The 2Nd Annual TSBE Engd Conference

The 2nd Annual TSBE EngD Conference

University of Reading

Whiteknights

July 2011

Abstract Papers of 2nd TSBE EngD Conference

Held at Henley Business School, Whiteknights Campus, Reading, RG6 6UD

5th July 2011

Organised by:

Technologies for Sustainable Built Environments Centre JJ Thomson Building Whiteknights PO Box 220 Reading Berkshire RG6 6AF

No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any of the methods, products, instructions or ideas contained in the materials herewith.

Preface

This is the second Engineering Doctorate (EngD) Conference hosted by the Industrial Doctorate Centre Technologies for Sustainable Built Environments (TSBE), University of Reading. The purpose of this Conference is to offer the opportunity for the Centre’s Research Engineers (REs) to present their research findings to University academics as well as an industry audience.

These proceedings include the abstracts of all the papers which will be presented at the Conference. The papers are prepared following the standard Conference format and have been reviewed by other academics in addition to the relevant supervisors. Each paper represents the current progress in the RE’s research project and a plan for continuing the research. The full papers will be published on data sticks and distributed to the Conference

The aim of this Conference is to develop the REs technical presentation skills to expert audience, encourage debate and respond to critique and advice for developing the research to the next phase. It is hoped that these papers could then be developed for publication in international conference proceedings and learned journals in the relevant fields.

I would like to express my gratitude to all those individuals who contributed to this Conference, without their dedication and enthusiasm it would not have been possible to hold this Conference. These include the REs who worked hard to prepare the papers (some of whom have only been working on their research project only for a few months), the project supervisors (from the University and the sponsoring companies) who gave encouragement and support for their researchers, the academics who reviewed these papers, and for the sponsoring companies who initiated the research projects and provided support throughout. I would also like to acknowledge the support and enthusiasm received from the Chartered Institute of Building (CIOB) for sponsoring the Conference and to my gratitude to Mr Alan Crane, CBE, CIOB Vice President for being our Keynote Speaker. My thanks also go to the Centre staff Jenny Berger, Emma Hawkins and Georgie Watson for their dedication and hard work in organising this Conference.

Finally, I hope that all the participants will find this event stimulating and enjoyable.

Professor Hazim Awbi Conference Chair TSBE Centre Director University of Reading

iii

Alan Crane CBE FCIOB

CIOB Vice President Qualifications: C.Eng; FICE;FCIOB; FCIM; FFB

Alan has spent 40 years in the Construction Industry, and has held CEO/COO appointments with Bovis International, Travers Morgan Consulting Group and Christiani & Nielsen Group.

He has had responsibility for a wide range of major projects including the Development & Construction of Canary Wharf in London, Eurodisney, and Petronas Towers in Malaysia. He has extensive international experience, throughout Europe, Middle and Far East, Asia and Australia.

He is a Chartered Engineer and Chartered Builder and he currently operates as an independent consultant providing input and advice to Clients, Designers and Contractors on procurement, construction, buildability, scheduling, Business Performance Improvement, Partnering, Supply Chain and Value Management.

He is a past Chairman of the Construction Confederation, was a member of the Construction Industry Board, the Strategic Forum of the Construction Industry, Chairman of the Institution of Civil Engineers Management Board, Privy Council appointed member of the Architects Registration Board and of a number of Government Task Forces.

A Trustee of the Chartered Institute of Building, immediate past Chair of the Education, Qualifications, Standards & Practice Board, Alan became Vice President in June 2009. Alan is also a regular conference speaker and conference chairman, both in the UK and internationally, he is the author of a number of industry and Task Force reports and “How To” guidance documents.

Following the issue of Sir John Egan's Task Force Report ''Rethinking Construction'' in late 1998 he was asked by Government and Industry to establish and Chair the Movement for Innovation (M4i) with the task of motivating industry to radically change the way in which it operates, and to adopt the recommendations and targets of the Egan report .

In April 2002 he became Chairman of Rethinking Construction, the overarching organisation taking forward the industry change programme which involved over 5000 industry and Government organisations who are leading the way in achieving significant performance improvements based upon both process and product, with modern off site methods being a key element.

He established and chaired the group responsible for development of the UK Construction industry Key Performance Indicators and Benchmark system. With the merger of Rethinking Construction and the Construction Best Practice Programme he became a Director of the combined organisation, Constructing Excellence.

He has provided “Rethinking Construction” support to overseas Industry/Government bodies including Australia, Hong Kong, Singapore, Denmark, Sweden, Greece, South Africa, India and Chile. He has undertaken a major lecture tours in many parts of the world including China where he was advising on behalf of CIOB on improvement to the construction industry. He is a Government appointed Construction sector member of the Building Regulations

Advisory Committee, a number of working groups, and was recently appointed Chair of the Building Control Performance Standards Group. :

“Sustainability and Action – over and under use – caused by myths?”

I will explore how we, built environment professionals, Construction Managers and managers of construction all, working in a fully integrated environment, can and will make a difference. Not just the environment, but all 3 dimensions to Sustainability – or is it 4?

Whether driven by the simple fact that UK law says that we must reduce Carbon by 50% by 2025; or by the moral imperative that the world of today is not ours to treat as we please, that we have in trust and must account for to those who come after; or even just the fact that people say as an industry that we can’t!

We have the skills, we have the technical knowledge, we now have the help of the CIOB CarbonAction2050 Toolkit.

Yes we should look at what other stakeholders, such as Government, are or more likely are not doing. Green Banks and other initiatives are and will be important but as Construction Managers it is time to stop looking for direction from others – or the excuses – we can and we must take Action and I will outline some of the Actions we can take Now.

Contents

SMEs And Smart Grids: What Are The Market Realities? 1 JM Rawlings1 The Impacts Of Renewable Energy Resource Variability On Conventional Thermal Generators M. L. Kubik Development Of A Virtual Pyranometer For Solar Energy Monitoring P.A. Burgess Reducing User Influence On Energy Consumption Through Improved Building And Control Design R.M. Tetlow Dynamic Stall For A Vertical Axis Wind Turbine In A Two-Dimensional Study R. Nobile Sustainable Procurement – Challenges For Construction Practice. R.J. Belfitt A Consumption And Emissions Model Of An RTG Crane Diesel Generator C. E. Knight Challenges In Intelligent Management Of Power And Cooling Towards Sustainable Data Centre S. Luong A Methodology To Quantify The Environmental Impacts Of The Microsoft Windows Operating Systems Daniel R. Williams Multifunctional, Adaptable Facades Bridget Ogwezi Heat Demand Analysis Of Residential Development In London Connected To District Heating Scheme R. Burzynski Considering Occupants’ Comfort In Sustainable Building Refurbishment Projects Michelle Agha-Hossein A Review Of Currently Available Standards And Software Tools For Assessing Life Cycle Greenhouse Gas Emissions From Buildings. H. J. Darby Sustainable Planning In The Era Of The Localism Bill T.Mcginley Review Of Factors Affecting Uncontrolled Ventilation In Food Supermarkets S. Sawaf Bats And Breathable Roofing Membranes: Do The Mechanics Of A Membrane Affect Mechanical Stability? Stacey Waring Potential Carbon Savings Through Hot-Fill Appliances: Field Test Data Validation D.Saker

Building Services Systems: Heating And Air Conditioning Design Approaches A. Barcellos The Use Of Sustainable Travel Planning Strategies Within Remote Cities M. H. Ismai1 Methods Used For Sustainable Management Of Time, Cost And Quality Throughout The London 2012 Olympic And Paralympic Programme - Interim Findings Using A Scoping Study J. M. Grossman Modelling Building Semantics: Providing Feedback And Sustainability H. H. Shah

vii

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

SMEs And Smart Grids: What Are The Market Realities?

JM Rawlings1*, PJ Coker2, AJ Doak3 and J Wallis4

1 Technologies for Sustainable Built Environments Centre, University of Reading, UK 2 School of Construction Management and Engineering, University of Reading, UK 3 School of Real Estate and Planning, University of Reading, UK 4 Reading Borough Council, Civic Centre, Reading, UK

* Corresponding author: [email protected]

ABSTRACT Small and medium sized enterprises (SMEs) constitute the vast majority of private sector businesses. However, their huge numbers and diversity present significant communication, engagement and implementation barriers to carbon reduction initiatives. Although individual company emissions may be low, collectively this group represents a significant and largely ignored portion of the UK total. (Vickers & Vaze, 2009).

The carbon saving challenges identified by SMEs point to their lack of money and time. This is made more difficult by these businesses not always being the bill payers and usually renting their premises. A number of possible solutions are reviewed that focus on improving the building fabric. These include loans for SMEs, a scheme to pay for simple energy efficiency improvements in buildings funded through energy bills and support to improve the quality of existing buildings.

