutral interface between communication and

application. They exist in the form of data structures in the RAM and EEPRON of

communication module. They basically comprise three parts:

communication object description

- object value

- communication flags.

The communication object description must contain at least the object type and transmission

priority. The communication flags contain the status of a communication object.

Communication objects are connected with the bus communication via group address. A

Case study

123

communication object can receive data via several group addresses, but only transmit data

via a single group address. The group addresses are unique with the network and are assigned

with local connection numbers within the devices.

Application program

Before the device can function, the application program must be loaded into its memory.

There may be more than one application program for a particular device, containing different

functions. The parameters define the concrete function of the application program and can be

set and loaded into the selected device with the ETS commissioning program.

Fig.3.2.3_ 5: Device configuration by parameters setting

The ETS 3 Professional program visualises projects via different editing windows:

• Building and Function view,

• Topology view,

• Group Addresses view

• Device view.

CHAPTER 3

124

The Building view is used to structure the project according to the physical structure of the

building and to insert the necessary KNX devices in the rooms of the building. Devices can

be inserted in rooms or in cabinets. A hierarchical view is useful in order to find one’s way

when working with large projects.

Fig.3.2.3_ 6: Device view of the case study program.

3.2.4 Supervision system configuration

The supervision software used in the case study presented is Gefasoft – Graphpic 7.1.In order

to let communicate it with KNX devices, the EIB OPC-server has been use.

OPC is an abbreviation for OLE for Process Control; OLE itself originates from the

Windows based term Object Linking and Embedding. Basically, OPC is a software concept,

which implements a unified interface between different installation bus technologies like

EIB on one hand and automation and visualization software on the other.

Like other drivers on PC – the EIB OPC-Server is equipped with an own user interface that

lets manipulate the driver's settings or import the ETS project data.

After having saved a configuration, these settings may be applied by any software with OPC

abilities. The software using this configuration is named OPC-Client or simply Client.

Case study

125

Usually the client software will have the ability to display the configuration of the

OPC-Server and let select the desired communication objects (in this case: the EIB group

addresses, see Fig. 3.2.4.4_1), that have to be operated.

Fig.3.2.4_ 1: EIB OPC server configuration

The communication module is the central application in GraphPic, both during projecting as

well as in the course of a GraphPic project. It coordinates all GraphPic modules and their

configuration, both on the local computer as well as on the other computers in the network in

case of distributed projects (Fig.3.2.4_2). The communication module manages the central

configuration of a project. It starts editor and runtime modules, provides information about

the project configuration, records runtime messages and transfers process variables from the

drivers to all modules that require them, both locally as well as in the network.

Two modules have been implemented in the case study supervision program: the

Visualization module and the Measurement Value Recording module.

Fig.3.2.4_ 2: Graphpic 7.1 Architecture

CHAPTER 3

126

With the Visualization editor, it is possible to create a user interfaces in 2D or 3D. An

extensive set of drawing and animation tools and the import of vector and bitmap graphics

accelerate display generation. The visualization process module connects the plant displays

to a dynamic user interface. All common input media like mouse, keyboard and touch-screen

are supported.

In the Visualization Window of the case study supervision it is possible to display for each

classroom, the following data, both as value and as dynamic image:

- the presence

- the instantaneous energy demand

- the inside illuminance

- the dimming percentage for each regulation channel

This dynamic interface allows setting these values directly operating on the visualization

windows.

Fig.3.2.4_ 3: Visualization interfaces of the case study.

Case study

127

The module Measuring Value Recording Pro saves process data as so-called measuring

channels in the integrated database and displays these in current or history format in different

types of graphics. The graphs can be displayed at wish as Y/t or X/Y diagrams.

The case study integrated data base is structured as follows: the monitoring results are stored

in the databank by the supervision software Gefasoft 7.1. For each day, a file is generated

with temporally ordered data. Each single message is labelled with a distinctive number.

Fig.3.2.4_ 4: Integrated database of the case study measuring module

The following diagrams are displayed in the supervision program of the classrooms analyzed

(Fig. 3.2.4_4):

- energy consumption comparison between automated and manually operated

classrooms (kWh)

- energy consumption in relation with detected presence in each classroom

- inside illuminance level (in lux)

- outside illuminance level (in lux)

CHAPTER 3

128

Fig.3.2.4_ 5: Graphs displayed by the case study supervision.

In order to elaborate and to analyze the data monitored, the data relating to each specific label

(variable monitored) have been divided and organized by means of a specific program. It

creates a folder for each day that contains one file for each variable recorded, named as the

identification label. Each file contains the values recorded for the relative variable in a

temporal sequence.

Fig.3.2.4_ 6: Example of the first elaboration data program.

Case study

129

Moreover the different data points have been synchronized using the specific program: all the

collected data for each variable, divided in daily folders, have been temporally expressed in

terms of a sequence of 3 minute intervals, calculating the average data in the temporal

interval. In this way it is possible to relate different events occurring at the same time interval

in a specific room.

Fig.3.2.4_ 7: Example of the second elaboration data program

CHAPTER 3

130

3.3 Evaluation phase

3.3.1 Software tools output comparison

SUPERLITE

A typical SUPERLITE simulation involves two steps:

1. to create an input file

2. to define selected sky and location conditions

The first step consists of creating an input file that contains a geometric description of the

building, as well as the solar and luminaire data to be used for the simulation.

The input file can be created in three different formats: a *.dwx file, a SCRIBE modeller

format or a SIMPLE INPUT model. Because of the simple geometry of the monitored class

rooms, in this research thesis an SIMPLE INPUT model has been created.

To define the input file it is necessary to set the following parameters and data:

• geometric base case

• parameters for the base case selected

• materials for walls, ceiling(s) and floor(s)

• openings (windows, roof lights) and their position

• outside obstructions for the windows defined

• luminairs

In this case study the shape of the room is defined as Shoebox geometry.

Room position and orientation have been defined as first, with reference to the building

geometry and dimensions. A facade larger than the room may be important when

interreflections between the building and visual obstructions are calculated, but it is not the

case of the rooms we refer to, because there is not any buildings closed to the university.

The work surface height defines the height of the virtual surface to calculate the values in

reference to the room’s floor height. For the classrooms the work desks are 0.8 high.

It is possible to define the material characteristics for each of the surfaces of the model in

Case study

131

question. A parametric material can be defined by typing the correct description for this

material. Possible identifiers for parametric materials are:

%TXX - a transparent material

%DXX - a diffuse translucent material

%GXX - a gray opaque material

XX stands for the transmittance or the reflectance of the material and is a two digit

percentage value.

The windows definition includes geometry, location and material setting. Two glazing types

may be selected: clear or diffuse. For each of the glazing types the Transmittance,

Reflectance and a Maintenance Factor (a value describing the cleanness of the window,

100% equals to a always clean window pane) have to be specified. All of these values are

percentage values. Each window may have an overhang and/or an obstruction.

Fig. 3.3.1_ 1: Input model (generated as simple input) of the classroom monitored

The options selectable to create the input data, especially for what concerns the geometrical

definition of the room, are almost the same for Relux and Dialux. Using these two tools it is

immediate to introduce and modify an indoor space with furniture: they permit an

instantaneous overview of the space simulation.

CHAPTER 3

132

The mean difference between Relux-Dialux and Adeline is the definition of the windows that

with Adeline can be defined with much more accuracy. It is possible with this program to

establish geometrical factors like:

• Glass recess, that refers to the distance that the window glass is recessed with respect

to the exterior wall surface.

• Overhang Data, that provides a simple means to take into account outdoor

obstructions that are close to the window and are of simple shape.

• Depth of the overhang, measured from the exterior edge of the overhang to the

exterior window surface.

and material parameters like:

• Glazing Type, clear window, clear window shaded by sheer curtains and diffusing

window (the case study has clear windows).

• Maintenance factor, that refers to the window cleaning and repair frequency. Any

value between 0 and 100 % is allowed (for the case study has been evaluated a

maintenance factor equal to 80%).

• Curtain transparency, fraction of light that is transmitted transparently by the sheer

curtains. The remaining fraction is assumed to be transmitted diffusely. Any value

between 0 and 100 % is allowed.

These parameters can not be selected in Relux or Dialux.

By SUPERLITE is possible to have a photorealistic rendering output of the indoor space,

running the RADIANCE program.

The graphic output of Relux and Dialux, based always on a radiance calculation, seems to be

very realistic as well and it is simple und fast to chance the point of view for the virtual

observer.

Fig. 3.3.1_ 2: Photorealistic rendering output generated by Adeline

Case study

133

Once an input file has been created, the Adeline program is ready to be run. It is possible to

type the sky definition, selecting among:

• CIE Overcast Sky

• CIE Clear Sky with sun

• Uniform Sky

• CIE Clear Sky without Sun

The solar and weather data input for the program can be supplied in three ways:

• Sun position and irradiance data,

• Geographical and atmospheric data,

• Sun position and atmospheric data.

The first option allows to supply specific irradiance and luminous efficacy data and also to

specify the appropriate sky condition.

The second option is based on the specification of geographic data such as latitude and

longitude, the time and date of the simulation under given sky conditions. This option

provides for a series of simulations for given times of day and year.

The case study geographic data setting is the follow:

- location: Trento

- latitude: 11°

- longitude: 46°

- elevation: 200m

- grand reflectance: 60%

- monthly turbidity and water content: from IGDG data file

The third option simply requires the solar position (altitude and azimuth) under specified sky

conditions. The sky models available in the program are the uniform sky, the CIE standard.

Finally it is possible to elaborate and compare the output.

Graphic output can be displayed: if the Isographics box is checked, an output plot file *.plt

will be generated during the simulation. Once this file has been produced, it is possible to

choose between various graphic representations of the results: contour plots of illuminance

(daylight and/or electric light) or daylight factor distribution on a work surface and section

plots or 3-D representations.

In Relux and Dialux it is possible to select only two sky conditions: CIE Overcast Sky and

CIE Clear Sky with sun.

The mean difference between these two software and Adeline, for what concerns the data

CHAPTER 3

134

definitions, is the setting of weather data. In fact it is not available the definition of these

parameters to qualified local irradiance value or luminous efficacies, neither it is selectable

the turbidity and the water content of air as monthly data.

The location definition is instead the same.

Two different types of graphic output are possible in Adeline:

• contour plots of illuminance (daylight and/or electric light) or daylight factor

distribution on a work surface and section plots

• 3-D representations

Six different types of results can be displayed:

1. diffuse daylight illuminance (only the diffuse components of the daylight distribution on

the selected work surface will be shown),

2. daylight factor (daylight factor distribution on the selected work surface will be shown),

3. daylight illuminance (diffuse and direct components of the daylight distribution on the

selected work surface will be shown),

4. electric lighting illuminance (only electric lighting components will be shown),

5. electric lighting and diffuse daylight illuminance (electric lighting and diffuse daylight

components will be shown),

6. illuminance from all sources, including direct daylight

7. Illuminance is displayed in klx, daylight factor is displayed in %.

For Relux and Dialux is it possible to visualize illuminance (daylight and/or electric light) or

daylight factor distribution on a work surface and section plots, as well as 3D representation

or contour illuminance. In this case it is not possible to choose between diffuse or direct

components. In the following graphs the second option has been set with different parameters.

0.00

0.31

0.62

0.94

1.25

1.56

1.87

2.18

Illuminance [klxdiffuse daylight

Projection against Y axis

Illum.[klx]

Fig. 3.3.1_ 3: Diffuse daylight output result for overcast sky conditions

Case study

135

0.00

1.42

2.85

4.27

5.69

7.12

8.54

9.96

Illuminance [klxdiffuse daylight


Illum.[klx]

Fig. 3.3.1_ 4: Diffuse daylight output result for clear sky with sun conditions.

0.00

6.39

12.79

19.18

25.57

31.97

38.36

44.76

Illuminance [klxdiffuse and direct daylight


Illum.[klx]

Fig. 3.3.1_ 5: Diffuse and direct daylight output result for overcast sky condition

CHAPTER 3

136

6.0

4.8

3.6

2.4

1.2

0.00.0

1.0

2.0

3.0

4.0

5.0

16.024.032.040.048.056.064.072.0

16.024.032.040.048.056.064.072.0

X Dimensionimension

0.008.5517.1025.6434.1942.7451.2959.83

daylightfactor[%]

daylightfactor

Fig. 3.3.1_ 6: 3D daylight factor calculation

6.0

4.8

3.6

2.4

1.2

0.00.0

1.0

2.0

3.0

4.0

5.0

2.08.014.020.026.032.038.044.0

2.0 8.014.020.026.032.038.044.0

X Dimensionimension

0.005.9311.8617.8023.7329.6635.5941.53

Illum. [k

diffuse and direct daylight

Fig. 3.3.1_ 7: 3D diffuse and direct daylight output result for sky with sun condition

Case study

137

0

100

200

300

400

500

600

700

800

900

6.0

0

7.0

0

8.0

0

9.0

0

10

.00

11

.00

12

.00

13

.00

14

.00

15

.00

16

.00

17

.00

18

.00

19

.00

20

.00

21

.00

22

.00

lx

Fig. 3.3.1_ 8: Typical day inside illuminance calculation in ML (October) and discrete

output data point of the Adeline calculation.

The Adeline output results have been compared with the monitored ones. As Fig.3.3.1_8

shows, the main difference between real data recorded for typical day and Adeline

calculation output in the afternoon hours has been recorded.

SUPERLINK

In a time where environmental and global issues are of high priority, an increased use of

daylight will provide a potential factor reducing energy produced from fossil fuels. The light

in modern commercial buildings is provided, mostly by daylight and partly by artificial

lighting. If we want to take full advantage of the potential energy saving, it is essential to

provide the building with control systems in order to adjust and reduce electric light output to

the available daylight. If the building is not supported by a control system for artificial

lighting, the lights tend to stay on and seldom switched off again when daylight alone is

sufficient.

The numerical prediction of daylighting effects requires a method taking several parameters

into account, namely

• time profile of the amount of daylight illumination available at a given location

• time profile of the amount of daylight illumination that enters the building

• requirements as to the workplace illumination level

• type and control of supplementary lighting installations

CHAPTER 3

138

To establish the effects of daylighting onto a building’s energetic performance it is necessary

to quantify the amount of energy supplied for supplementary lighting purposes in relation to

the amount of daylight available. In this context it is of interest whether and when daylight

supplies are falling below the required rated workplace illuminance (during occupation time),

as only during these periods supplementary lighting must be switched on or dimmed to keep

up the nominal illuminance level. Using an on/off supplementary-lighting control-strategy, it

is necessary to know the total time in the interval during which the rated workplace

illuminance cannot be maintained from daylight supplies. In case of a dimming strategy, it is

also necessary to quantify the daylight deficit per time step.

