PhD Thesis Jaime Arias 1

8/18/2019 PhD Thesis Jaime Arias 1

1/262

Energy Usage in Supermarkets

- Modelling and Field Measurements

Doctoral Thesis

by

Jaime Arias

Division of Applied Thermodynamics and Refrigeration

Department of Energy Technology

Royal Institute of Technology

2005


2/262

ii

Energy Usage in SupermarketsModelling and Field Measurements Jaime Arias

Trita REFR Report No. 05/45.ISSN 1102-0245.ISRN KTH/REFR/05/45-SE.ISBN 91-7178-075-0.

Doctoral Thesis by Jaime AriasDepartment of Energy TechnologyRoyal Institute of Technology

Jaime Arias 2005


3/262

iii

ABSTRACT

This thesis investigates a special type of energy system, namely energyuse in supermarkets through modelling, simulations and field studies. Auser-friendly computer program, CyberMart, which calculates the totalenergy performance of a supermarket, is presented. The modellingmethod described in this thesis has four phases: the first phase is the development of a conceptual model that includes its objectives, the envi-ronment and the components of the system, and their interconnections. The second phase is a quantitative model in which the ideas from theconceptual model are transformed into mathematical and physical rela-tionships. The third phase is an evaluation of the model with a sensitivityanalysis of its predictions and comparisons between the computer modeland results from field measurements. The fourth phase is the model ap-plication in which the computer model answers questions identified inthe beginning of the modelling process as well as other questions arisingthroughout the work.

Field measurements in seven different supermarkets in Sweden were car-ried out to: (i) investigate the most important parameters that influence

energy performance in supermarkets, (ii) analyse the operation of newsystem designs with indirect system implementation in Sweden duringrecent years, and (iii) validate the computer model.

A thorough sensitivity analysis shows a total sensitivity of 5.6 %, which isa satisfactory result given a 10% change in the majority of input parame-ters and assumptions, with the exception of outdoor temperatures andsolar radiation that were calculated as extreme values in METEO-NORM. Comparisons between measurements and simulations in fivesupermarkets also show a good agreement. Measurements and simula-tion results for a whole year were not possible due to lack of data.

CyberMart opens up perspectives for designers and engineers in the fieldby providing innovative opportunities for assessment and testing of newenergy efficient measures but also for evaluation of different already-installed system designs and components. The implementation of newenergy-saving technologies in supermarkets requires an extensive inte-grated analysis of the energy performances of the refrigeration system,HVAC system, lighting, equipment, and the total energy usage. Thisanalysis should be done over a long period, to evaluate and compare thereal energy performance with the theoretical values calculated by Cyber-Mart.

Keywords: Supermarket, Energy Performance, System Analysis, Model-ling, Simulation, Field Measurements, Refrigeration Systems, Indirect

Systems, Heat Recovery, Floating Condensing, TEWI, LCC.


4/262

iv


5/262

v

ACKNOWLEDGEMENTS

Many people have contributed to carry out this work and to whom I wish to express my gratitude. First of all, I would like to thank my super- visor Professor Per Lundqvist for his support, guidance and invaluableadvices.

I also want to thank my roommate Martin Forsén for long inspiring dis-

cussions and many useful ideas concerning systems, modelling, and pro-gramming in Delphi.

Special thanks go to Joachim Claesson, Cecilia Hägg, Anders Johansson,and Åke Melinder for many valuable comments and stimulating discus-sions about systems, secondary refrigerants, heat exchangers and this dis-sertation. I am also grateful to Jörgen Rogstam for useful commentabout the manuscript and to Dorothy Furberg who corrected my Eng-lish.

I extend my gratitude to all the staff at the Division of Applied Thermo-dynamic and Refrigeration, Professor Björn Palm, Professor Eric Gran-ryd, Benny Andersson, Klas Andersson, Getachew Bekele, Erik Björk,Carina Carlsson, Yang Chen, Inga du Rietz, Richard Furberg, Peter Hill,Hans Jonsson, Nabil Kassem, Rahmatollah Khodabandeh, Peter Kjaer-boe, Fredrik Lagergren, Susy Mathew, Teclemariam Nemariam, Jan-ErikNowacki, Shota Nozadze, Wimolsiri Pridasawas, Dimitra Sakellari, Ox-ana Samoteeva, and Benny Sjöberg for being part of a stimulating at-mosphere and an excellent place to work in. I would also to thank, Raul Antón, Primal Fernando, Claudi Martin, Wahib Owhaib, Samer Sawalhaand Branco Simanic to give me the possibility to play football with youin the dream team of the Department of Energy Technology.

Thank are also given to the financial support by the Swedish Energy Agency through the Swedish National Research Programme “eff-Sys -Efficient Refrigeration and Heat Pump Systems” in cooperation with the

companies COOP Sweden AB, ICA AB, AB Fortum Heat, Asarums In-dustry AB, Hydro Chemical AB and the Swedish National Testing andResearch Institute. Special thanks go to Lennart Bjerkhög and Gösta Andersson from COOP Sweden AB, and Per-Erik Jansson from ICA AB for support, ideas, discussions and encouragements.

I want to express my gratitude to my parents, Luis and Lucila, and to mysister Luisa and hers family for all support and understanding.

Regards are also given to my relatives and friends for their support andfriendships.

Finally, I would like to thank Birgitta, Sebastian and Veronica, my family,to whom this thesis is dedicated for their love, patience and understand-

ing


6/262

vi


7/262

1

TABLE OF CONTENTS

1 INTRODUCTION ................................................................. 5

1.1 B ACKGROUND..................................................................................7 1.2 PURPOSE ............................................................................................7 1.3 METHOD ...........................................................................................8 1.4 PUBLICATIONS..................................................................................9

1.4.1

Papers to Conferences and Journals............................ 9 1.4.2

Report........................................................................... 9

1.4.3

Other Publications ..................................................... 10

1.5 DISPOSITION OF THE THESIS ..................................................... 10

2 ENERGY USAGE AND ENVIRONMENTAL IMPACT INSUPERMARKETS.........................................................................13

2.1 INTRODUCTION ............................................................................ 13 2.2 ENERGY IN S WEDEN ................................................................... 21 2.3 ENERGY USAGE IN SUPERMARKETS......................................... 23 2.4 R EFRIGERANT EMISSIONS .......................................................... 29 2.5 TEWI AND LCCP......................................................................... 34 2.6 CONCLUSIONS ............................................................................... 35

3 REFRIGERATION SYSTEMS IN SUPERMARKETS .......37

3.1 DIRECT S YSTEM ............................................................................ 41 3.2 INDIRECT S YSTEM ........................................................................ 43

3.2.1

Completely Indirect System........................................ 43

3.2.2

Partially Indirect System............................................ 47

3.2.3

Indirect Cascade System ............................................ 49

3.3 C ARBON DIOXIDE........................................................................ 52 3.3.1

Cascade System with CO2 .......................................... 52 3.3.2 CO2 as the Only Refrigerant ...................................... 53

3.4 HEAT R ECOVERY AND FLOATING CONDENSING ININDIRECT S YSTEMS .................................................................................... 56

3.4.1

Heat Recovery Systems............................................... 56

3.4.2

Floating Condensing System...................................... 59

3.4.3 Heat Recovery and Floating Condensing System ...... 60 3.5 CONCLUSIONS ............................................................................... 61

4 FIELD MEASUREMENTS ..................................................63

4.1 O VERVIEW ..................................................................................... 63 4.1.1

Field Test 1................................................................. 64

4.1.2

Field Test 2................................................................. 65

4.1.3 Field Test 3................................................................. 66

4.1.4 Field Test 4................................................................. 67

4.1.5

Field Test 5................................................................. 67

4.1.6

Field Test 6................................................................. 68


8/262

2

4.1.7

Field Test 7................................................................. 69

4.1.8 Test Instruments ......................................................... 70 4.2 R ESULTS.......................................................................................... 71 4.3 CONCLUSIONS ............................................................................... 94

5 CYBERMART, SYSTEMS AND MODELS..........................95

5.1 INTRODUCTION ............................................................................ 95

5.2

THE MODELLING PROCESS........................................................ 99 5.3 CONCEPTUAL MODEL ............................................................... 104

5.4 QUANTITATIVEMODEL ............................................................109 5.4.1

Building Model......................................................... 109

5.4.2 Outdoor Climate....................................................... 115

5.4.3

HVAC System ........................................................... 117

5.4.4

Refrigeration System ................................................ 120

5.4.5 Life Cycle Cost (LCC) .............................................. 137 5.4.6

Total Equivalent Warming Impact (TEWI) .............. 138

5.5 CONCLUSIONS ............................................................................. 139

6 CYBERMART...................................................................... 141

6.1 M AIN W INDOW ........................................................................... 142 6.2

BUILDING ..................................................................................... 143

6.3 R EFRIGERATION S YSTEM DESIGNS ........................................150 6.3.1

Cabinets and Cold Rooms ........................................ 151

6.3.2

Pressure Drop of the Medium Temperature System. 155

6.3.3 Pressure Drop of the Low Temperature System....... 161 6.4 ENERGY C ALCULATION............................................................165