The UK Government has recently announced plans for smart meters to be installed in homes and small businesses, (DECC, 2011a). This introduces an interesting new route to engage SMEs in energy reduction through improved awareness of their usage and possible demand reduction tariffs.

Keywords: Energy saving, SMEs, smart grids, smart meters

1. INTRODUCTION Each local authority is mandated to influence organisations in their areas to reduce their CO2 emissions (DECC, n d). One practical example is The LoCus Project (Low Carbon understanding for SMEs), managed by Reading Borough Council. It aims to encourage SMEs in south east England to cut their carbon emissions. This paper looks at challenges SMEs face in reducing their CO2. It is estimated that SMEs in business premises emit around 47 MtCO2/yr (excluding transport) and that this could be reduced by 10-15% through cost-effective energy efficiency measures.

The status of SMEs with regard to energy use and especially property is discussed. This raises questions such as: To what extent are SMEs aware of their own energy consumption? What business reasons are there for SMEs to use less energy? This paper discusses a number of existing carbon reduction measures and then introduces the topic of smart meters which the UK Government expects will be installed in homes and small businesses. This makes energy use more visible, generates more accurate bills and has

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ the potential to engage energy users through network management. It is suggested that this measure may provide a new incentive for SMEs to engage in energy saving.

2. BACKGROUND

2.1 SMEs Small and medium sized enterprises (SMEs) account for over 99% of UK private sector enterprises (BIS, 2010). An SME has fewer than 250 employees, a turnover of less than €50 million or a balance sheet of less than €43 million (European Commission, 2011). There were an estimated 4.8 million SMEs in the UK in 2009, an increase of 1.1% over the previous year (BIS, 2010). This represents nearly 60% of all private sector employment and 49% of turnover. Over 90% of these companies have fewer than 5 employees. By contrast, only 0.1% of enterprises (or 5890) have 250 employees or more. 419,000 enterprises employed between 5 and 249 staff and had a turnover of £1,115 billion in 2009.

SMEs represent a hugely diverse sector, characterised by limited resources, the dominance of the owner manager, a preference for informal approaches, less influence over their operating environment and less awareness of the available support (compared with large businesses) (Vickers & Vaze, 2009, p.3). Some of the challenges facing SMEs in ‘going green’ are discussed in section 5. In order to understand SME energy use, this paper looks at their premises and then some characteristics of their energy consumption. This paper looks in particular at smart meters and how this relates to other carbon reduction initiatives.

2.2 Commercial Property

The CO2 emissions from commercial property were broadly flat from 1990 to 2008 (Committee on Climate Change, 2008; Committee on Climate Change, 2010). During 2009, indirect emissions (from electricity) fell by 14% and direct emissions (from fuel for heating) fell by 10%. This is understood to be largely due to the recession. Heating, lighting and ventilation account for 80% of the emissions related to the building fabric, (Carbon Trust, 2009).

2.3 UK Smart Meter Proposal Electricity and gas meters in all homes and small businesses will be replaced with smart meters according to The Government Implementation Plan. (DECC & Ofgem, 2011). Approximately 53 million electricity and gas meters in Great Britain will be changed in a programme scheduled to run from 2014 until 2019. Smart meters are expected to enable much greater investment in energy efficiency by providing near real-time information on energy consumption. Each meter collects data on energy use which is displayed for the benefit of the consumer and transmitted to the energy provider. This paves the way for bespoke energy advice and time of day tariffs. It replaces estimated bills and the requirement for regular meter reading with accurate bills. Smart meters are anticipated to facilitate greater competition by smoothing the process of changing suppliers. Energy providers will receive data on consumer behaviour at far higher granularity which will help them focus network investment. It also informs them immediately of supply failures. (DECC, 2011b).

The smart meter roll-out will include infrastructure to enable:

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/  Energy companies to communicate with their meters via a wide area network.  The display of energy consumption in a user accessible format.  Communication between the energy meters in the customer premises and potentially smart enabled devices and micro-generation systems.

For businesses, energy consumption data may be displayed on an in-home display (IHD) or via internet based applications (DECC, 2011b). The smart meter will communicate with smart enabled appliances, which can be switched off or delayed by signals from the meter. In a business this might consist of air-conditioning or ventilation.

3. COMMERCIAL PROPERTY

3.1 SME Property

A significant proportion of SME CO2 is unregulated by existing legislation. The Carbon Reduction Commitment (CRC), Climate Change Agreements and the EU Emissions Trading Scheme (EU ETS) control the emissions of the largest energy users. Most SMEs are below the thresholds of these measures. At the other end of the scale, at least half of all SMEs are based at home. Enterprise Nation in their survey of 7200 SMEs estimate this to be 2.8 million (Enterprise Nation, 2009)1. Their energy use is regulated by domestic energy efficiency measures (such as CERT and local authority insulation schemes) and so they are excluded from this evaluation. This paper considers businesses with 5 to 249 employees.

The culture in the [construction and development] industry is often risk-averse. (Government Office for Science 2008, p.126). This is partly because the capital is often invested by groups requiring steady growth and dividends. Almost half of UK property provides a return to UK pension funds and other personal saving schemes (British Property Federation, 2010). The largest group of owners are UK institutions (23%) responsible for long-term personal savings and pensions. Overseas investors increased their share rapidly during 2009 to 22%, although their ownership tends to ebb and flow in relation to global patterns of economic activity. The prestige end of the property market (occupied by large ‘blue chip’ companies) tends to be dominated by financial institutions. This leaves the public sector and smaller private sector landlords to own (higher risk) properties that SMEs occupy (Scrase, 2001).

Most commercial property (in the UK) is rented and the leases are becoming shorter. 62% of commercial property is rented (British Property Federation, 2010). The average length of all leases is becoming shorter: 5.9 years in 2008, compared with 8.7 years in 1999. Two thirds of new leases are for no longer than five years. New leases in 2007 were shorter for SMEs than for large companies (5.4 years compared with 8.7 years). SMEs in serviced accommodation usually occupy their space for less than one year. The typical license length was 7 months in Greater London in 2009 (OfficeBroker.com, 2010).

One reason that SME CO2 remains high is the complex interaction of commercial landlord, agent and tenant (Carbon Trust, 2009). This often creates a ‘circle of inertia’:

1 Databuild in their report for DECC estimate that there are 2 million SMEs in domestic premises, based on 400 interviews with SMEs in England in 2010 (Fawcett, 2010).

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ if the landlord were to invest in the building, the tenant would benefit but the landlord would not be able to recover the cost under a standard lease. Agents, along with building designers have an interest in over-specifying the building as this maximises their commission (Scrase, 2001).

3.2 SME Energy Consumption It is difficult to estimate the energy consumed by SMEs. No energy data organised by sector and size is publicly available. Research for the National Energy Efficiency Data Framework found problems in correlating energy meters with business premises (Bruhns & Wyatt, 2011).

SME energy bills are often low, so cost-effective energy savings are small. For many SMEs, especially those based in offices, energy represents only 1-2% of costs (The Carbon Trust, 2009, p.9). In some cases, energy spend is less than 1%. This was the case for between a quarter and a third of businesses surveyed in South East England in 2006 (BusinessLink, 2006). The Federation of Small Businesses (FSB) argue that SMEs have the potential to save 20% of their energy, compared with around 8% for large companies (Wood & Caro, 2010, p.4, citing Carbon Trust research). However this was based on data from companies that spend at least £50k/yr on energy. Defra evaluated the impact on SMEs of saving 15% on energy bills. Allowing £500/day for staff costs to implement savings, they found that enterprises with fewer than 10 staff, would save the equivalent of 0.2 person days per year. This was based on a typical energy spend of £600/yr. For enterprises with 11 to 50 employees, saving 15% was equivalent to 1.4 person days per year, (Defra 2006, p.13).

Some SMEs pay for energy indirectly (such as through the rent or service charge). A survey of 400 SMEs (published by DECC) found that 14% of sites pay for both gas and electricity indirectly (Fawcett, 2010). 43% of sites pay electricity (and not gas) directly to their supplier.