SUPERLINK mix hourly standard sky conditions to simulate real weather conditions and

produce data of illuminance on the work surface for three CIE-standard sky types: overcast

sky, clear sky without sun, and clear sky with sun. To simulate real sky conditions the hourly

sunshine probability derived from actual weather data is used for weighting the standard

conditions and obtain hourly work surface illuminance from daylighting. The work surface

illuminance is then compared to the required design illuminance and continuous recalculated

for the energy needed from artificial lighting.

Dialux does not calculate any prevision for the energy saving using lighting control systems.

Relux performs an estimation of the energy saving considering the natural light penetration

and the working hour plan, but it does not consider the implementation of smart devices. The

results of the economic Relux calculation are in the correspondent Annex.

The following sections report the Adeline calculation and the comparison between the

software tool output result and the monitored data.

Sunshine probability calculation

As reported in the section 2.1.2, for the case study location (Trento), it was not possible to run

the automatic calculation tool of Adeline in order to obtain the .SSP file. Indeed it is not

available a .TRY file that contains all the weather parameters for the calculation, neither it is

possible to get these data for the case study location for a statistically significant period, at

least 20 year.

The weather data available for S. Michele A.A. contains global, diffuse and direct radiation

separately only from 2000. Only the global radiation is recorded from 1990.

For this reason the global radiation has been calculated using the Bird/Hulstrom’s model and

the results have been compared with the monitored data available for the last 20 years,

Case study

139

combined in a test reference year calculation.

It is important to note that the coefficient for the surface albedo has a high value, equal to 0.6

(Fig.3.3.1_9), in order to model correctly the outside surface (white stone). This parameter

has been changed after the comparison (mean error calculation) between the diffuse radiation

obtained by Hulstrom’s model and the available weather data recorded. Using a coefficient

equal to 0.6 instead of 0.2 the mean error has been reduced of 30%.

The Fig.3.3.1_10 shows the Hustron calculation result obtained using the calculation sheet

described 2.2.1. Input latitude in decimal degrees (positive in northern hemisphere) 46,000 longitude in decimal degrees (negative for western hemisphere) 11,000 time zone in hours relative to GMT/UTC (PST= -8, MST= -7, CST= -6, EST= -5) 1 daylight savings time (no= 0, yes= 1) 0 start date to calculate solar position and radiation 1-Jan-06 start time 1.00 AM time step (hours) 1 number of days to calculate solar position and radiation 30 barometric pressure (mb, sea level = 1013) 995 ozone thickness of atmosphere (cm, typical 0.05 to 0.4 cm) 0,3 water vapor thickness of atmosphere (cm, typical 0.01 to 6.5 cm) 1,5 aerosol optical depth at 500 nm (typical 0.02 to 0.5) 0,1 aerosol optical depth at 380 nm (typical 0.1 to 0.5) 0,05 forward scattering of incoming radiation (typical 0.85) 0,85 surface albedo (typical 0.2 for land, 0.25 for vegetation, 0.9 for snow) 0,6

Fig. 3.3.1_ 9: Calculation of solar position based on NOAA's functions and solar

radiation based on Bird and Hulstrom's model: input data

Fig. 3.3.1_ 10: Calculation of solar position based on NOAA's functions and solar

radiation based on Bird and Hulstrom's model: output results.

CHAPTER 3

140

The difference between the theoretical data and the one recorded by the weather station of

S.Michele has been calculated day by day in an hourly computation. As Fig. 3.3.1_11 shows,

the Bird model radiation incident upon an horizontal surface for the case study location is

significantly higher than the recorded one. The mean difference in the working hours (8.30 –

18.30) for all the year has been calculated equal to 30%.

0,00

0,50

1,00

1,50

2,00

2,50

24:

00

02:

00

04:

00

06:

00

08:

00

10:

00

12:

00

14:

00

16:

00

18:

00

20:

00

22:

00

Bird model global radiation incident upon a horizontalsurface typical year for typical day

Fig. 3.3.1_ 11: Global radiation calculation obtained by Hulstron model and typical year

model; typical day in October.

The monthly typical day calculation, reported in Fig.3.3.1_12 and Fig.3.3.1_13, shows that

the main percentage difference has been detected at dawn and at sun seat. In these periods, in

winter semester, a percentage difference of 90-70% has been calculated. Between 11.00 and

16.00 the difference is lower than 40% (Fig. Fig.3.3.1_12).

In summer semester the difference falls down until the 20% during the middle working days

hours (Fig. Fig.3.3.1_13).

The working period contains the hours influenced by the main error, especially the morning

lessons hours, so that only the recorded data available from the S.Michele weather station

have been considered.

Case study

141

0

20

40

60

80

100

120

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

JANUARY FEBRUARY MARCH MEAN

Fig. 3.3.1_ 12: Percentage difference between global radiation calculation obtained by

Hulstron model and typical year model; winter months

0

20

40

60

80

100

120

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

APRIL MAY JUNE JULY AUGUST

Fig. 3.3.1_ 13: Percentage difference between global radiation calculation obtained by

Hulstron model and typical year model; summer months

In order to calculate the Sunshine Probability file, the real days for which the minimum

variance has been calculated from the total amount of the years sample (20 years from 1986

to 2006) have been selected. The calculation result is a vector that contains the amount of the

hourly sunshine seconds measured for each day of a reference virtual year (Fig.3.3.1_14).

CHAPTER 3

142

Fig. 3.3.1_ 14: Calculation results of the hourly sun seconds for each day of the test

reference year, the correspondent sunshine probability hourly calculation and the

calculation of the mean year

The results obtained by this calculation, referred to the IGDG data calculation method, have

been compared with the sunshine probability obtained as averaged data. The two approaches

have the following differences: the first vector data contains the more probable real days that

describes a real situation; the second vector data contains mean results, that have the

minimum square difference, but that can describe a situation never verified in real conditions.

In order to evaluate this distinction the two data series have been compared.

The Fig.3.3.1_15 shows the two trends for the first week of January: the real days with the

minimum standard deviation appear with a higher sunshine probability than the mean

calculation one, considering a sample of 20 year.

Fig. 3.3.1_ 15: Sunshine probability calculation for the January reference year month

0,00

0,10

0,20

0,30

0,40

0,50

0,60

0,70

0,80

0,90

1,00

01:

00

03:

00

05:

00

07:

00

09:

00

11:

00

13:

00

15:

00

17:

00

19:

00

21:

00

23:

00

01:

00

03:

00

05:

00

07:

00

09:

00

11:

00

13:

00

15:

00

17:

00

19:

00

21:

00

23:

00

01:

00

03:

00

05:

00

07:

00

09:

00

11:

00

13:

00

15:

00

17:

00

19:

00

21:

00

23:

00

01:

00

03:

00

05:

00

07:

00

09:

00

11:

00

13:

00

15:

00

17:

00

19:

00

21:

00

23:

00

01:

00

03:

00

05:

00

07:

00

09:

00

11:

00

13:

00

15:

00

17:

00

19:

00

21:

00

23:

00

01:

00

03:

00

05:

00

07:

00

09:

00

11:

00

13:

00

15:

00

17:

00

19:

00

21:

00

23:

00

01:

00

03:

00

05:

00

07:

00

09:

00

11:

00

13:

00

15:

00

17:

00

19:

00

21:

00

23:

00

Case study

143

Fig.3.3.1_16 shows the comparison between the day with the minimum variance in January

for the Typical Day Probability calculation and the Sun Shine Probability calculation. The

main difference appears on the central hours of the day: for the TDP calculation the sun hours

between 11.00 and 14.00 are more than the 85%; instead for the SSP calculation, which

described a Gaussian curve, the 85% is only the maximum value reached between 12.00 and

13.00 o’clock. The sun set is the other period in which the more considerable difference has

been detected: the absence of natural light has an hour difference in the two calculation.

C ompa r is on be t w e e n t he d a y w it h t he min imu m v a r in c e in J a n u a r y

0,00

0,10

0,20

0,30

0,40

0,50

0,60

0,70

0,80

0,90

1,00

01:

00

03:

00

05:

00

07:

00

09:

00

11:

00

13:

00

15:

00

17:

00

19:

00

21:

00

23:

00

SSP TDP

Fig. 3.3.1_ 16: Comparison between SSP and TDP calculation for the day with the

minimum variance in January

The Fig.3.3.1_17 shows the same comparison during the first working week of June, when

the main percentage difference has been verified once more at dawn and at sun seat. That

means that in summer months the period with the main variation are not included in the

working period.

CHAPTER 3

144

Fig. 3.3.1_ 17: Comparison between SSP and TDP calculation for the first working week

in June and percentage difference comparison

After the comparison between the summer and winter months (Fig. 3.3.1_18), the standard

deviation of the whole population has been calculated (global yearly calculation): the mean

difference between SSP and TDP during the working period is the 15%, considering the 24

hours is the 0.06%. The square difference is 0.028 (max=0.07 min= 0.004).

Fig. 3.3.1_ 18: Comparison between SSP and TDP calculation for winter and summer

months

Control of the lighting system

The most typical artificial lighting control system for common commercial buildings is a

manual on/off-switching system. With this type of "control system", the light is often

switched on when needed and seldom switched off again when daylight alone is sufficient.

Consequently, the light is often on during the whole working period. To take full advantage

of the potential energy saving, it is essential to provide the building with control systems to

reduce and adjust electric light output to the available daylight. The use of an automatic

Case study

145

lighting control strategy in buildings could reduce unnecessary energy consumption.

With Adeline the energy saving percentage in relation with the refer situation has been

calculated. The “Occupancy Schedule” has been set with a percentage consideration of the

weekend (5 workdays per week), for a working time schedule between 8.00 and 19.00.

For this parameter setting, the 100% operating hours corresponds to an energy demand of

7,128 kWh/working day, considering the luminaries installed (4X18W ceiling lights).

As first the refer situation so calculated has been compared with the real data monitored in

the traditional class rooms 2A, 2B, 2C (Fig3.3.1_18). The energy saving calculated with

Adeline has obtained by the comparison with this reference level.

As Fig.3.3.1_20 shows, the energy consumption estimated in November and December is

lower than the calculated one, instead in the other semester months the refer situation is

always higher. This difference in the operating time can be attributed to the operating time

after the working time schedule verified in ML (light forgotten turned on). It is important to

consider this result in the following sections, where the energy saving recorded and

calculated have been compared for the following lighting operation system:

- Lightswitch On/Off

- Continous Dimming For Reference Point

- Manual On Off Probability

September October November December March April Mai June

Real data monitored

in ML

54,3 151,7 169,3 143,2 77,6 30,9 51,0 40,5

Adeline reference

level

78,4 156,8 149,7 106,9 156,6 128,3 156,8 78,4

Fig. 3.3.1_ 19: Energy consumption (kWh/moth) for the refer situation and for the data

monitored in ML.

CHAPTER 3

146

-40,00

-20,00

0,00

20,00

40,00

60,00

80,00

100,00

Sep Oct Nov Dec Mrz Apr Mai Jun

kWh

Fig. 3.3.1_ 20: Percentage difference between refer situation and ML energy

consumption

Lightswitch On/Off

As the SUPERLINK/RADLINK Technical Manual, IEA SHC Task 21 / ECBCS Annex 29 ADELINE

3.0 Documentation refers, if the daylight illuminance profile Ev(t) has been provided in a time

step ∆t at the reference point for supplementary lighting control in sufficiently small time

steps, it is possible to compute the total switch-on time for on/off-controlled artificial lighting

in the interval from the daylight illuminance frequency distribution and a required rated

illuminance.

So it is necessary to establish how often daylight illuminance is failing to reach the level of

nominal illuminance at the point of reference in the time step. The relative frequency hrel is

determined as follows:

(3.3.1_1)

where

hrel relative frequency

nE<N number of times that daylight illuminance level fell below nominal illuminance at

reference points

nE number of daylight illuminance values at point of reference

Consequently, the total switch-on time of supplementary lighting in the time step ∆t is

obtained from the relative frequency of failures in the rated illuminance level

Case study

147

tB = hrel ⋅ ∆⋅ ∆⋅ ∆⋅ ∆t (3.3.1_2)

where

tB switch-on time of supplementary lighting in the time interval ∆t [ h ]

For on/offs in the space analyzed:

QB,a = tB ⋅ ⋅ ⋅ ⋅ PB (3.3.1_3)

where:

QB,a = energy release into the space due to supplementary lighting within ∆t for on/off

control[ Wh ]

The following graphs show the SUPERLINK output and the comparison between the

calculation results and the monitored data.

Fig. 3.3.1_21 shows the Adeline output result for a Lightswitch on/off control for two task

illuminance levels: 300 lx and500 lx. The total operating time amount in the 2 situations

differs of 257 hours, including mainly November and December. This data corresponds to a

yearly percentage difference of 8.9%.

The Fig. 3.3.1_22 shows the energy consumption comparison between data recorded for

Scenario 1 (classroom equipped with occupancy sensor) and Adeline simulation. This

scenario has been run only in winter semester. By the data analysis the energy saving for this

scenario has been calculated the 40% of the respective energy consumption in ML. On the

basis of this result the summer semester energy consumption has been calculated.

Fig. 3.3.1_ 21: Adeline output results for lightswitch on/off control system

CHAPTER 3

148

lightswith on/off

0

20

40

60

80

100

120

140

160

Sep

tem

ber

Oct

ober

Nov

embe

r

Dec

embe

r

Mar

ch

Apr

il

May

June

kWh

/mo

nth

data measured data calculated with Adeline

Fig. 3.3.1_ 22: Energy saving calculated by Adeline and recorded for Scenario1

0,0

10,0

20,0

30,0

40,0

50,0

60,0

70,0

80,0

90,0

100,0

Sep

tem

ber

Oct

ober

Nov

embe

r

Dec

embe

r

Mar

ch

Apr

il

May

June

%

Fig. 3.3.1_ 23: energy saving percentage difference between lightswitch on/off Adeline

calculation and recorded data.

Continuous Dimming

The use of continuous dimming gives a better controlled combination of artificial and natural

light, and can be almost unnoticeable in use. The saving potential using a dimming control

system is generally higher than switching control systems, because the interior illuminance

generated by electric light is varied in full response to the level of interior daylight.