6.4.1

Result Refrigeration System ..................................... 180

6.4.2

LCC .......................................................................... 182

6.4.3 TEWI ........................................................................ 184

6.4.4

Report....................................................................... 185

6.5 CONCLUSIONS ............................................................................. 186

7

EVALUATION OF CYBERMART..................................... 189

7.1 SENSITIVITY ANALYSIS..............................................................189 7.2 SIMULATION OF FIVE SUPERMARKETS AND COMPARISON

WITH FIELD MEASUREMENTS................................................................ 194 7.3 CONCLUSIONS ............................................................................. 209

8 APPLICATIONS OF CYBERMART .................................. 211

8.1 INTRODUCTION .......................................................................... 211 8.2 HEAT R ECOVERY VS. FLOATING CONDENSING..................211 8.3 DIRECT VS. INDIRECT S YSTEMS............................................... 218 8.4 CONCLUSIONS ............................................................................. 225

9 CONCLUSIONS, DISCUSSION, AND FUTURE STUDIES 227


9/262

3

9.1 CONCLUSIONS ............................................................................. 227 9.1.1

Energy Usage and Environmental Impact ............... 227

9.1.2 Field Measurements................................................. 228

9.1.3

Modelling and Simulations....................................... 229

9.2 DISCUSSION .................................................................................230 9.3 FUTURE S TUDIES ........................................................................ 231

NOMENCLATURE....................................................................233

Greek ........................................................................................ 236

Subscripts................................................................................. 237

APPENDIX A: INPUT DATA OF CYBERMART .................... 241

REFERENCES............................................................................247


10/262

4


11/262

5

1 Introduction

Information and communication technology as well as sustainabil-ity concerns are some of the driving forces of development in the world today. In a new era of rapid change, information and com-munication technology has influenced different sectors of our so-ciety. Personal computers, software, computer networks, internetand mobile phones are basic tools in both business and private life.Internet and mobile phones have transformed communication andchanged many aspects of our lives. In an increasingly intercon-nected world, the expansion and development of many industriesdepends on the efficiency of computers to access data and makedecisions. The energy sector has also been affected by informationand communication technology. Generation, distribution and utili-zation of energy have been influenced by personal computers,computer monitoring, control devices, communication equipment,internet, energy models, intelligent buildings, etc. These tools havebecome important instruments to assist researchers, designers, en-gineers, technicians and consumers in decision-making, mainte-nance and validation of new and old systems.

Sustainable development has also influenced the different sectorsof society. Sustainable development is a concept where environ-mental management, social equity and economic growth interact

both for the benefit of present and future generations. Energy in-fluences these three dimensions of sustainable development(International Energy Agency 2001). Energy is fundamental to in-crease productivity and industrial activities, to improve the stan-dard of living of people in the world, especially in developingcountries, and to reduce the emission of greenhouse gases. Never-theless, energy is associated in many aspects with unsustainabledevelopment. Global warming and ozone layer depletion are twoissues linked with energy production and energy utilization. Jo-hansson and Goldemberg (Johansson 2002) point out the chal-lenge of developing energy-related policies for sustainable devel-

opment:


12/262

6

It is imperative to find ways to greatly expand energyservices, especially to the two billion people who cur-rently rely on traditional forms of energy, as well as forgenerations to come; this expansion must be achievedin ways that are environmentally sound, as well as safe,affordable, convenient, reliable and equitable.

Stricter legislation to reduce global warming and ozone depletionas well as growing environmental awareness have brought about arevolution in energy generation, distribution and utilization. Newmore efficient systems and components have been developed toreduce energy consumption and emission of pollutants and green-house gases to the environment.

In Sweden, several policies and measures have been implementedto reduce emissions of greenhouse gases from the energy sector,improve efficiency in energy use, promote sustainable develop-ment, stimulate the use of renewable energy and increase interna-

tional cooperation. These measures include carbon dioxide emis-sion and energy taxation, new programmes to improve the effi-ciency of energy use, encourage the use of renewable forms of en-ergy, sustainable development and reduction in the amount ofheating provided by electricity (Swedish Energy Agency 2002).

Stricter environmental legislation to phase out CFC and HCFC re-frigerants has also been implemented in Sweden. The Swedish par-liament banned new installation with CFC refrigerants as of Janu-ary 1995 and prohibited the use of these refrigerants from January2000. For HCFC refrigerants, new installation was banned from1998. The phase out of CFC and HCFC refrigerants has led tonew refrigeration system designs with lower refrigerant charge andthe introduction of new refrigerants.

The supermarket sector has been affected by the replacement ofCFC and HCFC refrigerants and by a major reassessment of theuse of energy. New system solutions with indirect systems havebeen introduced in supermarkets in Sweden to minimize thecharge for refrigerant and refrigerant leakage. Supermarkets inSweden use approximately 3% of the total electricity consumed inthe country, which is equivalent to 1.8 TWh/year. The potentialfor increasing energy efficiency in stores is large.


13/262

7

Since the energy systems of a supermarket are relatively complexand new ideas and concepts have been introduced in supermarketsto decrease energy usage and minimize refrigerant charge, a com-puter model may assist designers and engineers with decisions re-garding and validation of energy measures that influence energyuse in supermarkets. The main objective of the computer model isto evaluate energy efficient measures in supermarkets with a focuson energy usage, environmental impact (TEWI) and economy(Life Cycle Cost) of refrigeration systems.

1 . 1 B a c k g r o u n d

In 1998 the Department of Energy Technology of the Royal Insti-tute of Technology of Sweden started a project in co-operation with the two major supermarket chains in Sweden, COOP Swedenand ICA AB - different companies engaged in refrigeration tech-nology and the Swedish Energy Agency. Modelling and fieldmeasurements of supermarket energy systems have been under-

taken in the research project known as “The Energy EfficientSupermarket” for more than six years. The results have been partof the Swedish contribution to the IEA Annex 26 (AdvancedSupermarket Refrigeration/Heat Recovery Systems) under theIEA Implementing Agreement on Heat Pumping Technologies(Baxter 2003).

The overall aim of the project “The Energy Efficient Supermar-ket” was to develop a simulation model where indoor climate,HVAC system, display cases, cooling and freezing rooms and therefrigeration system of a supermarket can be simulated for one

year.

1 . 2 P u r p o s e

The purpose of this thesis is to investigate and model the energyperformance in supermarkets. A user-friendly computer programthat can calculate the total energy consumption of a supermarket with reasonable accuracy has been developed. The model describesthe properties of the different components in the system when dif-ferent energy measures are compared. The investments and opera-tional costs are implemented through a Life Cycle Costing with fo-cus on the refrigeration system. The environmental impact from

the refrigeration system is characterized through the Total Equiva-


14/262

8

lent Warming Impact measure. Designers and engineers could usethe model to make decisions regarding energy improvement.

Field measurements in seven different supermarkets in Sweden were carried out to investigate the most important parameters thatinfluence energy performance in supermarkets, to analyse the op-eration of new system designs with indirect systems implementedin Sweden during recent years, and to validate the computer modeldeveloped.

1 . 3 M e t h o d

The main objective of this thesis is to investigate and model energyperformance in supermarkets.

The modelling method applied has four different phases. The firstphase is the development of a conceptual model that includes itsobjectives, the environment and components of the system and

their interconnections. The second phase is a quantitative model in which the ideas from the conceptual model are transformed intomathematical and physical relationships. The third phase is anevaluation of the model with a sensitivity analysis of its predictionsand with a comparison between the computer model and resultsfrom field measurements. The fourth phase is the model applica-tion in which the computer model answers the questions identifiedin the beginning of modelling process.


15/262

9

1 . 4 P u b l i c a t i o n s

1 . 4 . 1 P a p e r s t o C o n f e r e n c e s a n d J o u r n a l s Arias, J., & P. Lundqvist. “Innovative System Design in Supermar-kets for the 21st Century”. 20th International Congress of Refrigera-tion, Sydney, Australia, September 1999.

Arias, J., & P. Lundqvist. “Field Experiences in Three Supermar-kets in Sweden”. Workshop Annex 26, Stockholm, Sweden, Octo-ber 2000.

Arias, J., & P. Lundqvist. “Comparison of Recent RefrigerationSystems in Supermarkets”. VI Ibero-American Congress of AirConditioning and Refrigeration, Buenos Aires, Argentina, August2001.

Arias, J., & P. Lundqvist. “Heat Recovery in Recent Refrigeration

Systems in Supermarkets”. 7

th

Energy Agency Conference on HeatPumping Technology, Beijing, China, April 2002.

Arias, J., & P. Lundqvist. “A Computer Model that Compares Dif-ferent Refrigeration Systems in Supermarkets”. Eurotherm Semi-nar No. 72, Valencia, Spain, March 2003.

Arias, J., & P. Lundqvist. “Modelling and Experimental Validationof Advanced Refrigeration Systems in Supermarkets”. Journal ofProcess Mechanical Engineering 2004 (Accepted).

