Proceedings of the 5th International Conference on

Proceedings of the 5th International Conference on Improving Energy Efficiency in Commercial Buildings: IEECB Focus 2008 10 - 11 April 2008, Frankfurt am Main, Germany

Volume 2

Editors: Paolo BERTOLDI, Bogdan ATANASIU

EUR 24401 EN/2 - 2010

The mission of the JRC-IE is to provide support to Community policies related to both nuclear and non-nuclear energy in order to ensure sustainable, secure and efficient energy production, distribution and use.

European Commission Joint Research Centre Institute for Energy Contact information Address: Via E. Fermi, 2749, I-21027 Ispra (VA), ITALY E-mail: [email protected] Tel.: +39 (0)332 78 9299 Fax: +39 (0)332 78 9992 http://ie.jrc.ec.europa.eu/ http://www.jrc.ec.europa.eu/ http://re.jrc.ec.europa.eu/energyefficiency/

Legal Notice Neither the European Commission nor any person acting on behalf of the Commission is responsible for the use which might be made of this publication. Europe Direct is a service to help you find answers to your questions about the European Union Freephone number (*): 00 800 6 7 8 9 10 11 (*) Certain mobile telephone operators do not allow access to 00 800 numbers or these calls may be billed.

A great deal of additional information on the European Union is available on the Internet. It can be accessed through the Europa server http://europa.eu/

JRC 56926 EUR 24401 EN/2 ISBN 978-92-79-16017-2 ISSN 1018-5593 DOI 10.2790/21625 Luxembourg: Office for Official Publications of the European Communities © European Union, 2010 Reproduction is authorised provided the source is acknowledged Printed in Italy

Proceedings of the 5th International Conference on Improving Energy Efficiency in Commercial Buildings: IEECB Focus 2008 10 - 11 April 2008, Frankfurt am Main, Germany

Volume 2

Editors: Paolo BERTOLDI, Bogdan ATANASIU

EUR 24401 EN/2 - 2010

Preface The commercial non-residential buildings sector is one of the fastest growing energy consuming sectors. This is mainly due to the growth of commercial and public activities and their associated demand for heating, cooling ventilation (HVAC) and lighting. Moreover in the new economy, with a wide dissemination of information and communication technologies, information technology equipment is also an important source of electricity consumption. For the tertiary sector space heating is responsible for more than 50% of total consumption of the sector, while energy consumption for lighting and office equipment and "other" (which is mainly office equipment) are 14% and 16%, respectively. In its 2006 Action Plan on energy efficiency, the European Commission (EC) called for concrete measures to reduce growth in energy demand, mainly by promoting energy saving in buildings and the transport sector. According to the EU Green Paper, energy use in buildings could be reduced by at least a fifth by making greater use of available and economically viable energyefficient technologies. Such savings would also improve the energy supply security and the EU’s competitiveness, while creating job and raising the quality of life in buildings. Greenhouse gas reduction is a common denominator of many countries' environmental policies and programmes. Commercial buildings are a key area to achieve the EU 2020 20% energy savings target, and this makes economic sense for the building owners and occupiers. As a consequence of the EU 2020 commitment, all actors need to take all necessary steps to disseminate good practice, foster investment in energy efficiency and provide technical solutions for the commercial building sector. Other regions of the world are also exploring potential programme and policy options to reduce commercial building energy waste. Not only is every kWh saved avoiding pollution and CO2 emissions, but it is also reducing peak power requirements; a problem common to many countries. That is the reason why every achievement in the field of demand-side management (DSM), or more generally the improvement of energy efficiency has a direct effect on greenhouse gas emission reduction and on the security of energy supply. The European Directive on the Energy Performance of Buildings requires a major effort to improve building energy performance and will bring the energy performance of their buildings to the forefront of building market operators. This simultaneously presents an opportunity and challenge for energy efficiency. The recent liberalisation of the electricity and gas markets could be an additional opportunity in the development of these efforts, as the competition eventually developing between the key players in the electricity and gas industry will not be focused only on prices but also on the service. In the long term there is the possibility that energy services and renewable energy sources (RES) would enable greater differentiation among utilities, Energy Service Companies (ESCOs), etc. Many property managers are now offered the services of ESCOs and facility management companies to manage and reduce the energy consumption in their buildings. A number of local, regional and national policies and programmes have recently been implemented to achieve a long lasting market transformation. The Directive on energy efficiency and energy services shall further contributes to the establishment of an energy efficiency market.

Low consumption commercial (office) buildings have been constructed and operated in the EU and elsewhere and they have proven that it is feasible to reach low consumption targets. There are some very good examples of low consumption commercial (office) buildings, especially in Germany. A major result is, that the reduction of consumption of primary energy is not only some percent, but new buildings have reduced consumption by a factor of 3 to 4 ! In many cases low energy office buildings have lower investment cost than conventional ones, especially where supply efficiency can be integrated or natural cooling is used. Where the initial cost of the efficient is greater than the normal market practise, these additional investment costs invariably turn out to be economical within the expected lifetime of the buildings, even on the assumption of constant energy pricing, a totally unrealistic assumption. Furthermore energy efficient building owners and investors are happy. There is growing evidence on both sides of the Atlantic that the occupiers in high-efficiency buildings are happier, and significantly more productive. The value of the productivity normally outweighs the operating savings for the pure energy costs. Lower energy costs are combined with a good or even better comfort and substantially increased employee productivity. Thus investors and occupants are both happy with these buildings. The key question is why are such a kind of buildings still an exception and not the standard? And why cost effective building investments and retrofit do not take place. The EU GreenLight and GreenBuilding Programme (GBP) programmes help to overcome some of the barriers to energy efficiency - in particular the lack of interest and information - by providing public recognition and information support to companies and public organisations whose top management is ready to show actual commitment to adopting energy efficient measures in buildings. Following the success of the previous IEECB conferences (IEECB’98 in Amsterdam, IEECB’02 in Nice, IEECB’04 and IEECB’06 in Frankfurt), Messe Frankfurt with the scientific collaboration of the European Commission Joint Research Centre organised the fifth International conference on Improving Energy Efficiency in Commercial Buildings (IEECB’08) in conjunction with the Building Performance Congress (www.bp-congress.de). The IEECB’08 conference took place on 10 - 11 April 2008 in Frankfurt during Light+Building, the International Trade Fair for Architecture and Technology in Frankfurt, Germany. The IEECB conference brought together all the key players from this sector, including commercial buildings’ investors and property managers, energy efficiency experts, equipment manufacturers, service providers (ESCOs, utilities, facilities management companies) and policy makers, with a view to exchange information, to learn from each other and to network. At the conference key representatives of leading organisations and companies, institutions and equipment industry presented the overall picture and give details of policies, recent advancements and examples of best practice. The wide scope of topics covered during the IEECB’08 conference included: macro/micro approaches, state-of-the-art equipment and systems (lighting, HVAC auxiliary equipment, ICT & office equipment, miscellaneous equipment, BEMS, electricity on-site production, renewable energies, etc.) and the latest advances in R&D, tools, regulation & policy, demand-side and supply-side perspectives for all branches of activity (public and private sector, the commerce and retail sectors, hotels and restaurants, banks and insurance companies, local authorities,

civil services & public bodies, education, universities & laboratories, hospitals, airport and stations, etc.) We hope that the present proceedings could be a valuable contribution to disseminate information and best practices in policies, programmes and technologies to foster the penetration of highly efficient buildings in the commercial sector.

The Editors Paolo Bertoldi Bogdan Atanasiu

Contents Preface

Volume 1 Policies

1

Non-residential buildings for combating climate change: Summary of the findings of the Intergovernmental Panel on Climate Change Diana Urge-Vorsatz, Alexandra Novikova

3

Transforming UK non-residential buildings: achieving a 60% cut in CO2 emissions by 2050 Mark Hinnells, Russell Layberry, Daniel Curtis, Andy Shea Raising the efficiency of non-residential buildings through the GreenBuilding Programme: the experience of Spain Núria Quince, Cécile Bonnet, Joan Carles Bruno, Alberto Coronas The Worldwide Status of Energy Standards for Buildings: a 2007 Update Kathryn Janda Australia’s Path to Energy Efficiency in Commercial Buildings – ‘Your Building’ Best Practice Programme Jodie Pipkorn, Stephen Berry, Tony Stapledon

21

37

49

61

Reducing CO2 emissions through refurbishment of non-domestic UK buildings A. Peacock, P.F. Banfill, S. Turan D. Jenkins, M. Ahadzi, G. Bowles, D. Kane, M. Newborough, P.C. Eames, H. Singh, T. Jackson , A. Berry

73

Barriers to energy efficiency in buildings and some routes to overcoming them Constant Van Aerschot

83

Sustainability in commercial buildings – Bridging the gap from concept to operations Adam Hinge, Om Taneja, Michael Bobker

97

Life-Cycle assessment measures of energy efficiency projects Om Taneja, Adam Hinge Experiences on the programmatic approach under Joint Implementation – new opportunities for the building sector Markus Rothe

105

117

Operational rating vs asset rating vs detailed simulation Ljiljana Marjanovic-Halburd,Ivan Korolija, Rob Liddia, Andrew Wright

123

Energy Performance of Buildings Directive (EPBD)

133

Implementation of EPBD article 7.3 in Germany and the UK: comparison of methodologies and procedures and review of early outcomes R. Cohen, I Therburg, W. Bordass, J. Field

135

Towards an optimal approach to effectively incorporate feasibility studies of alternative energy systems (art. 5 EPBD) in the common building process Suzanne Joosen, Åsa Wahlström, Marjana Sijanec Zavrl,

149

A review of non-domestic energy benchmarks and benchmarking methodologies Rob Liddiard, Andrew Wright, Ljiljana Marjanovic-Halburd

163

Preparing the implementation of the EuP Directive for non-residential building ventilation Julia Oberschmidt, Peter Radgen

175

i

Renewable energies perspective to the energy performance of buildings Hans Bloem European Commission, DG JRC, Ispra, Italy

191

The seal of quality for the Energy Performance Certificate: A motor for energy modernisation! Thomas Kwapich

205

Behaviour & Investor Motivation

211

Analysis of the building owners' motivations for investing in energy efficiency Michaela Valentova , Paolo Bertoldi

213

Energy efficiency and performance of commercial real estate Anil Kashyap, Jim Berry, Stanley McGreal

227

Overcoming the commodity view Catherine Cooremans

235

Green Leases: an opportunity to develop a sustainable approach for tenanted commercial buildings in the UK Angela Langley, Lara Hopkinson, Vicki Stevenson

247

Is the client willing to pay to occupy a greener building? Sandra Gómez

257

Encouraging efficiency investments with an new energy risk management approach Jerry Jackson

265

Cold comfort for Kyoto: the link between air-conditioning in commercial buildings and consumer lifestyle choices Pedro Guertler, Jacky Pett

279

Programmes and Energy Services

291

High efficient circulation pumps for the building of Vienna Georg Benke, Edgar Hauer

293

Business hotel utility consumption and saving opportunities Paul Bannister

303

Efficient schools – Generation of best-practice examples all over Germany Felicitas Kraus, Nicole Pillen, Stefan Schirmer, Nana Doerrie

315

Volume 2 Successful Implementation

321

The new house of the region of Hannover: Using EPBD-strategies to improve energy efficiency in the building lifecycle Stefan Plesser, Ernesto Kuchen, Norbert Fisch

323

Design and simulation of an energy efficient glazed office building with a double skin façade Åke Blomsterberg

331

Designing a large office and laboratory building within the GreenBuilding Lorenzo Pagliano, Salvatore Carlucci, Tommaso Toppi, Paolo Zangheri

plus

project

Community energy planning for the city of Guelph - efficient buildings in an efficient city Peter Garforth, Karen Farbridge

ii

343

355

Design and control of a building with structurally incorporated phase change material P Bannister, P Taylor, C Ardren, M Schmidt

365

Monitoring and Building Energy Management Systems (BEMS)

377

Lessons learned from the implementation of metering and monitoring systems in public buildings in Europe – ENERinTOWN project Vasco Ferreira, Luis Alves, Guillermo Basañez-Unanue, Manuel Izquierdo González, NikosTourlis, Rodolfo Pasinetti, Antonio Siciliano, Paul Kenny

379

Determinants of energy use in the UK non-domestic stock Harry Bruhns

389

The analysis and interpretation of half hourly utility data in UK buildings Andrew Wright, Neil Brown

401

Monitoring electricity consumption in the tertiary sector as a basis for energy efficiency improvements Edelgard Gruber, Stefan Plesser, Markus Diem, Norbert Fisch

415

Non-invasive and cost effective monitoring of energy consumption patterns for electrical equipment N Brown, A J Wright

431

Can energy savings from actions promoting energy efficient behaviours in office buildings be accounted for? A French case study Broc Jean-Sébastien, Nogues Patrice, Baudry Paul, Bourges Bernard, Adnot Jérôme

439

Analysis of electricity consumption in the tertiary sector in Hungary Sonja Koeppel, Viktoryia Novikava, Benigna Boza-Kiss, Aleksandra Novikova, Steven Graning, Diana Ürge-Vorsatz

449

Heating, Ventilation and Air-Conditioning (HVAC)

463

A method to quantify and compare the performance of thermally activated buildings systems (TABS) with conventional air cooling units Olivier Pol, Anita Preisler, Hans-Joachim Kast

465

Utilization of a Deep Lake Water Direct Cooling Network (DLWDC) for cooling of a large administrative district. Energy and environmental demonstration and follow-up Pierre-Alain Viquerat, Bernard Lachal, Willi Weber, André Mermoud, Eric PampalonI

477

Prediction of the effectiveness of a chilled ceiling coupled bore hole heat exchanger system for cooling existing UK office buildings and reducing CO2 emissions Singh H., Eames P.C., Peacock A. D., Jenkins D.

491

Assessing the risks and likelihood of eliminating cooling loads in UK offices David Jenkins, Andrew Peacock, Ya Liu

501

Reversible air-conditioning potential in office buildings in Europe M.Caciolo, P. Stabat, D. Marchio

511

Proposal for the field identification of a short list of Energy Conservation opportunities related with air-conditioning equipments Julien Caillet, Jérôme Adnot, Philippe Riviere, Marco Masoero

525

Predictions of indoor CO2 levels in four UK school buildings with three selected ventilation scenarios Singh H., Eames P.C., Peacock A.D., Jenkins D.

535

iii

Using street surveys to establish Air-Conditioning incidence in UK’s commercial offices stock Jorge Caeiro, Hector Altamirano-Medina, Harry Bruhns, Hasim Altan

543

Impact of climate change in the tertiary sector of Europe (EU27+2) Martin Jakob, Giacomo Catenazzi, Eberhard Jochem Ashish Shukla

551

Allowing for thermal comfort in free-running buildings in the new standard EN15251 Fergus Nicol, Lorenzo Pagliano

565

A preliminary attempt to unify the different approaches of summer comfort evaluation in the European context Laurent Grignon-Massé, Jérôme Adnot, Philippe Rivière

575

How to help the unfortunate energy auditor of a HVAC system Cleide Aparecida Silva, Jules Hannay, Jean Lebrun, Vladut Teodorese, Cristian Cuevas

587

Lighting

601

Power density targets for efficient lighting: Practical examples Wouter R. Ryckaert, Sandra Herrebosch, Catherine Lootens, Stefaan Forment, Peter Hanselaer, Geert Deconinck

603

Concepts and techniques for energy efficient lighting solutions Eino Tetri, Wilfried Pohl

613

The use of high dynamic range luminance mapping in the assessment, understanding and defining of visual issues in post occupancy building assessments Steve Coyne, Gillian Isoardi, Michael Hirning, Mark B Luther

625

Integral approach to design building engineering systems: (lighting, heating, airconditioning) – as an effective way to Energy Saving Julian Aizenberg

635

Improvement of the energy efficiency of a distribution warehouse in Madrid (Spain) with special emphasis on daylight optimisation Cécile Bonnet, Joan Carles Bruno, Alberto Coronas

641

Market development of ESCO schemes for lighting refurbishment Mechthild Zumbusch

651

Data Centres and IT Equipment

659

Measuring the energy efficient performance of desktop and notebook computers Kevin Fisher

661

Qualitative Analysis of Power Distribution Configurations for Data Centers John Tucillo

667

iv

IEECB'08

Successful Implementation

321

322

The New House of the Region of Hannover: Using EPBD-strategies to improve energy efficiency in the building lifecycle Stefan Plesser,; Ernesto Kuchen, Norbert Fisch Institut für Gebäude- und Solartechnik, TU Braunschweig, Germany

Abstract The “New House of the Region of Hannover“ (NRH) is the first building in Germany that has been built according to the Standard „EnOB - Energieoptimiertes Bauen“ („Energy optimized building“) as defined by the German Ministry of Economics and Technology within a Public Private Partnership. This paper is elaborated in a research project on commissioning, operation and monitoring of the building carried out by the Institute of Building Services and Energy Design at the Technical University of Braunschweig and the University of Magdeburg. This paper documents x The integration of energy efficiency in the PPP competition brief using EBPD strategies, x The design competition and construction, and x Commissioning and monitoring results of the first year of building operation. The combination of a public private partnership for design, construction and financing with ambitious target values for energy efficiency is supposed to demonstrate the possibility of energy efficient buildings at low cost in a very competitive market situation. The results of 2007 show that the major goals for energy efficiency have been achieved. Figure 1 shows the front elevation of the building with the multifunctional copper-clad conference room next to the entrance.

Figure 1

New House of the Region of Hannover

323

The New house of the Region of Hannover The “New House of the Region of Hannover” provides 300 office workplaces for the public administration on a net floor area of 7.134 m²NGF1. The building includes a 700 m² conference area that can be divided into three sections. Using EPBD strategies in the Lifecycle of Buildings The PPP contract requests a high standard of energy efficiency. All calculations had to be carried out by the design and engineering team according to EnEV 2004 (Mandatory German Energy Efficiency Guideline) [1] and DIN 4108-6:2003-06 (standard calculation method) [2] for heating. For the electricity demand, the LEE Guideline [3] had to be applied using “power x hours/year” - calculations. Their combination was used to determine the overall annual primary energy demand of the building. The approach is similar to DIN 18599 [4] that defines the calculation method used to implement the EPBD in Germany. DIN 18599 had been introduced during the construction of the building. The following target values were defined in the competition brief: x The over all demand of primary energy was limited to 100 kWhPE/(m²NGFa) which is about 4050 % below the mandatory code requirements. x The primary energy demand for heating was limited to 40 kWhPE/(m²ANa) 2 (the use of district heating was mandatory) x The specific air leakage rate according to DIN EN 13829 [5] was limited to n50 < 1,5 1/h In addition, specific standards for construction and functional testing procedures had to be fulfilled: x Windows: Uw a * (CDD

SED C Base  ai * (CDD C Y ,WC ) 2  bi * (CDD C Y ,WC ) i

C

Y 0 , Base

2

)  bi * (CDD

C

@

Y 0 , Base

)

(7)

@

Table 4 Coefficients of Equation (7) ai -0.000230 -0.000015

Model base period (i=1) Time horizon of model 2050 (i=2) Source:

bi 0.5 0.2

Assumptions of the authors [last updated 22 July 2008].