There are two different continuous dimming control systems in SUPERLINK/RADLINK.

One is the ideal continuous dimming for all reference points; the other is the continuous

dimming system using one (or more) defined reference point. Continuous dimming systems

vary the interior illuminance generated by electric light in response to the level of interior

daylight down to a base load depending of the used ballast type. Conventional ballast has a

base load of 15 - 20%, whereas electronic ballast has base loads of nearly 0%.

In case of supplementary lighting being dimmed, it is necessary to know the difference

Case study

149

between the daylight illuminance level and the nominal illuminance in order to be able to

determine the energy content introduced to the space by permanent supplementary artificial

lighting.

As the SUPERLINK/RADLINK Technical Manual, IEA SHC Task 21 / ECBCS Annex 29 ADELINE

3.0 Documentation reports, in a given time interval ∆t this difference results from:

(3.3.1_4)

Where:

Ev(t) < Evn

Evn nominal illuminance [ lux = lm/m² ]

Ev ( t) illuminance at point of time t [ lux ]

QB,dg base load of dimmed supplement lighting within time step ∆t [ Wh ]

QB,d energy supplied to the space by supplementary lighting within time step ∆t, for

dimmed supplementary lighting [ Wh ]

AN reference area [ m² ]

h luminous efficiency of supplementary lighting [ lm/W ]

Scenario 2 has been implemented with a dimming system based on two detection reference

points and as been assumed as comparison term.

Fig. 3.3.1_ 24: Adeline output results for continuous dimming control system

CHAPTER 3

150

Progressing the same comparison presented above for the Lightswitch On/Off operation

system, the Fig.3.3.1_24 shows the Adeline output result for a Continous Dimming for

Reference Point control for two task illuminance levels: 300 lx and 500 lx. An energy

consumption difference of 2.8% has been calculated, considering the whole year calculation.

The total operating time amount differs of 183 hours, that corresponds to the 6.4% daylight

efficacy difference to support the minimum illuminace level required.

Fig. 3.3.1_ 25: Adeline output for reference illuminance level equal to 300 and 2 luminairs types:

4x14W T8 and 4x18W T5

Fig. 3.3.1_ 26: Adeline output for reference illuminance level equal to 500 and 2 luminairs types:

4x14W T8 and 4x18W T5

Case study

151

Fig. 3.3.1_25 and Fig. 3.3.1_26 show the difference output obtained simulating a continuous dimming system control, using different total power for the luminaries selected. In these cases the situation for the traditional luminaries installed in the traditional class rooms (fluorescent tubes T8, 4x18 W) is compared against to the other one for the automated classroom M2, where new luminaires have been installed (fluorescents tubes T5, 4x14). The total amount of operation time is the same in the two situations. In a first approximation it is possible to conclude, that the two graphs are translated for a saved electrical saving amount of 354 kWh for all the year. The total amount of operating hours is the same in both situations. The Fig. 3.3.1_28 shows the energy consumption comparison between data recorded for Scenario 2.

dimm

0

20

40

60

80

100

120

140

160

Sep

tem

ber

Oct

ober

Nov

embe

r

Dec

embe

r

Mar

ch

Apr

il

May

June

kWh

/mo

nth data measured data calculated with Adeline

Fig. 3.3.1_ 27: Energy saving calculated by Adeline and recorded for Scenario 2

dimm

0,0

10,020,0

30,0

40,0

50,060,0

70,0

80,090,0

100,0

Sep

tem

ber

Oct

ober

Nov

embe

r

Dec

embe

r

Mar

ch

Apr

il

May

June

%

Fig. 3.3.1_ 28: Energy saving percentage difference between Continous Dimming for

Reference Point Adeline calculation and recorded data

CHAPTER 3

152

These results can be improved setting the Shading System. Indeed the SUPERLINK

parameter file can be set as follow: none shading system, fixed shading system,

daylight-optimized shading system.

Fig. 3.3.1_ 29: Adeline output for reference illuminance level equal to 500 and

lightswitch on/off control system, setting the daylight-optimized shading system and

the fixed one with a shading coefficient equal to 0.7

Fig. 3.3.1_ 30: Adeline output for reference illuminance level equal to 500 and reference

point continuous dimming control system, setting the daylight-optimized shading

Case study

153

system and the fixed one with a shading coefficient equal to 0.7

If it is supposed a shading coefficient equal to 0.7 for fixed shadow, the percentage difference

between simulation results and real data follow sensibly as reported in Fig.31. The mean

difference percentage changes:

- from 20% to 13% for the Lightswitch On/Off Scenario;

- from 22% to 15% for the from , Continous Dimming for Reference Point Scenario.

A percentage error can be considered acceptable, considering the probabilistic calculation of

the weather conditions, as reported above.

-10,0

0,0

10,0

20,0

30,0

40,0

50,0

60,0

70,0

80,0

90,0

100,0

Sep

tem

ber

Oct

ober

Nov

embe

r

Dec

embe

r

Mar

ch

Apr

il

May

June

% lightswitch on/off dimming

Fig. 3.3.1_ 31: Energy saving percentage difference for Continous Dimming for

Reference Point and Lightswitch Adeline calculation: results for 0.3 shading factor

setting

Manual On/Off Probability is defined by Hunt and predicts the probability for use of

artificial lighting in a manually operated on/off-switching control system. The method is

based upon patterns of switching behaviour observed in field studies in England. Hunt found

out that the probability of someone switching on the artificial lights in a space is correlated

with the minimum daylight illuminance on the working plane. From the data set of the field

study, an empirical algorithm was defined.

CHAPTER 3

154

Fig. 3.3.1_ 32: Probability of somebody switching on the light on minimum daylight

illuminace level in working area [SUPERLINK/RADLINK Technical Manual, IEA

SHC Task 21 / ECBCS Annex 29 ADELINE 3.0 Documentation]

Running Adeline for the saved electrical energy calculation, the results reported in

Fig.3.3.1_33 have been obtained.

Fig. 3.3.1_ 33: Adeline output results for manual on/off lighting control system

These results do not describe correctly the real situation for the traditional classrooms, as

reported in Fig.3.3.1_34 and Fig.3.3.1_35. Indeed a difference between the 80% and the

160% in winter semester and between 30% and 50% in summer semester has been evaluated.

Case study

155

manual

-50

0

50

100

150

200

Sep

tem

ber

Oct

ober

Nov

embe

r

Dec

embe

r

Mar

ch

Apr

il

May

June

kWh

/mo

nth

data measured data calculated with Adeline

Fig. 3.3.1_ 34: Energy saving calculated by Adeline using a manual light system and

recorded for ML.

0,0

20,0

40,0

60,0

80,0

100,0

120,0

140,0

160,0

180,0

Sep

tem

ber

Oct

ober

Nov

embe

r

Dec

embe

r

Mar

ch

Apr

il

May

June

%

Fig. 3.3.1_ 35: Percentage difference for energy saving calculated by Adeline using a

manual light system and recorded for ML

Fig. 3.3.1_36 represents the occupancy level in the traditional classrooms as a function of the

inside illumination level measured. The typical day derived from the winter semester data

analysis shows that the occupancy level probability grows up significantly for an inside

illuminance level higher than 40 lx.

Fig. 3.3.1_37 shows the probability to have the light system operating as function of inside

illuminance. In particular it is still possible (probability>0,3) to have the light turn on with an

inside illuminance higher than 500 lx.

CHAPTER 3

156

Fig. 3.3.1_ 36: Occupancy level calculation versus inside illuminance for the typical day

Fig. 3.3.1_ 37: Operating time probability as function of the inside illuminance

Conclusion

• The inside illuminance simulated by Adeline is underestimated, because of the user’s

behaviour positioning the curtains. It has been verified that during the lessons the

windows can be partially covered (because they could have been forgotten close by

the previous lesson, because the lesson could include partially projections, sometimes

for any specific reason) so that it would be necessary to select a shading factor = 0,7

to simulate better the real situation

• The manual on/off probability used by Adeline doesn’t correspond to the real

condition monitored. The difference could be explained by the different use of the

room considered: the first one is an office, where a person works alone; the second

one is a classroom where more persons should attend a lesson.

0

20

40

60

80

100

120

0,0

5

0,0

8

2,2

5

3,0

2

5,7

3

6,4

9

12

,3

23

,1

40

,7

56

,4

63

,3

10

5

17

7

28

9

36

8

43

8

60

0

65

4

71

1

76

8

84

4

86

5

91

0

(%) occupancy level in 2B

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

39

45,

6

53,

3

54,

6

55,

5

73,

6

99,

2

125

152

184

207

261

356

428

498

559

591

627

647

659

695

728

740

747

757

764

778

Serie1

Case study

157

3.3.2 Data recording evaluation

All the 6 lecture rooms (traditional and automated) are equipped with occupancy sensors and power consumption meters in order to monitor these parameters separately for each lecture room. The energy saving percentage for each scenario has been calculated in comparison with a reference level of energy demand, hereafter named ML (maximum level). The energy consumption and the occupancy period for the 3 traditional classrooms have been recorded separately, so it has been possible to experimentally define the ML value using these data. In order to understand and model users' behaviour in traditional classrooms (ML) and to quantify the energy saving potentialities of automation systems for lighting control, a typical day data analysis has been carried out. In this case study, the working days for the following different period have been analyzed:

- September/December : winter semester (time A) - March/June : summer semester (time B)

With this division, it will be possible to have a general overview of the energy consumption in the classrooms with and without bus system, in relation with the two different weather conditions. The data results will be presented as monthly data, daily data and as typical day calculation, in order to have different observation levels for the normalization factors calculation. The following section reports a synthesis of the complete data collection and of the graphs elaboration that it is detailed in the ANNEX. To conduct this statistical analysis, it is necessary to consider the conditions operating in the monitored environment during a defined time interval over the course of the day. The data monitored by the input devices of the automation system are collected in an event based manner, so they do not have a regular frequency. Each data recorded is identified by a label. All the data recorded in a day have been separated by typology using this code number. Because of the great number of the data collected every day (more than 100.000) an elaboration data program has been produced to select the data typology. Each vector obtained for each day has been averaged considering a discrete temporal interval of 3 minutes and rewritten in an excel sheet by an other elaboration data program. In this way the data collection results are synchronized and it is possible to relate different factors occurring at the same time in a specific room. The comparison of the lighting energy used in the lecture rooms for the implemented

CHAPTER 3

158

scenarios has been carried out using the standardisation techniques described in the section 2.1.3. In particular the normalisation factors have been calculated as follow for the specific case study: - Occupancy level The presence is checked from the occupancy sensors with IR technology every 30 seconds. This measure is not precise because the bus might have not recorded presence during the 30 seconds even if the students were continuously at their working places. This is the reason why in the presence graphs of the origin data there is no continuous signal. However, for the research interest this detection interval is not significant. Absence period lower than 5 minutes has been considered as occupied, so that Dta,min=5 in (2.1.3_4). It is not possible to calculate or monitor the number of the students in the classroom using the installed smart. Therefore, the spaces are define as 0-occupied or 1-not occupied, independently on the number of students inside the room. The classroom 2M, equipped with 2 occupancy sensors (1 for each half room), has been classified as occupied if at least one half is occupied. - Outside illuminance factor The weather data during the occupancy period has been examined in each classroom in order to verify if it is possible to consider same boundary conditions, in particular with reference to illuminance, for the different rooms monitored. Indeed there is a different time table for the classes of each classroom. - Inside illuminance factor This factor defined in the (2.1.11) for the specific case study of lecture halls has been calculated as shown in equations (3.3.2_1) and (3.3.2_2), in order to quantify the visual comfort guaranteed in the traditional classrooms in relation with the inside illuminance level defined in compliance with the regulations UNI 10380/97 and UNI EN 12464-2/2004

(3.3.2_1)

(3.3.2_2)

−−= ∑ =

N

jjE

NDF1300 30030011

−−= ∑ =

N

jjE

NDF1500 50050011

Case study

159

OCCUPANCY LEVEL (winter semester) The scenarios implemented in Time A are in section 3.1.3. The PFi, defined in 2.1.3, derived by a monthly analysis have a value oscillation included between +7%, as reported in 3.3.2_1, but the value is not homogeneous during each month, so that any direct correlation between the lesson timetable and the monthly Presence Factor calculation has been detected.

0,85

0,9

0,95

1

1,05

1,1

SEPTEMBER OCTOBER NOVEMBER DECEMBER

PF1 PF2 PF3

Fig.3.3.2_ 1: Monthly value of the Presence Factor for the 3 scenario in winter semester

As for the monthly analysis, checking the daily calculation of the PFi, the energy consumption normalized by presence does not have a homogeneous distribution day by day. In this case, the maximum daily influence of the PF normalization could reach much higher values, as expressed in Fig.3.3.2_2: the maximum daily variation for the PF calculation can affect for a value included between the 40% and the 280% of the energy use recorded.

0

50

100

150

200

250

300

350


PF1 PF2 PF3

Fig.3.3.2_ 2: Modulus of the maximum percentage variation in the energy consumption normalization by PF value

CHAPTER 3

160

For instance in November for the scenario 2, the absolute energy consumption on 2 November (PFi=1.54) is lower than the one normalized by presence; on 22 and 27 November the energy consumption normalized by presence is lower than the absolute one (respectively PFi=0.79 and 0.77). In fact, the presence period on 2 November is 5.7 hours for scenario 2 and 8.2 for the ML; on 22 and 27 November there are around 3 hours presence difference. The daily occupancy hours do not correspond to the lessons timetable.

0,000,020,040,060,080,100,120,140,160,180,200,220,24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 18 19 27 28 29 30 day

kWh/

m2

day absolute normalized by presence

Fig.3.3.2_ 3: Energy consumption in November: scenario 2

0,00

2,00

4,00

6,00

8,00

10,00

12,00

14,00

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 18 19 27 28 29 30 day

hour

SCENARIO 1 SCENARIO 2 SCENARIO 3 ML

Fig.3.3.2_ 4: Occupancy period in November in the 3 automated classrooms and in maximum level

Case study

161

0

10

20

30

40

50

60

70

80

90

100

6:00

6:45

7:30

8:15

9:00

9:45

10:3

0

11:1

5

12:0

0

12:4

5

13:3

0

14:1

5

15:0

0

15:4

5

16:3

0

17:1

5

18:0

0

18:4

5

19:3

0

20:1

5

21:0

0

21:4

5

time

%

october november dicember mean

Fig.3.3.2_ 5: Presence level in percentage for traditional classrooms

Fig.3.3.2_5 shows, for the traditional (manually operated) classrooms, the occupancy results (presence level in %) as a function of the time of the day for 3 different months as well as the corresponding mean value.