Arias, J., & P. Lundqvist. “Heat Recovery and Floating Condens-

ing in Supermarkets”. Journal Energy and Building 2004 (Submit-ted).

1 . 4 . 2 R e p o r t

Arias, J., & P. Lundqvist. Final Swedish Report Annex 26: “Ad- vanced Supermarket Refrigeration/Heat Recovery Systems”, 2003.


16/262

10

1 . 4 . 3 O t h e r P u b l i c a t i o n s

Arias, J., & P. Lundqvist. ”Den Energieffektiva butiken i teori ochpraktik”. Klimat 21 - dagen, Stockholm, Sweden, December 1998.

Arias, J., & P. Lundqvist. ”Den Energieffektiva butiken i teori ochpraktik”. Klimat 21 - dagen, Göteborg, Sweden, November 1999.

Arias, J., & P. Lundqvist. ”Den Energieffektiva butiken i teori ochpraktik”. Klimat 21 - dagen, Eskilstuna, Sweden, November 2000.

Arias, J., & P. Lundqvist. ”Den Energieffektiva butiken”. 1:a eff-sys-dagen, Stockholm, Sweden, January 2002.

Arias, J., & P. Lundqvist. ”Den Energieffektiva butiken”. 2:a eff-sys-dagen, Göteborg, Sweden, January 2003.

Arias, J., & P. Lundqvist. ”Den Energieffektiva butiken”. 3:e eff-

sys-dagen, Eskilstuna, Sweden, January 2004, Slutrapporten.

1 . 5 D i s p o s i t i o n o f t h e T h e s i s

This thesis has been divided into nine chapters:

• Chapter 1 describes the background, purpose and methods ofthe thesis, as well as papers and reports presented in confer-ences, journals, and seminars;

• Chapter 2 gives an introduction to energy usage and environ-mental impact in supermarkets. An historical summary aboutsupermarkets is presented in this chapter;

• Chapter 3 presents an overview of different refrigeration sys-tem designs used in supermarkets, such as the direct system,completely indirect system and partially indirect system;

• Chapter 4 describes results from field measurements in sevensupermarkets in Sweden;


17/262

11

• Chapter 5 presents the different phases of the modelling proc-ess with a detailed explanation of the conceptual and quantita-tive models;

• Chapter 6 describes the different windows, input data, inter-face with users, calculation proceedings, and results from the

model;• Chapter 7 evaluates the model through sensitivity analyses and

comparisons between the model and field measurements;

• Chapter 8 describes two applications of the model, whichcompare heat recovery versus floating condensing and direct versus indirect refrigeration systems;

• Chapter 9 presents conclusions, discussion and future studies.


18/262

12


19/262

13

2 Energy Usage andEnvironmental Impact inSupermarkets

2 . 1 I n t r o d u c t i o n

The history of supermarkets began in the middle of the 18th cen-tury, when small food stores opened in different countries. A sin-gle person, normally the owner, who oversaw the purchasing andselling of products, ran the stores. At the end of the 18th century,some companies started chains of grocery stores to sell firstly their

products and later other kinds of everyday commodities. Thesestores offered a limited selection of dry goods. At the beginning ofthe last century, grocery stores carried a range of grocery items,such as dry goods, meat and dairy products. Customers made pur-chases every day in small quantities at the neighbourhood grocerystore, principally milk and bread. Some grocery stores offeredcredit and home delivery of everyday commodities to their cus-tomers (Mitchman 1995).

Around 1930, supermarkets emerged as a new concept of foodselling. Their main characteristics were low prices, self-service, im-

pulse buying, big sale areas, high volumes, cash sale and free park-ing. The emergence of supermarkets was also due to recently developed technologies such as mass communication, refrigerationsystems, cash registers and the automobile (Mitchman 1995). Theirappearance also transformed the wholesale sector. Wholesalersstarted groups of independent retailers to develop their own resaleproducts, and to take advantage of large-scale purchasing in orderto compete with national chains. Wholesalers were also influencedby the development of road infrastructure and availability of lorriesthat reduced the cost of transportation and moved wholesale pur-chasing to industrial areas with easy access to the transport system.


20/262

14

In Sweden, the first cooperative company that purchased and dis-tributed goods commenced around 1850 in Örsundsbro. The firstself-service store was introduced in Sweden in 1940. After the Sec-ond World War, structural changes of society influenced by devel-opment in the United States affected the food distribution system. There were approximately 30,000 grocery stores in Sweden at thebeginning of the 1950s. The structural reorganisation of the Swed-ish food industry reduced the number of stores to 12,000 in 1970,9,000 in 1980, 8,300 in 1990, and approximately 6,600 stores in1996. Structural changes in the Swedish food industry also affected wholesaling of merchandise. At the beginning of the 1950s, thethree dominant wholesalers in Sweden had 231 distribution storesand an average turnover of five million Swedish crowns per store.In 1986, the same wholesalers had 48 distribution stores and anaverage turnover of more than 120 million Swedish crowns perstore (Svensson 1998).

Development of the food industry has been influenced by some

factors that can be divided into three categories: competition, con-sumer preferences and technology (Mowery 1999). Competition inthe food industry has prompted creation of new strategies for sur- vival in an over-saturated market. Some supermarket chains havefocused on low prices, offering more store brands and less cus-tomer services. Other supermarket chains have extended the salearea with non-food merchandise lines and service departmentssuch as banking, nurseries, florists and pharmacies (Mitchman1998). Another driving force of supermarket development hasbeen consumer preferences that have influenced marketing strate-gies and food organization. Some aspects such as ethnic food,

more men buying in supermarkets, shorter time allotted for shop-ping and healthy eating trends have affected promotional strategiesand the form of supermarkets. The third influential factor is tech-nology with the incorporation of scanning registers, distribution ofinformation, product flow and information recorded about cos-tumers.

The interior design of supermarkets has also changed during re-cent years. The layout of supermarkets has been designed to makeshopping easier but also to increase “impulse shopping”. Shortershopping time and a growing demand for ready-to-eat food haveaffected the interior design and circulation pattern of customers in

supermarkets. The traditional design induced customers to buy ad-


21/262

15

ditional merchandise through long distances between the entranceand products and between products and check-out (see Figure2-1). Yet this layout can actually reduce sales since customers maybecome discontent and decide to shop at convenience stores. Dif-ferent user-friendlier interior designs have been implemented where products that customers purchase frequently are groupedtogether in one part of the sale area and frozen products are lo-cated close to check-out to avoid defrosting (Mowery 1999). Thegrowing demand for easy-to-prepare foods has increased the saleof frozen food. In Sweden, the purchase of frozen food has in-creased from about 172,927 metric tons in 1975 to 438,114 metrictons in 2002.

Checkout Area

Office

Dairy

Market

Green

Market

Machine

RoomMeat

Cooler

Cutting

Room F r o z e n F o o d s

Deli Cooler

Deli Cabinet

F r o z e n F o o d s

M e a t C a b i n e t

Produce

Produce Cooler

Entrance

Figure 2-1: Layout of a Traditional Supermarket

The introduction of online shopping in the late 1990s has also af-fected the supermarket sector. Customers’ growing interest ininternet may increase the potential of the electronic supermarketindustry. Online retail is also attractive from an economic point of view because e-trade does not require paying for checkout clerks,display cases or parking lots (Johnson 2000). In the U.S., online re-

tail is growing at 35% a year in comparison with about 5% in tradi-


22/262

16

tional retail (Jones 2004). In Sweden, only a few percent of the to-tal customers ordered goods through the internet in 2001. How-ever, a survey by the company SIFO shows that about 20% of cus-tomers are interested in online shopping. Three of the four com-panies that offered online shopping have discontinued this service. The reasons are poor sales and extra costs for order administra-tion, selection of products and transport (Supermarket 2002).

The appearance of superstores has also influenced the supermarketindustry. Hypermarkets have incorporated low price products, wider range of food and non-food items and additional servicessuch as banking, nurseries and restaurants. Superstores increaseproductivity and profitability of the selling space (Mitchman 1995).Hypermarkets have come to dominate the market during the lastten years.

In 1995, the 700 largest superstores in Sweden with a turnoverhigher than 50 million Swedish crowns per year (1 SEK is equiva-

lent to 0.07 Euro) had about 46% of the total turnover fromstores. In 2001, the number of superstores with a turnover higherthan 50 million SEK per year increased to 830 stores with about61% of the total turnover. On the other hand, the approximately3500 stores with a turnover less than 10 million SEK per year hadabout 12% of the total turnover in 2001 (see Table 2-1).