Note that in both cases, a techno-economic progress that increases the energy-efficiency of providing cooling services by 0.5%/year was assumed which results in an EE improvement of 20% up to 2050. Assumptions regarding the impact of WC on the specific heating energy demand Equations (3) and (4) and respective coefficients (Table 5) describe the adopted model that relates specific energy demand for heating purposes to HDD. In accordance to the results of the building physics simulation model it is assumed that the slope mEE is maximal in the case of non-retrofitted existing buildings and minimal in the case of buildings that comply with the German Passivehouse or the Swiss Minergie-P standard. The coefficients of buildings of sectors and countries with intermediate energy-efficiency (EE) are interpolated within these two boundary cases. Hence, as the average building stock is being retrofitted between 2005 and 2050, impact of warmer climate is steadily decreased as mEE decreases.

SEDWC

SEDPC  m EE * ( HDDWC  HDDPC )

(4)

Table 5 Coefficients of Equation (3) bEE EE = existing building stock without retrofit -70 EE = well insulated buildings -30 EE = best practice (equivalent to the German Passivehouse or the Swiss Minergie-P standard) -25.0 Source:

mEE 0.20 0.05 0.023

Assumptions of the authors [last updated 22 July 2008].

In the case of no climate change (base case, present climate), non-electricity SED decrease from the current levels in all sectors with the exception of the commerce/trade sector where a decrease is detected only after a period of growth (by 15% up to 2020, see Table 6). As a result of technical progress, non-electricity SED in 2050 is expected to be 20% to almost 30% below the level of 2005 (except commerce/trade: only 8% lower). In the case of warmer climate, non-electricity SED decreases significantly more, namely by about 25% to more than 40% (commerce/trade only by 21%). Hence, in 2050 non-electricity SED of the WC scenario is between 10% and 20% below the scenario for which no climate change was assumed. Note that the impact of warmer climate differs between sectors as there are structural differences between northern and southern European countries.

560

Table 6 Resulting non-electricity specific energy demand (in MJ/m2a) of the Service sectors of Europe (EU27+2), weighted average of EU 27+2) for 2005, 2035, and 2050 and relative change Base Case

Warmer Climate (WC)

WC/Base

2020/ 2035/ 2050/ 2020/ 2035/ 2050/ 2005 2020 2035 2050 2005 2005 2005 2020 2035 2050 2005 2005 2005 2020 2035 2050 Commerce, Trade 530 610 Finance 643 556 Hotels, restaurants 761 700 Education 308 262 Health Other Source:

569 505 654 232 678 596 553 458 415 386

486 467 608 217 520 359

1.15 0.87 0.92 0.85 0.88 0.91

1.07 0.79 0.86 0.76 0.82 0.84

0.92 0.73 0.80 0.71 0.77 0.78

577 523 674 241 565 387

510 448 609 198 500 337

408 388 548 171 446 292

1.11 0.83 0.90 0.80 0.85 0.86

0.98 0.71 0.81 0.66 0.75 0.75

0.79 0.62 0.73 0.57 0.67 0.65

0.95 0.90 0.84 0.94 0.89 0.83 0.96 0.93 0.90 0.92 0.85 0.79 0.95 0.90 0.86 0.93 0.87 0.81

Jochem et al.(2007), complemented and calculated by the authors [last updated 28 July 2008].

Results regarding the aggregate electricity demand of the Service sectors of Europe In the Base Case with present climate, electricity demand is expected to increase by about 50% to 60% up to 2050 (Table 7). In the warmer climate scenario, the increase is slightly higher, namely by about 10%-points. In 2050, electricity demand is 7% higher in the warmer climate scenario than in the base case (about 1% already occurred between the base case and the year 2005). A noticeable difference can be discerned between on the one hand the Mediterranean and South-Eastern European countries and the rest of the European countries on the other hand. For the former, the difference between the base and the WC scenario is 16% whereas for the rest of the EU27+2 countries it is only 3% in 2050. Table 7 Resulting total electricity demand of the Service sectors of Europe (EU27+2) for 2005, 2020, 2035, and 2050 (PJ) and relative changes Base Case

Warmer Climate (WC)

WC/Base

2020/ 2035/ 2050/ 2020/ 2035/ 2050/ 2005 2020 2035 2050 2005 2005 2005 2020 2035 2050 2005 2005 2005 2020 2035 2050 Mediterr., SE EU 659 894 1070 1141 1.36 1.62 1.73 960 1183 1321 1.43 1.76 1.97 1.07 1.11 1.16 Rest of EU27+2 2194 2806 3286 3302 1.28 1.50 1.51 2851 3368 3412 1.30 1.53 1.55 1.02 1.02 1.03 Total EU 27+2 2853 3700 4356 4444 1.30 1.53 1.56 3810 4551 4733 1.33 1.58 1.65 1.03 1.04 1.07 Source:

Calculations by the authors [last updated 26 July 2008].

The difference between the two scenarios is caused by the different development of electricity demand for cooling. Whereas in the base case, cooling electricity increases from 310 PJ to about 830 PJ, which represents an increase of +170%, it increases from 330 PJ in 2005 to slightly more than 1100 PJ which represents an increase of +240% (see Table 8). Table 8 Resulting electricity demand for cooling of the Service sectors of Europe (EU27+2) for 2005, 2020, 2035, and 2050 (PJ) and relative changes Base Case

Warmer Climate (WC)

WC/Base

2020/ 2035/ 2050/ 2020/ 2035/ 2050/ 2005 2020 2035 2050 2005 2005 2005 2020 2035 2050 2005 2005 2005 2020 2035 2050 Mediterr., SE EU 228 470 575 572 2.1 Rest of EU27+2 81 194 257 260 2.4 Total EU 27+2 310 664 832 832 2.1 Source:

2.5 3.2 2.7

2.5 3.2 2.7

536 688 751 2.2 238 339 370 2.7 774 1027 1122 2.3

2.9 3.9 3.1

3.1 1.14 1.20 1.31 4.2 1.23 1.32 1.42 3.4 1.17 1.23 1.35

Calculations by the authors [last updated 26 July 2008].

As already stated above there is a noticeable difference the Southern countries and the rest of the EU27+2 countries. First of all, the share of electricity demand for cooling as compared to total electricity demand is larger already today: it is 35% in the Southern countries, but only 4% in the other

561

countries and only 11% in the European average, see Table 9. Moreover the share of the Southern countries increases much more distinctively, namely by 15% to 20%-points whereas in the other countries it increases only by 5% to 7%-points. Table 9 Resulting electricity demand for cooling as share of total electricity demand of the service sectors of Europe (EU27+2)for 2005, 2020, 2035, and 2050 Base Case Mediterr., SE EU Rest of EU27+2 Total EU 27+2 Source:

Warmer Climate (WC)

2005

2020

2035

2050

2020

2035

2050

0.35 0.04 0.11

0.53 0.07 0.18

0.54 0.08 0.19

0.50 0.08 0.19

0.56 0.08 0.20

0.58 0.10 0.23

0.57 0.11 0.24

Calculations by the authors, based Table 8 and Jochem et al.(2007), [last updated 24 July 2008].

Finally, it can be stated that the share of electricity demand for cooling first increases, but then decreases again, particularly in the base case, but to a minor extent also in the WC scenario. First, this is due to stronger saturation phenomena in the case of cooling (particularly regarding the share of cooled floor area, see Table 3 in the previous section) as compared to other types of electricity demand. Second, a stronger techno-economic progress was assumed in the case of cooling (0.5%/year) as compared to other electricity services (0.2% to 0.5%, see Jochem et al. (2007). Results regarding the aggregate fuel energy demand of the Service sectors of Europe The aggregate energy demand of the Service sector as a whole is obtained from the sumproduct of the floor area per sector (Jochem et al., 2007) and the specific energy demand inputs (Table 6). In the base case, non-electricity fuel energy demand is increasing by 38% up to 2035, but only by 28% in the warmer climate scenario (Table 10). Due to technical progress, a decrease up to 2050 is then expected in both cases. In relative terms, the impact of warmer climate is slightly larger in the case of the Mediterranean and South-Eastern (SE) European countries as compared to the rest of Europe. This is due to a larger relative change of HDD (see Table 1 above). As can be expected the total of the EU27+2 countries is dominated by the non-Mediterranean and non-South-Eastern (SE) countries. Due to the warmer climate, total non-electricity fuel energy demand which is dominated by space heating in most sub-sectors, is reduced by 16% in 2050 as compared to the base case and by about 14% as compared to 2005. Table 10 Resulting total non-electricity fuel energy demand of the Service sectors of Europe (EU27+2) for 2005, 2020, 2035, and 2050 (PJ) and relative changes Base Case)

Warmer Climate (WC)

WC/Base

2020/ 2035/ 2050/ 2020/ 2035/ 2050/ 2005 2020 2035 2050 2005 2005 2005 2020 2035 2050 2005 2005 2005 2020 2035 2050 Mediterr., SE EU 599 765 877 843 1.28 1.46 1.41 710 780 679 1.22 1.34 1.16 0.93 0.89 0.81 Rest of EU27+2 3056 3771 4171 3784 1.23 1.36 1.24 3559 3799 3188 1.19 1.27 1.06 0.94 0.91 0.84 Total EU 27+2 3656 4535 5048 4627 1.24 1.38 1.27 4268 4579 3868 1.19 1.28 1.08 0.94 0.91 0.84 Source:

Calculations by the authors [last updated 28 July 2008].

Discussion and outlook Electricity and other energy demand of the tertiary sectors of the European countries are expected to increase considerably up to 2050 in both the base case and warmer climate, namely by more than 25% (base case) and by about 10% (warmer climate) in the case of non-electricity fuels and by more than 50% in the case of electricity. Due to the warmer climate, non-electricity fuel energy demand of Europe which is dominated by space heating in most sub-sectors, is reduced by 16% in 2050 as compared to the base case and by 14% as compared to 2005, and electricity demand is increased by

562

7% (by about 300 PJ). Depending on the future primary energy intensity of electricity generation, these results imply either a slight improvement or a slight worsening. Note that in the case of electricity the impact of WC in 2050 is lower than the “regular” demand increase between 2005 and 2050 due to cooling which is estimated to about 500 PJ in the base case (from 310 to 830 PJ). However, it should be noted that electricity demand due to cooling might increase considerably more if the share of cooled floor area approached saturation not only in the case Mediterranean and SE EU countries, but also in the case of other countries which in terms of energy demand are of higher relevance. Particularly, heat waves could accelerate the purchase of room air conditioners. Moreover, electricity could be increased due to the use of reversible appliances which are installed for cooling intentions, but would be also in their heat mode.

References ASHRAE (2002). International weather for energy calculations (IWEC Weather Files) User’s Manual, Version 1.1.2002, Atlanta, GA, USA. Adnot, J. et al. (2003). Energy Efficiency and Certification of Central Air Conditioners (EECCAC). Study for the D.G. Transportation-Energy (DGTREN) of the Commission of the E.U., Final report. Aebischer, B., M. Jakob, G. Catenazzi; The late G. Henderson (2007). Impact of climate change on thermal comfort, heating and cooling energy demand in Europe. Proceedings eceee Summer Study 2007, Colle sur Loup, France, June (ISBN 978-91-633-0899-4). Cartalsi C., Synodinou A., Proedrou, M., Tsangrassoulis A., Santamouris M., (2001). Modifications in energy demand in urban areas as a result of climate change: an assessment for the southeast Mediterranean region. Energy Conversion and Management 42 1647-1656. Christenson M., Manz H, Gyalistras, D. (2006). Climate warming impact on degree-days and building energy demand in Switzerland. Energy Conversion and Management 47 671–686. Frank, Th. (2005). Climate change impacts on building heating and cooling energy demand in Switzerland. Energy and Buildings 37 (2005) pp 1175-1185. IPCC ( 2007). Forth assessment report. Isaac, M. et al. (2008). Monthly average T per country for every 5 years between 1970 and 2100 as estimated by the IMAGE model. EU research project “ADAM, Adaptation and Mitigation Strategies: Supporting European climate policy”, Contract no 018476 (GOCE). Jochem E. (Work Package Leader), Barker T., Scrieciu S., Schade W., Helfrich N., Edenhofer O., Bauer N., Marchand S., Neuhaus J., Mima S., Criqui P., Morel J., Chateau B., Kitous A., Nabuurs G. J., Schelhaas M.J., Groen T., Riffeser L., Reitze F., Jochem E., Catenazzi G., Jakob M., Aebischer B., Kartsoni K., Eichhammer W., Held A., Ragwitz M., Reiter U., Kypreos S., Turton H. (2007). EU-Project ADAM: Adaptation and Mitigation Strategies: Supporting European Climate Policy - Deliverable M1.1: Report of the Base Case Scenario for Europe and full description of the model system. Fraunhofer ISI, Karlsruhe, November. Riviere, Adnot et al. (2007). Preparatory study on the environmental performance of residential room conditioning appliances (airco and ventilation), Tasks 1 to 5. Draft reports as of October 2007. Varga, M. Pagliano, L. (2006). Reducing cooling energy demand in service buildings. International Conference on Improving Energy Efficiency in Commercial Buildings (IEECB’06), Frankfurt, Germany, 26 / 27 April.

563

564

Allowing for Thermal Comfort in Free-running Buildings in the New European Standard EN15251 Fergus Nicol, Low Energy Architecture Research Unit (LEARN), School of Architecture, London Metropolitan University, UK Lorenzo Pagliano, end-use Efficiency Research Group (e-ERG), Building Engineering Faculty, Politecnico di Milano, Italy

Abstract This paper describes some of the thinking behind the thermal comfort provisions of the new European Standard EN15251 (CEN: 2007) which deals with all aspects on the indoor environment. The paper will present the evidence on which its provisions are based (focusing on thermal comfort) and the advantages they present for those concerned to design buildings which use the minimum of energy.

The adaptive approach to thermal comfort The thermal sensation of subjects is found in surveys where people are asked to reply to the question: how do you feel? They can choose as a reply one of the descriptors on the ASHRAE or the Bedford scale. In the Adaptive approach this surveys are conducted in real buildings, while in the Fanger-PMV approach they are conducted in a controlled room in a laboratory where people are subject to stationary thermal conditions for 3 hours. TABLE 1: Descriptors for the ASHRAE and Bedford scale of thermal sensation

The Adaptive Approach to thermal comfort (Humphreys and Nicol 1998) has been developed from field-studies of people in daily life. While lacking the rigour of laboratory experiments, field studies have a more immediate relevance to ordinary living and working conditions (deDear 1998, Humphreys, 1975, Auliciems, 1981). The adaptive method is a behavioural approach, and rests on the observation that people in daily life are not passive in relation to their environment, but tend to make themselves comfortable, by making adjustments (adaptations) to their clothing, activity and posture, as well as to their thermal environment. Over time people tend to become well-adapted to thermal environments they are used to, and to find them comfortable. Adaptation is assisted by the provision of control over the thermal environment to give people the opportunity to adapt. This ‘adaptive opportunity’ (Baker and Standeven 1996) may be provided, for instance, by fans or openable windows in summertime or by temperature controls in winter. Dress codes will also have consequences for thermal design, for services provision, and

565

consequently for energy consumption. A control band of ±2 K should be sufficient to accommodate the great majority of people (Nicol and Humphreys 2007). These customary temperatures (the ‘comfort temperatures’) are not fixed, but are subject to gradual drift in response to changes in both outdoor and indoor temperature, and are modified by climate and social custom. Field research can indicate the extent and rapidity of adaptation, and hence of the temperature drifts that are acceptable. During any working day it is desirable that the temperature during occupied hours in any day should vary little from the customary temperature. Temperature drifts much more than ±2 K in any day would be likely to attract attention and might cause discomfort. Clothing and other adjustments in response to day-on-day changes in temperature, will occur when a building is responding to weather and seasonal changes. These will occur quite gradually (Humphreys 1979, Nicol and Raja 1996, Morgan et al 2002) , and can take a week or so to complete. So it is desirable that the day-to-day change in mean indoor Operative temperature during occupied hours should not occur too quickly for the adaptive processes to keep pace. During the summer months many buildings in Europe are free-running (i.e. not heated or cooled). The temperatures in such buildings will change according to the weather outdoors, as will the clothing of the occupants. Even in air-conditioned buildings the clothing has been found to change according to the weather (deDear and Brager 2002). As a result the temperature people find comfortable indoors also changes with the weather (Humphreys 1981). Thus the temperature people find comfortable can vary quite considerably depending on the climate, but any change should occur sufficiently slowly to give building occupants time to adapt.

Comfort in buildings In buildings which are in free-running (FR) mode indoor conditions will follow those outdoors but will be modified to a greater or lesser extent by the physical characteristics of the building and the use which building occupants make of the controls (windows, shading devices, fans etc) which are available to them. In a successful building these actions, together with the changes which the occupants make to their own requirements – mainly through clothing changes – mean that occupants are able to remain comfortable most of the time. The function of a standard is to define the indoor conditions which occupants will find acceptable for any given outdoor condition. Humphreys (1979) showed that the temperature which occupants of FR buildings find comfortable (Bedford scale) or neutral (ASHRAE scale) is linearly related to the monthly mean of the outdoor temperature. Others researchers have since found similar results (e.g. deDear and Brager 2002). The SCATS survey based in 5 European Countries has increased the accuracy and applicability of the model by showing that it was the running mean of the daily mean outdoor temperatures which correlated best with indoor comfort (McCartney and Nicol 2002) and that for European offices the linear relationship is: Tc = 0.33Trm + 18.8

(1)

Where Tc is the optimal indoor operative temperature for comfort and Trm is the running mean of the daily mean outdoor temperature. The exponentially weighted running mean of the daily mean external air temperature Ĭed is such a series, and is calculated from the formula:

Ĭrm = (1- D).{ Ĭed -1 + D. Ĭed -2 + D2 Ĭed -3…..