0

10

20

30

40

50

60

70

80

90

100

6:00

6:45

7:30

8:15

9:00

9:45

10:3

0

11:1

5

12:0

0

12:4

5

13:3

0

14:1

5

15:0

0

15:4

5

16:3

0

17:1

5

18:0

0

18:4

5

19:3

0

20:1

5

21:0

0

21:4

5 t ime

%

ML scenario2 scenario 3 scenario 1

Fig.3.3.2_ 6: Mean presence value in the traditional and automated classrooms.

Fig. 3.3.2_6 compares the mean presence value in the traditional classrooms with the presence levels for the 3 automated control scenarios. As Fig. 3.3.2_5 and 3.3.2_6 demonstrate, the difference between the monitored occupancy levels of the traditional classrooms and those with automated control scenarios involves especially the afternoon hours. Thus, it was necessary to normalize the energy use values based on the occupancy data so as to make a comparison between various scenarios possible.

CHAPTER 3

162

OUTSIDE ILLUMINANCE LEVEL (winter semester) The IFi, defined in 2.1.3, derived by a monthly analysis have a maximum value of +11%, as reported in Fig.3.3.2_7, but its averaged value is about +5%.

0,8

0,85

0,9

0,95

1

1,05

1,1

1,15


IF1 IF2 IF3

Fig.3.3.2_ 7: Monthly value of the outside Illuminance Factor for the 3 scenario in winter semester

Checking the daily calculation of the IFi, the energy consumption, normalized by outside illuminance, has a homogeneous distribution day by day, excepted in some few days. For instance in December for the scenario 2, the energy consumption normalized only by presence on 4 December (IFi=1.16) is lower than the one normalized by presence and by outside illuminance. For the other days of the month examined, the difference is lower, around the 5%. By a daily data analysis, it is not possible to establish if the outside illuminance difference is to relate to a specific timetable during the day or if it is homogeneously distributed.

0,00

5.000,00

10.000,00

15.000,00

20.000,00

25.000,00

30.000,00

35.000,00

40.000,00

45.000,00

50.000,00

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 day

lux


Fig.3.3.2_ 8: Mean outside illuminance level during occupancy period in October

Case study

163

0,000,020,040,060,080,100,120,140,160,180,200,220,240,260,280,300,32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 day

kWh/

m2

day

normalized by presence normalized by out. illuminance

Fig.3.3.2_ 9: Energy consumption in December scenario 2

0

10000

20000

30000

40000

50000

60000

6.00

7.00

8.00

9.00

10.0

0

11.0

0

12.0

0

13.0

0

14.0

0

15.0

0

16.0

0

17.0

0

18.0

0

19.0

0

20.0

0

21.0

0

22.0

0 time

lx OCTOBER NOVEM BER DECEM BER

Fig.3.3.2_ 10: Outdoor illuminance levels during the occupancy period in the traditional classrooms

Fig. 3.3.2_10 illustrates the course of the outdoor illuminance levels for three reference days during the times where the traditional classroom where occupied.

CHAPTER 3

164

0

10000

20000

30000

40000

50000

60000

6:00

7:00

8:00

9:00

10:0

0

11:0

0

12:0

0

13:0

0

14:0

0

15:0

0

16:0

0

17:0

0

18:0

0

19:0

0

20:0

0

21:0

0

22:0

0

time

lx ML SCENARIO 1 SCENARIO2 SCENARIO 3

Fig.3.3.2_ 11: Outdoor illuminance levels for occupancy hours: ML, scenarios 1,2 and 3 in October

Fig. 3.3.2_11 shows, for the month of October, the outdoor illuminance levels for occupancy hours of ML as well as for occupancy hours of scenarios 1, 2 and 3. The occupancy period differs for each classroom, depending on the lecture schedule. A normalization was performed regarding the available outdoor illuminance levels (see Fig. 3.3.2_10 and 3.3.2_11), even though the variations of illuminance levels for various scenarios were not significant throughout the observation period. Indeed, the difference in the outside illumination conditions is concentrated in the period included between 12.00 and 15.00, when there is the main natural light contribution. INSIDE ILLUMINANCE LEVEL (winter semester)

0,000

0,100

0,200

0,300

0,400

0,500

0,600

0,700

0,800

0,900

1,000

1 4 5 6 7 11 12 13 14 15 18 19 20 21 22 day

kWh/

m2

day

inside illuminance normalisation factor-500 inside illuminance normalisation factor-300

Fig.3.3.2_ 12: Visual comfort index distribution in November

Case study

165

The DF300,500 derived by a monthly analysis grow up from September (DF500=11%) to December (DF500=33%), as reported in the calculation sheets reported in the ANNEX. Checking the daily calculation of the DFi, there is a distribution of this factor correlated with the outside illuminance condition. For instance in December, as reported in Fig.3.3.2_12, the darker days correspond to the higher value of DF: on 7 December (DF500=0.55 and DF300=0.71), when the mean outside illuminance level during the presence period is 2355 lx; on 11 December (DF500=0.75 and DF300=0.83), the mean outside illuminance level during the presence period is 15201 lx. The DF is the more significant normalization factor compared with the other one used in this calculation. However considering a daily data analysis, it is not possible define when this factor has the main incidence. By the analysis of the typical day, it has been verified that the higher influence of the DF on the energy calculation is concentrated in the early morning, during the first lesson hour, and in afternoon after 15.30, as the Fig.3.3.2_13 shows.

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1

6.00

6.45

7.30

8.15

9.00

9.45

10.3

0

11.1

5

12.0

0

12.4

5

13.3

0

14.1

5

15.0

0

15.4

5

16.3

0

17.1

5

18.0

0

18.4

5

19.3

0

20.1

5

21.0

0

21.4

5

i,300 i,500

Fig.3.3.2_ 13: Discomfort Index value in November

.

CHAPTER 3

166

0

100

200

300

400

500

600

700

800

900

1000

6:00

7:00

8:00

9:00

10:0

0

11:0

0

12:0

0

13:0

0

14:0

0

15:0

0

16:0

0

17:0

0

18:0

0

19:0

0

20:0

0

21:0

0

22:0

0

time

lx

OCTOBER NOVEMBER DECEMBER mean

Fig.3.3.2_ 14: Indoor illuminance levels during the occupancy period in the traditional

classrooms

Fig. 3.3.2_14 illustrates the course of the indoor illuminance levels for three reference days during the times where the traditional classroom where occupied. The inside illuminance level is lower than 300 lx before 10:00 and after 15:30, in correspondence with the results exposed before about the DF value, considering an average value for the whole winter semester, but it is important to notice that this value changes in relation with the outside illuminance condition, decreasing from October to December. At the same time, the inside illuminance level can be different for each classroom.

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

6:00

7:00

8:00

9:00

10:0

0

11:0

0

12:0

0

13:0

0

14:0

0

15:0

0

16:0

0

17:0

0

18:0

0

19:0

0

20:0

0

21:0

0

22:0

0 time

lx

2A 2B 2C ML simulation data

Fig.3.3.2_ 15: Indoor illuminance levels during occupancy period in the traditional classrooms in October

Fig. 3.3.2_15 shows, concerning October, the course of the indoor illuminance levels for occupancy hours of the three traditional classrooms (2A, 2B, and 2C) and the associated

Case study

167

mean value (ML). As Fig.s 3.3.2_14 and 3.3.2_15 show, the indoor illuminance levels in the three traditional classrooms differ from each other and are different in different months. Specifically, at certain times early in the morning or late in the evening the illuminance levels are rather low. This is due to the fact that the users either did not switch on the lights or did not open the shading devices. We thus considered this circumstance in the comparison of the respective energy use values based on the aforementioned normalisation procedure considering the actually prevailing illuminance levels. ENERGY CONSUMPTION (winter semester)

The energy consumption in traditional classrooms can be extremely different in the 3 traditional classrooms monitored, as it is possible verify for instance on 2 and 3 November in classroom 2C. Often this data is not related to the use period of the room, as in the previous example, for which has been detected respectively 10 and 9 hours presence on 2 November for the classrooms 2A and 2B, indeed only 6 in 2C; 6.5 and 8.5 hours presence on 3 November for the classrooms 2A and 2B, indeed 10 in 2C.

In these data analysis only the average value ML has been compared with the energy consumption in automated classrooms, but it is possible to check the energy use in each traditional classroom in the graphs reported in the ANNEX.

0,00

0,10

0,20

0,30

0,40

0,50

0,60

0,70

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 18 19 27 28 29 30 day

kWh/

m2 da

y

2A 2B 2C ML

Fig.3.3.2_ 16: Energy consumption in traditional classroom November

Considering the energy consumption value in the traditional classrooms month by month during the winter semester, a growing trend has been recorded (Fig.3.3.2_17), but this result is not verified in the data analysis of each single traditional classrooms. In fact in the classroom 2C, for instance, an improper use of artificial light is recorded in September, in comparison with the one used in the same classroom in October and in comparison with the

CHAPTER 3

168

energy consumption in the September in the classrooms 2A and 2B (Fig.3.3.2_18). Comparing the energy consumption for each traditional classroom during the same month

a considerable difference (included between 2 and 8 Wm-2) has been verified.

0

0,05

0,1

0,15

0,2

0,25

0,3

SEPT

EMBE

R

OC

TOBE

R

NO

VEM

BER

DEC

EMBE

R

kWhm

-2/d

ay

Fig.3.3.2_ 17: Daily energy consumption in ML per working day for each the winter semester months

0

5

10

15

20

25

30

35

2A 2B 2C

Wm

-2


Fig.3.3.2_ 18: Daily energy consumption in ML in the winter semester months

Case study

169

0

0,2

0,4

0,6

0,8

1

1,2

1,4

1,6

1,8

6:00

7:00

8:00

9:00

10:0

0

11:0

0

12:0

0

13:0

0

14:0

0

15:0

0

16:0

0

17:0

0

18:0

0

19:0

0

20:0

0

21:0

0

22:0

0 time

Wh/

m2

october november dicember

Fig.3.3.2_ 19: Mean lighting energy use in the traditional classrooms

Fig. 3.3.2_19 shows the mean lighting energy use in the traditional classrooms as a function of the time of the day for three different months. The energy consumption in traditional classrooms is not always directly correlated with outdoor illuminance levels. In fact, the energy consumption in the morning and early afternoon hours of October is higher than in the respective hours in November even if the outside illuminance conditions are more favorable in October. This value is confirmed by the energy consumption normalized by occupancy data analysis. Nonetheless, the mean energy use over the whole observation period is clearly related to outside illuminance level (see Fig. 3.3.2_19).

0

0,2

0,4

0,6

0,8

1

1,2

1,4

1,6

6:00

6:42

7:24

8:06

8:48

9:30

10:1

2

10:5

4

11:3

6

12:1

8

13:0

0

13:4

2

14:2

4

15:0

6

15:4

8

16:3

0

17:1

2

17:5

4

18:3

6

19:1

8

20:0

0

20:4

2

21:2

4 time

Wh/

m2

mean EC 2A EC 2B EC 2C

Fig.3.3.2_ 20: Energy use during the winter semester

CHAPTER 3

170

Fig. 3.3.2_20 shows, for the winter semester, the energy use in each of the mean of the 3

traditional classrooms and the associated mean value. There is a considerable difference in the energy use for scenario 1 and 3. It is possible to

verify a higher daily absolute energy use in scenaio1 than in ML, instead the scenario3 has always lower value (Fig.3.3.2_21).

0,00

0,05

0,10

0,15

0,20

0,25

0,30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

kWh/

m2

day

ML scenario 1 scenario 3

Fig.3.3.2_ 21: Absolute energy consumption in November

Comparing the energy demand normalized by the 3 normalization factors, the energy use in ML is higher than each automated scenarios (Fig.3.3.2_22). The only exception is verified when the occupancy period in one of the automated classroom is very low (for instance on 2 November in scenario 3).

0,00

0,05

0,10

0,15

0,20

0,25

0,30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 18 19 27 28 29 30 day

kWh/

m2

day


Fig.3.3.2_ 22: Daily energy consumption in November

Case study

171

0

0,2

0,4

0,6

0,8

1

1,2

1,4

6:00

7:00

8:00

9:00

10:0

0

11:0

0

12:0

0

13:0

0

14:0

0

15:0

0

16:0

0

17:0

0

18:0

0

19:0

0

20:0

0

21:0

0

22:0

0

time

Wh/

m2


Fig.3.3.2_ 23: Lighting energy use for scenarios 1 for three different months.

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1

6:00

7:00

8:00

9:00

10:0

0

11:0

0

12:0

0

13:0

0

14:0

0

15:0

0

16:0

0

17:0

0

18:0

0

19:0

0

20:0

0

21:0

0

22:0

0 time

Wh/

m2


Fig.3.3.2_ 24: Lighting energy use for scenarios 3 for three different months.

Fig.s 3.3.2_23 and 3.3.2_24 show (for scenarios 1 and 3 respectively) the lighting energy

use as a function of the time of the day for three different months together with the mean value.

The monitored energy use patterns in the classrooms with automated lighting control scenarios show a better daylight utilization level (Fig. 3.3.2_23 and 3.3.2_24). The energy use for scenario 1 is higher compared to scenario 3, especially during the classes in the evening, since maintaining the minimum illuminance level (in scenario 1) requires all luminaries to be switched on (Fig. 3.3.2_23).

CHAPTER 3

172

0102030405060708090

100110120130140150160170180190200210220230240250

6:00

7:00

8:00

9:00

10:0

0

11:0

0

12:0

0

13:0

0

14:0

0

15:0

0

16:0

0

17:0

0

18:0

0

19:0

0

20:0

0

21:0

0

22:0

0 time

W/m

2 ML SCENARIO2 SCENARIO 3 SCENARIO 1

Fig.3.3.2_ 25: Occupancy-normalized energy use for ML, scenarios 1, 2, and 3.

Fig. 3.3.2_25 illustrates the occupancy-normalized energy use in the traditional classrooms (ML) as well as in the automated classrooms (for scenarios 1, 2, and 3). Occupancy-normalized energy use in traditional classrooms is most of the time higher than those with automated scenarios (see Fig. 3.3.2_12). In morning and early afternoon hours available daylight is more efficiently used by the automated scenarios. After 19:00, the automated scenarios consider – in contrast to traditional classrooms – the absence of occupants and switch off the luminaires accordingly.