The Carbon Trust suggests that cost effective carbon savings of 15% can be achieved in existing buildings. For non- domestic buildings (excluding industrial process emissions), this was estimated to be 26 MtCO2/yr in 2005 (CTC765, p.87). This can be divided into businesses affected by the CRC and those not and buildings that will receive major interventions (construction and retrofit) and those in use. Each of these categories is roughly 6-7 MtCO2, represented by the quadrants in figure 1. Most SMEs occupy quadrant B, which is largely unregulated Source: Carbon Trust by carbon reduction policies. Figure 1 Relationship of Existing Policies to Potential Carbon Reduction

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 4. CARBON REDUCTION SOLUTIONS

A number of policies aim to reduce CO2 emissions from business but these have limited relevance to SMEs. Table 1 lists some existing and planned incentives: (Carbon Trust 2009; DECC 2011b; Carbon Trust n.d.; HMRC, 2010). Both buildings and occupants must be targeted. Some organisations have used ‘Green Leases’ whereby the landlord and tenant agree mutual obligations (such as sharing of energy data) in order to minimise the environmental impact of the building (Pinsent Masons, 2007). Existing Buildings – Refurbishment and New Build Building Regulations Part L specifies minimum standards for new build or renovation. DECs Identifies the actual energy consumed in public buildings during the previous year. EPCs Shows the expected energy consumption of the building.

SMEs and Landlords CRC Mandatory cap for large energy users. Only likely to affect SME landlords. EU Emissions Mandatory ‘cap and trade’ scheme affecting energy intensive Trading Scheme industries (such as electricity generation, vehicle production and glass making). Climate Change Levy A tax on business energy from fossil fuel sources. Charge is levied by supplier. Domestic supplies are excluded. Carbon Trust loans Interest-free loans, repaid through energy savings over four years. Feed-in Tariff Payment for electricity generated by small renewable systems. Green Deal Planned measure to allow businesses to invest in energy efficiency measures and re-pay through their energy bills. Enhanced Capital Tax break for investing in energy efficiency measures. Allowances

Table 1 Carbon Reduction Policies Affecting SMEs and Landlords

A new opportunity arises with the proposal for smart meters and smart grids, which are discussed in section 6.

5. CHALLENGES FACING SMES IN CUTTING CARBON EMISSIONS A number of reports analyse the challenges faced by small businesses on the path to greater environmental responsibility. These cover a broad range of topics, including social responsibility, waste reduction and energy efficiency. Table 2 lists the top four barriers given by small businesses in three surveys and shows that cost was the top factor (Lloyds Bank, 2010), (Connell, 2007), (National Audit Office, 2007).

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Lloyds Bank FSB NAO Cost 1 1 1 Lack of time 2 3 2 Impact of recession 4 Business is too small 2 4 Bureaucracy 4 Funding not available 3 3 539 SMEs with 1700 FSB 580 recipients of turnover up to members CT loans Sample £15M Publication date October 2010 December 2007 November 2007

Table 2 Summary of Top Barriers to Energy Saving

Lack of time and ‘business is too small’ (interpreted to mean: ‘lack of resources’) were high on the list. This also means that an owner/ manager considers that their impact is so small that there is no benefit to try to reduce energy as they are already a very light user. FSB noted that environmental legislation does not allow for the huge variation in SMEs and consequently imposes an unnecessarily large burden. 40% of the respondents in their survey had implemented at least one energy efficiency measure. The National Audit Office (NAO) survey of companies who had received Carbon Trust loans, is likely to be biased towards larger organisations and the public sector. Lloyds Bank found that only 38% of SMEs had sought to analyse their environmental risks.

Vickers and Vaze note the following characteristics of SMEs that are likely to affect any change (Vickers & Vaze 2009, p.11):  Limited resources in terms of finance, access to capital, knowledge, skills.  The dominance of the owner manager which can restrict the development of the business.  Preference for informal styles of management.  Less ability to influence their operating environment compared with larger businesses.  Low levels of awareness of available advice and a reluctance to access it.

They add: the sheer numbers and “low visibility” of smaller enterprises pose a particular challenge for those who wish to assist or otherwise have a positive influence on their growth and behaviour. (Vickers & Vaze 2009, p.12).

6. SMART METERS IN THE UK The smart meter roll-out will affect all businesses with unrestricted electricity meters (profile classes 3 and 4) or gas consumption of less than 732 MWh/yr (DECC, 2011a; Elexon, 2008). DECC estimates that the number of non-domestic meters affected will be: electricity- 2.14 million and gas- 1.5 million (DECC, 2011a, p.19). The estimated usage through each meter is 17,400 kWh of electricity and 79,800 kWh of gas. Assuming that all of these are supplying SMEs, SME emissions are 47.2 MtCO2e/yr. This excludes home-based SMEs. Databuild estimate that the SME sector contributes 20 – 40 MtCO2/yr, (Fawcett, 2010). This wide variation is indicative of the lack of data. A report by AEA for DECC proposes 49 MtCO2/yr based on a survey of 400 companies

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ that are not ‘constrained’ by existing policies such as the Emissions Trading Scheme (AEA, 2010).

The Government proposes that the smart meter roll-out will facilitate the development of smart grids which will enable the energy companies to manage the (electricity) distribution system more efficiently. According to The European Technology Platform SmartGrids, these are: electricity networks that can intelligently integrate the behaviour and actions of all Source: www.smartgrids.eu users connected to it - Figure 2 Elements of a Smart Grid generators, consumers and those that do both – in order to efficiently deliver sustainable, economic and secure electricity supplies. (“What is a Smart Grid?,” 2006). A smart grid is a collection of innovative technologies that will create a more responsive and integrated electricity network. Figure 2 indicates the relationship between smart meters and smart grids. Several other pressures on electricity supply companies account for their interest in smart grids:  An increasing proportion of electricity will come from variable sources (such as wind and solar). (At least 30% by 2020 in the UK Renewable Energy Strategy) (DECC, 2009).  Nearly 20 GW of generating capacity will be de-commissioned by 2020 (Dale, n d).  Cutting the CO2 intensity of electricity early on may encourage the take-up of electric heating (using heat pumps).  Electric vehicles could support this process, if they become popular.

The high voltage transmission network is actively managed but the lower voltage distribution network is a passive channel. It is thought that smart grids will integrate all parts of the network to enable active management of the customer and distribution network layers.

It is not yet understood the extent to which smart meters will facilitate engagement with SMEs in energy reduction. Darby looks at the scope for smart meters to engage house- holders and notes interest in simple messages about energy costs and direct comparisons over time (Darby, 2010). She argues that more effort is required to define what information is offered to the end user and in what format. For smart meters to achieve demand reduction, will require a clear focus and customer support.

7. DISCUSSION

Local authorities are required to strive for CO2 reduction in their areas, encompassing all organisations. Since the vast majority of businesses are SMEs, they have to contribute to carbon reduction and energy saving. Their sheer numbers mean they cannot be ignored but their diversity and small size require a variety of approaches.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Estimating the CO2 emissions from SMEs is difficult because of the lack of published data. One approach is to identify the data relating to business premises. Commercial property CO2 fell slightly during 2009, most likely due to the recession. Roughly half of SMEs are based at home, so their business CO2 is never considered. Individually this will be small though collectively this could be significant. Some SMEs occupy premises where the price of energy is not visible. This includes enterprises in serviced accommodation where the license period is typically less than one year. The trend is for commercial leases, especially those for SMEs, to shorten. This reduces any incentive for tenants or landlords to invest in fabric-related energy saving measures. Property owners want low risk investments that provide a steady return to pension or other investment funds.

The research described above highlights the energy saving potential for SMEs. For many, especially those in the service sector, the energy that could be readily saved, is a tiny sum. This does not compare well with the staff costs required, especially where many actions have to be repeated daily (such as switching off lights and computers). SMEs themselves recognise cost and lack of resources as key constraints in the carbon saving journey.

There are several existing carbon saving incentives. To be effective, these need to address the buildings and the tenant SMEs. As figure 1 shows, SMEs are largely unaffected by the CRC and Building Regulations.

Smart meters present new opportunities for SMEs to engage in cutting CO2 emissions. This is made possible through increased awareness of energy usage, time of day tariffs, possible financial incentives to reduce loads at peak times and measurement of renewable electricity. One example is Scottish and Southern Energy’s (SSE) Thames Valley Vision Project which aims to: revolutionise electricity distribution networks in and around Bracknell. (SSE, 2010). Renewable energy systems and smart grid infrastructure will provide more data on electricity flows and engage businesses and households.

8. CONCLUSION SME energy consumption is collectively significant but for most companies is low. Many SMEs are largely unaware of their energy costs either because they do not directly pay for it or because it is small. The cost effective energy saving for a small company (calculated in £) equates to a very small amount of human resource. Conflicting interests between landlord and tenant creates a ‘circle of inertia’ which prevents building refurbishment.

A number of possible solutions are reviewed that focus on improving the fabric of commercial buildings. These include loans for SMEs, a scheme to pay for simple energy efficiency improvements in buildings funded through energy bills and support to improve the quality of existing buildings.