(2)

This equation can be simplified to Ĭrm = (1- D)Ĭed -1 + D. Ĭrm-1

(3)

Where Ĭrm = Running mean temperature for today Ĭrm-1 = Running mean temperature for previous day Ĭed-1 is the daily mean external temperature for the previous day

566

Ĭed -2 is the daily mean external temperature for the day before and so on. D is a constant between 0 and 1. Recommended to use 0,8 The Standard offers the following approximate equation can be used where records of daily mean external temperature are not available: (4) Ĭrm = (Ĭed -1 + 0,8 Ĭed -2 + 0,6 Ĭed -3 + 0,5 Ĭed -4 + 0,4 Ĭed -5 + 0,3 Ĭed -6 + 0,2 Ĭed -7)/3,8 It should be mentioned that this relationship strictly applies to the subjects who took part in the SCATs surveys and the buildings they occupied, but it closely matches the relationship presented by deDear and Brager from their survey of buildings throughout the world and this suggests that it has general applicability. In Figures 1 and 2 below we present for two climates and typical years the evolution of external air temperature and the internal operative comfort temperatures. Adaptive Operative Comfort Temperature is calculated according to Eqn 1; Fanger Operative Comfort Temperature is calculated using the formulas presented in ISO 7730 and assuming the following values of the input variables: thermal resistance of the clothing = 0.5 clo metabolic rate = 1.4 met air velocity = 0.15 m/s relative humidity = 50% This corresponds to general practice, where building planners have to make an assumption on those values adopting an average reasonable value for all the season, and hence obtaining a constant value for the Comfort Temperature. Figure 1: Adaptive Operative Comfort Temperature (in blue) and Fanger Operative Comfort Temperature (in red) for standard summer outdoor temperatures in Milan (in grey), Italy

Milan - Operative Comfort Temperature 35

[ °C ]

30

25

20

15 1-Jul

8-Jul 15-Jul 22-Jul 29-Jul 5-Aug 12-Aug 19-Aug 26-Aug

Figure 2: Adaptive Operative Comfort Temperature (in blu) and Fanger Operative Comfort Temperature (in red) for standard summer outdoor temperatures (in grey) in Rome, Italy

567

Rome - Operative Comfort Temperature 35

[°C ]

30

25

20

15 1-Jul

8-Jul 15-Jul 22-Jul 29-Jul 5-Aug 12-Aug 19-Aug 26-Aug

Having defined an optimal comfort temperature Tc, the question arises of how far the temperature of a space can deviate from Tc before discomfort will occur. Nicol and Humphreys (2007) have analysed the data from SCATs to show that ‘the temperatures at which discomfort will not be unduly intrusive are up to ±2 K above or below the appropriate comfort temperature‘, which makes this a sensible limit for a comfort zone. Figure 3 is from CIBSE (2006) and includes the comfort zone for buildings that are heated or cooled (HC) as well as FR buildings. Figure 3: comfort zones for buildings in free running mode (continuous lines from equation 1 ± 2 K) and heated or cooled mode (dashed lines) from CIBSE (2006). Adaptive comfort zones for buildings

o

Indoor limiting temperatures ( C)

30 28 26 Free-running upper limit Free-running lower limit

24

Heated or cooled upper limit Heated or cooled lower limit

22 20 18 0

5

10

15

20

25

o

Outdoor running mean temperature ( C)

EN15251 and temperature limits in free running buildings1 The preamble of the European Energy Performance of Buildings Directive (EPBD) states: “(...) the displaying of officially recommended indoor temperatures, together with the actual measured temperature, should discourage the misuse of heating, air-conditioning and ventilation systems. This should contribute to avoiding unnecessary use of energy and to safeguarding comfortable indoor 1

The section of the Standard EN15251which deals with temperature limits in Free running buildings is given in an Appendix to this paper

568

climatic conditions (thermal comfort) in relation to the outside temperature.” Ensuring that both energy savings and a good indoor environment are targeted is essential (Varga and Pagliano 2006). The European Standard EN15251 Indoor environmental input parameters for design and assessment of energy performance of buildings- addressing indoor air quality, thermal environment, lighting and acoustics seeks to define minimum standards for the internal environment in buildings to complement the EPBD. A major consideration of this norm is to ensure a correct definition of thermal comfort. The revision of the International Standard EN ISO 7730 made in 2005 acknowledges the importance of adaptation mechanisms: “In warm or cold environments, there can often be an influence due to adaptation. Apart from clothing, other forms of adaptation, such as body posture and decreased activity, which are difficult to quantify, can result in the acceptance of higher indoor temperatures. People used to working and living in warm climates can more easily accept and maintain a higher work performance in hot environments than those living in colder climates.” But it does not provide explicit guidance on how to treat differences in comfort conditions in naturally ventilated (NV) and mechanically cooled (AC) buildings. For this reason it is important that EN15251 embodies the latest thinking about comfort in the variable conditions of real NV buildings, allowing designers to take advantage of occupants’ natural ability to adapt conditions to their liking. This not only optimises the interaction between occupants and the building to ensure comfort but also enables designers to maximise energy saving by allowing indoor conditions to track those out of doors. EN15251 makes a distinction between buildings which are HC and those which are FR. Thus NV buildings will be HC during the heating season and FR during the summer; AC buildings are HC throughout the year. In Standard EN15251, the comfort zone for HC buildings is defined in similar way as in EN ISO 7730 (2006) but with more differentiation of buildings in categories. EN15251 recommends values of PMV comprised within the interval -0,5 to +0,5 for new buildings and renovations (category II) and within -0,7 to +0,7 for existing buildings (category III); TABLE 2: Suggested applicability of the categories and their associated acceptable temperature ranges. Category

Explanation

I

High level of expectation only used for spaces occupied by very sensitive and fragile persons

II

Normal expectation (for new buildings and renovations) A moderate expectation (used for existing buildings) Values outside the criteria for the above categories (only acceptable for a limited periods)

III IV

Suggested acceptable range ± 2K ± 3K ± 4K

EN15251 uses the results of the SCATs survey to define the limits of temperatures in NV (or FR) buildings in the “summer” season, divided into categories defined as shown in table 1. The width of the acceptable zones allowed in each category is shown as a deviation from the value which is calculated from Eqn. 1. The applicability of the zones is assumed to be for values of Trm between 10oC and 30oC. EN15251 has also introduced (as ISO 7730) an allowance for air movement which can mean that the upper limit of acceptable temperature can be raised when substantial air movement is present such as might occur when a fan is in use.

Evaluation of thermal conditions for compliance with EN15251 There are two methods suggested in the EN for evaluating the thermal comfort conditions during an entire season: 1. Percentage outside range: the proportion of the occupied hours during which the temperature lies outside the acceptable zone during the season.

569

2. Degree hours criterion: The time during which the actual operative temperature exceeds the specified range during occupied hours is weighted by a factor depending on the number of degrees by which the range has been exceeded. Acceptability of the space on the ‘percentage’ criterion is on the basis that the temperature in the rooms representing 95% of the occupied space is not more than 3% (or 5% - to be decided on national level) of the occupied hours a day, week, month or year, outside the limits of the specified category. Subjective evaluation may also be used to evaluate existing buildings and methods, for assessing and reporting this are suggested.

Conclusions The new European Standard EN 15251 has been framed to allow the natural variability of the indoor climate in free running buildings to be matched to the natural ability of people in well designed buildings with adequate occupant control, to change their room conditions to suit their needs. This will mean that buildings can be designed which are both comfortable and can make full use of passive, low energy cooling and heating technologies.

Aknowledgments Evidence supporting the use of the Adaptive Comfort Model, simulations on its application and considerations on categories have been developed, summarised in the appropriate language and formally brought to the attention of the drafting group of EN 15251 by the authors. This was done within the work programme of the EIE projects KeepCool and Passive-on, and by means of ECOS, European Environmental Citizens' Organisation for Standardisation, which has Associate status with CEN. Some National Standardisation Bodies supported the presentation of parts of the amendments proposed. Further analysis of the Standard EN15251 is being undertaken in the EIE project Commoncense. We would like to acknowledge the role of Rev Prof Michael Humphreys in developing many of the ideas presented.

References ASHRAE Standard 55-04 (2004) Thermal Environmental Conditions for Human Occupancy, Atlanta GA, American Society of Heating Refrigeration and Air-conditioning Engineers Auliciems, A. (1981) Towards a psycho-physiological model of thermal perception, International Journal of Biometeorology, 25 pp 109-122 Baker, N.V. & Standeven, M.A. (1996). Thermal comfort in free-running buildings. Energy and Buildings Vol 23 CEN (2007) Standard EN15251 Indoor environmental parameters for design and assessment of energy performance of buildings- addressing indoor air quality, thermal environment, lighting and acoustics Brussels: Comité Européen de Normalisation. Error! Reference source not found.CEN, Brussels CIBSE (2006) Guide A: Environmental Design London, Chartered Institution of Building Services Engineers deDear, R.J. (1998) A Global database of thermal comfort field experiments. ASHRAE Transactions 104 (1) pp 1141-1152). deDear, R.J. and Brager, G.S. (2002) Thermal comfort in naturally ventilated buildings: revisions to ASHRAE Standard 55 Energy and Buildings, Volume 34,(6),pp 549-561 EN ISO 7730 (2006) Moderate thermal environments- determination of the PMV and PPD indices and specification of the conditions for thermal comfort. ISO, Geneva.

570

Humphreys, M.A. (1975). Field studies of thermal comfort compared and applied: J. Inst. Heat. & Vent. Eng. 44, pp 5-27. Humphreys, M.A. (1979) The influence of season and ambient temperature on human clothing behaviour In: Indoor Climate Eds: P O Fanger & O Valbjorn, Danish Building Research, Copenhagen Humphreys, M.A. (1981) The dependence of comfortable temperature upon indoor and outdoor climate In: Bioengineering, Thermal Physiology and Comfort Eds.: K Cena & J A Clark, Elsevier Humphreys, M.A. and Nicol, J.F. (1998) Understanding the Adaptive Approach to Thermal Comfort, ASHRAE Transactions 104 (1) pp 991-1004) McCartney K.J and Nicol J.F. (2002) Developing an Adaptive Control Algorithm for Europe: Results of the SCATs Project. Energy and Buildings 34(6) pp 623-635 Morgan, C.A., deDear, R. and Brager, G. (2002) Climate Clothing and adaptation in the built environment, Indoor Air 2002: Proceedings of the 9th International Conference on Indoor Air Quality and Climate Vol. 5, 98-103 ed. H. Levin, Indoor Air 2002, Santa Cruz, USA Nicol, J.F. and Humphreys, M.A. (2007) Maximum temperatures in European office buildings to avoid heat discomfort. Solar Energy Journal, (available online 09/06) Nicol, J.F. and Raja, I. (1996) Thermal comfort, time and posture: exploratory studies in the nature of adaptive thermal comfort. School of Architecture, Oxford Brookes University. Varga, M. & Pagliano, L. (2006). Reducing cooling energy demand in service buildings. In Proceedings of the International Conference Improving Energy Efficiency in Commercial Buildings (IEECB’06), Frankfurt, Germany, 26 - 27 April 2006.

571

Appendix Annex A2 from EN15251: Acceptable indoor temperatures for design of buildings without mechanical cooling systems. In figure A1 acceptable ‘summer’ indoor temperatures (cooling season) are presented for buildings without mechanical cooling systems. The operative temperatures (room temperatures) presented in figure A1 are valid for office buildings and other buildings of similar type used mainly for human occupancy with mainly sedentary activities and dwelling, where there is easy access to operable windows and occupants may freely adapt their clothing to the indoor and/or outdoor thermal conditions. Figure A1 Design values for the indoor operative temperature for buildings without mechanical cooling systems as a function of the exponentially-weighted running mean of the external temperature .

The figure will be extended to 30 C and start from 10 C . Please use external instead of outdoor and use symbols on the axis. The temperature limits only apply when the thermal conditions in the spaces at hand are regulated primarily by the occupants through opening and closing of windows. Several field experiments have shown that occupants’ thermal responses in such spaces depends in part on the outdoor climate, and differ from the thermal responses of occupants in buildings with HVAC systems, mainly because of differences in thermal experience, availability of control and shifts in occupants’ expectations. In order for this optional method to apply, the spaces in question must be equipped with operable windows which open to the outdoors and which can be readily opened and adjusted by the occupants of the spaces. There must be no mechanical cooling in operation in the space. Mechanical ventilation with unconditioned air (in summer) may be utilized, but opening and closing of windows must be of primary importance as a means of regulating thermal conditions in the space. There may in addition be other low-energy methods of personally controlling the indoor environment such as fans, shutters, night ventilation etc. The spaces may be provided by a heating system, but this optional method does not apply during times of the year when the heating system is in operation when the method of section A1 applies.

572

This optional method only applies to spaces where the occupants are engaged in near sedentary physical activities with metabolic rates ranging from 1,0 to 1,3 met. It is also important that strict clothing policies inside the building are avoided, in order to allow occupants to freely adapt their clothing insulation. The (summer) temperature limits presented in this Annex are primarily based on studies in office buildings. Nevertheless, based on general knowledge on thermal comfort and human responses, the assumption can be made that the limits may apply to other (comparable) buildings with mainly sedentary activities like residential buildings. Especially in residential buildings the opportunities for (behavioural) adaptation are relatively wide: one is relatively free to adjust metabolism and the amount of clothing worn dependant on outside weather conditions and indoor temperatures. Note that the field studies the temperature limits in this Annex are based upon diagram is based on comfort studies in offices and did not take account of work performance. In landscaped (open plan) offices most occupants have only limited access to operable windows and therefore poor control over natural ventilation. Therefore the temperature limits presented in this Annex may not always apply in such situations. The figure includes 3 categories of temperature limits for use as outlined in the introduction and section 5 to this standard. The allowable indoor operative temperatures of figure A1 are plotted against the external running mean temperature Ĭrm. This is defined as the exponentially weighted running mean of the daily outdoor temperature (see equations (2), (3) and (4) above for methods to calculate Ĭrm) Category I

upper limit: lower limit:

Ĭi max = 0,33 Ĭrm + 18,8 + 2 Ĭi min = 0,33 Ĭrm + 18,8 - 2

Category II

upper limit: lower limit:

Ĭi max = 0,33 Ĭrm + 18,8 + 3 Ĭi min = 0,33 Ĭrm + 18,8 - 3

Category III

upper limit: lower limit:

Ĭi max = 0,33 Ĭrm + 18,8 + 4 Ĭi min = 0,33 Ĭrm + 18,8 – 4

where Ĭi = limit value of indoor operative temperature, oC Ĭrm = running mean outdoor temperature. These limits apply when 10 80). Due to the combination of dimming according to daylight and occupancy control, the power curve of room G437 can be seen changing over short intervals. It uses full installed power only when the daylight is completely unavailable.

620

Measured values of illuminance, glare rating, installed power, and energy consumption Control system Manual Daylight and presence Presence EN12464-1

Average illuminance / lx Working plane Floor 580 380 670 390 700 500 500 300

UGR

W/m2

kWh/m2

11 16 12 19

14,1 16,9 16,3

33 20 24

UGR = Unified Glare rating W/m2 = Installed power for lighting per square metre of room, in W/m2 Wh/m2 = Annual energy consumption per square metre of the room, in kWh/m2 Room G437 has highest (16,9 kW/m2) and the room G435 has lowest (14,1 kW/m2) installed power for lighting but due to daylight based dimming and occupancy control in room G437, it consumes the least energy (20kWh/m2 per annum) compared to 24 kWh/m2 of rooms G438&439 (only occupancy control) and 33kWh/m2 of room G435 (manual control). On the other hand, as seen in the table, the working plane illuminance on the room with high energy consumption is less compared to the other rooms. The energy consumption for rooms except the one with manual control is well below the average annual energy use for lighting in Finnish offices, which is 31kWh/m2 (Korhonen et al., 2002). The average working plane illuminance levels of all these rooms are higher than the current recommendation level, so there is still space to reduce the annual energy consumption level below 20kWh/m2 without compromising on the quality. Detailed study of the quality aspects of lighting has also been done.

Recommendations for energy efficient lighting The European standard EN15193 ‘Energy performance of buildings – Energy requirements for lighting’ defines procedures for the estimation of energy requirements of lighting in buildings, and provides guidance on the establishment of national limits for lighting energy based on reference schemes. Limits for connected lighting power (in W/m2) according to EN15193 for different building types and quality levels.

621

Fontoynont and Escaffre have demonstrated that also lower values are achievable already with the technology available at the moment. They showed that power densities of about 6 W/m2 are achievable, with very high visual performances. (Fontoynont & Escaffre 2006) Baaijens showed that energy consumption can be reduced by 44 % in office rooms by occupancy and daylight sensors. (Baaijens 2006) Pertola estimated the savings through utilization of daylight to be about 50 % annually and further 10 % savings of cooling electricity. (Pertola 2007)

Summary To design energy efficient lighting solutions the designer should perform life-cycle cost evaluations and economic evaluations (including payback criteria desired by the building owner). The following rules should be kept in mind to reach or supersede these goals: 1. Intelligent architecture (use of daylight, artificial light only for supplementing, use of bright surfaces) 2. Energy efficient lighting installations (intelligent concepts, use of high quality luminaires and lamps) 3. Proper controls (on/off, daylight, occupancy) 4. Good maintenance.

Acknowledgement Both authors are Subtask leaders in IEA ECBCS Annex 45. International Energy Agency (IEA) is an intergovernmental body committed to advancing security of energy supply, economic growth and environmental sustainability through energy policy co-operation. IEA has Implementing Agreements (IA) to organize research. One of these IAs is Energy Conservation in Buildings and Community Systems (ECBCS). The function of ECBCS is to undertake research and provide an international focus for building energy efficiency. Tasks are undertaken through a series of annexes that are directed at energy saving technologies and activities that support their application in practice. Results are also used in the formulation of energy conservation policies and standards. The Executive Committee of the ECBCS program established a new Annex in June 2004 called Energy Efficient Electric Lighting for Buildings. The objectives of Annex 45 are to identify and accelerate the use of energy-efficient high-quality lighting technologies and their integration with other building system, to assess and document the technical performance of existing and future lighting technologies, as well as to assess and document barriers preventing the adoption of energy-efficient technologies, and to propose means to resolve these barriers.

References Baaijens 2006. Energy savings with modern lighting control system. IEA ECBCS Annex 45, Newsletter 4. http://lightinglab.fi/IEAAnnex45/publications/index.html EC 2002. Directive 2002/91/EC of the European Parliament and of the Council of 16 December 2002 on the energy performance of buildings (EPBD). EN12252 European standard ‘Indoor environmental input parameters for design and assessment of energy performance of buildings addressing indoor air quality, thermal environment, lighting and acoustics’ EN12464-1 European standard ‘Light and Lighting – Lighting of work places’. EU 2004. http://www.europa.eu.int/comm/energy_transport/atlas/html/lightdintro.html, accessed on 24.4.2004. Fontoynont & Escaffre 2006. More Comfort, Less Electrical Power for Office Spaces. IEA ECBCS Annex 45, Newsletter 3. http://lightinglab.fi/IEAAnnex45/publications/index.html IEA 2006. International Energy Agency. Light’s Labour’s Lost. IEA Publications, France. 360 p. Korhonen, Anne, Pihala Hannu, Ranne Aulis, Ahponen Veikko, Sillanpää Liisa. (2002). Kotitalouksien ja toimistotilojen laitesähkön käytön tehostaminen; Työtehoseuran julkaisuja 384. 158 s. Lehtonen M. et al. Talotekniikkaa kaikille. Asumisen talotekniikka – järjestelmät, palvelut ja asiakkuudet (ASTAT). Building services for everybody – systems. services and customership. Sähköinfo 2007. ISBN 978-952-5600-39-1.