0

10

20

30

40

50

60

70

80

6:00

7:00

8:00

9:00

10:0

0

11:0

0

12:0

0

13:0

0

14:0

0

15:0

0

16:0

0

17:0

0

18:0

0

19:0

0

20:0

0

21:0

0

22:0

0

time

W/m

2

energy consumption/occupancy period en.cons. noralized by i,300 en.cons. normalized by i,500

Fig.3.3.2_ 26: Energy use for scenario 2 normalized for occupancy level and for

occupancy and indoor illuminance levels.

Case study

173

Fig. 3.3.2_26 shows the energy use for scenario 2 normalised for i) occupancy level and ii) occupancy and indoor illuminance level. As mentioned earlier, the monitored energy use levels for the traditional classrooms and the ones with automated lighting control scenarios could not be directly compared, given differences in occupancy and the maintained illuminance levels. To illustrate the effect of maintained illuminance levels, Fig. 3.3.2_26 shows the monitored energy use for scenario 2 together with normalized curves for maintained illuminance levels of 300 lx and 500 lx. This normalization was performed based on equations 1 and 2 respectively. The effect of normalization is understandably more significant in the evening hours.

0

0,2

0,4

0,6

0,8

1

1,2

6:00

7:00

8:00

9:00

10:0

0

11:0

0

12:0

0

13:0

0

14:0

0

15:0

0

16:0

0

17:0

0

18:0

0

19:0

0

20:0

0

21:0

0

22:0

0 time

Wh/

m2

ML SCENARIO 2 SCENARIO 3 SCENARIO1

Fig.3.3.2_ 27: Comparison between ML energy use and that one for the three scenarios.

Fig. 3.3.2_27 compares the mean energy use in the traditional classrooms with the energy use for the three scenarios for a typical day.

0,00

10,00

20,00

30,00

40,00

50,00

60,00

70,00

80,00

90,00

100,00

MEAN SEPTEMBER OCTOBER NOVEMBER DECEMBER

% SCENARIO1 SCENARIO2 SCENARIO3

Fig.3.3.2_ 28: Energy saving of the three scenarios in the winter semester

CHAPTER 3

174

Fig. 3.3.2_28 shows the energy saving of the three scenarios for the three months as

percentage of the energy use in the traditional classrooms (ML). The overall energy use comparison between the traditional and automated classrooms

shows that, depending on the observation month in winter semester, a) scenario 1 requires 38% to 48% less energy for electrical lighting b) scenarios 2 requires 62% to 85% less energy for electrical lighting c) scenarios 3 consume 51% to 78% less energy for electrical lighting The percentage of energy saving has been converted in 2 values correlated to environmental and economic factors: CO2/m2 and €/m2. Considering for instance the Scenario 3 in October, the 76.36% of energy saving can be converted in KWh/m2 per month and in gram of CO2/m2

and €/m2 as follow:

ENERGY SAVING IN OCTOBER: SCENARIO 3

76,32%

61,05%55,21%

53,45%

0,00

0,50

1,00

1,50

2,00

2,50

3,00

3,50

absolute normalized bypresence

normalized by out.illuminance

normalized by ins.illuminance

Kwh/

m2 m

onth

SUSTAINABILITY EVALUATION IN OCTOBER: SCENARIO 3

€ 0,30 € 0,31€ 0,34

€ 0,43

0,00

500,00

1000,00

1500,00

2000,00

2500,00

absolute normalized bypresence

normalized byout.

illuminance

normalized byins.

illuminance

gram

ms

of C

O2/

m2 m

onth

€ /m

2 mon

th

Fig.3.3.2_ 29: Energy consumption in October, scenario 3

This means that for the winter semester the energy saving percentage for each scenario, if implemented in the complete building section automated (all the 3 classrooms), corresponds to the following values:

KWh/m2 gr CO2/m2 Kgr CO2 €/m2 € SCENARIO1 5,65 3277,86 524,46 0,82 131,91 SCENARIO2 9,98 5787,38 925,98 1,46 232,90 SCENARIO3 8,61 4996,09 799,37 1,26 201,06

RESULT COMPARISON BETWEEN WINTER AND SUMMER SEMESTER The scenarios implemented in Time B are reported in section 3.1.3. It is important to notice that in this second detection period the scenario 1 is changed (dimming regulation in three rows for three channels distinctly); on the contrary the scenario

Case study

175

2 and 3 are maintained the same. For the summer semester, the energy saving in KWh/m2 for each scenario, considering the implementation of each system in the complete building section automated (all the 3 classrooms) corresponds to the following values: KWh/m2 gr CO2/m2 Kgr CO2 €/m2 € SCENARIO1bis 3,59 2079,36 332,70 0,52 83,68SCENARIO2 2,93 1698,81 271,81 0,43 68,36SCENARIO3 3,62 2101,55 336,25 0,53 84,57

As expected, there is a reduction of the total amount of energy demand both in traditional and automated classrooms and, as consequence, an absolute diminution of the saved energy in KWh/m2.

This depends on two factors: the occupancy level decreases and the outside illuminance level increases. Fig.3.3.2_31 shows that the mean presence level in summer semester is lower than the mean value in winter semester. The presence value in April affects mainly this result; indeed the presence percentage in March and May is lower than the mean winter value only during the period between 16.00 and 18.00, during the other part of the day the presence value for the 2 semesters is comparable. The presence value for automated classrooms is significantly lower than the one for ML: from 13.30 until 16.30 there is a difference of 10% for each scenario, this value reaches the 20% for the scenario1 in the morning hours.

0

10

20

30

40

50

60

70

80

90

100

6.00

6.45

7.30

8.15

9.00

9.45

10.3

0

11.1

5

12.0

0

12.4

5

13.3

0

14.1

5

15.0

0

15.4

5

16.3

0

17.1

5

18.0

0

18.4

5

19.3

0

20.1

5

21.0

0

21.4

5

t ime

%

medium in w inter semester medium in summer semestermarch aprilmay

0

10

20

30

40

50

60

70

80

90

100

6.00

.00

6.45

.00

7.30

.00

8.15

.00

9.00

.00

9.45

.00

10.3

0.00

11.1

5.00

12.0

0.00

12.4

5.00

13.3

0.00

14.1

5.00

15.0

0.00

15.4

5.00

16.3

0.00

17.1

5.00

18.0

0.00

18.4

5.00

19.3

0.00

20.1

5.00

21.0

0.00

21.4

5.00t ime

%

medium in summer semester scenario1 scenario 2 sceanrio3

Fig.3.3.2_ 30: ML presence levels in percentage in summer semester months and mean presence value in ML and in each scenario.

The second correlation confirms as well the results obtained in the winter data analysis about the outside illuminance and the energy demand, as reported in the graphs in the ANNEX. In summer semester, as in winter semester, the correspondence between low illuminance

CHAPTER 3

176

level and high-energy demand is respected in traditional classrooms only considering the mean value in the whole observation period (Fig. 3.3.2_31) Looking month by month the correlation between energy consumption and outside illuminance level, there is not always a direct correspondence. Indeed the energy demand in April is lower than the one in May (Fig.3.3.2_33). In this case, the detected difference in summer semester is not only concentrated in the morning/early-afternoon hours, but also involves the evening hours.

COMPARISON OUTSIDE LUX AND ENERGY DEMAND

0

10000

20000

30000

40000

50000

60000

6:00

6:36

7:12

7:48

8:24

9:00

9:36

10:1

2

10:4

8

11:2

4

12:0

0

12:3

6

13:1

2

13:4

8

14:2

4

15:0

0

15:3

6

16:1

2

16:4

8

17:2

4

18:0

0

18:3

6

19:1

2

19:4

8

20:2

4

21:0

0

21:3

6

lx

0

0,2

0,4

0,6

0,8

1

1,2

W/m

2ML energy demand

outside illuminance

Fig.3.3.2_ 31: Correlation between outside illuminace level and energy demand, for traditional classrooms in winter semester

0

10000

20000

30000

40000

50000

60000

70000

6.00

.00

7.00

.00

8.00

.00

9.00

.00

10.0

0.00

11.0

0.00

12.0

0.00

13.0

0.00

14.0

0.00

15.0

0.00

16.0

0.00

17.0

0.00

18.0

0.00

19.0

0.00

20.0

0.00

21.0

0.00

22.0

0.00

time

lux

SEPTEM BER OCTOBER NOVEM BER DECEM BER M ARCH M AY

Fig.3.3.2_ 32: Outside illuminace level in each detected month.

Case study

177

0

0,2

0,4

0,6

0,8

1

1,2

1,4

6.00

7.00

8.00

9.00

10.0

0

11.0

0

12.0

0

13.0

0

14.0

0

15.0

0

16.0

0

17.0

0

18.0

0

19.0

0

20.0

0

21.0

0

22.0

0 time

Wh/

m2

march april may mean in w inter semester mean in summer semester

Fig.3.3.2_ 33: ML energy consumption in summer months and mean winter semester value

0

5

10

15

20

25

30

35

40

45

50

6.00

7.00

8.00

9.00

10.0

0

11.0

0

12.0

0

13.0

0

14.0

0

15.0

0

16.0

0

17.0

0

18.0

0

19.0

0

20.0

0

21.0

0

22.0

0 time

W/m

2

march april may

Fig.3.3.2_ 34: Occupancy-normalized energy use for ML in summer months

By the analysis of the typical day energy use normalized by occupancy, the power used in May is higher than the one in April between 14.00 and 18.00. Indeed the values are comparable between 10.00 and 14.00 (Fig. 3.3.2_34). The inside illuminance level is connected to this result: indeed values for May are lower than those for April (Fig.3.3.2_35). The shatter system used is the parameter that affects the indoor natural light level.

CHAPTER 3

178

INSIDE ILLUMINANCE IN OCCUPANCY PERIODE: SUMMER SEMESTER

0

100

200

300

400

500

600

700

800

900

time

lux

MARCH

APRIL

MAY

Fig.3.3.2_ 35: Indoor illuminance level during the occupancy period in the traditional classrooms in summer semester

In summer semester, the influence of the DF is less relevant than in winter semester (Fig.3.3.2_36).

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1

6.00

6.45

7.30

8.15

9.00

9.45

10.3

0

11.1

5

12.0

0

12.4

5

13.3

0

14.1

5

15.0

0

15.4

5

16.3

0

17.1

5

18.0

0

18.4

5

19.3

0

20.1

5

21.0

0

21.4

5DF,300 DF,500

Fig.3.3.2_ 36: Discomfort Index value in April

Case study

179

0,00

10,00

20,00

30,00

40,00

50,00

60,00

70,00

80,00

90,00

100,00

MEAN MARCH APRIL MAY JUNY

% SCENARIO1 SCENARIO2 SCENARIO3

Fig.3.3.2_ 37: Energy saving of the three scenarios in the summer semester

The overall energy use comparison between the traditional and the automated classrooms shows that, depending on the observation month in summer semester, a) scenario 1 requires 55% to 75% for electrical lighting less energy b) scenarios 2 requires 40% to 65% for electrical lighting less energy c) scenarios 3 consume 48% to 82% for electrical lighting less energy. Comparing winter and summer trends, the mean energy saving for scenario 3 is above the 70% for the whole year. Checking the difference between scenario 1 and 3 in summer semester, there is not any significant improvement of the system efficiency. This means that any effectiveness is detected in the lecture room controlled in terms of two separate spatial zones (front and back) with dedicated occupancy sensors. The mean difference between winter and summer semester performance is mainly significant for the scenario 2: indeed the energy saving percentage for this scenario deceases from the 75% in winter semester to the 55% in summer semester.

CHAPTER 3

180

0

0,1

0,2

0,3

0,4

0,5

0,6

6.00

7.00

8.00

9.00

10.0

0

11.0

0

12.0

0

13.0

0

14.0

0

15.0

0

16.0

0

17.0

0

18.0

0

19.0

0

20.0

0

21.0

0

22.0

0 time

Wh/

m2

mean in summer semester scenario1 scenario2 scenario3

Fig.3.3.2_ 38: Comparison between ML energy use and the one for the three scenarios

0

5

10

15

20

25

30

35

40

45

50

6.00

6.36

7.12

7.48

8.24

9.00

9.36

10.1

2

10.4

8

11.2

4

12.0

0

12.3

6

13.1

2

13.4

8

14.2

4

15.0

0

15.3

6

16.1

2

16.4

8

17.2

4

18.0

0

18.3

6

19.1

2

19.4

8

20.2

4

21.0

0

21.3

6

ML scenario3 scenario2 scenario1

Fig.3.3.2_ 39: Occupancy-normalized energy use for ML, scenario 1, 2 and 3

The energy consumption of the scenario 3 differs mainly in the first hour lesson from the scenario 1 and 2 (Fig.3.3.2_38-39). This depends again on the use of the shadow system that is improper and inefficient especially for this classroom. The April energy demand in scenario 1 affects primary this result, as the Fig. 3.3.2_40 shows.

Case study

181

0

20

40

60

80

100

120

140

6.00

7.00

8.00

9.00

10.0

0

11.0

0

12.0

0

13.0

0

14.0

0

15.0

0

16.0

0

17.0

0

18.0

0

19.0

0

20.0

0

21.0

0

22.0

0time

W/m

2 march april may

Fig.3.3.2_ 40: Occupancy-normalized energy use for scenario 3 in winter semester months

Conclusion Considering the whole observation period the energy consumption percentage is:

• 40%, if all three rows of luminaries are switched on when occupancy is detected and the room illuminance level (measured in the middle of the room) is below a predefined minimum level (scenario 1- winter semester)

• 65%, if three rows of luminaries are switched on when occupancy is detected and

dimmed (in two separately controlled circuits) so as to provide predefined minimum illuminance levels, measured at two points in the room, each corresponding to a circuit (scenario 2). This lighting system has a better performance in winter than in the summer.

• 65%, if the three rows of luminaries are switched on when occupancy is detected and

dimmed (in three separately controlled circuits) so as to provide predefined minimum illuminance levels (measured in three points in the room, each corresponding to a circuit). There are any significant different results if the lecture room is controlled in terms of two separate spatial zones (front and back) with dedicated occupancy sensors.