Table 2-1: Percentage of Total Turnover for Four Different Groups ofStores in Sweden (Source: Supermarket 2000, Supermarket 2002 andSupermarket 2003 )

Turnover:more than

50m SEK

Turnover:20-50m

SEK

Turnover:10-20m

SEK

Turnover:less than

10m SEK1995 46% 27% 11% 16%

1996 47% 27% 11% 15%

1997 49% 26% 10% 15%

1998 53% 24% 9% 14%

1999 56% 23% 8% 13%

2000 57% 23% 8% 12%

2001 61% 20% 7% 12%

2002 63% 19% 6% 12%


23/262

17

The number of supermarkets with a turnover less than 10 millionSEK per year decreased by about 440 stores during the period1997–2002 (Supermarket 2000; Supermarket 2002; Supermarket2003).

Supermarket chains cover the market with different store formats.Superstores, supermarkets, low price stores, conveniences storesand grocery markets compete to attract customers from differentlocal areas. Store formats and average surface of stores are signifi-cantly different even in neighbouring countries.

The definition of a superstore is a self-service store with a widerrange of food and non-food items. In Sweden, the sale area of asuperstore is greater than 2500 m2 (Supermarket 2002), in Ger-many, the area of a superstore is greater than 4000 m2 (Harnisch2003) and in the U.S., the average surface area of hypermarkets is11500 m2 (UNEP, 2002). Typical surface areas of supermarketsand hypermarkets in different countries are presented in Table 2-2

(UNEP, 2002).Table 2-2: Typical Surface Areas of Supermarkets in Different Countries(Source: UNEP, 2002)

Brazil China France Japan USA

Average Surface Areaof Supermarkets (m2 )

680 510 1500 1120 4000

Average Surface Areaof Hypermarkets (m2 )

3500 6800 6000 8250 11500

The number of supermarkets and hypermarkets are also differentdepending on the country (see Table 2-3). The economic growthof China in recent years has also influenced the supermarket sec-tor. As an example, the number of small stores, with an averagesurface area of about 380 m2, has increased six times during thelast four years (UNEP, 2002).


24/262

18

Table 2-3: Number of Supermarkets and Hypermarkets (Source: UNEP,2002)

Number ofSupermarkets

Number ofHypermarkets

EU 58134 5410

Other Europe 8954 492

USA 40203 2470

Other America 75441 7287

China 101200 100

Japan 14663 1603

Other Asia 18826 620

Africa, Oceania 4538 39

Total 321959 18021

In Sweden, the number of stores in 2003 was approximately 6100.Four supermarket chains dominated the market until 2003 whenLidl and Netto opened new stores in Sweden (Supermarket 2004). The Swedish grocery market in 2003 is presented in Table 2-4.

Table 2-4: Swedish Grocery Market 2003 (Source: Supermarket, 2004)

Swedish Grocery Market 2003 Stores Turnover(mil. SEK)

Percent Turnover

ICA 1791 74913 45.0%

Kooperationen 879 36709 22.0%

Axfood 890 36556 21.9%

Bergendahl Group 129 5303 3.2%

Netto 28 700 0.4%

Lidl 28 265 0.2%

Other Grocery Stores 2310 12187 7.3%

Total 6060 166633 100%


25/262

19

The supermarket chain ICA has 45% of the total turnover in theSwedish grocery market. Kooperationen and Axfood both haveabout 22% of the turnover. The other grocery stores, i.e. trafficand services stores, have 7.3% of the turnover.

The majority of the new supermarkets constructed in Sweden upuntil 2002 were superstores with a sale area greater than 2000 m2. This tendency changed in 2003 when Lidl and Netto appeared onthe Swedish market (Supermarket 2004). During 2002, the numberof new supermarkets constructed was 20 with an average sales areaof 2400 m2. In 2003, the number of new supermarkets constructed was 59 with an average sales area of 1600 m2 (Lidl and Netto con-structed 38 of those 59 new supermarkets). Lidl and Netto are lowprice stores with a limited sale area (the average sales area of Lidlstores is 1600 m2 and 700 m2 for Netto stores) (Supermarket2004). Figure 2-2 illustrates the number of new supermarkets inSweden and their sales areas.

New Supermarkets in Sweden

0

500

1000

1500

2000

2500

3000

1997 1998 1999 2000 2001 2002 2003

Year

S a l e s A r e a m 2

0

10

20

30

40

50

60

70

N e w S u p e r m a r k e t s b y Y e a r

Average of Sales Area Number of New Supermarkets

Figure 2-2: New Supermarkets in Sweden

An analysis of the overall cost structure of a typical supermarket inSweden including the product costs and the profit (Lundqvist2000) is shown in Figure 2-3. Product costs are 76%, 11% are sal-ary costs, 3% are rent costs, 2% are marketing costs, 4% are othercosts, 1% is energy cost and 3% is profit.


26/262

20

Cost Structure and Profit

Profit

3%

Products

76%

Others

4%

Energy

1%Marketing

2%

Rent

3%

Salaries

11%

Figure 2-3: Cost Structure and Profit of a Typical Supermarket

Figure 2-4 presents the overall cost structure of a typical super-

market in Sweden excluding the product costs and the profit. Sal-ary costs are 52 %, 11% are rent costs, 10% are marketing costs,19% are other costs, and 5% is energy cost.

Cost Structure, Products Excludes

Salaries

52%

Rent

11%

Marketing

10%

Others

19%

Energy

5%

Figure 2-4: Cost Structure of a Typical Supermarket excluding ProductCosts


27/262

21

The cost of energy for this supermarket is only 1% of the totalturnover. Since the profit is 3% of the turnover, a 50% reductionof energy consumption gives a 15% increase in profit.

2 . 2 E n e r g y i n S w e d e n

National and EU decisions determine, among other things, theconditions of the energy market in Sweden. The objective of thecurrent energy policy in Sweden is to ensure a reliable supply ofelectricity and other forms of energy on a competitive market andto create the right conditions for efficient use and cost efficientsupply of energy with minimal influence on climate, health and theenvironment. The Swedish energy balance, presented in Figure2-5, shows that the total energy supplied in 2002 was about 616 TWh (including 5 TWh electricity import). The energy suppliedfrom heat pumps to the energy system is the output heat. The totalfinal energy use in 2002 was about 401 TWh spread over threeuser sectors: Industry, Transport, Residential and Services (see

Figure 2-6). The total losses were about 215 TWh. The losses innuclear power are 132 TWh. Electricity and bio-fuels are the mostimportant energy carriers for the industrial sector, oil products arethe most important for transport, and district heating are the mostimportant for the residential and services sector (Swedish Energy Agency 2003).

Total Energy Supply in Sweden 2002

Crude Oil and

Oil Products 199 TWh

Natural Gas

9 TWh

Coal and Coke

29 TWh

Biofuels, Peat, etc.

98 TWhHeat Pumps

7 TWh

Hydro Power

67 TWh

Nuclear Power 201 TWh

Wind Power

0.5TWhElectricity Import

5 TWh

Figure 2-5: Total Energy Supply in Sweden 2002 (Source: Swedish En-ergy Agency, 2003)


28/262

22

The electricity markets in the Nordic countries and the EU havebeen transformed from national monopolistic companies into in-ternational markets during the last few years.

Total Energy Use 2002

Industry

152 TWh

Residential, Commercial,

and Service Sector, etc

155 TWh

Domestic Transport 94 TWh

Figure 2-6: Total Energy Use in Sweden 2002 (Source: Swedish Energy Agency, 2003).

Sweden was the second Nordic country after Norway to set up acompetitive electricity market in 1996. Today, all the Nordic coun-tries except Iceland are integrated in the Nordic electricity market,known as Nord Pool. Germany, Poland and some Baltic states arealso active on the Nordic electricity market. The price of electricityin the Nordic countries is determined by the availability of hydro-power in Norway and Sweden, nuclear power in Finland and Swe-den, and fuel prices.

Electricity and district heating are the most important energy carri-ers for the residential and services sector. Electricity production inSweden in 2002 was about 143 TWh. Hydropower and nuclearpower produce most of the electricity in Sweden. 46% of the totalelectricity production, which is about 66 TWh, was supplied byhydropower in 2002. Nuclear power production amounted to 65.6 TWh, which is also 46% of the total electricity production. In addi-tion to nuclear and hydropower, Sweden also operates wind andcombustion-based power production. Wind power production hasexpanded during the last ten years and at the end of 2002, electric-

ity production from wind power was 0.56 TWh, less than 0.4% of


29/262

23

the total electricity production for that year. Combustion-basedpower production supplied in 2002 was 11 TWh, which is about8% of the total electricity production (Swedish Energy Agency.2003).

District heating is centralised production and distribution of hot water to multiple buildings. The heat can be supplied in different ways from boilers, heat pumps, waste heat from industry, geo-thermal or co-generation plants. In Sweden, district heating provided over 40% of the heating requirements for residential andcommercial buildings. The total energy provided to the districtheating sector was 55 TWh in 2002. Heat production from bio-fuels was 35 TWh, which is more than 60 % of the total energyprovided. Sweden has about 13000 km of distribution mains thatsupplied 49 TWh of heating in 2002. District cooling is a similarenergy system to district heating with a centralised production anddistribution of cold water. District cooling is produced in chillers,heat pumps, heat-driven absorption chillers and free cooling from

the bottom of the sea or lakes. In Sweden, the length of distribu-tion mains amounts to over 220 km, which supplied about 600GWh of district cooling in 2002 (Swedish Energy Agency. 2003).