622

Mills E. 2002. Why we’re here: The $320-billion global lighting energy bill. Right Light 5, Nice, France. pp. 369-385. Moore ,T., Carter, D.J., and Slater, A.I. 2003. A qualitative study of occupant controlled office lighting. Lighting Research and Technology, Vol 35, No. 4, 297-317. Pertola 2007. Possible energy savings of electric lighting by using redirected daylight. IEA ECBCS Annex 45, Newsletter 6. http://lightinglab.fi/IEAAnnex45/publications/index.html

623

624

The Use of High Dynamic Range Luminance Mapping in the Assessment, Understanding and Defining of Visual Issues in Post Occupancy Building Assessments Steve Coyne, Light Naturally, Queensland, Australia Gillian Isoardi and Michael Hirning Queensland University of Technology, Queensland, Australia Mark B Luther Deakin University, Victoria, Australia

Abstract The international focus on embracing daylighting for energy efficient lighting purposes and the corporate sector’s indulgence in the perception of workplace and work practice “transparency” has spurned an increase in highly glazed commercial buildings. This in turn has renewed issues of visual comfort and daylight-derived glare for occupants. In order to ascertain evidence, or predict risk, of these events; appraisals of these complex visual environments require detailed information on the luminances present in an occupant’s field of view. Conventional luminance meters are an expensive and time consuming method of achieving these results. To create a luminance map of an occupant’s visual field using such a meter requires too many individual measurements to be a practical measurement technique. The application of digital cameras as luminance measurement devices has solved this problem. With high dynamic range imaging, a single digital image can be created to provide luminances on a pixel-by-pixel level within the broad field of view afforded by a fish-eye lens: virtually replicating an occupant’s visual field and providing rapid yet detailed luminance information for the entire scene. With proper calibration, relatively inexpensive digital cameras can be successfully applied to the task of luminance measurements, placing them in the realm of tools that any lighting professional should own. This paper discusses how a digital camera can become a luminance measurement device and then presents an analysis of results obtained from post occupancy measurements from building assessments conducted by the Mobile Architecture Built Environment Laboratory (MABEL) project. This discussion leads to the important realisation that the placement of such tools in the hands of lighting professionals internationally will provide new opportunities for the lighting community in terms of research on critical issues in lighting such as daylight glare and visual quality and comfort.

625

Introduction Several equations and indices have been proposed as a means of quantifying the glare experienced by an observer in a daylit environment; and yet there is no significant agreement amongst lighting researchers on a reliable method for predicting discomfort glare from daylight. The Daylight Glare Equation (DGI or Cornell Equation) developed by Hopkinson [1] has been an accepted standard for predicting glare conditions, but there is a large body of research that demonstrates the DGI unreliable when applied to real, complex lighting environments [2-4]. This stems from the testing protocol used to develop this model, which was conducted with artificial light sources and using simple lighting setups. Indeed the difficulty in successfully applying most glare index formulae to situations involving daylight and windows lies largely in discrepancies between results obtained from contrived and controlled laboratory experiments and the experiences of observers in real situations. It is generally accepted that people respond differently to glare associated with electric sources than that arising from the sun and sky. It is evident that the perception of glare in laboratory environments cannot always be used as a predictor of glare experienced in field situations, where the environment is visually complex and changing temporally, the observer is interacting with their surroundings and interesting visual stimuli (e.g. view windows) are provided [5, 6]. Most existing glare formulae are based on experiments using uniform light sources and should therefore not be applied when discomfort glare is caused by non-uniform light sources. Efforts to collect comprehensive data on actual instances of daylight glare in the past have been restricted by the difficulty in measuring and recording all of the information necessary to fully account for all potential influences on the subject’s visual assessment. Point-by-point measurement of luminances for a full field of view assessment of complex and temporally changing visual environments is impractical. However, recent advances in CCD or camera-based luminance mapping techniques have demonstrated the capacity for furthering our understanding of discomfort glare by permitting the relatively quick and simple collection of detailed luminance information [7-9]. If it could be applied to field assessment, particularly as a routine element of a lighting practitioner’s postoccupancy evaluation, HDR imaging could be an ideal method for the accumulation of a substantial database of luminances in real glare-producing environments. If this physical data could be collected in conjunction with additional information on a range of possible influences on observers’ visual assessment, it could contribute substantially to our understanding of discomfort glare.

Taking Discomfort Glare Research to the Field As the empirical data suggest, the correlation between indices derived from highly-controlled test procedures, particularly those that involve electric light sources, and actual glare perception in real and visually complex situations is limited. This is likely to be due in part to the variability of the daylight source, and is certainly influenced by confounding factors that are difficult to quantify and predict: visual amenity of the environment, view provided by windows, personal preference and physical condition of occupants, task difficulty, test conditions. A recent study by Tuaycharoen and Tregenza on glare from view windows [5] concluded that discomfort glare cannot be predicted from physical variables alone: the four factors common to most glare formulae – source luminance, source size, surround luminance and a position index – are not enough. In reality, identifying and parameterising all of the influences on daylight glare may well be impossible. However, identifying and understanding some of the most critical parameters in gross terms (i.e. yes/no) may be just as beneficial. In light of this, we are proposing a glare data monitoring procedure – the purpose of which is to accumulate an extensive database benefiting lighting research – that relies on the input and experience of lighting practitioners and takes place in a broad range of actual glare/glare-free environments. By placing data collection in the hands of a multitude of lighting practitioners, we hope to collect a significant amount of data on real glare situations. The proposed monitoring procedure includes; physical measurement of the lit environment (luminance maps using HDR cameras), information on factors identified as influential to glare perception as reported by the lighting practitioner (e.g. visual interest), and results of a subjective glare assessment from the occupants involved. This paper posits the development of a uniform collection method with respect to these 3 major data inputs, and highlights the pathway to initiating this project.

626

Physical measurement tools for site assessment The development of High Dynamic Range (HDR) imaging has enabled digital cameras to record the broad range of luminance data necessary to describe how a scene might appear to the human eye. For lighting assessment, HDR photography has been demonstrated as a useful tool that can capture luminance values overall within 10% accuracy over a wide range of luminances [10]. The technology provides a great potential for improved understandings of the relation between measured lighting conditions and user response. CCD camera-based luminance mapping technology has already been applied to the measurement of luminance distributions within the field of view of a test subject [8] and to the calculation of glare indices [11]. Of particular significance to this project is the relative ease with which a moderately priced digital camera can be calibrated to capture images that permit the creation of HDR images or luminance maps of a scene. When equipped with a fisheye lens, these cameras become tools that can produce detailed information on the visual environment of a subject in a fraction of the time required by pointby-point luminance measurements. Their affordability and ease-of-use combine to make the calibrated digital camera an appropriate tool for lighting practitioners to record HDR images of actual glare environments for research purposes. The first step in this project is to identify and engage a group of (approximately 50) consultants and companies willing to participate in the collection of data. In order to achieve a uniform methodology for site measurements, it is intended that participants purchase digital cameras and equipment appropriate for the task that will then be centrally calibrated in order to generate appropriate camera calibration curves and corrections for lens and vignetting effects. Having provided a calibration service and instructions on how to collect images for the later construction of HDR images, the lighting practitioner is able to proceed with data collection. Of particular importance is ensuring that an adequate exposure range is captured for each scene, camera settings are correct and consistent; and that images are correctly framed and centred in order to represent the subject’s field of view. Raw data, in the form of any images recorded in site measurements, are then returned to the central agency where HDR images can be created using the calibration data for that camera and appropriate software (e.g. Photosphere (Anyhere Software) or Photolux (Coutelier and Dumortier) [12]). The management of contact and file transfer between the participating lighting practitioners and the central agency for calibrations, data analysis and data storage will occur through a web-based interface. A web site will function to allow upload of raw image data, then the subsequent HDR production and data analysis can be conducted remotely. The results of this analysis can then be passed back to the data provider, which can then inform their understanding of glare and lighting quality and add to the professional service they provide. At the same time, results are stored centrally (and de-identified if appropriate) and a database of images is formed. The growth of this database – fed by real case studies and post-occupancy assessments – is the key to developing an understanding of complex glare situations. Analysis of the raw data begins with the production of HDR images and luminance maps. example of such an image (in false colour) is presented below.

627

An

Figure 1: HDR image of office with glare conditions present Rapid analysis of HDR images for glare information is conducted using a program developed in MatLab® (MathWorks, Inc.). The program overlays an angular matrix on the image in order to reference the location of any glare sources according to the subject’s central visual axis. This angular web (overlain on figure 1) is shown in the figure 2 below.

Figure 2: Office image overlain with angular matrix (centred on subject’s central visual axis) After calculating the average luminance of the scene (excluding potential glare sources defined by a basic threshold), the program can then locate all potential glare sources that exceed a predefined multiple of the average luminance. This isolation of glare sources is illustrated in figure 3 below.

Figure 3: Potential glare sources located and isolated – identified glare sources circled The precise location and solid angle subtended by the potential glare sources found using this method is recorded. Now a glare threshold can be specified in the program in order to identify actual

628

glare sources. For example, in the situation referenced in figure 3, if the glare threshold is specified as any source with a luminance 100 times greater than the average luminance, then the program identifies 2 glare sources. These are circled in the figure. However, if the program fails to indentify all of the relevant glare sources (i.e. too few sources are identified when comparing with the occupant’s evaluation) then the threshold may be lowered until a defined level of correlation has occurred. Similarly, the threshold can be raised should the program identify more glare sources than were perceived present. By adjusting the glare thresholds to correlate with subjective assessment, and incorporating site information about influential factors, the process hopes to refine our understanding of the interplay between factors that affect discomfort glare perception. Furthermore, this analysis program is designed to automatically calculate the values each scene achieves according to conventional glare indices. For example using the information obtained from figure 2, a DGI index of 42.7 was calculated, and the glare rated ‘intolerable’. Using this function, the glare predicted by new and existing indices can be confirmed or refuted.

Subjective glare assessment tools for the occupant The correlation between the glare predictions made by the software (in terms of both the potential glare sources identified and the existing indices calculated) and the glare experienced by the subject is explored with the use of a questionnaire for occupants. The template from which the proposed questionnaire is to be derived is demonstrated below (in a similar questionnaire produced for postoccupancy surveys in the Mabel project). Of particular note is the similarity between the matrix applied to the HDR image collected onsite and the region where survey subjects are asked to identify glare sources. The occupant is asked for an assessment of any glare sources in gross terms only – rather than ranking on a larger numerical scale. In this way it is hoped that a meaningful comparison will be possible between glare regions identified by the automated program and by the subject, with minimal confusion to the person completing the questionnaire. Ultimately it is the combination of this physical assessment and the occupant’s response to the questionnaire, to be investigated on a large scale across a broad range of real daylit environments that stand to offer the greatest opportunity to increase our understanding of glare and to ultimately validate and refine new or existing glare indices. However, it is expected that some additional information will be required in order to completely capture all potential influences on the perception of glare in each environment. This information, regarding issues such as visual interest of the environment, could impact on the subjective assessment and should therefore be included via a survey conducted by the lighting practitioner. Figure 4: Sample indicating style of the suggested questionnaire for subject glare assessment

629

Survey tools for the lighting practitioner One of the greatest logistical difficulties in undertaking large scale light-technical assessments across many sites in many locations (viz countries) is the access to suitably qualified people to conduct the measurements (ie to correctly acquire the digital images required for producing the HDR luminance images). The utilisation of lighting professionals could resolve this issue and at the same time provide a greater opportunity for data collection. It is well accepted that factors beyond quantitative light-technical parameters contribute to an individual’s perception of the visual amenity of a space. These are factors such as relative mix of daylight and electric light, visual interest of the space, visual link to the exterior, quality of the view, an individual’s access to some form of (day)lighting controls, work practices and the social and political dynamics within the office. They cannot be recorded by a digital camera. The effect of the confluence of these factors should however be able to be interpreted, to at least to a basic informative level, by an experienced lighting practitioner. Through their experience they should be capable of empathising with occupants within “successful” and “unsuccessful” office environments when conducting pre-refurbishment or post occupancy assessments. The nature of the information required for this research and the level of detail sought from the lighting practitioner during a site assessment is yet to be finalised but it is expected to relate to factors similar to those discussed above. Developing an assessment methodology which entails relatively simple forced answer questions and a limited range within scales will help confine the variability likely to be experienced by the differing sensitivities of the lighting practitioners. It is envisaged that the documenting of a lighting practitioner’s empathy with building occupants (collected across a large database of buildings) could prove to be a critical element in helping to understand (and maybe quantify) the sensitivity of discomfort glare to some of these factors. Some post occupancy assessments in Australia have utilised this practice in an informal way and preliminary investigation of the results have begun to provide an insight into what needs to be developed in a formal research methodology.

630

An example using data from a MABEL post-occupancy assessment of a building in Australia Occupants of the building offered for assessment in this example had experienced glare problems, the MABEL assessment included the capture of HDR images (collected using a calibrated Nikon Coolpix and produced using Photolux software), and a site survey. Formal questionnaires were not completed by subjects in this case, however information about glare conditions were informally obtained and recorded by the lighting practitioner conducting the site survey. Figures 5a and 5b, below show a low dynamic range image (ie standard photograph), captured using a fisheye lens of a workstation adjacent to a northern (equatorial) window wall. The photograph on the left was taken at 10:00am, the right shows the same workstation at 3:30pm – it should be noted that the exposure values for these images are different. Figures 6a and 6b show the false colour luminance maps produced from data collected at these locations.

Figures 5a: Level 2 Position 6 - 10:00am (left) and 5b: 3:30pm (right)

Figures 6a: False colour HDR Level 2 Position 6 - 10:00am (left) and 6b: 3:30pm (right) Applying the analysis software to these images yields the following results, shown below in figures 7 and 8. In each case, arbitrarily defining glare sources using the glare threshold of 100 times the average scene luminance, one glare source was identified. The DGI for the scene at both times of day is approximately 26.2, and the glare rated as ‘uncomfortable’. In terms of how this physical

631

assessment corresponds with the subjective glare assessment, the glare source identification agreed with the subject’s assessment. However, according to the subject, the rating ‘uncomfortable’ was considered an understatement of the glare problem occurring in the morning and afternoon.

Figures 7a: 10:00 am HDR image with polar mesh and 7b: Threshold set to 100 times average scene luminance. In particular at 10:00am, shown in figures 7a and 7b, reflections off a anodised aluminium sill at the extreme left of the subject’s field of view exceeded 12,000 cd.m-1. This is approximately 50 times greater than the luminance of the central task (the screen at 235 cd.m-1) and 800 times the average scene luminance (which is ~13 cd.m-1). The relatively small solid angle the source subtends (0.02 sr) combined with its location in the periphery of the subject’s field of view have accorded this glare source a DGI that underestimates how it is perceived by this subject.

Figures 8a: 3:30 pm HDR image with polar mesh and 8b: Threshold set to 100 times average scene luminance. At this location in the afternoon, figures 8a and 8b, the glare identification software fails to detect a band of intense sunlight that moves across the workstation (seen on the subject’s shoulder in figure 5b) that was identified as a glare source in the subjective assessment. This area was not identified as a glare source as it was only 80 times greater than the measured average scene luminance. This may suggest that the central (task area) location of this glare source and its distracting (diurnal) movement across the task region give rise to a lower glare threshold as perceived by this subject. The lighting practitioner noted a number of issues through discussions with occupants during the assessment. The occupants of this building appear to be acutely sensitive to glare issues. This has manifested from ongoing frustrations with the facility manager’s failure to demonstrate reasonable attempts at rectifying the issues. And when an attempt was made to reduce window glare, a film was only applied to the upper half of the window to save costs. This has resulted in lower adaptation levels for the eyes of the occupants and consequently increasing the glare effect of the lower window section (which has maintained its high luminance).

632

Ultimately the occupants have been forced to deal with glare by whatever means are available to them. This has included disabling light fixtures within an occupant’s field of view (Figure 5a) resulting from an inability to control (dim) the electric lighting system or have it adjusted. These are all issues that would have not been captured by light-technical measurements (such as a HDR image) alone. Therefore quantitative glare analysis of the HDR luminance images would not provide an insight into the heightened glare sensitivity.

Conclusion This paper outlines the development of a glare analysis tool that employs HDR imaging techniques. It also highlights a pathway to accumulating a substantial database of light technical parameters and site environmental and social data for real office environments with glare and non-glare scenarios. Such a large and broadly sampled database is seen as being of significant value in advancing the field of glare research. The process of acquiring the information for the database, using lighting practitioners, is also seen as a healthy way of developing the skills and understanding of the lighting professional community on glare. This has become a more significant issue in recent times as light sources are trending to smaller sizes with higher lumen packages and with the popular shift back to more environmental and sustainable (and difficult) practices of daylighting design as a core solution to lighting.

References 1. 2. 3. 4. 5. 6. 7.

8.

9. 10. 11. 12.

Hopkinson, R.G., Glare from daylighting in buildings. Applied Ergonomics, 1972. 3(4): p. 206215. Chauvel, P., Collins, J.B., Dogniaux, R. and Longmore, J., Glare from windows: current views of the problem. Lighting Res. Technol., 1982. 14(1): p. 31-46. Chauvel, P., Perraudeau, M.,. Daylight as a source of visual discomfort. in Daylighting Atlas. 1995. Lyon. Iwata, T., et al., Discomfort caused by wide-source glare. Energy and Buildings, 1990. 15(34): p. 391-398. Tuaycharoen, N., Tregenza, P.R., View and Discomfort Glare From Windows. Lighting Res. Technol., 2007. 39(2): p. 185-200. Tuaycharoen, N., Tregenza, P.R., Discomfort Glare from Interesting Images. Lighting Res. Technol., 2005. 37(4): p. 329-341. Osterhaus, W.K.E., Discomfort glare assessment and prevention for daylight applications in office environments. Solar Energy CISBAT '03: Innovation in Building Envelopes and Environmental Systems, 2005. 79(2): p. 140-158. Wienold, J. and J. Christoffersen, Evaluation methods and development of a new glare prediction model for daylight environments with the use of CCD cameras. Energy and Buildings Special Issue on Daylighting Buildings, 2006. 38(7): p. 743-757. Nazzal, A.A., A new evaluation method for daylight discomfort glare. International Journal of Industrial Ergonomics, 2005. 35(4): p. 295-306. Inanici, M.N., Evaluation of high dynamic range photography as a luminance data aqcuisition system. Lighting Res. Technol., 2006. 38(2): p. 123-136. Dumortier, D., et al. PHOTOLUX: a new luminance mapping system based on Nikon Coolpix digital cameras. in Lux Europa. 2005. Coutelier B., Dumortier D. (2003). Luminance calibration of the Nikon Coolpix 990 digital camera. In Proceedings of the 25th Session of the CIE, San Diego, 2003, Vol. 1, D3-56

633

634

Integral Approach to Design Building engineering systems: (lighting, heating, air conditioning) – as a effective way to Energy Saving Julian B. Aizenberg, VNISI, Moscow, Russia

Abstract The presentation demonstrates the method of substantial decrease in energy consumption in buildings achieved owing to the comprehensive way of all engineering systems design: lighting, heating, air conditioning. All results are illustrated by examples of several successfully implemented projects with different types of Hollow Light Guides (HLG). A large production area in Moscow was illuminated by mirror slit HLG. Thanks to the fact that heat from high wattage lamps in HLG igniters was generated outside the working area it allowed decreasing the wattage of air-conditioning systems by 600 kW (energy saving of 5,160 MWh per year). A cardinal solution of integrating natural and electrical illumination and energy saving was achieved in the «Heliobus» lighting installation in a 4-storied school in St. Gallen (Switzerland) with the help of long vertical HLG and a static heliostat. The other method was applied in a 2-storey school building in Schiers (Switzerland). This installation has 2 dynamic mirror heliostats.