A further solution may arise with the delivery of smart grids. This offers the potential for SMEs to engage in energy and CO2 reduction through greater energy awareness and possible demand side tariffs to persuade consumers to switch off loads at peak times.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 9. REFERENCES AEA. (2010). Assessing the Carbon dioxide Emissions and Cost-effective Carbon Savings Potential for Organisations not Covered by EU ETS , CCAs or CRC (CESA 0903). AEA: Harwell. BIS. (2010). SME Statistics for the UK and Regions 2009. Sheffield: BIS. Retrieved from http://stats.bis.gov.uk/ed/sme. British Property Federation. (2010). Property Data Report. London. Retrieved from www.bpf.org.uk Bruhns, H., & Wyatt, P. (2011). Constructing a Data Framework for Measuring the Energy Consumption of the Non-domestic Building stock. Reading: University of Reading. BusinessLink. (2006). Fuel, Energy and Water Costs. South East Business Monitor, (2), 12. Carbon Trust. (2009). Building the Future, Today. London: Carbon Trust. ref: CTC765. Carbon Trust. (n.d.). Enhanced Capital Allowances. Retrieved May 6, 2011, from www.eca.gov.uk. Committee on Climate Change. (2008). Building a low-carbon economy – The UK’s contribution to tackling climate change. London: The Stationery Office. Retrieved from www.theccc.org.uk/pdf/TSO-ClimateChange.pdf Committee on Climate Change. (2010). Progress Reducing Emissions from Buildings and Industry. Meeting Carbon Budgets – ensuring a low-carbon recovery. London: Committee on Climate Change. Retrieved from www.theccc.org.uk/reports/2nd-progress-report Connell, N. (2007). Social and Environmental Responsibility and the Small Business Owner. London: Federation of Small Businesses. Retrieved from www.fsb.org.uk/policy/archivepubs. Dale, L. (n.d.). Access & Network Investment. Business. London: National Grid. Retrieved from www.cigre-uk.org/Events_files/joint09/33.pdf. Darby, S. (2010). Smart metering: what potential for householder engagement? Building Research & Information, 38(5), 442-457. doi: 10.1080/09613218.2010.492660. DECC. (2009). The UK Renewable Energy Strategy. Incentive (1st ed., p. 238). London: TSO. Retrieved from www.decc.gov.uk/ DECC. (2011a). Smart Metering Implementation Programme: Overview Document (p. 64). London. URN 11D/676. DECC. (2011b). Smart meter Rollout for the Small and Medium Non-domestic Sector: Impact Assessment. London: DECC. doi: DECC0010. DECC. (2011c). The Green Deal. Retrieved May 6, 2011, from www.decc.gov.uk/en/content/cms/what_we_do/consumers/green_deal/green_deal.aspx. DECC. (n.d.). NI 186: Per capita reduction in CO2 emissions in the Local Authority area. Retrieved May 9, 2011, from www.decc.gov.uk/en/content/cms/statistics/indicators/ni186/ni186.aspx. DECC & Ofgem. (2011). Smart Metering Implementation Programme. London: DECC & Ofgem. Defra. (2006). Policy Options to Encourage Energy Efficiency in the SME and Public Sectors. London: Defra. Retrieved from www.decc.gov.uk Elexon. (2008). Allocation of Profile Classes & SSCs for Non-half hourly SVA Metering Systems Registered in SMRS. Retrieved from www.elexon.co.uk/pages/bscps.aspx.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Enterprise Nation. (2009). Home Business Report 2009. Retrieved April 19, 2011, from www.scribd.com/doc/43355563/Home-Business-Report-2009. European Commission. (2011). SME Definition. Retrieved May 2, 2011, from http://ec.europa.eu/enterprise/policies/sme/facts-figures-analysis/sme- definition/index_en.htm. Fawcett, J. (2010). Department of Energy and Climate Change: Unconstrained Sector Research. London: DECC. Retrieved from www.decc.gov.uk Government Office for Science. (2008). Powering our Lives: Sustainable Energy Management and the Built Environment. London. URN 140-08-Fo/b. HMRC. (2010). A General Guide to Climate Change Levy. Ref: CCL1. Lloyds Bank. (2010). Small Businesses Suffering “Environmental Inertia”. London: Lloyds TSB Bank plc. National Audit Office. (2007). The Carbon Trust Accelerating the Move to a Low Carbon Economy. London. Retrieved from www.nao.org.uk OfficeBroker.com. (2010). Serviced Office Review: Greater London. Available at: www.officebroker.com. [Accessed May 2011] Pinsent Masons. (2007). The Development of Green Leases. London: Pinsent Masons. Retrieved from www.pinsentmasons.com/PDF/PinsentMasonsSustainabilityToolkit_Part-3.pdf. Scrase, J. I. (2001). Curbing the growth in UK commercial energy consumption. Building Research & Information, 29(1), 51-61. doi: 10.1080/09613210010001150. SSE. (2010). Funding for Low-carbon Electricity Networks. Retrieved May 4, 2011, from www.sse.com/Templates/NewsAndPressReleases.aspx?id=6702. Vickers, I., & Vaze, P. (2009). SMEs in a Low Carbon Economy. London: BERR. URN 09/574. What is a Smart Grid? (2006). Retrieved April 18, 2011, from www.smartgrids.eu/?q=node/163. Wood, F., & Caro, D. (2010). Making Sense of Going Green. London. Retrieved from www.fsb.org.uk.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

The Impacts Of Renewable Energy Resource Variability On Conventional Thermal Generators

1* 2 3 4 M. L. Kubik , P. J. Coker , C. Hunt and H. B. Awbi 1 Technologies for Sustainable Built Environments Centre, University of Reading, United Kingdom 2 School of Construction Management and Engineering, University of Reading, United Kingdom 3 AES, Richmond upon Thames, United Kingdom 4 Technologies for Sustainable Built Environments Centre, University of Reading, United Kingdom

* Corresponding author: [email protected]

ABSTRACT The Republic of Ireland and UK governments have put forward an ambitious target of 40% of electricity generation to be supplied with renewable sources by 2020. The dominant source of this energy is anticipated to come from wind power, as this is the most mature renewable technology. However, wind generation is inherently variable in its output, and this introduces significant challenges for the System Operator when balancing supply and demand. Although demand side management, energy storage and greater interconnection are all anticipated to help with dealing with the challenge of variability, conventional thermal generators will have a very significant role to play in balancing supply and demand.

Running conventional generation more flexibly in order to cater for a wind led regime reduces the efficiency of the plant, as well as shortening its lifespan and increasing O&M costs. The link between variability and the impacts on conventional generation is not well addressed in current literature, but is of vital importance for informing the development of the generation mix. This paper introduces some of the potential impacts of greater variability on conventional generators, the past work that has gone into modelling these impacts and identifies areas of future work that need to be addressed.

Keywords: Variability, intermittency, balancing, conventional generation, Ireland

1. INTRODUCTION

Recognising a global consensus of the need to limit future carbon emissions, the Irish and UK governments, along with other EU-27 member states, agreed in 2008 an EU Climate and Energy Package (European Commission 2008). In contributing towards this legislation, the UK and Irish governments have set an ambitious target of 40% of annual electricity consumption to be met by renewable sources by 2020 within the Irish all island electricity market2. Ireland is currently heavily dependent on conventional

2 Northern Ireland and the Republic of Ireland have a single common electricity market for wholesale electricity.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ fossil fuels (Howley et al. 2009), but is amongst the most gifted in Europe in terms of renewable wind, wave and tidal resource (Rourke et al. 2009). Energy forecasts by Walker et al. (2009) suggest that under the legislation set out by Ireland’s 2007 Energy White Paper and subsequent energy targets the single largest renewable will be from wind (making up 65% of the 40% target alone). Although wind generation output is to an extent predictable, wind is inherently a variable resource, and this presents additional challenges for system balancing (Laughton 2007). There are a number of technological developments available to help address the impacts of a more variable generation pattern, but conventional generation plant is expected to play a significant role in smoothing out the generation profile when there is a shortfall of wind by running more flexibly. As many of the existing thermal generators will still be operational in 2020, it is particularly important to understand the impacts of a wind led regime on their performance. Such an understanding will inform the roadmap towards the integration of the levels of wind required to meet the ambitious goals set by the government. This paper introduces the current electricity market regime and the characteristics of conventional thermal generation. The specific challenges of variability for conventional thermal generation in Northern Ireland are highlighted, a research area that existing literature does not address. A need for more research into the impacts of variability on the operation of conventional plant is identified and proposed as a future direction for research.