2 . 3 E n e r g y U s a g e i n S u p e r m a r k e t s

Supermarkets are intensive users of energy in all countries. Elec-tricity consumption in large supermarkets in the US and France isestimated to be 4% of the national electricity use (Orphelin 1997).In the US, typical supermarkets with approximately 3700-5600m2 of sales area consume about 2-3 million kWh annually for total

store energy use (Baxter 2003). The national average electricity in-tensity (the annual electricity use divided by the size of the facility)of a grocery store in the US is about 565 kWh/m2 per year (EnergyStar 2003). In Sweden, approximately 3% of the electricity con-sumed is used in supermarkets (1,8 TWh/year) (Sjöberg 1997). Asurvey made by one of the Swedish supermarket chains shows thatthe average energy consumption in 256 supermarkets is about 421kWh/m2 per year. The total energy consumption in a hypermarket(about 7000 m2 ) is about 326 kWh/m2 per year while the total en-ergy consumption in small neighbourhood shops (about 600 m2 ) isabout 471 kWh/m2 per year (Olsson 1998).


30/262

24

Typical Electric Energy Usage in A Grocery Store in U.S.

WaterHeating

2%

Miscellaneous

3%

Refrigeration

39%

Lighting

23%

Cooling

11%

Ventilation

4%

Heating

13%

Cooking

5%

Figure 2-7: Typical Electricity Use of a Grocery Store in the US

A typical supermarket in Sweden uses between 35-50% of its total

electricity consumption for refrigeration equipment (Lundqvist2000). Typical electricity usage of a grocery store in the US, pre-sented in Figure 2-7, shows that 39% is used for refrigeration, 23%for lighting, 11% for cooling, 4% for ventilation, 13% for heating,3% for miscellaneous, 2% for water heating and 5% for cooking(Energy Star 2003).

Energy Usage in A Supermarket in Sweden

Refrigeration

47%

Lighting

27%

HVAC

13%

Kitchen

3%

Outdoor

5%

Others

5%

Figure 2-8: A Breakdown of Energy Usage in a Supermarket in Sweden


31/262

25

A breakdown of the energy usage from a medium-sized supermar-ket in Sweden (see Figure 2-8) shows that 47% of the energy isused for medium and low temperature refrigeration, 27% for illu-mination, 13% for fans and climate control, 3% for the kitchen,5% for outdoor usage and 5% for other uses (Furberg 2000).

There is a great potential for improvement of energy systems insupermarkets. Typical efficiency improvements may involve refrig-eration systems, illumination and HVAC system. Energy-savingtechnologies such as heat recovery, floating head condensing pres-sure, defrost control, energy efficient lighting, high efficiency mo-tors, efficient control and energy efficient display cases have beenimplemented in several supermarkets to reduce energy consump-tion.

Utilization of heat rejected from condensers to heat service wateror the premises in cold climates is a good measure to improve en-ergy usage in supermarkets. Heat recovery leads to a reduction in

costs and in the usage of fossil fuels for heating. A drawback withheat reclaiming is the higher condensing temperature that increasesenergy consumption from refrigeration systems.

The floating condensing system has been implemented in somesupermarkets to reduce energy usage by refrigeration systems. Theintroduction of electronic expansion valves operating over a widerange of pressure drops allows for a low condensing temperatureat low ambient temperatures. A reduction of condensing tempera-tures increases the coefficient of performance of refrigeration sys-tems.

Illumination accounts for about 25% of total electricity used in su-permarkets. Cost savings between 25-35% of the electricity con-sumed for lighting are possible by using the most energy efficientlamp and control system, maximising the use of daylight, paintingthe surfaces of the room with matt colours of high reflectance(Irish Energy Centre, 1995).

Display cases commonly carry large refrigeration loads, especially vertical open display cabinets. The reason is that this kind of cabi-net displays a large amount of food on a small surface in the store with a large open front area (see Figure 2-9). The heat and

moisture exchanged between the products in the cabinet and thestore environment affect the refrigeration load, defrost and


32/262

26

environment affect the refrigeration load, defrost and condensa-tion on walls and products. Infiltration causes about 60-70% of thecooling load for a typical open vertical display cabinet (Axell,2002). Installing glass doors in display cases reduces the infiltrationand energy consumption of cabinets. The reason for the absenceof glass doors in display cases is to avoid placing an obstacle be-tween the customer and the product, which may hinder the cus-tomer’s impulse to purchase a new product. Results from a labora-tory test that evaluated glass doors on an open five-deck displaycase show a reduction of the total cooling load of the case by 68%(Faramarzi, 2002).

Figure 2-9: Open Vertical Display Cabinet

High relative humidity increases the frost accumulation in cabinets, which deteriorates the coil performance and reduces the refrigera-tion capacity of the coil (Datta 1998). To remove this frost, fre-quent defrost cycles are necessary, which increase energy use. Highrelative humidity produces more condensation on products andcabinet walls, which increase energy consumption by anti-sweatheaters. A low relative humidity in the supermarket decreases theoperating costs of display cases but increases the costs of the airconditioning system (Howell 1993). Dehumidification technologiesapplied in supermarkets are desiccant cooling, heat pipe-assistedair conditioning systems and dual path systems (Henderson 1999).


33/262

27

Different actors influence the implementation of new energy-saving technologies (Lundqvist 2000). The question is how to ad-dress these issues and furthermore, how to identify the key playersin the decision process. It is important to understand that theoverall objectives of these actors differ considerably. In Figure2-10, different actors that influence decisions concerning energy-efficient supermarkets are presented.

Shop-Owner

Customers

Manufacturers

Energy

Consultans

Researchers

Supermarket

Chain

Employers

Suppliers

Energy-LabellingInstitutions

Energy-Efficient Supermarket

Services

Companies

Figure 2-10: Different Actors that Influence Decisions concerning En-ergy Efficient Supermarkets

The profile and structure of a supermarket chain influence deci-sions about investment and operation of new equipment in su-permarkets. A green profile facilitates the introduction of energyefficiency technologies. A centrally-oriented structure, such as thesupermarket chain COOP in Sweden (member of Euro Coop),may make decisions with rationales that are different from a moredecentralized organization such as ICA Group.

The shop owner is responsible for the operation of and investmentin the supermarket. The owner in supermarket chains with cen-trally-oriented structures is the owner of the supermarket company while in supermarkets with decentralized organization, the owneris a private company or person. The owner usually makes the prin-cipal decisions, choosing between many different potential invest-ments. Since the cost of energy is a low percentage of the overallturnover (see Figure 2-3), a more attractive interior decoration ofthe shop or better marketing may promise shorter payback figures. Yet energy efficiency is a driving force for a green profile that may

attract customers. Relatively short paybacks on simple energy effi-


34/262

28

ciency measures such as night lids/curtains, timers, cleaning of fancoils, etc. are attractive today.

Energy consultants designing supermarkets are making selectionsamong various overall solutions and concepts for the supermarketowner. The consultants are relatively knowledgeable in various sys-tem solutions and techniques and the way these affect the energyefficiency of a supermarket, but they cannot handle the complexityon HVAC and refrigeration systems in a changing climate. The in-fluence on the owner is thus considerable. Low life cycle cost is a weak argument in these discussions. Low first cost is often givenhigh priority. Manufacturers of components or sub-systems andservice companies are also important players. These companiesdevelop new technologies. Energy-efficient display cases, low-charge central chillers, energy-efficient illumination systems, etc.are typical examples.

Suppliers of foodstuffs such as beverages, milk products, etc. are

offering concepts such as the “milk market” - a special low-temperature zone within a supermarket. Customers’ awareness ofenergy efficiency issues and long-term climate problems are grow-ing and may lead to competitive advantages for those supermar-kets that manage to adopt, and present, an energy-efficient profilefor the future.

Other key players are energy-labelling institutions such as Envi-ronmental Protection Agencies, governmental research councils,refrigerant manufacturers, etc. Education of employers and re-search support for new energy-efficient technologies are importantactions that need to be undertaken.

It is clear that several important actors need to co-operate in orderto attain more energy-efficient supermarkets in the future. Sincethe energy systems of a supermarket are relatively complex, im-provements in one subsystem affect other systems. An example isthe heat recovery system that influences the design and operationof refrigeration and HVAC systems. Here the cooperation be-tween different actors or divisions in supermarket chains is impor-tant to implement and operate energy-saving technologies. Unfor-tunately, the overall objectives of these divisions or actors differconsiderably. A systems approach is necessary where the whole

supermarket is considered. The organization of supermarket chains


35/262


36/262

30

sponsible for the certification of companies and qualification ofpersonnel who work with refrigeration systems. In Germany, twoarticles report emission rates in the range of 5-10% from central-ised systems with direct systems in supermarkets. In Denmark,strong initiatives to phase out HFC refrigerants completely havebeen undertaken by the Danish Environmental Protection Agency. The annual emissions from new commercial refrigeration systemshave decreased from 20-25% of refrigerant charge to about 10%of charge. In Norway, supermarket chains show annual emissionsof about 14% of the refrigerant charge from 220 supermarkets.