Introduction A substantial decrease in energy consumption in buildings can only be achieved if all engineering systems (natural and artificial lighting, ventilation, air conditioning and heating) are designed in a comprehensive way, integral. It can be shown by three case studies of lighting installation with different types of Hollow Light Guides (HLG) – main line, with heliostate and with short suntube (HLG skylights) [1-7].

Method 1. A 20000 sq.m. thermo-constant cooper plate production area (at day and night, summer and winter temperature must be (20±1)ºC) was illuminated by mirror slit HLG. The construction of building (Fig. 1) was designed with special electrical engineering corridors to decrease the heat input in the production area. In these corridors outside the working premises in maintenance bridges were installed the injectors with HID mirrorlike lamps 3×700 W each. That alone (thanks to the fact that heat from high wattage lamps was generated outside the working area) allowed to decrease the wattage of the air–conditioning systems by 600 kW. The resulting energy saving in an 24/7 production cycle amounted to 5,160 MW hours per year. The other results: - number of luminaires (in comparison with first design) was decreased by a factor of 30; - extension of the electro net was reduced many time; - the maintenance costs were greatly decreased. 2. Integration of natural and electrical illumination is an important problem of modern lighting engineering. Good solution of the problem gives benefits not only in high single-story buildings (workshops, supermarkets, exhibition halls, museums, etc.), but also in the deep buildings glazed along the perimeter, with windows located far from the central zones, and in the buildings having no daylight at all (underground trade centers, car parks, offices, stores, metro stations, etc.).

635

Fig. 1.The constraction of termo-constant building with main HLG line: 1 – Ingector 3x700W HID Mirrorlike + 1 IL500W; 2 – Maintenance Bridges; 3 – Electrical engineering corridor Figure 2 shows how to improve the solution and provide more balanced integration of daylight and artificial lighting systems. There is no wiring inside the premises and maintenance of the lighting system is run from outside. Besides, the system is better esthetically, since lamps and luminaries are not seen from below, while appearance of the integral system is nearly the same in the evening and at daytime.

Fig.

2. A suggestion concerning integration of lighting systems on the base of HLG skylights

More cardinal solutions of how to lead sunlight into a room lacking daylight were achieved using heliostats and long HLG [3-5]. The "Heliobus" lighting installation in the 4-storied school in St. Gallen (Switzerland) was the next significant step in the development of the integrated lighting systems. For the first time the systems for transporting and distributing solar light and electric light were combined in a single long vertical hollow light guide (Fig. 3). A similar solution was realized in the lighting installation of the 2-storied school in Schiers (Switzerland) (Fig. 4). The two systems differ only in the design of heliostats. "Heliobus" installation comprised tightly sealed stationary heliostat (with a shape being the intersection between two cylinders with equal diameter) and transitional device with light sources located at the upper part of it. The system in Schiers school had open mirror heliostats with two axes of rotation that were directed to the Sun permanently, the lamps were located at the bottom of the light guide, hence the luminous flux was passing in reverse direction with respect to solar light. We suggest three types of integral systems: for high one-story buildings, for underground rooms, and for wide buildings with perimeter glazing. Making the proposition we understand the need to do the following: - to use the simplest sealed heliostats (i.e. without open optical systems) in the buildings located in industrial sites; - to use unified units for transportation and distribution of solar or electric light; - to take electric circuits and units that require maintenance out of the illuminated room; - to eliminate dissipation of heat (radiated by the Sun or electric lamps) in the room in order to reduce the power of air-conditioning systems; - to reduce the dimensions of the openings, through which the light is being introduced in the room.

636

Fig. 3. Integrated lighting system "Heliobus" Left: Vertical section of system showing: 1 – collector (helliostat); 5 – HLG with SOLF internal prismatic film; 10 – specularly reflecting film; 11, 12 – internal extractors; 13 – diffuser Right: light collector (helliostat) at root level and HLG view

Fig. 4. Integral lighting system in Schiers: 1 – heliostat mirrors; 2 – sunlight input secondary optical system; 3 – plane transparent silicate glass; 4, 5, 12 – connection group of the light guide system; 6, 7, 11, 13, 15 – construction assembly of fastenings; 8 – cone mirror's reflector; 9 – circular extractor; 10 – light guide system; 14 – filter for thermo protection; 16, 18 – basement with electronic power supply system; 17 – four mirror's halide lamps

Figure 5 show the schemes of the suggested systems that meet the requirements stated above. Figure 5a shows the integral system for high one-story building. It comprises heliostats 1, transitional units 2 (with light sources) running through roof or walls, and hollow light guides - lightdistribution devices having the shapes of luminous disks, rectangles, or bands. The calculations give evidence that for a 8-10 m high building located in the Central Europe the single system (Fig. 5a) is

637

able to provide the illuminance of 300-400 lx within the area of 150-250 m2 (depending on heliostat dimensions and the number of light sources or luminaries). In underground rooms the latter system shall be completed with a vertical light guide 3 with different height and diameter (see Fig. 5b, dotted line).

ɚ)

b)

Fig. 5. The integral lighting system suggested for high one-story buildings (a), for underground rooms (b), for deep rooms (c)

c) Figure 5 (c) shows lighting systems designed for the rooms with a depth of 10-20 m (from windows). The calculations show that said system with a hollow light guide (Fig. 5c) provides the illuminance level of 300 lx in the office having minimal height and the length of 17 m. The integrated system presented by us (in cooperation with A.A. Korobko and V.M. Pyatigorsky) meets nearly all above-stated requirements. The reduction of electric power consumption (often by 3-5 times, like in "Heliobus" installation) is the main economic effect that makes the integral systems most promising. 3. The third group are the short HLG skylights. The simplest skylight systems provide the solar light and prevent the premises from overheating that can occur in summer months and limit the heat loss that occurs in winter months. Skylights HLG in Cottedg 15 are shown in Fig. 6a (companies name – Solarspot or SunPipe) [6, 7]. This effect is perhaps one of the most significant advantages of HLG, particularly in southern European countries and the Middle East. With temperatures externally often being 40°C, the prevention of solar gain to the inner space is vitally important. The HLG skylight creates a sealed tube of still air, which acts as an excellent insulator against heat transfer. Due to the relatively small diameter of lightpipes, IR is unlikely to penetrate the HLG except very short systems and so, as a result, the skylight remains unaffected by the external temperature conditions. This factor alone has a significant effect on the application of HLG skylight as an alternative to conventional rooflights or skylights.

638

a) b) Fig. 6. Skylights in Cottedg In winter months the situation is reversed but again, the HLG tube of a relatively small diameter results in minimal heat loss. The most popular sized HLG skylight 250 mm – 530 mm diameter have been assessed to have a U-value as follows: U = 0.31 – 1.51W/K. These values show a considerable improvement on standard double glazing conventional skylights. A simple area calculation shows that these figures are equivalent to 1.59-1.72 W/m2/K. The UK Building Regulations states a value of 1.9W/m2/K for argon filled triple glazing with 16 mm spacing, which demonstrates the performance of the excellent HLG skylight. 4. Input daylight in basement premises. One of new options of using natural light for low level and basement premises is a new systems also called «Heliobus» which is implemented in Switzerland, by Signer Company (Fig. 7 a, b). Their device is a sealed rainproof box maintained on to the low level window. The inner surfaces of the box covered with mirror reflected material of Alanod Co type «Miro». The top of the box covered by flat clear hard silicat glass. This provide input box natural light: direct sunlight and global light from outside.

Results HLG skylights give the possibility of: - reduced energy consumption (lighting energy cost for office space could be reduced by as much as 60%); - reduced demand at peak demand times; - daylighting measures cost less than 1% of the construction budget and achieved a payback in less than 2-3 years.

Conclusion In conclusion, we have to point out (along with high economic and technical advanteges) one very important feature of suggested integral lighting systems with the light guides. Unlike other known electrical lighting systems that provide steady-state lighting conditions, the new integral system operating with solar light provides dynamic, permanently changing illumination. It reflects everything that takes place in the sky: either rainbow or changes of sky brightness, e.g. shadows run along the light guide as clouds move in the sky. All these make integral systems look lively and optimistic. The author is very thankfull to your colleagues and co-authors of investigations G. Bukhman, W. Buob, G. Brackale, V. Pjatigorsky, A. Korobko, R. Signer.

References: 1. Hollow Light Guide technology and applications. Technical report CIE 164:2005. 2. Tubular daylight guidence systems. Technical Report CIE 173:2006. 3. Aizenberg J.B., Bukhman G.B., Pyatigorsky V.M., Korobko A.A. Development and application of Slit Lightpipes. Present and Future. Proceedings of IESNa Conferenc, 1992. 4. Aizenberg J.B., Korobko A.A., Pyatigorsky V.M., Buob W, Signer R. Daylighting and artificial lighting of central zones of multi-storey buildings with Hollow Light Guide System «Heliobus». Proceedings of Lux-Europa-97 Conference, Amsterdam, 1997. 5. Aizenberg J.B. Integral lighting systems for rooms with insufficient daylight. Light and Engineering V 11, ʋ 1, 2003.

639

6. Bracale G. Building daylight by Solarspot. A new passive Hollow Light Guidance system. Light and Engineering V 13, ʋ 4, 2005. 7. Payn T. The development of lightpipes in the UK. Light and Engineering, V 12, ʋ 2, 2004.

640

Improvement of the Energy Efficiency of a Distribution Warehouse in Madrid (Spain) with Special Emphasis on Daylight Optimisation Cécile Bonnet, Joan Carles Bruno, Alberto Coronas Universitat Rovira i Virgili - CREVER, Group of Applied Thermal Engineering

Abstract This paper presents a technical study which has been realised by CREVER – Group of Applied Thermal Engineering on an industrial building to be built in the province of Madrid with a total area of 33,795 m2. The main part of the building is used as warehouse and is thus neither heated nor cooled, so that the energy demand is basically due to lighting. The building also includes two air-conditioned office areas of 985 m2 each one. As reference, another existing building with the same constructive characteristics, use and location has been considered in order to evaluate the potential energy savings. In this building, it has been evaluated that almost 84 % of the electricity demand is due to the warehouse lighting, 9 % to offices (heating cooling, domestic hot water, lighting and other electrical equipments) and 7 % to battery load of handling machines. First, the efficiency of the planned lighting installation has been assessed, according to the references of the Spanish Building Technical Code. Then, with the help of the daylight simulation software DAYSIM, the saving potential through the use of a photocell controlled dimming system has been evaluated. Also the increase of the daylight penetration through the use of roof glazing with higher transmittance has been contemplated. The effect of the increased roof glazing transmittance on the summer comfort conditions has been analysed with the help of the building energy simulation software EnergyPlus (DesignBuilder interface). The study concludes that energy savings of up to 25 % could be reached through the optimisation of the warehouse lighting system implementing a photocell controlled dimming system and increasing the sky glazing transmittance up to 30 %. The effect of the increased glazing transmittance on the summer comfort conditions is considered to be reasonable. Additionally, the installation of PV panels on the roof of the building is planned, which should permit to reduce the non-renewable energy consumption of the building by additionally 25 %, reaching a global primary energy saving of almost 50 %.

1. Introduction and objectives of the study The GreenBuilding Programme is a programme of European Commission which aims at encouraging the implementation of energy efficiency measures in non-residential buildings. Both existing and new non-residential buildings can take part to the programme. The condition for owners of new buildings to become partners of GreenBuilding is to reduce the primary energy consumption of the building by at least 25 % with respect to a reference building of similar characteristics. [1] The objective of this study is to evaluate different alternatives to reduce the energy consumption of a new industrial building for logistic use with respect to an existing building presenting the same characteristics, use and location, in order to become a partner of the Programme. A special focus of this study will be the reduction of the lighting electrical demand. This study has to be considered as a preliminary study evaluating possible measures from an energy saving point of view. No economical considerations are contemplated. This paper is structured as follows. In section 2, it is presented the studied and reference building used to evaluate the potential savings and also it is determined the corresponding energy demand of the reference building. In section 3, it is presented the method used to evaluate the lighting efficiency and the impacted associated with different improvement measures proposed. Finally, in section 4, are summarised the benefits of the proposed energy improvement measures.

641

2. Presentation of the studied and the reference buildings 2.1. Presentation of the studied building The object of this study is an industrial building to be built during 2008 in a logistic park located in the province of Madrid. The building, to be used as distribution warehouse includes a storage area of 31,824 m2 mainly occupied by storage racks, as well as two 3-floor office sections of 985 m2 each one located on the north-west and south-east facades (see Fig. 1), including offices and changing rooms. Due to the use made of the building, the occupancy of the warehouse section is foreseen to be very low and variable. Also in the office section, it is predicted a relatively low occupancy of about 20 pers / building corresponding to about 0.02 pers / m2. The warehouse section will neither be heated nor cooled, for which the energy demand of this section representing almost 94 % of the total building area, is exclusively related to lighting and battery load of the handling machines. However, the office sections are air-conditioned with reversible air handling units and additionally present usual energy demand for domestic hot water (DHW), lighting and other electrical equipments. Figure 1: General view of the studied building showing one of the office areas

2.2. Presentation of the reference building To enable comparisons and thus facilitate the assessment of the considered energy saving measures in the considered building, a building with almost identical constructive characteristics, use and location (both building being from the same real estate developer) is considered as reference building in this study. This building is in operation since October 2006. Real electricity consumption data for this reference building is available from the electricity bills of the seven first month of operation of the building. The buildings, besides being both located in the province of Madrid, present similar configurations (location of office sections on the opposite façades), floor areas as well as roof inclinations and roof glazing proportions. Figure 2: General plan of the reference building

642

2.3. Repartition of the energy consumption by use (reference building) 2.3.1. Estimation of the annual energy consumption of the reference building Both reference and studied buildings exclusively present electrical energy consumption. The warehouse section has no thermal demand. In the office area, the thermal demand (heating and cooling) is covered by an electric driven air handling units. According to the seven first electricity bills of the reference building, an annual energy consumption of 2 1,323 MWh/a has been estimated, corresponding to 38 kWh/m a. This electricity consumption is due to warehouse lighting, battery load of handling machines, and office requirements for heating, cooling, DHW, lighting and office equipment.

2.3.2. Estimation of the electricity consumption due to warehouse lighting In order to evaluate which part of the electricity consumption is due to the warehouse lighting, a daylight simulation of the warehouse has be realised using the daylight simulation tool DAYSIM [2]. The corresponding electricity demand for artificial lighting is largely dependent on the user behaviour. Therefore, two user profiles have been considered in the simulation: -

-

an intermediate user profile with 50 % active users and 50 % passive users. o Active users operate the electrical lighting according to the ambient daylight conditions o Passive users do not operate the electrical lighting according to the ambient daylight conditions and keep the electrical lighting on throughout the working day. a passive user profile with 100 % passive users. [3]

A building occupation from 7 AM to 11 PM has been considered. An illuminance of 200 lux is required in the warehouse. According the plans of the lighting installation of the reference building, a lighting 2 power of 8 W/m is installed in the warehouse.

643

As there are no windows on the building façade, the daylight only enters the building through translucent roof glazing (polyester) which represents 17 % of the total roof area. The roof tilt is around 7 %. The results of the simulation for the warehouse lighting are presented in table 1. Table 1: Estimation of the annual electricity consumption for warehouse lighting User profile

Intermediate

Passive

kWh/m .a

33.4

36

MWh/a

1,100

1,186

2

Estimated annual electricity consumption

2.3.3. Estimation of the electricity demand of the office sections (reference building) To evaluate the annual energy demand of the office buildings for heating, cooling, DHW, lighting and office equipments, a thermal simulation has been performed using the simulation software EnergyPlus [4] with the DesignBuilder interface [5]. The results of the DesignBuilder simulations are presented in table 2. Table 2: Annual final energy demand of the office areas (reference building) North section

TOTAL

South section 2

Total office sections 2

kWh/a

kWh/m a

kWh/a

kWh/m a

kWh/a

kWh/m2a

59,753

65.1

58,818

64.1

118,570

64.6

2.3.4. Estimation of the electricity consumption for battery charging of the handling machines For battery charging, an annual electricity consumption of 93600 kWh has been estimated considering a power of 60 W for 6 hours charging per day, 5 days a week and 52 weeks per year.

2.3.5. Conclusions - repartition of the energy demand by use Table 3: Distribution of the energy consumption by use (reference building) Real Estimated energy demand according to simulations consumption Lighting user profile (elect. bills) Intermediate Passive kWh/a % of total kWh/a % of total kWh demand demand Warehouse 1,100,602 83.8 1,186,278 84.8 lighting Offices 118,570 9.0 118,570 8.5 Battery load 93,600 7.1 93,600 6.7 Total 1,312,772 1,398,448 1,323,713 Total /m2 37.7 kWh/m2a 40.2 kWh/m2a 38.05 From these results, it can be concluded that the intermediate lighting user profile is the most coherent with the real consumption data according to electricity bills. For instance, an intermediate user profile will be considered as parameter in the following simulations to evaluate the energy saving potential through daylight optimisation. The results also confirm the hypothesis that the warehouse lighting represent the major part of the electricity demand of the building and thus that the study has to concentrate on the reduction of this demand.

644

3. Results 3.1. Assessment of the energy efficiency of the planned lighting installation First of all, it is necessary to assess the energy efficiency of the planed lighting installation of the considered building. For this purpose, the energy efficiency index of the warehouse and of each zone of the office areas will be determined. According to the new Spanish Building Technical Code (Código Técnico de la Edificación) part HE 3 [6], the energy efficiency Index IEE is defined as the lighting power installed per m2 lighted surface and per 100 lux illuminance maintained in this zone.