Figure 1 - Schematic of Northern Ireland power stations and transmission network, adapted from Kennedy (2007).

2. BACKGROUND

2.1. THE IRISH ALL ISLAND ELECTRICITY MARKET

Since November 2007, a single electricity market (SEM) has operated for the whole island of Ireland, combining the two previously separate Northern Ireland and Republic

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ of Ireland systems. The SEM consists of a gross mandatory electricity pool, which all generators bid into (Pöyry 2007). All generators that make themselves available to the system are paid a capacity payment, designed to cover the capital costs of constructing the unit in the first place. The system operators, SONI for Northern Ireland and EirGrid for the Republic of Ireland, select the most cost effective plant3 to satisfy an unconstrained schedule of demand from this pool based upon the operating characteristics submitted by the generator units in the market. All such merit order units are paid a system marginal price (SMP), determined by the cost of the most expensive (marginal) unit required to meet demand. System constraints, such as plants being non-operational for maintenance, the need for voltage regulation, or limits to the amount of electricity that can be carried by certain transmission lines, are then imposed and the necessary changes are made to construct a constrained schedule of demand. The key difference is that generators asked to run due to system constraints instead of on merit are only reimbursed their operating costs, and make no operating profit. This already presents a challenge for operators in Northern Ireland, as although they are part of a whole island market, there is limited interconnection across the border into the Republic of Ireland (Figure ). This bottleneck prevents many merit order plant in the Republic of Ireland from satisfying demand in Northern Ireland. Further challenges emerge with the introduction of more variable generation; these will be addressed later in this paper (see Figure ).

2.2. CONVENTIONAL THERMAL GENERATION

In conventional power stations, mechanical power is produced by a heat engine that transforms thermal energy, obtained from the combustion of a fuel, into kinetic energy. This kinetic energy is used to drive a generator and produce electricity that can be exported for use elsewhere. Thermal power plants are normally classified by their prime mover (usually a gas or steam turbine, or a combined cycle of both) and their fuel source (e.g. nuclear, fossil fuel, geothermal, biomass). Although there are basic similarities between many variants of conventional generation, and generally their performance is based upon similar characteristics, there are also some important differences which influence their operation, efficiency, flexibility and cost. This section of the paper identifies the key types of conventional generation and their characteristics. Error! Reference source not found. summarises the main conventional generation plant in Northern Ireland.

3 The term “unit” and “plant” are used interchangeably in this paper; however, strictly speaking a power plant may be made up of multiple generator units. The difference is clarified in Table 1.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Table 1 - Summary of conventional generation capacity in Northern Ireland

Location Plant typei Fuel Units Capacity/unit MSGiii/unit Ballylumford Steam plant Gas 3 180MW 60MW Single shaft CCGT Gas/oil 1+1 100MW (combined) 65MW CCGT/OCGTii Gas/oil 2+1 2x160MW + 2x68MW 180MW +113MW OCGT (quad aero Oil 2 58MW derivative) Coolkeeragh CCGT Gas/oil 1+1 240MW (combined) Kilroot Steam plant Coal/oil 2 220/260MW 110/70MW OCGT (twin aero Oil 2 29MW derivative) OCGT Oil 2 44MW i. CCGT is an acronym for combined cycle gas turbine, OCGT for open cycle gas turbine. ii. Ballylumford’s 2+1 CCGT was designed to operate flexibly with or without the HRSG units. iii. Minimum stable generation (not applicable to OCGT as these only run at peak loads).

2.2.1. STEAM POWER PLANTS

In these power plants a steam turbine is used to produce electricity, based upon the thermodynamic Rankine cycle. The basic principle of a subcritical steam plant is shown in Figure (a); fuel is combusted in a boiler, the heat is used to turn water into steam, which drives a series of turbines “T”, before the steam is condensed back into water and the cycle is completed by pumping the water back to the boiler. Despite the simplicity of its basic operation, a large number of auxiliary systems are required in order to deliver the fuel, maximise the efficiency of this process and clean up the exhaust emissions. The exact set up of a power plant depends on the nature of the fuel it uses; both in terms of optimisation of design and the equipment required.

For example, a pulverised coal4 power plant requires a stockpile to store the coal, a conveyer mechanism to deliver coal to the plant, mills to grind the coal into a fine dust before injecting it into the boiler, electrostatic precipitators to remove particulates from the flue gas and a facility to dispose of the ash. This list is by no means exhaustive, but serves to illustrate that although the core operation of a steam plant is identical, the additional processes to facilitate this with a certain type of fuel are complex. Similarly, the boiler in a coal fired power plant has a different optimal size and geometry to that of oil or gas fired boiler, so while a plant can be designed to run flexibly on multiple fuels (desirable from a security of supply as well as fuel pricing point of view) it is at the expense of efficiency and operating costs.

4 This is further complicated if different types of coal are considered; for example, lignite coal produces very large quantities of ash compared to bituminous coal, or if different types of coal combustion are used; such as Circulating Fluidised Beds (CFB) rather than Pulverised Coal (PC). To keep this paper focused, only the technologies present in Northern Ireland are discussed.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Water Condenser Combustion Exhaust Chamber gas

T Fuel C T Fuel Pump Steam Boiler Air inlet (a) (b)

Figure 2 - Simplified schematic of (a) A steam plant (b) An open cycle gas turbine.

2.2.2. GAS TURBINE POWER PLANTS

Gas turbines operate under the Brayton Cycle. A simplified Open Cycle Gas Turbine (OCGT) schematic is shown in Figure (b). Air is drawn into the compressor “C” and compressed, combined with a fuel and ignited. The combustion passes through a gas turbine “T” that is used to drive a generator and produce an electrical output. Modern OCGTs can be more efficient than older steam plant, are very reliable and are able to respond very quickly to a need for generation, taking only minutes to be brought to full load from an off state, and can even be controlled autonomously without anyone manning the plant. However, OCGTs generally have quite high running costs, hence their use is primarily as peaking plant on an electricity system; turning on only for short periods to meet spikes in demand. Furthermore, their performance is influenced by the weather, as the ambient temperature and pressure influences their efficiency and output capacity. The most flexible gas turbines are aeroderivative; these are based on jet engine designs which can handle load changes even faster than industrial OCGT machines but do this at the expense of efficiency and operating costs. A wide variety of fuels can be used to power a gas turbine. Natural gas is commonly used in land-based gas turbines while light oil distillates (e.g. kerosene) can be used as an alternative fuel and to power aero derivative gas turbines. Diesel oil or specially treated residual oils can also be used, as well as combustible gases derived from blast furnaces, refineries and the gasification of solid fuels such as coal, wood chips and bagasse (Langston & Opdyke 1997).

2.2.3. COMBINED CYCLE POWER PLANTS

Combined cycle gas turbines (CCGT) combine the features of a gas turbine with a steam plant. Rather than rejecting the hot exhaust gases as in an OCGT (Figure (b)), these are passed into a Heat Recovery Steam Generator (HRSG) unit, using the exhaust heat to convert water to steam and use this to produce further mechanical work in a very similar manner to a conventional steam plant (Figure (a)). A CCGT plant yields even higher efficiencies than an OCGT, but at the loss of some operational flexibility. Multiple gas turbines may be coupled to the same steam turbine for efficiency of design; such CCGTs are described as “+1” units, so, for example, two gas turbine units and one steam turbine are referred to as a “2+1” CCGT. Other configurations also exist outside of Northern Ireland.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 2.3. KNOWLEDGE GAP

Despite the importance of operating conventional plant with high levels of integrated wind, outside of the industry very little research into these aspects has been carried out. A large number of energy modelling tools have been developed to model the integration of more variable renewable resources; however, none have been designed to examine the impacts on conventional generators (Connolly et al. 2010). Oswald et al. (2008) recognise that volatile swings in renewable energy output affect loading cycles for conventional plant and that the impacts on generators are not well recognised. However, these are only discussed qualitatively. Pöyry (2009) modelled the market (i.e. financial) impacts of high levels of variable wind generation, using broad high-level characteristics of plant generation categories. However, there was no focus on the technical impacts or consideration of lower efficiencies, increased wear and other such aspects discussed in Section 3. SKM (2008) carried out some modelling work on the costs and carbon savings of future plant mixes, but again assumed no reduction in efficiency of conventional plant. Meibom et al. (2009) is one of the few academic studies to have looked directly at the impacts of increased part loading and more flexible operation of power plants using stochastic optimisation modelling. However, this study focused on the costs and only in Germany and Scandinavia under 2010 levels of wind. Although there is some research in this area, there is a lack of literature addressing the impacts of many of the issues that are described in Section 3. There is a need to understand what characteristics are most desirable from conventional plant to facilitate the high levels renewables legislated for in Ireland, as well as better understanding the impacts that a flexible operating regime will have on the existing conventional generation plant.