Emission rates from supermarkets in the US were not measuredearly on, but rates between 30% to 50% of the total refrigerantcharge were values used by industry (Gage 1998). 349 stores re-ported annual emission rates between 8-15% in the time period1993-1994. Two supermarket chains located in the western andeastern parts of the US reported annual refrigerant loss rates in therange of 18-22% of refrigerant charge for 2001 and 2002 (Bivens

2004).

In Sweden, regulation pertaining to CFC, HCFC and HFC refrig-erants (presented in Table 2-5) banned new installations with CFCrefrigerants from January 1995 and stopped the use of these refrig-erants from January 2000. For HCFC refrigerants, new installa-tions were banned from 1998.

Table 2-5: Regulation of CFC, HCFC and HFC Refrigerants in Sweden

ASHRAENumber

Type ofRefrigerant

Import or NewInstallation

Banned

RefillBanned

UseBanned

R12, R502 CFC 1-Jan-1995 1-Jan-1998 1-Jan-2000

R22 HCFC 1-Jan-1998 1-Jan-2002

R134a,R404A

HFC

Refrigerant distribution in supermarkets in Sweden has varied con-

siderably because of the phase-out of CFC and HCFC refrigerants.


37/262

31

This has been verified in a Master’s Thesis carried out by Engstenand Lindh, who studied refrigerant distribution and emission ratesfrom a supermarket chain in Sweden (Engsten 2002).

Refrigerant Distribution 1996

R134a

18%

R404A17%

Others

0% R1210%

R502

15%

R22

40%

Figure 2-11: Refrigerant Distribution in 1996 in Sweden (Engsten 2002)

In 1996 the dominant refrigerant in supermarkets was R22, makingup 40% of refrigerants used in 488 stores (see Figure 2-11). CFCand HCFC refrigerants were 65% of total refrigerant used.

In 2003, refrigerant distribution was completely different. Thedominant refrigerant in supermarkets is now R404A, making up70% of refrigerants used in 371 stores (see Figure 2-12). CFC re-frigerants have disappeared from the supermarkets in the study

and R22 was only 4 % of total refrigerant used. The other refriger-ants used in the 371 stores were R407C, R417A and R507.

Emission rates from about 450 stores (average of 508 stores in1996 and 408 in 2003) belonging to two supermarket chains inSweden, COOP and ICA, are presented in Table 2-6. The amountof COOP supermarkets included in the report decreased from 488in 1996 to 371 in 2003. The ICA supermarkets are from thegreater Stockholm area. The amount of ICA stores in the study was 20 in 1996 and 32 in 2003.


38/262

32

Refrigerant Distribution 2003

R404A

70%

R134a

24%

R22

4%Others

2%

Figure 2-12: Refrigerant Distribution from a Supermarket Chain in Swe-den 2003 (Engsten 2002)

The total refrigerant charge from 508 supermarkets in 1996 wasabout 65 metric tonnes. The total refrigerant charge from 403stores in 2003 decreased to about 47 metric tonnes. The refrigerantcharge per store decreased from about 128 kg in 1996 to 117 kg in2003.

Table 2-6: Emission Rates from about 450 Stores from 1996-2003 inSweden (Engsten 2002)

Year StoresCOOP

StoresICA

TotalStores

TotalCharge

[Kg]

Leakage[Kg]

Leakage[%]

Charge/Store[kg]

Leakage/Store[kg]

1996 488 20 508 65181 7934 12.2 128.3 15.6

1997 496 28 524 65589 9278 14.1 125.2 17.7

1998 465 31 496 60556 7986 13.2 122.1 16.1

1999 452 36 488 57477 7215 12.6 117.8 14.8

2000 451 37 488 61479 7674 12.5 126.0 15.7

2001 417 39 456 55545 4784 8.6 121.8 10.5

2002 389 38 427 50404 4325 8.6 118.0 10.1

2003 371 32 403 47210 5288 11.2 117.2 13.1


39/262

33

The emission rates, from the stores in the study, decreased from14.1% of the total refrigerant charge in 1997 to 8.6% of totalcharge in 2001 (see Figure 2-13 and Table 2-6).

Total Leakage (~ 450 Stores) 1996-2003

12.2%

14.1%13.2%

12.6% 12.5%

8.6% 8.6%

11.2%

0%

2%

4%

6%

8%

10%

12%

14%

16%

1996 1997 1998 1999 2000 2001 2002 2003

Year

L e a k a g e p e r T o t . I n s t a l l e d A m o u n t

Figure 2-13: Emission Rates from about 450 Stores from 1996-2003(Engsten 2002)

The reason for the large decrease between 2000 and 2001 might bea new contract between COOP and the service companies. In thecontract, the service companies are responsible for the refilled re-frigerant and prevention of leakages. In 2003, emissions increasedto 11.2% of the total charge. The reason for the increase is unclear(Engsten 2002).

The study also shows that a few stores have large leakage prob-

lems. About 50% of all COOP stores in the study have a leakageof 4% or less and 67% of COOP stores have a leakage of less than10%. On the other hand, the five percent of the stores with thehighest leakage have a leakage over 40% and 70-80% of the annualleakage from 1999-2003 was caused by 25% of the stores. Five se-lected stores with the highest leakage have, or have had during themajor part of the studied period, a centralized direct system. Fourof the five stores have one refrigeration system for the cabinetsand storage in the medium and low temperature levels. When thereis a leak, it is not unusual that the total charge of the system leaksout (Engsten 2002).


40/262

34

2 . 5 T E W I a n d L C C P

Global environmental impacts such as ozone layer depletion andglobal warming due to greenhouse gas emissions to the atmos-phere have been associated with refrigeration systems during re-cent years. The influence on the ozone layer from CFC and HCFCrefrigerants is well known and stronger legislation has been im-

plemented to decrease the use of these refrigerants. RefrigerantsCFC and HCFC are also recognised as greenhouses gases sincethese primarily substitute the refrigerant HFC. The concept of TEWI is a useful tool when studying the influence of a refrigera-tion system on global warming. The TEWI combines the directemissions of CO2 due to refrigerant leakage and refrigerant lossesat the end of the system’s life and the indirect emissions of CO2 associated with energy consumption and generation.

The TEWI calculation of a refrigeration system is based on the fol-lowing relation:

( )( )losses ref ref TEWI M N M 1 GWP

RC E N

= ⋅ + ⋅ − κ ⋅ +

+ ⋅ ⋅ (2.1)

Where Mlosses is the refrigerant leakage, N is the lifetime of the re-frigeration system, Mref is the refrigerant charge, κ is the recyclingfactor, GWPref is the Global Warming Potential of Refrigerant, RCis the Regional Conversion Factor, which is the emission of CO2 per unit of energy delivered, and E is the annual energy consump-tion of the equipment.

The two first parts of equation (2.1) are the direct impact, whichtake into consideration the refrigerant leakage during the lifetimeof the system and the refrigerant losses at the end of the system’slife. The second part of the equation is the indirect impact thattakes into account the energy used during the lifetime of the re-frigeration system and the CO2 emissions from the production ofelectricity. The CO2 emissions from electricity generation is calcu-lated with a Regional Conversion Factor RC, which is the emissionof CO2 per unit of energy delivered in kg CO2/kWh (Sand 1997).

The regional conversion factor varies from country to country due

to the efficiency of power plants and regional fuel mix. The aver-


41/262

35

age of CO2 emissions from a carbon power plant is about 1.11 [kgCO2/kWh], from an oil power plant is about 0.77 [kg CO2 /kWh],from a gas power plant is about 0.55 [kg CO2/kWh] and from anuclear and hydroelectric power plant is 0.00 [kg CO2 /kWh]. Thebest value of the regional conversion factor for Sweden is about0.04[kg CO2/kWh], for Denmark it is about 0.84[kg CO2/kWh],for Norway about 0.00[kg CO2/kWh] and for Finland about0.24[kg CO2/kWh] (Sand 1997).

Life Cycle Climate Performance is a concept that takes into con-sideration the effect of production of the refrigerant in the systemin addition to the direct impact of leakage and indirect impact ofenergy used.

Values of Global Warming Potential (GWP) (which is the contri-bution to global warming of the refrigerant as compared to anequivalent amount of CO2 ) and values of emissions and energyused in the production of refrigerant are presented in Table 2-7 for

different refrigerants (Spatz 2003).Table 2-7: Refrigerant Global Warming Potential

RefrigerantR-22 R-410A R-290 R-404A

GWP 1700 1975 3 3784Manufact 390 14 2 18

Total 2090 1989 5 3802

2 . 6 C o n c l u s i o n s

Supermarkets are intensive users of energy in several countries.Electricity consumption in large supermarkets in the US and inFrance is estimated to be 4% of national electricity use. In Sweden,approximately 3% of total electricity consumption is used in su-permarkets.