IEE

P ˜ 100 S ˜ Em

in W /m2·100 lux

where -

P: installed lighting power [W] S: lighted surface [m2] Em: medium horizontal illuminance maintained [lux]

The Spanish Building Technical Code establishes maximum values for this index according the use made of the considered area. These values will be used as reference to evaluate the efficiency of the installation. From the results (Table 4), it can be concluded that the energy efficiency index of the planned lighting installation in the warehouse lies well below the critical value established by the Spanish Building Technical Code for warehouse. Indeed, the selected lamps (fluorescent tubes and high pressure sodium lamps) present a high luminous efficiency [7]

Table 4: Evaluation of the efficiency of the planned lighting installation in the warehouse Type of lamps

Unit power Number of lamps Installed power Corresponding area Required illuminance Calculated energy efficiency index Max. energy efficiency index according to CTE

W W m2 lux

Fluorescent tubes

High pressure sodium lamps

58 464 26,912

400 544 217,600

Total

244,512 32,952 200

W m2 lux

W/m2100 lux

3.71

W/m2100 lux

W/m2100 lux

5

W/m2100 lux

3.2. Evaluation of energy saving measures for warehouse lighting

645

The previous results showed that the energy efficiency of the planned installation is good and no significant energy saving potentials can be associated to the change of the type of lamps. Therefore the study has been focused on saving potentials in the lighting use. A major way to reduce the lighting energy demand is to optimise the use of daylight by increasing the daylight penetration and particularly by adapting the artificial light to the daylight availability through the use of daylight sensors and dimmers. In section 2.3.5, it has been determined that a large part of the users probably do not operate the electrical lighting according to daylight conditions and often keep the light on throughout the working day. However, considering the important area which is considered in this study (over 30,000 m2), the low density of occupation, and the activities realised in the building (load and unload of goods between storage racks and trucks implying almost constant movement), the use of occupancy sensors in the warehouse section has not been contemplated as an adequate alternative. On the other hand, the following alternatives will be simulated using the daylight simulation tool DAYSIM [2] and compared with the reference building: -

Control of the artificial lighting system using daylight sensors (photocells) and dimmers Increase of the relative roof glazing area (with and without daylight control) Use of roof glazing with higher daylight transmittance (with and without daylight control)

These alternatives or measures are presented in detail in table 6. Table 6: Considered alternatives to reduce the energy consumption for warehouse lighting Case studied

Roof glazing percentage

Roof glazing transmittance

Daylight control

Description

Ref. Case

17

20

no

Reference building

Case1pc

17

20

yes

Case 2 Case2pc

21.3 21.3

20 20

no yes

Case 3

17

30

no

Case 3pc

17

30

yes

Ref. case + photocell controlled dimmer Increased roof glazing surface 21.3 % Case 2 + photocell controlled dimmer Use of glazing material with higher transmittance (30 %) Case 3 + photocell controlled dimmer

In order to evaluate these measures, the daylight availability will be estimated using the following factors: -

Daylight factor: it is defined as the ratio of the indoor illuminance at a point of interest to the outdoor horizontal illuminance under the overcast CIE sky. [3] Daylight autonomy: it is defined at a point in a building as the percentage of occupied hours per year, when the minimum illuminance level (lux) can be maintained by daylight alone. In contrast to the more commonly used daylight factor, the daylight autonomy considers all sky conditions throughout the year. [3]

The simulation will be realised for an intermediate user profile as it has been determined in section 2.3.5. The simulation includes an optimized regulable lighting control system with photocells. The regulation of the artificial lighting with photocells permits to adequate the light intensity of the lamps according to the quantity of daylight available. Once the lamps have been manually activated through an on/off switch, the dimming system regulates the artificial light so that the total illuminance (natural and artificial) reaches the minimum required illuminance (here 200 lux). In case that the minimum required illuminance level is reached only with daylight, the artificial light is automatically switched off [2]. The photocell stand-by power is 2 W. A ballast loss factor of 20 % has been considered.

646

The results are shown in Table 7. It can be concluded that the implementation of a daylight control with photocells and dimmers should contribute to significant energy savings for the lighting of the warehouse. With this measure and without any modification of the building envelope characteristics, such a system should allow a reduction of the warehouse lighting energy consumption by 22.5 %.

Table 7: Evaluation of the energy saving potential for warehouse lighting Ref. Case

Case 1pc

Case 2

Case 2.pc

Case 3

Case 3pc

%

17.0

17.0

21.3

21.3

17.0

17.0

%

20

20

20

20

30

30

Manual

Photocells

Manual

Photocells

Manual

Photocells

%

2.06

2.06

3.15

3.15

3.14

3.14

%

57

57

63.4

63.4

62.7

62.7

kWh/m2.a

33.4

25.9

31.9

23.4

32.5

24.3

MWh/a

1,101

853

1,051

771

1,071

801

MWh/a

0

247

49

330

30

300

7.50

1.50

10.00

0.90

9.10

22.5

4.5

29.9

2.7

27.2

Case studied Roof glazing percentage Roof glazing transmittance Daylight control Daylight factor Daylight autonomy Estimated lighting energy demand Savings with respect to ref. case

2

kWh/m .a %

0

If this measure is combined with a higher daylight penetration through the use a roof glazing material with higher daylight transmittance or the increase of the roof glazing surface, savings of up to 27 % and 30 % respectively, could theoretically be reached. However, an increase of the daylight penetration without daylight control system only produces limited savings of only 4.5 % in case 2 and 2.7 % in case 3. This is due to a non-optimal user profile where 50 % of the users are considered to operate the artificial light in a passive way.

3.3. Influence of an increased daylight penetration on summer comfort conditions The increase of the roof glazing area or the use of a material with higher daylight transmittance leads to higher daylight penetration in the building and should thus permit to reduce the electric lighting requirements. However this also means higher solar heat gains in summer which can lead to a sensible increase of the operative temperature inside the warehouse, which is not air-conditioned. Hence a thermal analysis of the warehouse using the EnergyPlus software [4] with DesignBuilder interface [5] will be performed in order to analyse the summer comfort conditions for the case studied (Ref. Case, case 2 and 3) in the previous section and for two additional cases with higher increase of the daylight penetration through higher percentage of roof glazing (case 4 and 5). This will be achieved by determining the number of hours per year by which the operative temperature of the building is higher than 28 ºC.

Table 8: Effect of the different measures on the summer comfort conditions Case

Percentage of

Transmittance

647

Percentage of

Increase of the

studied

roof glazing

of the roof glazing 20

occupied hours per year with Top > 28ºC 12.8

number of hours per year with Top > 28ºC 0

17

Case 2

21.3

20

13.5

5.4

Case 3

17

30

13.3

3.7

Case 4

25.6

20

14.2

11.2

Case 5

29.8

20

15.0

17.5

Ref. Case

In the reference case (Table 8), the percentage of occupied hours with an operative temperature over 28 ºC lies by 12.8 %. In the case with an increased roof glazing surface from 17 to 21.3 % (case 2) and the case using a roof glazing material with higher daylight transmittance (case 3), the effect on the comfort conditions is limited: the augmentation of the number of hours of discomfort increases respectively by 5.4 and 3.7 % which could be considered as acceptable. A higher increase of the translucent roof surface up to 25.6 % and 29.8 % would be not acceptable from the indoor comfort conditions point of view since it would represent an augmentation of the number of hours of discomfort of 11.2 and 17.5 %, respectively.

4. Conclusions A general assessment of the saving potential of the different measures has been compiled in table 10. Table 10: Evaluation of the energy saving potential of the different measures Energy saving measures kWh Warehouse lighting Case 1pc Photocell control Increased roof glazing area Case 2 (21,3 %) Case 2pc Case 2 + photocell control Higher transmittance of roof Case 3 glazing Case 3pc Case 3 + photocell control

Predicted energy savings % of total estimated demand kWh/m2

238,681

7.5

18.5

47,736

1.5

3.7

318,241

10.0

24.7

28,642

0.9

2.2

289,600

9.1

22.5

Daylight control with photocells constitutes an attractive alternative to reduce the energy consumption of the building. This measures could contribute to reduce the electricity consumption of the whole building by 18 %, corresponding to 238.7 MWh/a or 108.7 tons CO2/a (considering a CO2 emission factor of 0.4556 kg CO2/kWh electricity). Associated with an increase of the daylight penetration through an augmentation of the roof glazing area from 17 % of the roof area to 21 %, savings of up to 24.7 % (318.2 MWh, 145 tons CO2/a) could be reached without causing significant worsening of the summer comfort conditions. The use of roof glazing materials with a higher daylight transmittance combined with daylight control would represent savings of 22.5 %. Additionally, the installation of a solar photovoltaic plant on the roof of the building is foreseen. This plant with a nominal power of 261 kWp should produce almost 346 MWh/a electricity and would represent an energy saving of 26.8 % of the total energy demand of the building.

648

By implementing a daylight control system, increasing the roof glazing area up to 21.3 % and installing the planned photovoltaic plant it could be reached a primary energy savings of up to 51.5 % corresponding to 664.2 MWh/a or 302.6 tons CO2/a. If the use of roof glazing materials with a higher transmittance (30 %) is preferred instead of the increase of the roof glazing area, a primary energy savings of 49.3 % could be reached corresponding to 584.7 MWh/a or 266.4 tons CO2/a. These measures would mean that the energy saving goals of the GreenBuilding programme would be reached. However additional considerations on the economic viability of the measures are required.

Acknowledgement The authors would like to acknowledge the collaboration of the technicians of Coperfil Inmobiliaria in providing data for this work and the funding support of the project EIE/04/057/S07.38638 and EIE/07/109/SI2.466268.

References [1]

GreenBuilding website: www.eu-greenbuilding.org

[2]

Daylighting analysis software DAYSIM, http://irc.nrc-cnrc.gc.ca/ie/lighting/daylight/daysim_e.html

[3]

Tutorial on the Use of Daysim Simulations for Sustainable Design, Dr. Christoph F. Reinhart, Institute for Research in Construction, National Research Council Canada, Ottawa, Canada, 2006

[4]

EnergyPlus website: http://www.eere.energy.gov/buildings/energyplus/

[5]

DesignBuilder website: http://www.designbuilder.co.uk/

[6]

Codigo Técnico de la Edificación, Documento básico HE – Ahorro de Energía, Sección HE3 – Eficiencia energética de las instalaciones de iluminación, 2006

[7]

GreenLight website: www.eu-greenlight.org

[8]

DAYSIM Case Studies, Dr. Christoph F. Reinhart, The Lighting Group – Institute for Research in Construction, National Research Council Canada, Ottawa, Canada, 2004

649

650

Market Development of ESCO Schemes for Lighting Refurbishment Mechthild Zumbusch, Berliner Energieagentur GmbH

Abstract Lighting in buildings and working places fulfils different important tasks. Good lighting shall increase working quality, improve the condition of work, and prevent accidents. Additionally, lights can highlight and accentuate objects and areas in buildings. While a bad quality of lighting causes arises the costs for energy and maintenance - which are usually not recorded separately – stay unrecognised. Hence, rising operation costs usually do not lead to necessary and economic refurbishment measures. Additionally, the public sector is often missing the financial capabilities. For companies the willingness to invest in measures other than their core business is usually low. So called Public-Private-Partnerships (PPP) offers the possibility to outsource investments in efficient technologies to a third party. In addition, the Energy service company (ESCO) provides optimised operation during the complete contract duration. Thus, contracting is not just a financing instrument, but includes essential elements of operation optimisation and management. PPP models like contracting are successful tools to save sustainable amounts of energy and maintenance costs. The Berliner Energieagentur developed in 2007 two guidelines as practical manuals to establish successful contracts with regard to street lighting and indoor lighting. They guide through all the necessary steps of action to be taken from project identification and development to tendering procedure and finally to the conclusion of the contract. Furthermore the guidelines explain the different concepts and possibilities of contracting, discuss possible barriers and problems and the respective suggestions for solutions. The ESCO has to assure that appropriate laws and standards are fulfilled and that saving measures will not fall below these values during the contract period. A good practice example for lighting contracting was the project of the city of Mechernich. By a contracting project it was possible to reduce the energy costs for lighting of the local school significantly. Additionally, the public street lighting was refurbished in the same call for tender and the budget disburdened by another 20,000 Euro per year. The project of the city of Mechernich was awarded as “GreenLight Partner” in May 2006. Two other good practice examples for lighting contracting will be introduced. One of them is the refurbishment of the street lighting of the city of Kempten. The other one is the maintenance and operating of the traffic light system in Berlin.

Introduction About 10 percent of the German electricity consumption is used for lighting – in office buildings about 20 percent. By the use of modern and efficient lighting technologies and user specific saving potentials consumption decreasing about 30 to 50 percent can be gained. Especially street lighting often is characterised by out-dated technology and a high amount of maintenance and fault clearance. In Germany for example more than 4,000 million kWh are used for street illumination every year, which costs the communes about 500 million Euro annually. This results electricity costs for every citizen about 4 € per year – including service and maintenance costs rising up to 10 € per year. Unfortunately the high electricity costs are accompanied by an insufficient illumination.

651

Lighting in buildings and working places fulfils different important tasks. Good lighting shall increase working quality, improve the condition of work, and prevent accidents. Additionally, lights can highlight and accentuate objects and areas in buildings. The lighting of our night time environment stands for safety and security implying life quality. Modern public lighting fulfils even more than societal goals. Its purposes are guidance of traffic, highlighting historic buildings and constructions and enhancing economic development. While a bad quality of lighting is obvious the costs for energy and maintenance - which are usually not recorded separately – stay unrecognised. Hence, rising operation costs usually do not lead to necessary and economic refurbishment measures. Additionally, the public sector is often missing the financial capabilities. For companies the willingness to invest in measures other than their core business is usually low.

Models of Third Party Financing for lighting refurbishment Contracting is a well established tool to finance measures for energy efficiency. Especially Saving Contracting has become an instrument to realise economic CO2 emission reductions. Furthermore Contracting offers: x x x

Modernisation, e.g. lighting Reduction of operation costs Lean and efficient management

A main distinguishing feature of contracting is, that the service company obliged under the contract bears the economical risk (or major parts of the risk) of the project which includes energy consumption as well as maintenance. At the same time the ESCO is given the chance to gain its own appropriate profit if the intended improvement in efficiency is achieved or even exceeded. A contracting project can be financed in three different ways: Table 1: Financing of several contracting models self-financing -

The building owner provides the financing for the energetic measures himself

debt financing -

The financing is provided by a financial institution

third party financing -

The ESCO finances its energy measures

Also hybrid solutions are possible, e.g. a third party financing by the ESCO with an investment grant by the owner. Although in recent years the most varied models have emerged, a basic structure can still be determined which has led to the following widely recognised classification. Depending on the system approach or aim of contracting, the following basic forms can be distinguished: x x x

Technical Plant Management (also Operation Management Contracting or Technical Building Management Energy Supply Contracting (also Facility Contracting or Energy Delivery Contracting/delivery of useful energy) Saving Contracting (also Performance- or Energy Saving Contracting)

The basic models (especially Supply Contracting) are widely-used in the sector of energy services for lighting. Basically, three contracting models can be differentiated:

652

x x x

Lighting Contracting Light Supply Contracting Saving Contracting

Thereby, it can be differentiated between pure service models, in which the lighting system remains in the ownership of the public authority, a complete transfer of the system to the private company or rather a combination of both models. Table 2 shows the volumes of the three models. Generally, energy services offered in the lighting environment range from project development to operation, from servicing and maintenance up to complete reconstruction combined with the respective financing.

Table 2: Different models of contracting in the lighting sector Lighting Contracting Optimising operations and if Application applicable refurbishment measures of the lighting devices Operational management and if applicable financing, Services planning of the refurbishment, installation and maintenance

Light Contracting

Supply

Optimising operations and if applicable refurbishment measures of the lighting devices Operational management and if applicable financing, planning of the refurbishment, installation and maintenance Additionally: energy purchasing and supply

Saving Contracting Measures to achieve energy and maintenance costs savings

Financing, Planning, Installation, maintenance and implementation of saving measures

Contracting rate as Remuneration for the Financing Contracting rate as remuneration for the services operating cost savings remuneration for the services and energy costs achieved For single refurbishment Remarks measures including maintenance

Currently not relevant in indoor lighting1

Analogue to: Technical Plant Management Energy Supply Contracting

For complex solutions with high saving potential Performance Contracting

Contract models Basically, saving contracts can be divided into two groups based on the time-relation of your participation in the saved operating costs: duration model and participation model. Both models have in common that the contract period is divided into two phases: the preparatory phase and the phase of the ESCO’s main obligation to perform.

Preparatory phase:

Implementation of investments and optimisation of lighting system

1

In indoor lighting a Light Supply Contracting is so far unknown because energy consumption for light is usually not measured separately from other electricity consumptions and operating hours of the lamps can not exactly be determined. Modern building control systems are able to determine operating hours and energy consumption per lamp and, hence, make Light Supply Contracting theoretically possible.

653

Warranty of operating costs savings on basis of the measures taken in the preparatory phase and further optimised operation

Main obligation to perform:

Figure 1: Principle of duration model

Duration model Operating costs (€/a)

Actual costs

Actual costs - reduced costs = saved costs = contracting rate which during the contract period goes to the contractor as remuneration for the services

Reduced costs

Commencement of main obligation to perform

Time (a) Useful life of saving measure

End of contract

In the case of the duration model contract, the ESCO is entitled to the cost savings to be achieved for the entire duration of the main obligation to perform, i.e. the ESCO receives as remuneration for the investments and the services a so-called contracting rate which corresponds to the saved operating costs. Thus, owners can enjoy the cost savings only after the end of the contract period. Increase in comfort and environmental relief can be profited from right from the start. Figure 2: Principle of a participation model Participation model Operating costs (€/a)

Actual costs

Portion of saved operating costs which during the contract period goes to the client as his share

Portion of saved operating costs = contracting rate which during the contract period goes to the contractor as remuneration for the services

Reduced costs

Commencement of main obligation to perform

End of contract

Time (a) Useful life of saving measure

654

In the participation model, the owner participates financially from the guaranteed savings with the commencement of the main obligation to perform. The amount of your participation is to be stipulated in the contract and usually is at least 10% of the savings achieved. The participation results in longer contract durations, however, at the same time, your budget is immediately unburdened during the phase of the main obligation to perform.

Motivations, Barriers and Solutions Motivations The consequently operator of the lighting system can profit from several advantages of Saving Contracting. First of all the authority or the owner is – after signing the contract – relieved from both organizational implementation and financial burden. Both are interesting incentives for owners to perform such a contracting project. Due to the specific knowledge, financial incentives and the legal commitment of the ESCO, economical saving potentials are particularly efficiently used and saving measures are implemented much faster. Besides significant reductions of operating costs implemented measures can lead to better light quality, fewer failures, upgrading of value and attractiveness and also to a reduction of greenhouse gases.