CHAPTER 3

182

From October to December the energy saving percentage decreases, indeed it increases from March to April, in accordance with the outside illumination level. The disability and discomfort glare influence this trend, because an improper use of the shadow system often causes the unnecessary use of artificial light. The minimum illuminance level value maintained in the traditional classrooms is low, especially in winter semester in the early morning and late in the afternoon. The use of automation system guarantees a more efficient use of the artificial light and a higher visual comfort level. TOTAL RESULT KWh/m2 gr CO2/m2 Kgr CO2 €/m2 € SCENARIO1 10,67 6190,59 990,50 1,56 249,13 SCENARIO2 20,28 11764,01 1882,24 2,96 473,42 SCENARIO3 19,23 11153,44 1784,55 2,81 448,85

0

20

40

60

80

100

120

140

160

180


kWh/

mon

th monitored DIN18599 EN15193

0

10

20

30

40

50

60

70

80

Jan

Feb

Mrz

Apr

Mai

Jun

Jul

Aug

Sep Okt

Nov

Dez

LIG

HTS

WIT

CH

ON

/OFF

% DIN18599 EN15193 monitored

Fig.3.3.2_ 41: Energy consumption comparison: calculation data and monitoring results (left); energy saving percentage: monthly data for dimming control and yearly datum

for lightswitch control (right).

It is important to note the difference between the energy consumption calculated by DIN18599 and pr EN15193 in comparison with the data monitored: both the standards estimate an energy demand too low in winter months and too high in summer (Fig.3.3.2_41 left). Moreover, the energy saving calculated in compliance with the two standards is lower than the one monitored for the winter semester, instead it is much higher in the summer semester (Fig.3.3.2_41 left).

3.3.3 Visual comfort test evaluation

The visual comfort sheet defined in the section 6.1 has been tested on a population of 135 students. In order to have a complete overview of different lesson managements and environmental boundary conditions, a specific timetable for the sheet distribution has been organized, including different inside and outside illuminance conditions. For this reason classes during the 1st hour (8.30-9.30), during the 5th, 6th hour (12-30-14.30) and during the last hours (16.30-18.30) have been selected. Both the classrooms typology (with the automated lighting system and the traditional one) have been included in the investigation. In the following paragraphs the results of the visual comfort test have been presented and discussed.

02/05/2007 03/05/2007 8.30 2A 2M 9.30

10.30 2B 11.30 12.30 2M 13.30 2N 14.30 15.30 16.30 17.30 2M 18.30 2C 19.30

Fig. 3.3.3_ 1: Time table definition for the visual comfort evaluation

1. Boundary condition definition The 43.3% of the students sample analyzed does not have any focusing problems diagnosed. The 37.7% is affected by Myopia (34%<6 dioptres, 3.7%>6 dioptres), the 17.5% by Astigmatism and only the 1.5% by Hyperopia. This data is to be compared with the detail detection prove, to verify that there is not any eyes pathology not diagnosed.

FOCUSING PROBLEMS

43,3%

37,7%

1,5%

17,5%1

2

3

4

CHAPTER 7

184

n°

STUDENTS IN % n° STUDENTS IN % n°

STUDENTS IN % IN %

1.no corrections 58 43,28 0 0 43,3 2.Myopia 0 45,5 33,95522 5 3,731343284 37,7 3.Hyperopia 0 1 0,746269 1 0,746268657 1,5 4.Astigmatism 0 20,5 15,29851 3 2,23880597 17,5

Fig. 3.3.3_ 2: Focusing problems

The students sample analyzed is mostly 23 years old, so that, in conformity with the calculation of the AF factor (2.3.11), there is no significant influence of the user’s age in the visibility performance for the case stady presented.

STUDENTS AGE

67,2%

17,9%3,7%

11,2%

1

2

3

4

n° STUDENTS IN % n° STUDENTS IN % n° STUDENTS IN % <23 90 67,16 0 0 24-25-26(or more) 15 11,19 24 17,91 5 3,73

Fig. 3.3.3_ 3: Student age

The 44% of the students occupies the first 3 desks, and the 23% the farther part. It is important to note that this is the mean data including both the classrooms sites: 6m length (in which there are only 6 desks lines) and 12m length. Looking the single classroom report, it is possible to observe that the same number of students is distributed in the first and in the second half of the space, even if there are many free places closed to the blackboard, considering the number of the students present in the room and the number of the places available. See for instance the evaluation in the classroom 2M at 8.30 o’clock (ANNEX…), where the 38% occupies the first section (violet in Fig. 3.3.3_4) and the 42% the last section (yellow in Fig. 3.3.3_5), even if there are only 29 students in a room for 72 study places. For the calculation of the Visibility, the maximum distance has been assumed, as critical position (10m and 5m).

Case study

185

DISTANCE FROM THE TASK AREA

44%

33%

23%1

2

3

Fig. 3.3.3_ 4: Student position: distance from the blackboard

The student distribution in relation with the window distance is uniform, so that it is important to consider the disability glare possibility for the place closed to the part influenced directly by the sunshine (Fig. 3.3.3_5).

DISTANCE FROM THE WINDOW SITE

28%

33%

39% 1

2

3

Fig. 3.3.3_ 5: Student position: distance from the windows

2. Critical detail definition Considering the detail definition for the boundary conditions present in the indoor environment during the lesson, the following results have been recorded: only the 50% of the students can read without difficulties the target that represents the 9/10 definition, for a white background. A quite similar result (49%) has been obtained for a white background. That means that the critical dimension of the characters written on the blackboard should be not smaller than 24mm for a reading distance of 10m. This corresponds to a detail dimension of 7.3 mm (2.3.1). Calculating the visual angle of the target following the (3.3.3_1), the critical a results 8.3 min.

5.0

2cosarctan2

⋅⋅=l

d ϑα (3.3.3_1)

This result is quite similar for different lesson times. From the analysis of this question, it is possible to conclude that a significant percentage of the student sample (10%-17%) could have some visual disfuntionality not diagnosed.

CHAPTER 7

186

7/10 white backgroud

83%

14% 3%

1

2

3

9/10 white background

50%

40%

10%

123

YES % PARTIALLY NO %

7/10 111,0 82,8 19,0 14,2 4,0 3,0

9/10 66,0 49,3 54,0 40,3 14,0 10,4

Fig. 3.3.3_ 6: Critical detail definition for white background

7/10 black backgroud17%

79%

4%

1

2

3

9/10 black backgroud

49%

38%

13%

1

2

3

YES % PARTIALLY % NO %

7/10 106,0 79,1 23,0 17,2 5,0 3,7

9/10 65,0 48,5 51,0 38,1 18,0 13,4

Fig. 3.3.3_ 7: Critical detail definition for black background

Case study

187

3-4. Visual discomfort evaluation: veiling luminance and eyes sickness indicators The main eye disturb recorded in this data analysis is the “burning eyes” perceived by the 34% of the population. The “blurred vision from far” interests the 29% of the population. It is important to observe that only the 64% of the total sample answered to this questionnaire section.

VISUAL DISCOMFORT EVALUATION

0102030405060708090

100

1 2 3 4 5 6 7

nopartiallyyes

YES % PARTIALLY % NO % 1 42 31 30 22 16 122 71 53 14 10 2 13 67 50 14 10 6 44 67 50 15 11 4 35 67 50 14 10 5 4

6 69 51 11 8 0

7 49 37 21 16 17 13

Fig. 3.3.3_ 8: Visual discomfort evaluation

The disability glare has been calculated in compliance with the (2.3.2) for the more critical student’s position, in the work places closed to the window site for each illuminance condition detected. The contribution of each glare source has been evaluated separately. It is important to observe that the multiplication factor to calculate the disability glare is 9.2 in the (2.3.2), instead of 4.16 (1981 IESNA Lighting Handbook).

Fig. 3.3.3_ 9: Illuminance from each glare source at the eye in lux

1 burning eyes 2 periocular pain 3 headache 4 eyes itch (eyesrain

discomfort) 5 lachrymation 6 blurred vision (closed

vision) 7 blurred vision (far vision)

CHAPTER 7

188

We considered the veiling luminance contribution as relevant factor for the VL calculation especially for the work desk sector closed to the windows. For this reason in the results reported in the calculation sheet in Fig. 7.4.1_17-20, 2 different luminance conditions have been considered:

- VL calculated considering the veiling luminance contribution (yellow column) for the violet sector (Fig. 3.3.3_5).

- VL calculated considering just the background illuminance level (grey column) for the yellow and red sector (Fig. 3.3.3_5).

5. Visibility evaluation The detail definition in different illuminance levels has been defined for the fixed conditions expressed above, concerning the population age, the critical detail dimension and the background/target reflection factors. In particular 3 different illuminance settings have been evaluated:

a) inside illuminace between 200 lx and 300 lx, in order to describe the inside illuminance conditions in the traditional class room, when the artificial light is not turned on in low outside illuminance conditions;

b) inside illuminance between 300 and 500 lx, in order to describe the inside

illuminance level in compliance with the regulation UNI 10840 and UNI 12464-1; c) inside illuminance >500 lx, in order to describe the inside illuminance level for

possible glare conditions. The illuminance levels have been measured in different points on the first and on the last work desk of the classroom, during the test compilation. The result of the measurement has been compared with the data calculated by the simulation software on the work surface, as reported in the following figures. The illuminace range for the inside illuminance that define the 3 cathegories (a), (b), (c) has been measured in the middle of the work desk. However it is important to notice case by case the difference between the illuminance levels for different position in the work desk. Using the shading system and the artificial light it was possible to regulate and verify the inside illuminace in compliance with the level definition (a), (b), (c). For the illuminance level (a) the percentage of the students that perceived a good detail definition is 38%, with black background, 52% with white background (Fig.3.3.3 _15 and

Case study

189

Fig.3.3.3 _16). This means that the percentage difference of the satisfieds is 14%. Fig. 7.4.1_10/11 shows the inside illuminance condition as output results of the Relux calculation. Similar conditions have been measured for instance in the class room 2A at 8.30, only with the natural light and the shatter completely open for the work places closed to the inner wall. On the inside work positions (almost a middle of the classroom surface) there is an illuminance level inferior to 200 lx. The main percentage of unsatisfied users in the illuminance condition (a) are placed in this classroom position, as results from the partial data analysis curried out subdiving the test answers by student position (Fig 3.3.3_5). This result confirms the visibility calculation reported in Fig. 3.3.3_17 “E=200”, where the VL for positive contrast is between 4.81 and 6.11 (considering the veiling luminance or not) for positive contrast, instead of the 7.38-9.52 calculated for negative contrast.

Fig. 3.3.3_ 10: Inside illuminance calculation by Relux: overcast sky acc. CIE; date: 02.05.2007; time: 8:40; location: Trento


CHAPTER 7

190

For the illuminance level (b) the percentage of the students that perceived a good detail definition is 82%, with black background, 85% with white background (Fig.3.3.3 _15 and Fig.3.3.3 _16). This means that the percentage difference of the satisfied is 3%. Fig. 3.3.3_12 shows the inside illuminance condition as output results of the Relux calculation. Similar conditions have been measured for instance in the classroom 2A at 8.30, with artificial light operating and the shatter completely open or in the classroom 2B at 10.30, with only natural light and the shatter completely open. On the inside work positions (almost a middle of the classroom surface) there is an illuminance levels superior that 300lx. The percentage of unsatisfied users in the illuminance condition (b) is homogeneously placed in the whole classroom surface that means that there is no sensible variation in the subjective perception of an illuminace of 300 lx instead of 500 lx. This result confirms the visibility calculation reported in Fig. 3.3.3_18 “E=300”, where the VL for positive contrast is between 6.97 and 5.72, instead of the 10.68-8.62 calculated for negative contrast, and in Fig. 3.3.3_19 “E=500”, where the VL for positive contrast is between 8.08 and 6.61 for positive contrast, instead of the 11.96 – 9.54 calculated for negative contrast. It is important to notice that the difference in the VL calculation is higher comparing positive and negative contrast than comparing the 2 illuminance levels (300 lx and 500 lx).


For the illuminance level (c) the percentage of the students that perceived a good detail definition is 69%, with black background, 73% with white background (Fig.3.3.3 _15 and Fig.3.3.3 _16). This means that the percentage difference of the satisfied is 4%. The Fig. 3.3.3_13 and Fig. 3.3.3_14 show the inside illuminance condition as output results of the Relux calculation. Similar conditions have been measured for instance in classroom 2B at

Case study

191

10.30, with artificial light operating and the shatter completely open. On the work positions closed to the windows (almost a middle of the classroom surface) there is an illuminance levels superior to 700 lx that could reach the 1500 lx too. The main percentage of unsatisfied users in the illuminance condition (c) are placed in this classroom position, as results from the partial data analysis curried out subdiving the test answers by student position (Fig. 3.3.3_5). This result confirms the visibility calculation reported in Fig. 3.3.3_20 “E=700”, where the VL for positive contrast is between 8.82 and 5.88, instead of the 10.68-8.62 calculated for negative contrast. It is important to notice that the VL calculation is higher for E= 500 than for E=700. Moreover the VL calculated for E=700 lx is 8.82 for positive contrast, without veiling luminance, and 7.97 for negative contrast considering the veiling illuminance.


Fig. 3.3.3_ 14: Inside illuminance calculation by Relux: clear sky acc. CIE; date: 02.05.2007; time: 10:30; location: Trento

CHAPTER 7

192

Detail definition with illuminance 200 <lx<300, black background

�38%

48%

14%1

2

3

Detail definition with illuminance 300<lx<500,

black background

82%

17% 1%1

2

3

Detail definition with illuminance lx>500, black background

69%

28%

3%1

2

3

good weak bad

Detail definition with illuminance included between 200 lx and 300lx 51 38,0597015 64 47,76119 19 14,17910448

Detail definition with illuminance included between 300 lx and 500 lx 109 81,3432836 23 17,16418 2 1,492537313 Detail definition with illuminance > 500 lx 93 69,4029851 37 27,61194 4 2,985074627

Fig. 3.3.3_ 15: Detail definition for black background

Case study

193

Detail definition w ith illuminance 200 <lx<300, white background

52%39%

9%1

2

3

Detail definition with illuminance 300<lx<500,

white background 13% 2%

85%

1

2

3

Detail definition with illuminance lx>500,

white background

73%

22%5%

1

2

3

Detail definition with illuminance included between 200 lx and 500lx 70 52,238806 52 38,80597 12 8,955223881 Detail definition with illuminance included between 300 lx and 500 lx 114 85,0746269 17 12,68657 3 2,23880597 Detail definition with illuminance > 500 lx 97 72,3880597 30 22,38806 7 5,223880597

Fig. 3.3.3_ 16: Detail definition for white background

CHAPTER 7

194

Conclusion Considering each different illuminance level described above and the correlated test results, it is possible to conclude that:

• black task on white background is the preferred contrast condition (negative contrast), especially for low illuminance level, often verified in the classrooms during the first and last classes;

• there is a lower difference in the detail definition between the inside illuminance condition of 300 lx instead of 500 lx than between posite and negative contrast;

• the main visual discomfort causes is the disability glare in the classrooms, perceived especially in the places closed to the windows. For this reason the shadow system is often not properly closed, so that the illuminance level could result lower than what aspected.