There is great potential for improvement of energy systems in su-permarkets. Typical efficiency improvements may involve betterrefrigeration systems, illumination and HVAC system. Energy-saving technologies such as heat recovery, floating head condens-ing pressure, defrost control, energy-efficient lighting, high effi-ciency motors, efficient control and energy-efficient display cases


42/262

36

have been implemented in several supermarkets to reduce energyconsumption.

Different, sometimes conflicting, interests influence the implemen-tation of new energy-saving technologies. The cooperation be-tween different actors or divisions in supermarket chains is ofgreat importance in implementing and operating energy-savingtechnologies. Unfortunately, the overall objectives of these divi-sions or actors differ considerably. A system approach is necessary where the whole supermarket is considered. Supermarket chainsmust take into consideration the supermarket as a system whendifferent energy measures are discussed.

Stricter environmental legislation to phase-out CFC and HCFC re-frigerants and the impact of refrigerant leakage on the environ-ment have affected the refrigeration system in supermarkets. InSweden, refrigerant distribution in supermarkets has varied consid-erably following the phase-out of CFC and HCFC refrigerants. In

a study carried out by Engsten and Lindh at the Department ofEnergy Technology at KTH, the CFC and HCFC refrigerants usedin 488 stores in 1996 were about 65% of the total refrigerant in-stalled. In 2003, CFC refrigerants disappeared and HCFC refriger-ants were only 4% of the refrigerant used in 371 stores. On theother hand, the emission rates from 524 supermarkets in 1997 wasabout 14.1% of the total refrigerant charge, while in 2003 theemission rate in 403 stores was about 11.2% of the total refrigerantcharge. The study also shows that a few stores have large leakageproblems. About 50% of all COOP stores in the study have aleakage of 4% or less and 67% of COOP stores have a leakage of

less than 10%. The five percent of stores with the highest leakage,however, have a leakage over 40% and 70-80% of the annual leak-age from 1999-2003 was caused by 25% of the stores.


43/262

37

3 Refrigerat ion Systems inSupermarkets

The purpose of refrigeration systems in supermarkets is to providestorage of and display perishable food prior to sale. Food is storedin walk-in storages before the transfer to display cases in the salearea. There are two principal temperature levels in supermarkets:medium temperature for preservation of chilled food and lowtemperature for frozen products. Chilled food is maintained be-tween 1°C and 14°C, while frozen food is kept at -12°C to -18°C,depending on the country. The evaporation temperature, for a

medium temperature system, varies between –15°C and 5°C, andfor a low temperature system, the evaporation temperatures are inthe range of –30°C to -40°C. Variations in temperature are de-pendent upon products, display cases and the chosen refrigerationsystem (UNEP, 2002).

Three main types of refrigeration systems are used in stores: stand-alone equipment, condensing units and centralised systems(UNEP, 2002). Stand-alone or plug-in equipment is often a displaycase where the refrigeration system is integrated into the cabinetand the condenser heat is rejected to the sales area of the super-market. The purpose of plug-in equipment is to display ice creamor cold beverages such as beer or soft drinks.

Condensing units are small-size refrigeration equipment with oneor two compressors and a condenser installed on the roof or in asmall machine room. Condensing units provide refrigeration to asmall group of cabinets installed in convenience stores and smallsupermarkets.

Centralised systems consist of a central refrigeration unit located ina machine room. There are two types of centralised system: directand indirect system. In a direct system (DX), racks of compressors

in the machine room are connected to the evaporators in the dis-


44/262

38

play cases and to the condensers on the roof by long pipes with re-frigerant. In an indirect system, the central refrigeration unit coolsa fluid1 that circulates between the evaporator in the machineroom and the display cases in the sales area.

The quest for increased energy efficiency and the phase-out ofozone depleting substances have affected refrigeration system de-sign for supermarkets considerably. The traditional CFC andHCFC refrigerants are replaced today with R404A, R134a, etc. Arenewed interest in natural refrigerants such as ammonia, propaneand CO2 has resulted in charge minimisation and relatively leakproof systems being placed in machine rooms. Supermarkets ap-pear in all kinds of sizes from small neighbourhood shops tohypermarkets and the choice of overall system solutions vary con-siderably.

New system solutions with completely or partially indirect systemshave been developed and introduced in recent years in supermar-

kets. This has been done in order to lower the refrigerant chargeand, at the same time, minimize potential refrigerant leakage. Forexample, in a supermarket in the city of Lund in the south of Swe-den, a refrigeration system with 500 kg CFC refrigerant was re-placed with a new one that uses 36 kg ammonia as refrigerant andCO2 as a secondary refrigerant (Arias 1999).

Water solutions of glycols, alcohols and chlorides have long beenused as secondary refrigerants. The increasing interest in indirectsystems has led to the development of some new secondary refrig-erants based on potassium formate and potassium acetate alone ormixed, which have good heat transfer and pressure drop character-istics. Thermo-physical properties, material compatibility, envi-ronmental pollution and toxicity, flammability and handling secu-rity and finally cost are the aspects to take into consideration whendetermining which secondary refrigerant is to be used in a particu-lar application such as in supermarkets (Melinder 2003).

1 This fluid is known by different names, such as secondary refrigerant, secondaryfluid, secondary coolant, heat transfer fluid, or brine. In this thesis, the fluid is referred

to either as secondary refrigerant or as brine.


45/262

39

The temperature profile and most important components of thesecondary refrigerant circuit are presented in Figure 3-1.

Evaporator

Compressor

Condensor

Secondary Refrigerant (Brine) Circuit

Cabinet

Cooling Coil

Evaporating Temperature

Air Temperature (Average)

Brine Temperature (Average)

Brine Temperature Outlet

Brine Temperature Inlet

Temperature

A i r - B r i n e

B r i n e - E v a p

Pump

A i r

Figure 3-1: Temperature Profile and the Components of the SecondaryRefrigerant Circuit (Melinder 1997)

Other very promising developments are phase-changing secondaryrefrigerants such as CO2 and ice slurries.

Technology with CO2 as secondary refrigerant for low temperature

has been implemented since 1995 in the Nordic countries. In Swe-den, by the year 2000, 40 supermarkets were using CO2 systems with capacities ranging from 10 to 280 kW (Sawalha 2003). CO2 systems require much lower tube diameters and the pressure dropis negligible when compared to conventional systems. CO2 is a veryinteresting heat transfer fluid for low temperature systems becauseof its transport properties and low viscosity at temperatures below-20°C. The problem with CO2 is the high pressure required (19.7bar at -20°C).

Ice slurry is another interesting secondary refrigerant. Ice slurry is a

mixture of fluid and ice particles that is produced when a water so-


46/262

40

lution of an antifreeze substance, e.g. propylene glycol or ethanol,is cooled below its depressed freezing point (Brandon 2003). Thissystem offers additional advantages with enhanced thermal capac-ity and thermal storage in the system. Ice slurry has been tested indisplay cases with promising results (Tassou 2001); (Field 2003). Tests have also been carried out in supermarkets (Ben Lakhdar2001); (End 2001). Results from a supermarket with ice slurry inFrance show higher working temperature (–5°C to -4°C instead of-10°C to -8°C for the same display cabinet), better defrost andgood distribution of ice slurry. The drawback of ice slurry systemsis the high cost of ice generation and accumulation, for which thesavings in distribution and accumulation do not compensate (Rivet2001).

Secondary systems and the minimisation of refrigerant charge maylead to an unwanted increase in the overall energy used. Compari-son between the energy consumption of a primary and secondaryrefrigeration system have been carried out in different studies.

Theoretical calculations and practical experiences confirm the in-crease in energy consumption due to the extra temperature differ-ences and pumps introduced into the system (Clodic 1998; Yunting 2001). Recent theoretical and experimental studies, how-ever, indicate that indirect systems are more energy efficient thandirect systems, if properly designed (Lindborg 2000); (Horton2002); (You 2001); (Faramarzi 2004). More efficient defrosting sys-tems, better part load characteristics and more reliable systems arebelieved to contribute to energy savings. An evaluation using aconcept like Total Equivalent Warming Impact (Sand 1997) maybe used to estimate the overall environmental impact of a system.

Calculations made for the two extremes: Sweden (nuclear and hy-dropower) and Denmark (coal and gas) show that the impact fromlarge refrigerant leakages always dominates over CO2 releases re-lated to the production of electricity (Arias 1999).


47/262

41

3 . 1 D i r e c t S y s t e m

The most traditional refrigeration system design in supermarkets isthe direct system (Figure 3-2). In direct systems, the refrigerant cir-culates from the machine room, where the compressor is found, tothe display cases in the sales area where it evaporates and absorbsheat. The system requires long pipes to connect the compressors

to display cases and to the condensers on the roof. This implies very large refrigerant charges.