Barriers Experience has shown some general barriers for the use of contracting models exist in the building sector. These are, of course, also valid in the lighting sector. Some of the main barriers are: x x x

Technical and organizational barriers: (data collection; baseline calculation; increased organizational efforts) Legal barriers: (provision of a secure and fair contract; uncertainties in e.g. public budget, municipal law, procurement) Human barriers: (lack of confidence in the largely unknown model and external service provider; fear of staff reduction)

Suggestions for Solutions Technical / organisational barriers: The effort of compiling data or calculating a baseline depends on the complexity of the project and the available records. Available tools like data entry forms, model contracts or evaluation tools for awarding the most economic offer assist to enhance and simplify the processes of a contracting project. Furthermore the possibility exists to contact energy agencies or consulting companies for further assistance. Legal barriers can be solved by standardised and legally examined model contracts. The provision of such contracts can guarantee fair and secure project procedures. Usually only small adaptation is needed for the concrete project conditions. Human barriers are sometimes shown in the opinion: “We can realize refurbishments the same way but even cheaper!” Theoretically the possibility for a solution in own-direction exists, virtually these projects are rarely realized successfully. Therefore the following questions should be considered before deciding for a solution in own-direction: x x x x

Can the amount of investment be provided? Is the necessary know-how available? Can the desired savings be guaranteed for several years? Do sufficient economical incentives exist in achieving these savings?

655

The fear of staff reduction caused by external energy services can be reduced if the responsible personnel are involved in the process right from the start. Usually the ESCO works in the field of energy saving and energy management and therefore takes over responsibilities which have been sparsely, if at all, been focused on before. In general, Saving Contracting does not lead to staff reduction but rather to a shift in duties.

Best practice examples Best Practice Mechernich The realisation of the energy saving project of the city of Mechernich was consulted by the Energieagentur NRW. The lights of several schools and of the street lighting were planned to be refurbished by a lighting contracting. Following an European wide call for tender the contractor euroluxAG was assigned with a full-service contract. The complete lighting constructions of the schools were replaced by highly efficient energy saving lighting systems. Even more energy savings of 30% were realised with the refurbishment of the street lighting. The investment costs will be amortised within 10 years. As a result of the lighting refurbishment of the school and the street lighting the city of Mechernich saves energy costs of around 70,000 Euro per year. These savings equal a mitigation of 750 tonnes of greenhouse gas emissions. Additionally, the operating costs are 6,000 Euros less per year.

A more detailed version (only in German) of this Best Practice Example is available at http://energiesparende-beleuchtung.de/pdf_files/enerlin_good_practice_mechernich_2.pdf

Best Practice Kempten With the intent to modernise the street lighting of the city and to save money and energy at the same time the city of Kempten decided to implement a street lighting contracting project. The most convincing offer was handed in by the contractor Luretec. The measures were carried out by the newly founded Kestra GmbH. The inefficient lights were replaced by sodium vapour lamps and power control units were integrated into the lighting system. A large amount of energy was already saved by the exchange of the lamps. The power control units are dimming the light to the mandatory lighting levels and by this even more energy is saved at the same light. The city of Kempten saves around 106,000 Euros resulting from the refurbishment of the street lighting. The investment of the contractor is repaid via a contracting rate for 9 years resulting from the savings of the project. Although more than 400 lighting points were additionally set up the energy savings are more than one million kWh per year. This equals a CO2 reduction of 630 tonnes per year. A more detailed version (only in German) of this Best Practice Example is available at http://energiesparende-beleuchtung.de/pdf_files/enerlin_good_practice_kempten_3.pdf

Best Practice light signal system of Berlin The state of Berlin had successfully handed over the operation of the street lighting to a third party. Based on this experience the senate decided also to contract out the operation of the light signal system. The Nuon Stadtlicht GmbH – also operating the light signal system of Amsterdam - was successful with the tender and signed the contract. Since 2006 Nuon Stadtlicht GmbH is responsible for the management, refurbishment, maintenance, operation, replacement and retrofitting of the light signal system. The contract has a volume of 126 Mio. Euro and the contract duration is 10 years.

656

According to the contract 618 light signal units (about one third) have to be modernised within the first five years. Mainly, this is the refurbishment from common light signals with light bulbs to LED types (Light Emitting Diodes). The installation of LED technology has several advantages: the energy consumption is around 80% less, the light intensity is much higher, sun reflection is almost none, and they have a 20 times higher life expectancy and have, hence, lesser maintenance costs. In the course of the contract between the state of Berlin and Nuon Stadtlicht GmbH more than 30% of the existing light signal units will be replaced by energy saving LED signal units by 2011. More than 100 units have already been refurbished by 2007. According to the plan more than 2.75 Mio. kWh electricity are saved per year. The result is that the state of Berlin saves more than 1 Mio. Euro per year. This equals a CO2 saving of about 1.650 tonnes per year. A more detailed version (only in German) of this Best Practice Example is available at http://energiesparende-beleuchtung.de/pdf_files/enerlin_good_practice_berliner_ampeln_2.pdf

GreenLight The European GreenLight Programme aims at reducing electricity consumption by promoting energy efficient technologies in the lighting sector. Enterprises and public authorities voluntarily commit to undertake measures in energy efficient lighting as a partner of the European Commission. In return the Commission supports the commitment of their partners with public-effective representation. The Berlin Energy Agency supports the European Commission with the promotion and promulgation of the GreenLight Programme and serves as National Contact Point in Germany. The GreenLight Programme has currently more than 430 partners within the European Union, around 40 of them in Germany. More information to the GreenLight Programme an all GreenLight Partners is available on the website http://www.eu-greenlight.org The city of Kempten became GreenLight Partner 2006, the city of Mechernich and the senate department for urban development in Berlin became GreenLight Partner in 2007.

657

658

IEECB'08

Data Centres and IT Equipment

659

660

Measuring the Energy Efficient Performance of Desktop and Notebook computers1 Kevin Fisher – Convenor Ecma TC38 – TG2 and Intel Corporation

Abstract Ecma TC38 has set up a Task Group (TG2) to develop standards for ICT and CE products to enable the measurement of the energy consumption of the device whilst taking into account its performance and capabilities. The initial work is to develop a standard for notebook (in AC mode) and desktop computers. Whilst the standard can be used by anyone, the initial user of the standard is targeted to be the US EPA’s Energy Star® programme Tier 2 implementation in 2009. Driven primarily by concerns over global warming along with desires to decrease the demand for both domestic and foreign oil resources, regulators worldwide are paying ever increasing attention to the energy efficiency (EE) of ICT & CE products. Due to the complexity and vast number of possible uses for products in the ICT and, to some extent, CE sectors, establishing standardized test procedures poses a great number of challenges. As such, regulators and industry have historically focused on limiting power consumption for each of a variety of singular power states, those states having been either well defined or in some cases simply inferred, rather than a true energy efficiency measure. This approach is far from optimal for the environment, for technological development or for the consumer as it fails to take into account the breadth and depth of usage patterns users of these devices employ them under. Simply measuring the energy consumption of a computer is not sufficient. The performance and capability of the product must also be taken into account. And so, the term “Energy Efficient Performance” (EEP) has been adopted by the Ecma task Group. Using the car as an analogy, for example, you would not buy a two seat highly fuel efficient car if you had a family of 6 to transport. You also would not use a motorbike if you were a salesman needing to transport large product samples in the trunk. Likewise when buying an ICT/CE product, fitness for purpose must be part of the decision making process alongside energy efficiency. To fill out the corollary, what is being proposed is to create a metric that goes beyond simply Kilometres per Litre or even Passenger Kilometres per Litre and to take into account also how fast those passengers arrive at their destination together with the capabilities they experience along the way. Ecma will therefore produce a standard which will describe how to measure the EEP of desktop and notebook computers. The standard will provide a standardised way of reporting results. These results will include the performance, capabilities and energy consumption of the machine.

1

The opinions expressed in this paper are those of the author and do not necessarily reflect full consensus within the Ecma TG.

661

History Ecma developed and published the world’s first environmentally conscious design standard (ECD) for the ICT & CE industries in 2003 as ECMA-341. The standard is aimed at the designer and provides pragmatic advice on how to reduce the environmental footprint of a product at the design stage. ECMA-341 was offered to the IEC (International Electro-technical Committee) for conversion into an IEC standard. IEC TC108 set up a Project Team (PT62075) to complete this work. The standard is now available as IEC 62075. Whilst ECMA-341 includes the definitions of energy saving modes and generic energy saving guidance for designers of ICT & CE products, the work of TG2 within Ecma complements that guidance by defining a methodology on how to measure the energy efficiency of a product whilst taking into account its performance and capabilities.

Today’s Energy Regulation for Computers There are a great number of voluntary and mandatory regulations and labels for computer energy consumption across the globe. These initiatives typically focus on applying fixed limits on the energy consumed in different power states such as “sleep” and “idle”. Whilst such limits clearly have good intent, they may not be ideal for two reasons: 1. They can hinder innovation. By limiting energy consumed in sleep, technological solutions become restricted. Solutions that may encourage a device to enter sleep mode more often for example may be precluded by strict limits. 2. User behaviour is crucial. A user is more likely to switch energy save modes off if they are difficult to use or hinder in any way his use of the product. Limiting energy consumption in lower power modes may well hinder technology that makes switching between energy save states seamless and invisible to the user. It is therefore clear that a more holistic approach is required.

The Computer vs. the Fridge and the Motorbike Many people argue that white goods manufacturers have had energy labels for many years and so why doesn’t the ICT industry have similar labels? Let’s take a look at a fridge as one example. How do you measure its energy efficiency? A fridge typically has just one function in life (internet linked fridges are not yet widely available) – get cold and keep cold. So a simple way to measure its energy efficiency could be as described in Figure #1. Figure #1 Of course you could introduce a duty cycle complication on how often the door is opened, which may well vary based on demographics, but the concept remains relatively simple.

A computer on the other hand is infinitely more complicated. The performance (e.g. how fast it does the work) of the machine will vary, the capabilities will also vary considerably and the application of such a device is many orders of magnitude greater than that of a fridge. A great analogy for this concept is the car and motorbike. We can talk about the vehicles energy consumption in litres / kilometre. However, is this enough? There is the concept of performance – how fast it accelerates and what is its top speed. Additionally, a vehicle has a

662

range of capabilities, for example, the number of seats, air conditioning, electric windows, heated seats etc. So what of fitness for purpose; a motorbike with a small engine is very fuel efficient, but would it be of any use to a family of 5 or a salesperson (pizza sales aside)? Which vehicle has the highest quality: a hybrid energy efficient family saloon or a Land Rover when driving off road? The point is most people buy the machine that fits their purpose. Of course some people buy a 4 x 4 and use it solely for the school run in already congested cities – however, they have the data and this is called informed choice. Finally, whilst our highways have speed limits, the information super highway has no such limits making performance more relevant to ICT products than to the automobile industry.

Measuring the Energy Efficient Performance of a Computer Five key factors must be considered in making this measure. The first three factors are the intersection of the Performance, Capabilities and Energy Consumption of the device as described in figure 2. Figure #2 Then the classification of the device must be considered – e.g. is it a notebook or a desktop, was it designed primarily as an office productivity device or a rich media machine? Finally, the duty cycle must be taken into account. For example, how long is the machine doing work and how long is it in energy save mode? There are two key vectors to be taken into account when measuring the energy consumption of a computer, a POWER vector and a TIME vector. The historical emphasis has tended to focus only on the POWER vector – e.g. capping power consumption in sleep mode. The Ecma methodology will place more emphasis on the TIME vector – e.g. how long a device is in idle or sleep mode. Figure #3

663

Which one is more energy efficient? Hint: Add up the total area of the graphs

Power

Today

Max power regulated

Future Metric on the total energy consumed in all states, adjusted for performance

active Max power regulated

off

idle

active idle

sleep

sleep

off

Time Note: This is simplified conceptual example and represents a sum of activities in each state

Figure 3 shows an over-simplified representation of the energy consumed by two different computers over time. The left computer is the result of today’s regulations focusing on maximum power in sleep and off states. The right graph shows a hypothetical scenario whereby the capability of the computer during its sleep state has been increased, thereby allowing the computer to move out of idle more often. Not surprisingly, in this scenario, much less energy overall is consumed by the computer. Therefore, it can be seen that a focus on the overall energy consumption of the computer and not on setting limits for each power state, will benefit all constituents (innovation, the user and the environment).

Ecma TC38-TG2 Membership In order to ensure maximum transparency and wide adoption of the work, Ecma has agreed to the participation of a number of non-Ecma member companies and organizations. The Task Group currently includes industry participation from Ecma members IBM, HP, Apple, Microsoft, Toshiba, Lexmark, Ricoh, Sony and Intel, and non-Ecma members AMD, Dell, nVidia, VIA Technologies, Novell and ATI. At a regulatory consultant level, Ecos, ICF, Terra Novum, LBNL and the UK Market Transformation program also participate. The US Environmental Protection Agency (EPA) has been consulted from the outset. The objective is for the work to support the “Tier 2” Energy Star® criteria to be enforced in 2009 (see milestones toward the end of this paper). However, the Standard is being designed for general use, with Energy Star being one potential user. The European Commission and the China Standardisation Centre are also monitoring the work of the Task Group. The Task Group meets every two weeks via teleconference and on an ad-hoc basis for face to face meetings.

Summary of the Ecma Standard The initial focus of the Ecma work is at a system level for both desktop and notebook (AC mode) computers. The Ecma standard, currently under development, will provide a consistent methodology and reporting structure for the measurement of the Energy Efficient Performance (EEP) of these devices. To support the standard, a benchmark company (BAPCo) is developing a benchmark suite called EECoMark™. There are three key components to the standard:

664

x x x

The Unit Under Test (UUT) workload classification: E.G. Rich Media or Office Productivity The Compute performance and Capability measurement Measurement of the energy consumption of a defined workload / duty cycle per classification. Figure #4

1. Select the appropriate Benchmark tool (Annex A)

4. Select the appropriate meter (Annex B)

2. Decide on the Classification of the UUT

5. Set up the test (Annex C)

3. Decide on the duty cycle (Annex E)

6. Record the results (Annex D)

Figure 4 describes the standard flow. First, the user will select the appropriate benchmark suite. Initially this will be the BAPCo EECoMark suite, however over time additional benchmarks meeting the requirements of the standard may well be developed by other vendors. Secondly, the user decides on the classification of the UUT as noted in the bullets above. Next, the user decides on the duty cycle he wishes to apply to the machine. Examples will be provided in an annex of the standard, but it is down to the user to decide, based on data, how long the machine will be in active mode, a low power state, switched off etc. This data will be required in order to measure the energy consumption of the machine. The user will then be required to select an appropriate meter. The standard will provide the specifications such a meter must meet. Once a meter has been selected and made available, the test will need to be set up. This is described in one of the annexes of the standard. Finally, after running all tests, the user will record the results in a standardised form as described in an annex. The results will give details about the system (product code, manufacturer, etc), the name and version of the benchmark suite used, the version of the OS installed, etc. It will then report the performance score as defined by the benchmark suite the energy consumed and the capabilities of the system. By reporting the results in a standardised manner, a customer of the product will be able to make an informed choice. For example, he may look at two machines, both with exactly the same performance, but one with capabilities he may well desire whilst the other has lower energy consumption.

Data Collection challenges To measure the overall energy consumption of a computer, some assumptions on duty cycles must be made. The assumptions used are down to the user to decide. However, in support of Energy Star Tier 2, the EPA has developed a data collection programme to help them decide on a typical duty cycle to be applied to specific machines in alternate use scenarios. The challenges associated with the accurate collection of this data cannot be underestimated and to this end, the US EPA has recently requested industry and other sector help on collecting this data.

665

Key Definitions used in the standard It is important to agree on key definitions at a very early stage in the development of a standard. Otherwise, participants and observers fail to speak the same language and missunderstandings occur. The following definitions are currently part of the Ecma standard: Energy Consumption: The amount of energy consumed by an UUT measured from the power source over a given period of time and measured in kWh Performance: The amount of useful work accomplished by the UUT compared to time Capabilities: Is a feature or set of features that enhances usability and/or experience of a Product Duty Cycle: The annual time per activity state (active, idle, sleep, off) UUT: Unit Under Test Workload: A set of activities performed by the UUT during the active state. Productivity workload: For office (home or business) applications such as word processing, internet access, accounting, etc. Media Rich workload: For entertainment purposes such as listening to music, watching videos, playing games, editing audio, pictures, video, etc.

Key Milestones and users of the standard The development of the standard and supporting software suite is linked directly to the implementation of the US EPA Energy Star Tier 2 program which will come into force in July 2009. The key milestones are therefore: x x

June 2008: The Ecma standard completed June 2008: BAPCo issue the first release of their benchmark suite. This release will support Productivity and Rich Media workloads on Windows Vista and Mac O/S desktop and notebook computers.

Whilst the key customer of the work is initially the US EPA, the methodology is not targeted only at this user. It could be used in support of other regulatory activities for example or used by companies about to make a large (in volume) purchase of computers – by using this methodology the customer could make a more informed choice as to the performance and capabilities he desires in a computer vs. the energy costs associated with the decision. Due to the cost of test hardware such as the power meter, it is not envisaged that this methodology would be used by individuals or for small purchases; rather such individuals will use the results as reported by the OEM to help make an informed choice.

Conclusions Moving from setting arbitrary limits on power consumption to a more holistic model is good for innovation, good for the customer and good for the environment. The development of the Ecma methodology together with the supporting benchmark suites will provide the necessary tools to enable the ICT industry to give clear comparable data on energy consumption, performance and capabilities of a given product, thereby allowing the user to make an informed choice when deciding which machine to purchase.

666

Qualitative Analysis of Power Distribution Configurations for Data Centers By The Green Grid

Executive Summary Many different power distribution configurations exist today that can be used to power a data center. Each of these configurations has its own advantages and disadvantages that can have a major impact on all aspects of the facility. This paper discusses the qualitative differences between seven possible configurations that can either be found in the United States or Canada today, or could be used in the future.

Introduction This paper will examine a variety of power distribution configurations for data center applications and compare the advantages, disadvantages and the future outlook of each configuration. The emphasis for the analysis will be on energy efficiency. The discussion here is strictly qualitative; any quantitative discussion is beyond the scope of this document. For this paper, alternating current is abbreviated “AC”, direct current is abbreviated “DC” and volts is abbreviated “V”. The power distribution configurations discussed are: 1) 480V AC 2) 600V AC 3) 277V AC 4) 400V AC 5) 48V DC 6) 550V DC 7) 380V DC The advantages and disadvantages of each of the following attributes are discussed for the power distribution configurations: x Current usage and availability x Efficiency x Reliability x Equipment x Standardization and acceptance

Background and Assumptions The power distribution configurations are described based on what is needed to build a new data center in the United States or Canada with 480V AC or 600V AC input at the building service entrance. It focuses on the power delivery path to the compute equipment and does not include powering of other loads, e.g. the cooling plant. This paper will analyze a single power path. Many data centers utilize parallel, redundant power paths to achieve higher availability which mitigates the effect of single failure points, improving the availability of the system. The specific voltage used for each configuration is merely a nominal voltage that was chosen for the analysis purposes of this paper. The specific voltages are representative of the possible ranges that could be used; for example, configuration 6 (550V DC) could theoretically range anywhere from 500600V DC. In general, fewer conversions within a configuration will inherently result in a higher efficiency. This higher efficiency will result in electrical cost savings and will reduce cooling requirements. Based on published efficiency data and assuming best in class components are used, at 30 percent and higher loading, efficiencies of the various configurations are within 5-10 percent of each other.