3. IMPACTS OF VARIABILITY ON CONVENTIONAL GENERATORS

The impacts of higher levels of variable generation are broadly accepted to be twofold; the issue of day-to-day balancing of supply and demand and the longer term planning of sufficient capacity from the right mix of plant. Much topical work has gone into understanding the characteristics of variable resources, particularly wind, but broadly this has focused on lower penetrations (< 20% levels) of wind specifically and very few studies look particularly at the impact on conventional generators (Kubik, Coker & Hunt 2010b). Further issues are the occurrence of low probability high swings in renewable generation output (described here as “events”) that may require conventional plant to operate outside their usual operational characteristics (“excursions”). This section of the paper introduces some of the ways in which the conventional generators described in Section 2.2 may be impacted by the integration of a high penetration of variable renewable energy generation.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

£/MWh Unconstrained schedule to £/MWh Unconstrained schedule to meet demand meet demand SMP Wind SMP Wind

Capacity Capacity (a) (b)

Figure 3 – Visualisation of plant ranked by merit order showing how increasing wind on a system pushes conventional plant off the unconstrained schedule.

3.1. IMPACT ON MERIT ORDER

The concept of the all island electricity market and how it operates was introduced in Section 2.1. The issue of “bottlenecking” due to a lack of interconnection between Northern Ireland was described, which requires generators in Northern Ireland to run “constrained on” and asked to generate in preference to more cost effective merit order plant. The introduction of wind has further impacts on merit order, as illustrated in Figure . Wind is a near zero operating cost generation technology, so increasing levels of wind capacity from (a) to (b) pushes existing conventional plant further down the merit order. For a constant system demand, this reduces the system marginal price (SMP) for the constrained schedule, meaning merit order generators make smaller profits on the electricity they produce, and also pushes more conventional generators off the merit order schedule, meaning they make no profit on operation, and have to remain viable on their capacity payment alone. Although the intention of this is to drive down system costs, it creates an unattractive market for investment in the right mix of conventional generation plant to facilitate the operation of a wind led regime. The Northern Ireland System Operator (SONI) has a “3 generator” rule in Northern Ireland for system security purposes. At any one time, three units must be kept running, so that if one were to suffer an unplanned outage, the other two could ramp up their production to meet the system demand. As power plants have a minimum stable generation level (MSG) below which they cannot be securely synchronised with the grid, an increased level of wind generation presents a challenge in keeping enough generators online to meet this requirement. A particular time period of concern is night time, where wind generation is typically stronger and demand is at its lowest. Traditionally, generator operation instructions from the system operator are driven by system demand. However, as wind forms a significant contribution to the energy mix, the swings in wind output are likely to become the dominant driver dictating when conventional generation will be asked to run. This is problematic for conventional generators, as although demand patterns are relatively predictable into the future, wind can only be accurately forecast in the short term. Conventional generators try to time their planned outages for maintenance (typically at the timescale of weeks to months) to minimise the impact to their revenue and this is much harder to achieve in a wind led regime.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

3.2. OPERATIONAL IMPACTS

In the future conventional generation displaced by wind will have to operate more flexibly. However, running plant more flexibly typically reduces efficiency (a power plant tends to achieve its best efficiency at or near its capacity output), increases operating costs and increases emissions per unit of electricity generated. In the past, conventional steam generators and CCGT plants were typically designed to operate as baseload for the system; providing a predictable and continuous output of power, while OCGT units provided the flexibility of dealing with peak loads. Having to more frequently ramp up and down generation to balance out fluctuations in wind generation will increase wear on units, shorten their lifespans and increase the costs of maintenance. This also places a different emphasis on which characteristics are important in a generator. Plant with a large turndown (able to operate at a minimum stable generation that is a low percentage of its rated capacity) becomes desirable in a variable wind led regime, and plant start-up characteristics from cold, warm and hot states also become important. Depending how long it has been since a unit was last synchronised to the grid there are different start up profiles. The rate of start-up has to be limited due to a number of factors, but a significant limitation is differential expansion rates in the steam turbine between the stator and rotor; the rotor heats up and expands much faster than the stator because it is physically a much smaller mass of metal. This compromises the fine tolerances that are used to improve efficiency. Operational two-shifting, where a unit operates cyclically on a two eight-hour shift on, one eight-hour shift off principle, may become more common, as this keeps most generators in a hot start up state that they can be quickly brought online from. The value of ancillary services as wind penetration increases on a system will also rise. Conventional generation is able to provide voltage regulation (controlling reactive power levels on the grid; important for efficient electricity transmission and distribution) and restorative system inertia to keep the system frequency at 50Hz when a unit trips and the generation it provides is suddenly lost, something that existing installations of wind in Ireland cannot provide at present. Conventional generators also tend to consume a large amount of their own power (particularly at lower loads) to run auxiliary systems such as fans, pumps and mills. As these do not all need to be operational all of the time, there is possible scope for smarter control of this load as an ancillary service to help balance supply and demand. Future volatility of fossil fuel prices may increase the value of fuel diverse conventional generation. For example, Kilroot power station (Table 1) is able to operate on coal or oil. Running on fuel oil is more expensive, but allows a minimum stable generation of 70MW per unit rather than 110MW with coal, as oil burns in a much more stable manner than coal. This may be of increasing value as the system operator has to free up capacity for more wind on the system. At present, the market rules requires strictly monotonically increasing price bids, so the operators at Kilroot are not permitted to start operating on oil (at higher cost) and switch to coal when the 110MW threshold has been reached.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Ballylumford’s 2+1 CCGT plant (Table 1) has been designed so that it can flexibly operate as an OCGT by isolating its HRSG units. This is something that CCGT plants are not normally designed to do. This OCGT/CCGT flexibility is rarely used in the current market schedule, but may be of future value and interest.

4. CONCLUSIONS

This paper has introduced the nature of the significant challenge faced by the Irish Electricity market if intends to meet its goals for a wind penetration of 40% by 2020. The variability of this energy source will have an impact on the operation of conventional plant, particularly in Northern Ireland where there are physical and market constraints in play. A number of generating plant characteristics become increasingly significant with rising penetrations of wind generation, and conventional plant will need to change their operating regimes to maintain commercial viability. These characteristics include:

 The ability to plan maintenance periods to minimise commercial losses.  Plant turn down and start-up profiles.  The ability to two-shift plant operation.  Fuel flexibility.  Plant ancillary system load management.  CCGT/OCGT flexibility.

Past research by the authors (Kubik et al. 2011) has considered the nature of wind resource characteristics against the background of the Irish Electricity Market (Kubik et al. 2010a, Kubik et al. 2010b), but further work is needed to link these characteristics to those of generating plant. This is an area that the authors intend to pursue in future research.

ACKNOWLEDGEMENTS

The authors would like to extend their thanks to David Bothwell, Roger Graham and Brian Mongan all of whom provided data, insight and guidance that made this paper possible. They also wish to acknowledge the funding of the EPSRC, without whom this research would not have taken place.

REFERENCES

Connolly, D. et al., 2010. A review of computer tools for analysing the integration of renewable energy into various energy systems. Applied Energy, 87(4), pp.1059-1082. European Commission, 2008. The EU Climate and Energy Package. Available at: http://ec.europa.eu/clima/policies/package/index_en.htm [Accessed April 19, 2011].