So it is possible to define the following design guidelines for classrooms: to introduce a proper shadow system, in order to control the disability and veiling glare and at the same time in order to use the natural light contribution and to replace the old blackboard with the white, that permits a higher visibility level in different illuminance condition.

Case study

195


Description Symbol Value m.u. Symbol Value m.u. Symbol Value m.u.Illuminance E 200,0 lux 200,0 lux 200,0 lux

Reflection factor ρρρρ Nero 3 % Bianco 94 %Luminance L 1,91 cd/m2 59,84 cd/m2

Veiling luminance Lve 0,8 cd/m2 0,0067 cd/m2 0,2094 cd/m2Detail d 24,0 mm

Distance D 10,0 mDetail αααα 8,25 min

Esposure time t (esp.) 2,0 sec 2k factor k 2,6

age age 23 years 23 years

∆∆∆∆Leff/E0 0,91C=∆∆∆∆Leff/Lb 0,30

k 9,20Af1 0,997Af 1,000

Visula luminance Lv 2,73 cd/m2Log(Lb+6) 6,4357

a(Lb) 0,1667Log(αααα+0.523) 1,4395

a(αααα) 0,1981(a(αααα, Lb)+t)/t 1,0000 1,0616

φφφφ1/21/21/21/2 0,9124 0,9936L1/2 0,0804 0,0949

∆∆∆∆Lpos (t>=2) 0,0948 0,1206 cd/m2∆∆∆∆Lpos (t,Af) 0,0948 0,1206 cd/m2VLpos (t,Af) 6,11 4,81

Ccr 0,0496 0,0631TI 21,4 %m 0,2574 0,3005ββββ 0,5449 0,5168

Fcp 0,6418 0,6511∆∆∆∆Lneg (t) 0,0609 0,0785 cd/m2

∆∆∆∆Lneg (t,AF) 0,0609 0,0785 cd/m2VLneg (t,Af) 2,58 9,52 7,38

Ccr 0,0319 0,0411TI 22,5 %

Background TargetScene

CALCULATIONScene Positive NegativeContrast

Negative Contrast

Age

Esposure time

Positive Contrast

Fig. 3.3.3_ 17: Visibility calculation for E=200

CHAPTER 7

196








∆∆∆∆Leff/E0 0,91C=∆∆∆∆Leff/Lb 0,30

k 9,20Af1 0,997Af 1,000


a(Lb) 0,1609Log(αααα+0.523) 1,4395

a(αααα) 0,1981(a(αααα, Lb)+t)/t 1,0000 1,0608

φφφφ1/21/21/21/2 1,0059 1,0808L1/2 0,0971 0,1107


Ccr 0,0435 0,0530TI 17,9 %m 0,3069 0,3458ββββ 0,5130 0,4920

Fcp 0,6527 0,6640∆∆∆∆Lneg (t) 0,0814 0,1008 cd/m2


Ccr 0,0284 0,0352TI 19,3 %

Scene Background Target

CALCULATIONContrast Scene Positive Negative

Age

Esposure time

Positive Contrast

Negative Contrast


Case study

197








∆∆∆∆Leff/E0 0,91C=∆∆∆∆Leff/Lb 0,30

k 9,20Af1 0,997Af 1,000


a(Lb) 0,1527Log(αααα+0.523) 1,4395

a(αααα) 0,1981(a(αααα, Lb)+t)/t 1,0000 1,0596

φφφφ1/21/21/21/2 1,1495 1,2410L1/2 0,1232 0,1399


Ccr 0,0375 0,0459TI 18,2 %m 0,3803 0,4237ββββ 0,4755 0,4566

Fcp 0,6758 0,6926∆∆∆∆Lneg (t) 0,1211 0,1518 cd/m2


Ccr 0,0254 0,0318TI 20,2 %



Age

Esposure time

Positive Contrast

Negative Contrast


CHAPTER 7

198








∆∆∆∆Leff/E0 0,91C=∆∆∆∆Leff/Lb 0,30

k 9,20Af1 0,997Af 1,000


a(Lb) 0,1439Log(αααα+0.523) 1,4395

a(αααα) 0,1981(a(αααα, Lb)+t)/t 1,0000 1,0583

φφφφ1/21/21/21/2 1,2642 1,4851L1/2 0,1441 0,1841


Ccr 0,0344 0,0516TI 33,3 %m 0,4342 0,5238ββββ 0,4523 0,4182

Fcp 0,6970 0,7381∆∆∆∆Lneg (t) 0,1602 0,2544 cd/m2


Ccr 0,0240 0,0381TI 37,0 %



Age

Esposure time

Positive Contrast

Negative Contrast


Case study

199

Parameter:Background luminance [cd/m2] (t=2 sec)

0,0100

0,1000

1,0000

10,0000

100,0000

1,0 1,6 2,5 4,0 6,3 10,0 16,0 25,0 40,0 63,0 100,0

alfa

Log(

DL)

1,9100

2,8600

6,6800

9,5500

Fig. 3.3.3_ 21: Positive Visibility

Parameter:Background luminance [cd/m2] (t=2 sec)

0,0100

0,1000

1,0000

10,0000

100,0000

1,0 1,6 2,5 4,0 6,3 10,0 16,0 25,0 40,0 63,0 100,0

alfa

Log(

DL)

1,9100

2,8600

6,6800

9,5500

Fig. 3.3.3_ 22: Negative Visibility

CHAPTER 7

200

Chapter 4

Conclusions

Based on the literature overview, it is becoming increasingly important to establish a realistic baseline of the actual lighting energy consumption in buildings for the different scenarios nowadays used (both manually and automatically operated), which incorporates occupant behaviour. At the same time, it is strategic to explore and quantify the benefits of typical energy saving design measures (automated systems) compared with a traditional operation system (manual system). The features of lighting simulation tools nowadays available underline the importance of defining suitable reference cases for benchmarking the performance of automated lighting control. In the referred research thesis a new design methodology has been developed and verified, that relates energy saving and visual comfort. The research method includes, as referred in the summary, the definition of the methodological approach for the energy efficiency evaluation (design methodology, software tool outputs evaluation methodology, data analysis method) [section 2.1] and the development of specific innovative design tools for both lighting system design [sections 2.2] and visual comfort evaluation [section 2.3]. The analysis of the energy saving potential of automated lighting scenarios have been monitored in real use conditions and not in controlled laboratory environment, in order to prove the automation systems efficiency in operating time. The theoretic efficiency and the functionality of the automation system are indeed affected in field by the occupants’ behaviour, when they can operate manually on the indoor environmental factors (e.g. operating the blind system), in comparison with a standard and controlled state. The case study analyzed allows specifically the simultaneous comparison between manually operated systems and automatically operated systems, because of the symmetry of the building examined [section 3.1.1]. The case study evaluation has been carried out in the following steps:

1- Analytical phase: monitoring and analysis of existing situation

CHAPTER 4

202

2- Programmatic phase: definition of project objectives and system requirements 3- Propositional phase: development of system solution 4- Evaluation phase: evaluation and verification

By the case study analysis, the methodology and the design tools developed have been applied and verified. During the case study development, the following weak points have been pointed out:

- the lack of blind position data, because specific sensors were not installed in the rooms monitored

- the lack of weather data format file required by the simulation software and moreover the absence of complete weather databank for a significant monitoring period for the case study location

- the influence of the orography location (presence and dimension of the mountains) is not correctly modelled by the software simulation

- the improper use of the automation system by the users, during the first operating period, because they don’t have enough familiarity with the new lighting system operation mode.

The research outputs have the following main implications. 1. Practical/technical implication: the definition of new automation lighting scenarios and their potentialities for the specific case study analyzed.

The research results demonstrate that the use of automated systems for the artificial light control can yield considerable energy savings, roughly between 40% and 60% depending on the complexity of the implemented system and of the parameters controlled, in the observed classrooms during the academic year 2005/2006. The automated control system of the experiment could be further improved by incorporating shading control functionality. In fact, the improper use of the shading system can negatively influence the automation system efficiency and the daylight utilisation. Specifically, considering the whole observation period, the energy saving percentage for the analyzed scenarios is [section 3.3.2]:

- 40%, if the luminaries are switched on when occupancy is detected and the room illuminance level (measured in the middle of the room) is below a predefined minimum level (scenario 1); - 65%, if the luminaries are switched on when occupancy is detected and dimmed (in two separately controlled circuits) so as to provide predefined minimum illuminance

Conclusions

203

levels, measured at two points in the room, each corresponding to a circuit (scenario 2); - 65%, if the luminaries are switched on when occupancy is detected and dimmed (in three separately controlled circuits) so as to provide predefined minimum illuminance levels (measured in three points in the room, each corresponding to a circuit). There are any significant different results if the lecture room is controlled in terms of two separate spatial zones (front and back) with dedicated occupancy sensors (scenario 3).

The monitoring data results could support the effective and proactive operation of building service systems for indoor environmental control. During the operation period of the automated systems, the following considerations have been pointed to a useful operation mode: - the luminaries efficiency (use of fluorescent tubes T5 instead T8) could be wiped up by the improper use of the blind system - the occupancy detection in different spatial zones could bring any contribution for the energy saving, if the student are spread in classrooms even if they are a scan number in comparison with the available workplaces - the use of the dimming regulation in classroom, where power point show has been used, could improve the students performance; it is indeed possible to maintain a balanced illuminance level that could allow simultaneously the slides vision and the writing action

2. Theoretic/cognitive implication: the new functions that it is possible to implement control-oriented user behaviour software tools, in order to calculate energy consumption estimations more realistic.

Several studies have been and still are developed internationally to collect data on building users’ interactions with building control systems and devices [section 1.4]. Such data can bring about a better understanding of control-oriented user behavior in buildings and thus support the development of corresponding behavioral models for integration in building performance simulation applications. The monitored data analysis has pointed out that the manual on/off probability used by Adeline doesn’t correspond to the real condition in the observed lecture halls: specifically with an inside illuminance level higher than 500 lux it is still possible to have the light turn on (probability equal to 40%) [Fig. 3.3.1_37]. Moreover the inside illuminance simulated by Adeline is underestimated, because of the user’s behaviour positioning the curtains: the use of a shading factor equal to 0.7 better estimate the monitored condition but this approach do not consider the stochastic use of the blind in classrooms. It has been verified that opening/re-positioning the blinds at the

CHAPTER 4

204

beginning of the each lesson is not the more common practice. In a typical day, the occupancy distribution is not directly correlated to the timetable of the referred classrooms but has a similar trend for each one [Fig. 3.3.2_5]. The energy consumption in traditional classrooms doesn’t always directly correlate with outdoor illuminance levels, because of the improper use of the shading system [Fig.3.3.2_19]. Nonetheless, the mean energy use over the whole observation period is clearly related to outside illuminance level [Fig.3.3.2_33]. The obtained information would support the assessment of energy saving potential due to consideration of occupancy and behavioral patterns specifically in university buildings.

3. Comfort and efficiency evaluation: the relation between visual comfort perceived and the lighting system operational standard required.

Improving the energy-efficiency of lighting systems should include better use of daylight, but it will require the development of control systems that result in luminous conditions and that are suitable to occupants. Such the section 1.5 shows, what control system features would be most acceptable is not yet know, nor what range of luminous conditions the system should permit. Moreover the evaluation of visual comfort and visual ergonomic has been related mainly with the office/working places especially in reference to the use of VDT. The analysis carried out in this research would include another application field for these concepts: the educational building, specifically the lecture halls of university building. By the literature analysis, it is important to evaluate the correlation between productivity and satisfaction of the users on one hand and the efficiency of the lighting system on the other. Considering the results of the subjective visual comfort analysis sheet [section 3.3.3], it is possible to conclude that: - black task on white background is the preferred contrast condition (negative contrast), especially for low illuminance level, often verified in the classrooms during the first and last classes; - there is a lower difference in the detail definition between the inside illuminance condition of 300 lx instead of 500 lx than between positive and negative contrast; - the main visual discomfort causes is the disability glare in the classrooms, perceived especially in the places closed to the windows. For this reason the shadow system is often close not properly, so that the illuminance level could result lower than what expected.

Conclusions

205

In conclusion, the use of automation system guarantees a more efficient use of the artificial light and a higher visual comfort level: in fact it has been recorded that the minimum illuminance level value maintained in the traditional classrooms is low, especially in winter semester in the early morning and late in the afternoon, while on the contrary it is always guarantee in the automated classrooms. Moreover the energy saving percentage is mainly related with the outside illuminance level; even if the disability and discomfort glare influence this trend (an improper use of the shadow system often causes the unnecessary use of artificial light). The referred research evidences some significant results in order to better understand the effectiveness application of automated lighting systems. Moreover, the following future steps could be developed, also in order to improve the limitations indicated above: - the general patterns definition of user control behaviour as a function of environmental parameters (indoor and outdoor), in order to develop proved models than can implement the reliability of computational building performance simulation application; - the application of the new design tools in different building typologies: the methodological approach and the scientific rigor defined in this research could be a general praxis to be used in other case studies.