667

Alternative energy sources can be connected with any system. Alternative energy sources are DC and would generally be easier to integrate into a DC system. Any alternative energy source could be used to reduce the overall load on either an AC or DC configuration. For the purposes of this paper, it is assumed that a facility will generate far less power than it consumes, resulting in no net power flow back to the grid. The following is a brief description of each of the equipment types in the distribution path: Building Entrance The building entrance, also called a service entrance, is the boundary between the data center’s electrical system and the utility grid. It is the origin for each of the configurations discussed. Input Switchgear and Distribution This encompasses all equipment needed to provide the proper electrical protection for the data center per the National Electric Code (NEC). UPS The typical uninterruptible power supply (UPS) in an AC system uses a double conversion topology that will rectify the input power from AC to DC and invert back to AC output again. The UPS contains batteries that can supply power via the inverter to the load when utility power is unavailable. This study will also briefly discuss alternate UPS technologies for certain configurations. Any equipment labeled “DC UPS/Rectifier” will have an AC voltage as an input, contain batteries and then output a DC voltage; the AC inversion will not take place. In a redundant UPS system, multiple power paths are possible. For simplicity, only the 3 phase line to line voltage is listed until the rack input. Static Transfer Switch A static transfer switch is a solid state device used by UPS systems to transfer loads between two independent AC power sources. Transformers/PDU Isolation transformers and power distribution units (PDUs) are often integrated as a single unit in many data centers and this approach is used for this analysis. A typical PDU in an AC distribution configuration contains an isolation transformer to step the AC distribution voltage down to the desired voltage. The other main function of a PDU is to house circuit breakers that are used to create multiple branch circuits from a single feeder circuit. These branch circuits then supply power to the IT equipment throughout the data center. Because PDUs are generally rated at a small fraction of the UPS’s rating, a data center typically has numerous PDUs. Note that for simplicity, only one PDU is shown in each figure. Autotransformer Because it does not supply isolation and because it has a single winding, an autotransformer is much smaller, lighter, cheaper and more efficient than a typical isolation transformer. DC/DC Converter The converter essentially does the DC work equivalent to a transformer, stepping the DC voltage down or up. Power Supply For the purposes of this analysis, it is assumed that the output voltage of all power supplies is 12V.

Distribution Configuration Analysis The distribution analysis of 480V AC, 600V AC, 277V AC, 400V AC, 48V DC, 550V DC and 380V DC begins below. Distribution Configuration 1: 480V AC

668

Building Entrance

480V AC

Input Switchgear and Distribution

480V AC

UPS

480V AC

208V/120V AC

Transformers/ PDU

Power Supply

Highlights: x This is the most common configuration in use in the United States (US) today and will be the basis for all other comparisons. Table 1 - 480V AC Configuration Advantages and Disadvantages Advantages Disadvantages This configuration is very common and Historically, this configuration has not often Current usage usage is widespread. been optimized for efficiency. and availability All system components are available off the shelf. Multiple power conversion stages between AC and DC voltages give inherent inefficiencies in the system. Because the load cannot directly utilize the High efficiency equipment is becoming distribution voltage, this configuration more and more available throughout the Efficiency requires numerous isolation transformers industry. resulting in added losses and lower efficiency. High efficiency equipment can be more expensive than “typical” efficiency equipment. Many components and higher complexity Extensive knowledge in the industry has than other configurations give a higher created the highly reliable and serviceable possibility of failures. Reliability configuration, which has been proven to Incremental increases in reliability by meet reliability objectives. adding additional equipment, such as static switches or parallel UPS systems, can be very expensive. The number and relative size of the components in this configuration requires Because of high volumes, equipment such more floor space. as PDUs, UPSs or AC circuit breakers are Paralleling AC UPSs is more complex than Equipment common, widely available and relatively paralleling DC UPSs. low cost. Transformer/PDU take up space, add to weight on the raised floor and add heat load. There are many standards already in place for this configuration. It is accepted and known by many groups Standardization involved with the data center, including but and acceptance not limited to users, consulting engineers, inspectors, architects and electrical contractors. Future outlook: Because it is so commonly used, this distribution is likely to be used well into the future. Many equipment manufactures are striving to increase efficiencies on their products and educate consumers on ways to use their products as efficiently as possible. Distribution Configuration 2: 600V AC

669

Building Entrance

600V AC

Input Switchgear and Distribution

600V AC

UPS

600V AC

Transformers/ PDU

208V/120V AC Power Supply

Highlights: x This configuration is currently used by most Canadian data centers. x There are several sites in the United States that have been using this system for many years. Table 2 - 600V AC Configuration Advantages and Disadvantages Advantages Disadvantages Current usage and This configuration is possible in the US This configuration is not commonly availability without any major changes to used in the US. infrastructure. Multiple power conversion stages between AC and DC voltages give inherent inefficiencies in the system. Because the load cannot directly utilize the distribution voltage, this configuration requires numerous isolation transformers resulting added Efficiency losses and lower efficiency. High efficiency equipment can be more expensive than “typical” efficiency equipment. Some vendors create 600V AC UPSs by utilizing isolation transformers in conjunction with a 480V AC UPS. When this is the case, UPS efficiency is significantly reduced. This configuration has the same This configuration has been proven Reliability number of components as reliable in field use. configuration 1 and will give the same probability of failures. Nearly all, if not every, manufacturer already makes equipment for this distribution. The equipment is less common and 600V AC will be slightly less cost than Equipment has smaller install base. configuration 1 because smaller Paralleling AC UPSs is more complex amounts of copper are needed. than paralleling DC UPSs. Allows building of larger data center while using common 4000 amp distribution equipment. Standardization acceptance

and

No additional product safety regulations would be needed for 600V in the United States.

Future outlook: Because 600V AC is not readily available for US electrical distribution systems and because higher efficiency alternatives exist, it is unlikely that this system will ever become popular in the US. Distribution Configuration 3: 277V AC

670

480V AC Building Entrance

Input Switchgear and Distribution

480V/ 277V AC

480V AC

PDU

UPS

277V AC

Power Supply

Highlights: x This is the US version of a 400V/230V AC configuration. Table 3 - 277V AC Configuration Advantages and Disadvantages Advantages Disadvantages This configuration is not common in data Current usage and centers. availability System efficiency should be higher than for configuration 1 because of elimination of the isolation transformers in the numerous PDUs. High efficiency equipment can be more High efficiency equipment is becoming expensive than “typical” efficiency Efficiency more and more available throughout the equipment. industry. There is an increase in efficiency when running IT equipment at 277V AC versus 208/120V AC. Incremental increases in reliability by adding additional equipment, such as Reliability static switches or parallel UPS systems, can be very expensive. Very little high volume equipment available to run on 277V AC, which No transformer is needed in this reduces purchasing options and configuration. increases cost. This configuration reduces wire size. Because most connectors and breakers Equipment At high volumes, the cost of 277V AC are only rated to 250V AC, very limited power supplies should be comparable to computing equipment is available that standard AC equivalents. can run on 277V AC. Paralleling AC UPSs is more complex than paralleling DC UPSs. The current US standards already Because 277V AC is above the 250V AC include the voltages in this configuration. rating of nearly all IT equipment, a new Standardization and Many consulting engineers and standard similar to that used in 277V acceptance contractors are familiar with this system. emergency lighting would need to be adopted within the IT industry.

Future outlook: Because 277V AC is above the 250V AC rating of nearly all IT equipment, changes would have to be implemented in the power supplies of IT equipment for this system to become practical. Higher voltage rating connectors and breakers will also be required. Because of these required changes, 277V AC systems may be less prevalent than 400V AC systems in the near term. It is believed that potential efficiency gains are comparable to other architectures. If this proves to be the case, 277V AC may become more of a mainstream offering in the future. Distribution Configuration 4: 400V AC

671

Building Entrance

480V AC

Input Switchgear and Distribution

400V/230V AC

480V AC

UPS

230V AC PDU

Power Supply

Highlights: x This system is a hybrid approach that brings most of the efficiency and all of the equipment compatibility advantages of the end-to-end 400V AC systems commonly used in the rest of the world, to the US, in a manner that is simple to retrofit into an existing 480V AC environment or deploy in a new 480V AC data center x With the correct transformer at the service entrance, this system could also be built as an end to end 400V AC system, Table 4 - 400V AC Configuration Advantages and Disadvantages Advantages Disadvantages This setup is not very widespread in the Current usage and United States, but it is very common availability outside of North America.

Efficiency

Reliability

Equipment

Standardization acceptance

and

System efficiency should be higher than configuration 1 because of elimination of the isolation transformers in the numerous PDUs. There is an increase in efficiency when running IT equipment at 230V AC versus 208/120V AC. Reliability for this configuration is comparable to or slightly better than the 208/120V AC distribution, because it has fewer components. AC power is taken directly to the rack power supply without an additional isolation transformer. All power equipment is readily available today. Computing equipment needs no changes to run from 230V AC. Configuration would be less costly than the 208/120V AC system because of the elimination of multiple isolation transformers and smaller branch circuit conductors. All US standards already apply to this configuration.

Need an autotransformer in the bypass path of the UPS. Paralleling AC UPSs is more complex than paralleling DC UPSs.

Unfamiliarly with the system may cause confusion with local building authorities.

Future outlook: Because this configuration is more efficient, more reliable and less costly than configuration 1, as well as being readily deployable, it could be expected to be widely used in the coming years. Distribution Configuration 5: 48V DC

672

480V AC Building Entrance

Input Switchgear and Distribution

480V AC

DC UPS/ Rectifier

48V DC

PDU

48V DC

Power Supply

Highlights: x This configuration has existed for many years in the worldwide telecommunications market and is very well understood in that industry. Due to its negative polarity, it is sometimes known as -48V DC. Table 5 - 48V DC Configuration Advantages and Disadvantages Advantages Disadvantages This system is well known and widely Current usage and used in the US telecom industry, as This system is not very widely used well as in several smaller data in large data centers. availability centers. The configuration should give a In extremely large plants, there is a higher efficiency than configuration 1 requirement for large copper runs; it Efficiency because one or more conversion requires “distributed” infrastructure stages are eliminated. versus a centralized infrastructure. The system has proven to be very reliable in the telecom industry. Reliability Static switches at the UPS output are eliminated, reducing reliability concerns about static switch failure. Larger gauge power cables and/or bus bars are needed to keep Equipment for a 48V DC distribution losses low, increasing configuration is widely available. cabling costs. This may make the DC power plants are generally solution impractical for large, smaller than comparable AC power centralized, high density data plants because they have fewer centers. conversion stages. In a distributed system, many The voltage used in this configuration Equipment rectifiers and battery strings are is suitable for direct backplane needed for this setup to keep the low distribution; therefore it has the voltage transport distance as short as potential to eliminate chassis level possible. power supplies. This configuration is better suited to Paralleling DC UPSs is less complex an environment where rectifiers can than paralleling AC UPSs; therefore have shorter cable runs to the the system can more easily be scaled equipment than with long cable runs with varying data center capacity. from a power room at one end of a very large facility. There are worldwide standards already in place for this distribution. There is a lot of expertise in the Standardization and Safety and grounding practices are telecom industry with this well understood. acceptance configuration; however it is not as 48V DC is considered a Separated familiar to data center operators. Extra Low Voltage (SELV), though it can still present an energy hazard. Future outlook: This is a common configuration in the telecom industry. This high efficiency configuration, which eliminates some of the intermediate conversions common in most AC systems, is available for data centers today. Concerns over the use of larger gauge cables or bus bars may be mitigated by using distributed DC systems over shorter distances. However, 48 V DC distributions may be impractical for large, centralized, high density data centers.

673

Distribution Configuration 6: 550V DC Building Entrance

480V AC

Input Switchgear and Distribution

480V AC

DC UPS/ Recifier

550V DC

550V DC

PDU

DC/DC Converter

48V DC

Power Supply

Highlights: x While this configuration concept may be new when applied to the design of data centers, 500600V DC is used in several other areas such as renewable energy generation and transportation markets. x From a standards and safety viewpoint (UL and ANSI), 550V DC is low voltage, even though 550V DC is often referred to as high voltage. Table 6 – 550V DC Configuration Advantages and Disadvantages Advantages Disadvantages Current usage and This system is not widely used today. availability When using today’s products, this This configuration should be less efficient Efficiency configuration should be more efficient than a 380V DC distribution because of than some of the 480V AC configurations. the extra DC/DC conversion step. This configuration has fewer components as a whole when compared to Reliability configuration 1, which makes this inherently more reliable. With this configuration, static switches are not needed. Multiple converters are needed to keep transporting voltages high, varying from one per row to one per rack Only limited commercial rectifiers This configuration reduces the size of the designed for this application available for main distribution buses compared to 48V today. DC, making it comparable to 480V AC, Circuit breakers, fuses and hotplug and allows long distances between the circuits may be more expensive, are Equipment power room and IT equipment. larger and have a more limited selection 48V DC input power supplies are than AC rated equipment. available today Extra floor space and heat loading space Paralleling DC UPSs is less complex than are needed because of the multiple paralleling AC UPSs. converters, when compared to other DC configurations. Current equipment is generally more expensive because of lower volumes. There is no standardization for other products affected by 550V DC power, The National Electrical Code, NFPA, UL such as connectors. Standardization and and IEC cover the requirements for AC There are no product safety standards or acceptance and DC voltages less than 600V in building electrical standards written current specifications. specifically for 500-600V DC. Unfamiliarly with the system may cause confusion with local building authorities. Future outlook: This configuration shows promise as a means to provide higher efficiency and reliability than an AC system through the reduction in components. With facility level distribution at a high voltage and because standard 48V DC power supplies already exist, this configuration could be expected to find application in data centers in the future.

674

Distribution Configuration 7: 380V DC

Building Entrance

480V AC

Input Switchgear and Distribution

480V AC

DC UPS/ Recifier

380V DC

PDU

380V DC

Power Supply

Highlights: x This is a relatively new configuration concept. Table 7 - 380V DC Configuration Advantages and Disadvantages Advantages Disadvantages Similar DC voltages have been used Current usage and successfully internal to several mainframe This system is not widely used in data computer systems, but have not been centers today. availability commonly used for building distribution. Fewer power conversion stages than Efficiency configuration 1 should give a higher overall efficiency. This configuration has fewer components as a whole when compared to other Reliability systems, which should result in higher reliability. Static switches are not needed at the output of the UPS. This configuration reduces the size of all 380V DC compatible IT equipment is not distribution buses compared to 48V DC, available today because 380V DC power making it comparable to 480V AC, and supplies have not been agency approved. allows long distances between the power However, 380V DC power supplies could room and IT equipment. easily be built. Generally DC conversion equipment will Only limited commercial rectifiers be smaller in size than AC equipment. Equipment designed for this application are available The system can more easily be scaled today. with varying (capacity) size equipment. Circuit breakers, connectors, fuses, inrush At high volumes, the cost of 380V DC circuits and hotplug circuits may be more UPSs and power supplies should be expensive, are larger and have a more comparable to AC equivalents. limited selection than AC rated Paralleling DC UPSs is less complex than equipment. paralleling AC UPSs. A standard DC voltage into the rack has not yet been defined. A standardized connector needs to be The National Electrical Code, NFPA, UL established for the power supply input. Standardization and IEC cover the requirements for AC There are no product safety standards or and acceptance and DC voltages less than 600V in current building electrical standards written specifications. specifically for 300-400V DC. Unfamiliarly with the system may cause confusion with local building authorities. Future outlook: There is a great deal of research being done on the 380V DC distribution system. While most of the equipment is not available for purchase today, the technology necessary is already available and it is likely that support equipment manufacturers could implement if the demand is there. The time it takes to implement is debatable. However, it will likely be a few years before 380V DC is a viable option for the typical data center because all power supplies for computing equipment would need to be modified, safety standards would need to be established, and electricians and end users would need to be trained.

675

Summary The common existing and proposed data center power distribution configurations have been discussed. 480V AC and 600V AC are the standard data center alternating current power distribution systems in the USA and Canada respectively, the baseline or status quo for current data centers. They are expected to continue to experience only incremental efficiency improvements. 277V AC would offer an improved efficiency, however the changes to the power supplies in IT equipment and requirement for higher voltage breakers and connectors makes the system less likely to be become prevalent. 400V AC can be implemented today, is compatible with a wide variety of power distribution and IT equipment, and has the potential to increase system efficiency. 48V DC is the standard power distribution system for the telecom industry. It offers improved efficiency over common AC systems, is available today and has a wide variety of compatible power distribution and IT equipment. However, in very large data centers with centralized distribution systems, 48V DC will have higher cable and/or bus bar costs associated with high currents and long runs. 550V DC overcomes the distribution losses associated with centralized 48V DC. It maintains good efficiency and compatibility with existing 48V DC power distribution and IT equipment. However, the 550V DC would likely not be the most efficient choice. 380V DC appears to promise the highest efficiency but will require the introduction of new products, including UPSs, and changes to IT equipment power supplies. Future publications from The Green Grid will address each configuration in a more detailed, quantitative manner.

676

European Commission EUR 24401 EN/1– Joint Research Centre – Institute for Energy Title: Proceedings of the 5th International Conference on Improving Energy Efficiency in Commercial Buildings: IEECB Focus 2008, 10-11 April 2008, volume 2 Authors: Paolo BERTOLDI, Bogdan ATANASIU Luxembourg: Office for Official Publications of the European Communities 2010 – 367 pp. – 21 x 29,7 cm EUR – Scientific and Technical Research series – ISSN 1018-5593 ISBN 978-92-79-16017-2 DOI 10.2790/21625

Abstract

This book contains the Proceedings of the 5th International Conference on Improving Energy Efficiency in Commercial Buildings - IEECB’08 which was held in Frankfurt, Germany, 10 - 11 April 2008. The IEECB’08 conference has been very successful in attracting a large international audience, representing a wide variety of stakeholders involved in policy implementation and development, research and programme implementation, investments and property management of energy efficient commercial buildings. IEECB’08 has provided a unique forum to discuss and debate the latest developments in energy and environmental impact of commercial buildings and the installed equipment and lighting. The presentations were made by the leading experts coming from virtually every corner of the world. The presentations covered policies and programmes adopted and planned in several geographical areas and countries, as well as technical and commercial advances in the dissemination and penetration of energy efficient commercial buildings.

How to obtain EU publications Our priced publications are available from EU Bookshop (http://bookshop.europa.eu), where you can place an order with the sales agent of your choice. The Publications Office has a worldwide network of sales agents. You can obtain their contact details by sending a fax to (352) 29 29-42758.

LD- NB- 24401- EN C

The mission of the JRC is to provide customer-driven scientific and technical support for the conception, development, implementation and monitoring of EU policies. As a service of the European Commission, the JRC functions as a reference centre of science and technology for the Union. Close to the policy-making process, it serves the common interest of the Member States, while being independent of special interests, whether private or national.