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Howley, M., Gallachóir, B.Ó. & Dennehy, E., 2009. Energy in Ireland Key Statistics 2009. Available at: http://www.seai.ie/ [Accessed April 19, 2010]. Kennedy, A., 2007. Pinch points. Available at: http://www.soni.ltd.uk/wind.asp [Accessed April 20, 2011]. Kubik, M.L., Coker, P.J., Barlow, J. F., & Hunt, C., 2011. A study into the accuracy of using meteorological wind data to estimate turbine generation output. Renewable Energy Journal, [Paper Submitted]. Kubik, M.L., Coker, P.J. & Hunt, C., 2010a. Adopting high levels of renewable electricity: an international perspective on approaches. In 1st Annual TSBE Conference. Reading University. Kubik, M.L., Coker, P.J. & Hunt, C., 2010b. An overview of the current status of research into adopting high levels of renewables in Ireland. In World Renewable Energy Congress XI. Abu Dhabi: Elsevier. Langston, L. & Opdyke, G., 1997. Introduction to Gas Turbines for non-Engineers. Global Gas Turbine News, 37(2). Available at: http://files.asme.org/IGTI/101/13001.pdf [Accessed April 21, 2011]. Laughton, M., 2007. Variable renewables and the grid: an overview. In G. Boyle, ed. Renewable electricity and the grid: the challenge of variability. Earthscan, pp. 1-30. Meibom, P. et al., 2009. Operational costs induced by fluctuating wind power production in Germany and Scandinavia. Renewable Power Generation, IET, 3(1), pp.75-83. Oswald, J., Raine, M. & Ashraf-Ball, H., 2008. Will British weather provide reliable electricity? Energy Policy, 36(8), pp.3212-3225. Pöyry, 2009. Impact of intermittency: how wind variability could change the shape of the British and Irish electricity markets, Available at: http://www.uwig.org/ImpactofIntermittency.pdf. Pöyry, 2007. Trading and Settlement Code - Helicopter Guide. Available at: www.allislandproject.org [Accessed April 20, 2010]. Rourke, F.O., Boyle, F. & Reynolds, A., 2009. Renewable energy resources and technologies applicable to Ireland. Renewable and Sustainable Energy Reviews, 13(8), pp.1975-1984. SKM, 2008. Growth scenarios for UK renewables generation and implications for future developments and operation of electricity networks, BERR Publication URN 08/1021, Sinclair Knight Merz. Walker, N. et al., 2009. Energy Forecasts for Ireland to 2020, Sustainable Energy Ireland.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Development Of A Virtual Pyranometer For Solar Energy Monitoring

P.A. Burgess*, M.M. Vahdati*, D.D. Davies†, S.K. Philip‡ * – University of Reading, United Kingdom † – Solarcentury, United Kingdom ‡ – SSE Contracting, United Kingdom

*Corresponding author [email protected]

ABSTRACT The primary aim of this project is to develop a robust method for applying remotely gathered radiation data to a solar array at any point in the UK for the purpose of system performance monitoring and validation. Central to this is the development of a virtual pyranometer (radiation sensor) which takes radiation data from the Met Office radiation network and information about a given photovoltaic (PV) system (location, orientation, tilt) and estimates the radiation received in the plane of the PV system. In this paper we report on progress towards an operational virtual pyranometer. The main challenges in the development of the virtual pyranometer are (i) decomposition of global irradiance data into diffuse and direct components (ii) transposition of the diffuse component into the plane of the array which is dependent on both the diffuse fraction of the global irradiance and the model of the sky-dome used (iii) interpolation between radiation measurement sites. Several models have been established for the decomposition and transposition of irradiance. This work seeks to determine how accurate these models are in a UK context and find the simplest effective combination of models. This paper does not address the interpolation between measuring sites.

Keywords: Solar energy, irradiation, transposition, virtual pyranometer, remote monitoring

NOMENCLATURE Kt - Clearness index

Id - Diffuse Horizontal Irradiance I - Global Horizontal Irradiance

IT - Global In-plane Irradiance ψ - Hour Angle S - Tilt Angle Z - Zenith angle a & b - dimensionless coefficients

INTRODUCTION

Complete monitoring of the performance of a PV system requires knowledge of the irradiation incident on the panels. This requires that the system be fitted with radiation sensor

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ and a datalogger. Radiation sensors may be either a thermopile pyranometer or a calibrated reference cell. A pyranometer provides higher quality data as it does not have the same issues of spectral mismatch or temperature coefficients as a reference cell. The relatively high cost of installing and maintaining in good condition a radiation monitoring system reduces the circumstances in which they are installed. It would be of great value to the PV industry if a robust system could be developed to allow the use of radiation data widely collected for meteorological purposes, i.e. hourly global (i.e. total) horizontal irradiation which is measured by the Met Office at 91 sites across the UK. The solar radiation at the Earth’s surface is not uniform. In the process of passing through the atmosphere, solar radiation is scattered out of the beam (i.e. the solar disc) by two processes; forward (Mie) scattering and molecular (Rayleigh) scattering. The amount of scattering which occurs depends on atmospheric conditions, principally the quantity of water vapour. The anisotropic distribution of diffuse radiation makes the transposition of horizontal diffuse irradiation data into a tilted plane non-trivial. The other question that must be resolved when using global horizontal irradiation data is how is it split between the diffuse and direct components. The ultimate objective of this project is to derive the irradiation in the plane of a PV system located at any point in the UK. The process which will need to be followed for this is shown in Figure 2.

Figure 2: Steps in the Virtual Pyranometer process

This paper describes efforts to develop the horizontal to in-plane irradiance process (i.e. steps two and three). There are several models available for both component separation and diffuse component transposition, it is necessary to investigate which models will offer an accurate estimate of the in-plane irradiation. Models to separate the diffuse and direct irradiance components will be compared to the measured diffuse and direct irradiance from the weather station at the University of Reading. Transposition models will be tested for conversion of horizontal irradiance components into an inclined plane tilted at a range of angles.

METHODOLOGY

Decomposition of global irradiance into direct and diffuse components is generally achieved through a clearness index model. In simple terms, these models are based on the correlation between the overall level of radiation at the Earth’s surface and the fraction of this irradiance which is diffuse. The technique was originally developed by Liu & Jordan [1] whose model was limited to clear sky conditions and only validated on a monthly or annual basis. Several models have evolved from this technique to cover a wider range of sky conditions and shorter (hourly) time steps. Batlles et al.[2] evaluated the most reliable of these models for a Spanish climate. Other authors [3] have specifically recommended the Reindl-2 model for the British climate. In this paper we will evaluate the performance of both the Reindl-1 (R1) and Reindl- 2 (R2) models [4].

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Both Reindl models have three equations for three sky conditions; overcast skies where the clearness index, Kt < 0.3; intermediate skies where 0.3 < Kt < 0.78 and clear skies where Kt > 0.78.

R1: R2: Equation number Kt < 0.3 Eq. 1

0.3 < Kt < 0.78 Eq. 2

Kt < 0.78 Eq. 3

For transposition of diffuse irradiance into a tilted plane, there are three commonly used models, the Hay model [5], the Klucher model [6] and the Perez model [7]. All three are based on an anisotropic sky dome comprised of an isotropic background superimposed with regions of enhanced or diminished brightness about the solar disc (the circumsolar region) and the horizon. Of the three, the Perez model is the most complex and is used by PVSYST modelling software for detailed analyses at the design stage of a PV system. The Klucher model is a simpler model, similar to the Hay model which PVSyst recommends for use when making use of modelled diffuse irradiance data. In this paper we present the results of the Perez and Klucher transposition models on diffuse and direct components derived from both the original observations and from the Reindl decomposition models.

Perez: ] Eq. 4

Klucher: Eq. 5

Data (including inter alia the global horizontal irradiance, diffuse horizontal irradiance and the sunshine duration) was taken from the University of Reading weather station on a 5 minute basis from October 2009 to February 2011. This has been converted to hourly values and subjected to a series of cleaning steps as follows. All readings were removed if the radiation was a. negative, b. non-zero between sunset and sunrise c. greater than the extraterrestrial irradiance Points where the diffuse irradiance exceeded the global irradiance were also removed.

The separation of diffuse and direct irradiance and the transposition process require several geometric parameters, namely the Zenith angle (Z), Extraterrestrial irradiance (ETR), air mass (m) and sunrise and sunset times. These were taken on an hourly basis (ETR daily) from the SMALLDISC model [8] published by NREL [9].

The third component of in plane irradiance is the ground reflected irradiance calculated as per Iniechen 1988 [10] using a fixed albedo of 0.2.

The results of the decomposition and transposition models calculated using recent global horizontal irradiance data will be compared to the results of a PVSyst model which uses 1990

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ meteorological data to provide a broad indication of the effectiveness of the models used to transpose the modern data.

RESULTS

Decomposition Modelling

The modelled diffuse fraction from the Reindl models were calculated and used to derive the absolute values of the diffuse and direct horizontal irradiance. Both models were compared to the proportion of diffuse irradiation as measured. Figure 3 shows the results of the diffuse fraction modelling. Both models clearly offer a good estimate of the diffuse fraction across all three conditions though the R2 model offers a better representation of the diffuse fraction under clear skies which is important as these conditions correspond to times of high generation from PV so any inaccuracy will be amplified when energy is considered.

(a) (b)

Figure 3 Diffuse fraction (Y-axis) vs. Clearness index (X-axis); measured values and (a) R1 (b) R2 The root mean square and mean bias errors for both models were calculated with a range of maximum zenith angles. Due to geometric effects such as the large increase on relative optical air mass at low sun angles, the uncertainty around these times is significantly higher. Table 2 shows the effect of increasing the maximum zenith angle from 75° to 90°; the amount of data available nearly doubles but at the cost of increased uncertainty with the RMS error rising from around 11% to 25% and 24% for the R1 and R2 models respectively.