Reference

Ashdown, I. 1996 - dream.leeds.ac.uk/ cuddles/rover/abs-ian2.htm Adraian W., 1993, The physiological basis of the visibility concept, Proceeding of the 2nd International symposium on Visibility and Luminance in Road Lighting, Orlando USA Adrian W., 1989, Visibility of targets: model for calculation, Lighting Res. Technology, 21:181-188 Ahumada A., Scharff L., 2003, Letter Identification: Contrast Polarity and Speed-Accuracy Trade-Off Strategies, Abstracts of the Psychonomic Society 44th Annual Meeting, Vol. 8, p. 67 Ashdown, I. 1996. Lighting for Architects, Computer Graphics World, 19 (8):38-46 Ashford P., 1999, The cost implications of energy efficiency measures in the reduction of carbon dioxide emissions from European building stock. Caleb management services Aulhorn E., 1964, Graefe’s Archiv für Klinische und Experimentelle Ophthalmologie, 1964: 167: 4 Balci R.; Aghazadeh F., 2003, The effect of work-rest schedules and type of task on the discomfort and performance of VDT users, Ergonomics, Volume 46, Number 5, April 15, , pp. 455-465(11), Taylor and Francis Ltd Baron R.A., 1994, The physical environment of work settings: Effects on task performance, interpersonal relations, and job satisfaction. Research in Organizational Behavior, 16, 1–46 Bellantani S., 2004, Manuale della domotica, il sole 24ore editore Berson D.M., Dunn, F.A. Motoharu Takao, 2002, Phototransduction by retinal ganglion cells that set the circadion clock, Science, February 8 Bertoldi P., 2002, The European Programmes and Policies to Save Energy in Lighting -

REFERENCE

208

European Commission DG Joint Research Centr; http://www.eu-greenlight.org/What-to-do/what.htm Booth J., 1998, The User Interface in Computer-based Selection and Assement: Applied and Theoretical Problematics of an Evolving Technology, Volume 6, Number 2, Blackwell Publishers Bourgeois D., Reinhart C., Macdonald I., 2005, Assessing the total energy impact of occupant behavioural response to manual and automated lighting systems Boyce P., 2003, Human Factors in Lighting, London: Taylor & Francis Brennan J., 2007, Effect of Light on Humans, ‘Best Written Paper’ award in the Society of Light & Lighting Young Lighters Capolla M., 2004, Progettare la domotica, Maggioli editore Carter D., Slater A., Moore T., 1999, A study of occupier controlled lighting systems, in: Proceedings of the 24th Session of the CIE, Warsaw, Poland, Vienna, Austria: CIE, pp. 108–110, 1(2) CENELEC 2005 SmartHouse Code of Practice CWA 50487:2005 Cuttle C., 1983, People and windows in workplaces, in: Proceedings of the People and Physical Environment Research Conference,Wellington, New Zealand, pp. 203–21 D. Lgs. 19/09/1994 n. 626, “Attuazione delle direttive 89/391/CEE, 89/654/CEE, 89/655/CEE, 89/656/CEE, 90/269 /CEE, 90/270/CEE, 90/394/CEE, 90/679/CEE, 93/88/CEE, 97/42/CEE e 1999/38/CE riguardanti il miglioramento della sicurezza e della salute dei lavoratori durante il lavoro” De Concini M., Betta A., 1997, Salute e Sicurezza al Video Terminale, Collana Salus, Servizio Rapporti con il pubblico dell’Azienda Provinciale per i Servizi Sanitari Dennis R. Ankrum CIE Ankrum Associates, 1996, Visual Ergonomics in the Office, reprinted from WorkPlace Ergonomics, available at www.office-ergo.com/setting.htm Directive 2002/91/EC of the European Parliament and of the Council of 16 December 2002

Energy efficiency and saving on lighting systems in existing building: intervention strategies

209

on the energy performance of buildings Directive 2006/32/EC of the European Parliament and of the Council of 5 April 2006 on energy end-use efficiency and energy services and repealing Council Directive 93/76/EEC Dubiel J., Wilson M., Jacobs A., 1995, The Light Fantastic, Building Services, 17 (8): 28-29 Dubrous, 2004, The lightswitch wizard – reliable daylight simulations for initial design investigation, IRC-institute for research and construction EIB Konnex Association, 1999, Project Engineering for EIB installations – Principles and applications Environmental Change Unit, 1998, University of Oxford for the European commission DG-XVII, final report EPBD Concerted Actions - 5° meeting, Budapest, 8-9 May 2006 European Climate Change Program, final report 2001 Galasiu A. D., Veitch J. A., 2006, Occupant preferences and satisfaction with the luminous environment and control systems in daylit offices: a literature review, Energy and Buildings Volume 38, Issue 7, Pages 729-742 Hajkowicz S., McDonald G., Smith P., 2000, An Evaluation of Multiple Objective Decision Support Weighting Techniques in Natural Resource Management http://www.ergonomics.org.uk/ http://www.iea.cc/index.php Humanic Ergonomics, What is Ergonomics, available at http://www.humanics-es.com/def-erg.htm Hunt D.R.G., 1979, The use of artificial lighting in relation to daylight levels and occupancy, Building and Environment, 14, 21-33. IEA, 2000, Daylight in Building –Solar Heating and Cooling Programme Task 21, section 6

REFERENCE

210

Design Tolls IEA,1998, Applicability of daylighting computer modeling in real case studies: comparison between measured and simulated daylight availability and lighting computing-IEA ECBCS Annex 29 IESNA, 2000, The IESNA Lighting Handbook, 9th ed. New York:IESNA IESNA, 2003, Advanced Lighting Guidelines Edition PB-221-03 IESNA. 199, 1997 IESNA Software Survey, Lighting Design + Application, 11 (7): 41-50 International Energy Agency, 2000, Radlink User's Manual and Technical Manual, Documentation of the Software Package ADELINE, Version 3 International Energy Agency, 2000, Superlink User's Manual and Technical Manual, Documentation of the Software Package ADELINE, Version 3 ISO (1998). ISO 9241-5. Ergonomic requirements for office work with visual display terminals (VDTs) Part 5: Workstation layout and postural requirements Jaschinski-Kruza, W. ,1990, On the preferred viewing distances to screen and document at VDU workplaces. Ergonomics, 33, 8 , 1055-1063 Mahoney, D. P., 1994, Walking through Architectural Designs, Computer Graphics World, 17 (6) Mills E., 2002, The $230-bilion global lighting energy bill - expanded from version published in the proceeding of the 5th international congerence on energy-efficienct lighting, May 2002, Nice, France Mohammadi A., Kabir E., Mahdavi A., Pröglhöf C., 2007, Modeling user control of lighting and shading devices in office buildings: an empirical case study, proceeding of the Building Simulation International Conference, 3-6 September 2007, Bijing, China

Mahdavi A., Mohammadi A., Kabir E., Lambeva L., 2007, An empirically-based approach toward user control action models in buildings, proceeding of the 24th W78 Conference, 2007, 26.-29.6.2007, Maribor


211

Newsham, G.R., Mahdavi, A., Beausoleil-Morrison, I., 1995, Lightswitch: a stochastic model for predicting office lighting energy consumption, inthe Proceedings of Right Light Three, the 3rd European Conference on Energy Efficient Lighting, Newcastle-upon-Tyne, 60-66 Nijkamp P., Rietveld P. & Voogd H., 1990, Multicriteria Evaluation in Physical Planning (Amsterdam, North Holland) Novitski, B.J., 1993, Energy Design Software, Architecture, 83 (6): 125-127 Novitski,B.J., 1992, Lighting Design Software, Architecture, 81 (6): 114-117 PrEN 15193-2006, Energy performance of buildings — Energy requirements for lighting Pritchard, R.D., 1992, Organizational productivity. In Dunnette, M.D. and Hough, L.M. (eds), Handbook of Industrial and Organizational Psychology, 2nd edn. Palo Alto, CA: Consulting Psychologists Press, 443–471 Quaranta G., Mongiovì P., 2004, L’ABC della domotica, il sole 24 ore editore Reinhart C., Selkowitz S., 2006, Daylighting—Light, form, and people, Energy and Buildings Volume 38, Issue 7, Pages 715-717 Reinhart, C.F. 2004. Lightswitch 2002: a model for manual control of electric lighting and blinds, Solar Energy, 77 (1), 15-28. Richard E. Bird, Roland L.Hulstro, 1981, A Simplified Clear Sky Model for Direct and Diffuse Insolation on Horizontal Surfaces, SERI Rubin, A., 1987, Office Design Measurements for Productivity: A Research Over- 220 Veitch view (NBSIR 87–3688). Washington, DC: United States Department of Commerci S.Marsigliante, Adattamento al buio ed alla luce, Appunti di Fisiologia oculare, Uni Lecce – CD Ottica e Optometria –A.A. 2006/2007 - available at http://www.biologia.unile.it/docs/docenti/marsigliante/fisiologia_oculare0607/Fis%20Oculare%2010.pdf

REFERENCE

212

S.Marsigliante, A.A. 2006/2007, La Retina-Struttura e Funzione, Appunti di Fisiologia oculare, Uni Lecce – CD Ottica e Optometria available at http://www.biologia.unile.it/docs/docenti/marsigliante/fisiologia_oculare0607/Fis%20Oculare%203.pdf Saaty, R.W., 1987, The analytic hierarchy process—what it is and how it is used, Mathematical Modelling, 9(3), pp. 161–176 Sauter T., Dietrich D., Kastner W., 2001, EIB installation bus system, publics; ISBN 3895781754 Servizio Metereologico del Trentino, Istituto Agrario di S.Michele A.Adige, Rete Agrometeorologica IASMA, http://217.222.71.209/meteo/datimeteo/ricercadati.php Sharifi A., Herwijnen M., Toorn W., 2004, Spatial Decision Support System, ITC (international institute for geo-information science and earth observation) Sullivan, A. C.,1996, Photorealistic Light Simulation, Architecture, 85 (10): 177-79 Sustainable Building 2006-2007 International Conference Proceeding U.S. Department of Energy, Energy Efficiency and Renewable Energy, Weather Data, available in: http://eere.energ.gov/buildings/energyplus/weatherdata_sources.html#IGDG Ubbelohde, 1999, Comparative Evaluation of Four Daylighting Software Programs UNEP, 2007, Report on Building and Climate Change UNI EN 12464-1, 2004, Light and lighting - Lighting of work places - Part 1: Indoor work places Veitch J.A., 2006, Lighting for high-quality workplaces, National Research Council Canada, http://irc.nrc-cnrc.gc.ca Veitch J.A., Gifford R., 1996, Assessing beliefs about lighting effects on health, performance, mood, and social behavior, Environment and Behavior 28 (4) 446–470 Veitch J.A., Hine D.W., Gifford R., 1993, End users’ knowledge, beliefs, and preferences for


213

lighting, Journal of Interior Design 19 (2) 15–26 Wells B.W.P., 1965, Subjective responses to the lighting installations in a modern office building and their design implications, Building and Environment 1 57–68 Wolska A. Switula M., 2003, Luminance of the surround and visual fatigue of VDT operators, International Journal of Occupational Safety and Ergonomics (JOSE), Vol.5, pp. 553-81 Wolska A., 2003, Visual strain and lighting preferences of VDT users under different lighting systems, International Journal of Occupational Safety and Ergonomics (JOSE), Vol.9, pp. 431-40 www.abb.it www.hager.de www.iguzzini.com www.merten.de www.osram.com www.siemens.it www.theben.de www.tiflosystem.it www.zumtobel.com Zalesak M., 2006, Possible impacts of intelligent technologies application in residential buildings on energy efficiency, proceeding of KNX scientific Conference 2006, 9-10 November 2006, Vienna

215

Acknowledgement

I would like to thank everyone that have been with me, guiding and supporting me, during these study years and for the elaboration of this work, particularly: Professor Antonio Frattari, for his availability with which he has sustained me, for his enthusiasm and passion in the teaching activity and the role of innovation researcher, because he has been an example for me; CUnEdI staff: Rossano, for the precious suggestions and the time spent for me; Michela, for her sincere friendship; Paolo, for his daily joviality; Professor Mahdavi and the research group of the Technical University of Vienna – Department of Building Physic and Building Ecology for the support and the suggestions during my stay in Vienna; especially Mahijid and Elam for the constrictive comparison about the common research field and Lechleitner for the development of the Labview program; Dr. Eng. de Boer and the technical staff of the Fraunhofer-Institute for Building Physics in Stuttgart, that helped with the introduction to the Adeline program system; Dr. De Concini, top clinician of the Ophthalmology Department of the S.Chiara Hospital in Trento and Dr. Eng. Zancarli of Stainer studio, for the support in the visual comfort and visibility analysis; My brother Gabriele, for his fundament contribution in the implementation of the calculation program in FORTRAN and for his labor limae in the review of the English text, and above all to have developed together the love for the research activity; My mother and my father, always present as guide and support during all my live, to have taught me to work with perseverance and care; My husband Lorenzo, to have been with me every day, to have believed in me and to have encouraged me to ambitious purposes, sharing with me troubles and satisfactions of this iter studiorum, one more time together.

Desidero ringraziare tutti coloro che mi hanno accompagnato, guidato e sostenuto in questi anni di studio e nella preparazione di questo lavoro, in modo particolare: il Professor Antonio Frattari, per la disponibilità e l’incoraggiamento con cui mi ha supportata e l’entusiasmo e la passione con cui svolge il suo lavoro di docente e ricercatore dell’innovazione, perché ha rappresentato per me l’esempio da seguire; lo staff del CUnEdI: Rossano, per i preziosi consigli ed il tempo dedicatomi; Michela, per l’amicizia sincera che mi ha dimostrato; Paolo, per avermi consentito di lavorare in un ambiente sereno e per la sua giovialità quotidiana il Professor Mahdavi e tutto il gruppo di ricerca della Technical University of Vienna – Department of Building Physic and Building Ecology, per la competenza con cui mi hanno seguita e la disponibilità accordatami; in particolare Mahijid e Elam per il confronto costruttivo sul nostro ambito di ricerca e Lechleitner per il supporto nello sviluppo del programma in LabView; l’ing. Jan de Boer ed il personale del Fraunhofer Institute di Stoccarda, per avermi permesso di lavorare con loro ed avermi dato tutti gli strumenti ed il supporto per sviluppare la modellazione e l’analisi del mio caso di studio con Adeline; il dottor De Concini, primario del reparto di oftalmologia dell’ospedale S.Chiara di Trento e l’ingegner Zancarli dello studio Stainer, per avermi indirizzato e fornito gli stumenti per condurre l’analisi sul benessere visivo degli studenti; mio fratello Gabriele, per avermi aiutata nello strutturare il programma di elaborazione dati in Fortran e per il labor limae nella revisione del testo inglese della tesi, ma soprattutto per aver coltivato e cresciuto insieme la curiosità e l’amore per la ricerca; mia madre e mio padre, da sempre presenti come guida e sostegno in tutto ciò che ho fatto nella mia vita, per avermi insegnato a lavorare con perseveranza e impegno ed a sapere raccogliere i frutti della propria fatica; mio marito Lorenzo, per essermi accanto ogni giorno, per aver creduto in me ed avermi spronato a puntare sempre alle mete più ambite, condividendo i sacrifici di questo percorso ed assaporando la gioia e la soddisfazione dei traguardi raggiunti, ancora una volta insieme.

Documents

Energy efficiency and saving on lighting systems in existing buildings