Sales Area

Deep-FreezeDisplayCabinet

Display

Cabinet

Racks of Compressors

Direct system

Evaporator Evaporator

Condensers

Machine Room

Figure 3-2: System Design 1: Direct System

The most common direct system in supermarkets is the multiplexrefrigeration system, which consists of a rack of compressors op-erating at the same saturated suction temperature with commonsuction and discharge refrigeration lines (Baxter, 2003). Theamount of refrigerant in a centralised direct system is typically 4-5kg/kW of refrigeration capacity (Baxter, 2003). Another direct re-frigeration system used in supermarkets is the single-compressorcondensing system, which provides refrigeration to a small set ofdisplay cases (Horton, 2002). The benefits and drawbacks of thedirect system are summarized in a strengths, weaknesses, opportu-nities and threats (SWOT) analysis (see Table 3-1).


48/262

42

Table 3-1: SWOT of Direct Systems

Strengths

- Good efficiency

- Less components than indirectsystems

- Lower investment cost thanindirect system

Weaknesses

- Large refrigerant charges

- Built on site: Leakage

- No possibility to use ammonia

or HC refrigerants

Opportunities

- New techniques to make tightsystems

- Mechanical subcooling 1

Threats

- Stiffer legislation for HFC re-frigerants leading to charge lim-its and environmental taxes forleakage

Another variation of the direct system is the distributed system with a separate rooftop condenser (see Figure 3-3).

Deep-Freeze

Display

Cases

Display

Cases

Refrigerant

Distributed System DX

Sale Area

Condensers

Racks of Compressors

Evaporator

Figure 3-3: Distributed System DX

1 Mechanical subcooling means that a vapour compression cycle is used for the purpose

of providing subcooling to the main refrigeration system (Thorton, 1994).


49/262

43

The distributed system consists of several small compressor rackslocated in boxes near the display cases. The refrigerant circuitlengths in a distributed system are less and the total refrigerantcharge will be about 75 % of multiplex systems (Bivens, 2004).

3 . 2 I n d i r e c t S y s t e m

3 . 2 . 1 C o m p l e t e l y I n d i r e c t S y s t e mRefrigeration with indirect systems has been introduced in super-markets to decrease the refrigerant charge and to minimize poten-tial refrigerant leakage. Indirect systems have many forms; one ofthem is the completely indirect system. A design with a completelyindirect system is presented in Figure 3-4. In this system design,there are two refrigeration systems (chillers) with different brinesand levels of temperature.

Sales Area

Deep-Freeze

Display

Cases

Display

Cases

Chillers

Completely Indirect System

Secondary

Refrigerant 1

Dry Coolers

Coolant Fluids

Machine Room

Pump

Secondary

Refrigerant 2

Figure 3-4: System Design 2: Completely Indirect System

The secondary refrigerant in the medium temperature level oftenhas an approach temperature around -8°C and a return tempera-ture around -4°C. A typical value of secondary refrigerant tempera-

ture going to deep-freeze display cases is about -32°C and the re-


50/262

44

turn temperature is about -29°C. Secondary refrigerants based onpotassium formate, potassium acetate, glycols, alcohols and chlo-rides are used as secondary refrigerants. CO2 vapour-liquid mightbe used as a secondary refrigerant in the low temperature system.

One or two other secondary loops (coolant fluids or dry coolerfluids) are used in the system to transport the heat rejected fromthe condensers, in the machine room, to two different dry coolerslocated on the roof of the supermarket. Typical out-going tem-perature of the coolant fluid is about 32°C and the return tempera-ture is about 36°C.

The waste heat from the condenser can be recovered during the winter with substantial energy savings in cold climates. The bene-fits and drawbacks of completely indirect systems are summarizedin a SWOT analysis in Table 3-2.

Table 3-2: SWOT of Completely Indirect System

Strengths

- Lower refrigerant charges

- Simple and cheaper service

- Use of natural refrigerantspossible

Weaknesses

- Risk for low energy efficiency

- Pump work

- Risk for corrosion

- Pipes need to be insulated

Opportunities

- Phase-changing brines: iceslurries and CO2

Threats

- More efficient direct systems

A variant of the design of the completely indirect system that im-proves its performance is presented in Figure 3-5. In this case, therefrigerant in the low temperature unit is sub-cooled with the brineof the medium temperature unit (Thorton 1994). The temperatureof the refrigerant after sub-cooling can be about +5°C.


51/262

45

Sales Area

Deep-Freeze

Display

Cases

Display

Cabinet

Mechanical Subcooling

Secondary

Refrigerant 1

Dry Coolers

Coolant Fluid

Evaporator

Machine Room

Pump

Secondary

Refrigerant 2

Figure 3-5: Completely Indirect System with Mechanical Sub-cooling

The effect of sub-cooling on the low temperature system is pre-sented in the P-h diagram for R404A in Figure 3-6.

-50 0 50 100 150 200 250 300 3510

2

10

3

10 4

h [kJ/kg]

P [kPa]

40°C

5°C

-40°C

0,9

1 kJ/kg-K

R404A

Subcooling

30% of

About

Q2

Figure 3-6: P-h Diagram for R404A with the Refrigeration Cycle of theLow Temperature System


52/262

46

The influence of sub-cooling on the refrigerating effect depends,among other things, on the kind of refrigerant. For R404A, the in-crease in refrigeration capacity for the low temperature system canbe approximately 30%, as shown in Figure 3-6. Heat from the sub-cooling process is rejected to the medium temperature system, which has a higher coefficient of performance COP2 than does thelow temperature system. This implies lower energy consumptionthan in a completely indirect system.

Deep Freeze

Display

Cases

Display

Cases

T1

Coolant Fluid

District Cooling

Return Pipe District Cooling

Sales Area

Figure 3-7: System Design with District Cooling

Another variation of the completely indirect system is presented inFigure 3-7. The system design uses district cooling to cool the heatrejected from condensers. In this installation, the display casescontain compressors, condensers and evaporators.

The district cooling return pipe cools the condenser from the dis-play cases through a secondary fluid. The coolant fluid tempera-tures fluctuate between 12°C and 16°C, which are lower thannormal. The condensing temperature of the refrigeration system isabout 20°C, which implies lower compressor power in comparison with the other system designs. One disadvantage with this solution


53/262

47

is the price of district cooling that might increase the total energycost.

3 . 2 . 2 P a r t i a l l y I n d i r e c t S y s t e m The most common partially indirect system in supermarkets inSweden is shown in Figure 3-8. The heat from the condensers is

rejected by a dry cooler on the roof of the supermarket to the environment. The low temperature system has a direct system be-tween the compressors and the deep-freeze display cases, and themedium temperature system has an indirect system between thecabinets and the chiller.

Sale Area

Deep-freeze

display casesDisplay

cases

Partially Indirect System

Brine

Refrigerant

Coolant Fluid

Dry Coolers

Evaporator

Machine Room

Figure 3-8: System Design 3: Partially Indirect System

The benefits and drawbacks of the partially indirect system aresummarized in a SWOT analysis in Table 3-3.


54/262

48

Table 3-3: SWOT of Partially Indirect System

Strengths

- Good efficiency

- Lower charge, less leakage

- Higher reliability

Weaknesses

- Refrigeration charge largerthan completely indirect system

- Pump work

- Risk of corrosion- Pipes need to be insulated

Opportunities

- Mechanical sub-cooling of lowtemperature system

Threats

- Stiffer legislation for HFC re-frigerants leading to charge lim-its

Another variation of the partially indirect is the distributed system, which has a secondary loop for heat rejection (Baxter, 2003). The

distributed system consists of several small condenser units locatedin boxes near the display cases. The heat from the condenser is re-jected by a coolant fluid loop, which circulates between the con-densers and a dry cooler on the roof (see Figure 3-9).

Deep-Freeze

Display

Cases

Display

Cases

Coolant Fluid

Distributed System Indirect

Sales Area

Dry Cooler

Condenser Units

Evaporator Evaporator

Figure 3-9: Distributed System Indirect


55/262

49

3 . 2 . 3 I n d i r e c t C a s c a d e S y s t e m The cascade system, shown in Figure 3-10 is a favourable solutionthat avoids the large pressure ratio in the low temperature systemobtained in the completely indirect system. The installation oper-ates with two different temperature levels and secondary loops. The temperature of the secondary refrigerant in the medium tem-perature unit has, as in completely indirect systems, an out-goingtemperature of about -8°C and a return temperature of about -4°C. The out-going temperature of the secondary refrigerant in the lowtemperature system is about -32°C and the return temperature isabout -28°C.

The condenser heat from the low temperature system is rejected tothe secondary refrigerant with the medium temperature. The con-densing temperature of the low temperature system is about 0°C, which increases the coefficient of performance of the refrigerationcycle and decreases the energy consumption of the low tempera-ture system. The drawback with this system is the increase of re-frigeration capacity and compressor power of the medium tem-perature system due to the condenser heat from the low tempera-ture system.

Deep-FreezeDisplayCases

DisplayCases

Chiller

SecondaryRefrigerant 1

Cascade System A

Sales Area

Dry Cooler

Documents

PhD Thesis Jaime Arias 1