Table 2 Change in RMS and Mean Bias error with increasing maximum zenith angle Reindl-1 Reindl-2 Maximum zenith Number of RMSE MBE RMSE MBE angle datapoints 75 661 10.90% 4.00% 11.10% 3.90% 80 966 13.60% 5.40% 16.80% 5.00% 85 1158 17.20% 6.20% 19.20% 6.00% 90 1294 25.30% 12.80% 23.70% 14%

Transposition Results

The diffuse and direct horizontal irradiance components described in section 0 have been used as inputs to the Klucher and Perez transposition models. Transposed irradiance was calculated on a half hourly basis and has been aggregated to monthly results. Planes tilted at 0°, 10°, 20°,

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 30°, 45°, 60° and 90° were evaluated. Table 3 shows the results of the transposition models for a plane inclined to 30°. The ratio of global to inclined irradiance for a PVSyst model using meteorological data for 1990 was used to give an indication of expected in plane irradiance. In both cases, the modelled transposed irradiance was well below what would be expected. In the case of the Klucher model, there was a clear difference between the results using the observed and modelled irradiance components with much better results, 48% and 10% lower than estimated using PVSyst. The bias in the results of the Klucher model varies from month to month with winter 2009-2010 well below the expected level but winter 2010-2011 being very close to the expected values. The Perez model is more consistent than the Klucher model, both across the seasons and across the different input components (observed and modelled irradiance) however it gives a stronger underestimate of the transposed irradiance; 18% below the PVSyst estimate using observed diffuse and direct irradiance and 23% using both the R1 and R2 models.

Table 3 Klucher & Perez Model Transposed Irradiance for a 30° tilt Klucher Transposition Perez Transposition In Horiz In Plane In-Plane In-Plane In Plane In-Plane In-Plane Plane ontal (PVSyst) (Obs) (R1) (R2) (Obs) (R1) (R2) 2009 Oct 53 71 18 21 20 52 51 51 Nov 28 47 12 9 9 31 30 30 Dec 19 34 18 7 7 21 22 22 2010 Jan 22 33 17 7 8 24 24 25 Feb 33 48 4 38 38 35 33 33 Mar 78 93 34 96 96 74 68 69 Apr 131 142 45 132 132 112 113 113 May 150 149 135 135 135 142 125 125 Jun 176 170 177 186 186 166 144 144 Jul 146 144 66 141 141 127 119 119 Aug 89 93 31 79 79 79 76 76 Sep 78 91 41 95 95 72 69 70 Oct 57 76 44 94 94 56 55 56 Nov 27 45 19 50 50 29 28 29 Dec 15 27 8 34 34 17 17 17 2011 Jan 21 32 14 43 43 23 23 23 Feb 27 39 13 40 40 29 28 28 Mar 1 1 0 1 1 1 1 1 Tot. 1151 1335 697 1207 1207 1089 1025 1032

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Figure 4 Monthly Irradiation for (a) Klucher and (b) Perez transposition for a non-tilted plane

DISCUSSION

Both Reindl decomposition models have been shown to be effective in calculating the diffuse horizontal irradiation using the global irradiance, the R2 model which incorporates geometric terms is more accurate than R1 under clear sky conditions. as PV generation will be disproportionately high under these conditions, future developments will use the R2 model for decomposition.

Both the Klucher and Perez transposition models significantly underestimated the in plane irradiance.

In the Klucher model, the observed and modelled irradiation components give substantially different variation in the first and second terms of the governing equation as well as the F term. There are periods where the first and second terms show good agreement between the observed and modelled irradiation inputs. There are also periods where both terms diverge concurrently usually with the first term underestimated for the modelled irradiation and the second term being overestimated, for the F term there are no times where the observed and modelled irradiation values are in good agreement, nor is there a consistent bias, high or low.

Excepting the geometric terms which are common to both the observed and modelled input irradiance, the Perez model is dependent on the F coefficients. Comparing those from our analysis to those of Perez in the original paper shows the same general shape, the semi-

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ analytic nature of the F coefficients leads to a stepped function which may be inhibiting the model’s performance. A fully analytic function (or an increased number of ε bins) would mitigate this potential source of error. The Perez model is recommended in PVSyst for use when accurately measured irradiation components are known which suggests it may not be appropriate for use with modelled component data as presented here.

Figure 5 Perez Variation of F1 and F2 coefficients with sky brightness (45° < Z < 55°)

Figure 6 Perez's original variation of (a) F1 and (b) F2 with sky brightness [7]

While all the models discussed in this paper have been validated for an extensive range of climatic conditions, it is possible that the UK climate has some characteristics which require adjustment of the coefficients used in the models. Ensuring the validity of the models in the UK will be a necessary piece of future work.

For both the Klucher and Perez transposition models, there are clearly unresolved issues which are limiting their effectiveness. Further investigation to reveal the source of these problems and eliminate them is a priority. CONCLUSIONS

We have compared the effectiveness of two irradiation decomposition models, Reindl 1 and Reindl 2. We have also examined two commonly used transposition models for converting horizontal irradiation data into a tilted plane, the Klucher and Perez models.

Both R1 and R2 decomposition models have been shown to provide an accurate estimate of the diffuse horizontal irradiation with the R2 model performing slightly better under clear sky conditions. Due to the relatively high proportion of PV generation under clear skies, this finding reaffirms previous work which suggests that the R2 model is a better choice for PV monitoring applications.

The transposition models assessed in this paper have both significantly underestimated the in- plane irradiation across all possible tilt angles. The models were evaluated using irradiation data capped at Z = 90° which is the least accurate in terms of the decomposition models

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ however this was necessary to allow comparison to PVSyst estimates of the in-plane irradiation as this is only available on a monthly basis so the complete sunrise to sunset irradiation data is required. The performance of the models may be improved by restricting the irradiance data to a lower maximum zenith angle. To verify the effectiveness of transposition models in these circumstances requires in-plane global irradiation data. This data is being collected for several PV systems installed at SSE sites as part of this EngD project and will be available shortly.

REFERENCES

[1] B.Y.H. Liu and R.C. Jordan, “The interrelationship and characteristic distribution of direct, diffuse and total solar radiation,” Solar Energy, vol. 4, 1960, pp. 1-19.

[2] F. Batlles, “Empirical modeling of hourly direct irradiance by means of hourly global irradiance,” Energy, vol. 25, Jul. 2000, pp. 675-688.

[3] J. Roy, T.R. Betts, and R. Gottschalg, “Validation of the Energy Yield Prediction of Photovoltaic Modules,” Proceedings of PVSAT-7, Edinburgh: 2011, p. 197.

[4] D.T. Reindl, W.A. Beckman, and J.A. Duffie, “Diffuse fraction correlations,” Solar Energy, vol. 45, 1990, pp. 1-7.

[5] J.E. Hay, “Calculation of monthly mean solar radiation for horizontal and inclined surfaces,” Solar Energy, vol. 23, 1979, pp. 301-30.

[6] T.M. Klucher, “Evaluation of models to predict insolation on tilted surfaces,” Solar Energy, vol. 23, 1978, pp. 111-114.

[7] R. Perez, P. Ineichen, R. Seals, J. Michalsky, and R. Stewart, “Modeling daylight availability and irradiance components from direct and global irradiance,” Solar Energy, vol. 44, 1990, pp. 271-289.

[8] E. Maxwell, “A Quasi-Physical Model for Converting Hourly global Horizontal to Direct Normal Insolation,” 1987.

[9] NREL, RReDC: DISC Model, rredc.nrel.gov/solar/models/disc accessed April 2011

[10] P. Ineichen, A. Zelenka, O. Guisan, and A. Razafindraibe, “Solar radiation transposition models applied to a plane tracking the sun,” Solar Energy, vol. 41, 1988, pp. 371-377.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Reducing User Influence On Energy Consumption Through Improved Building And Control Design R.M. Tetlow1*, C. P. Beaman2,3, A.A. Elmualim4 and K. Couling5 1 Technologies for Sustainable Built Environments, University of Reading, UK 2 School of Psychology & Clinical Language Sciences, University of Reading, UK 3Centre for Integrative Neuroscience & Neurodynamics, University of Reading, UK 4 School of Construction Management and Engineering, University of Reading, UK 5 AECOM, Bristol, UK

* Corresponding author: [email protected]

ABSTRACT

Considerations of energy and CO2 abatement by building designers currently focus heavily on compliance with Part L Building Regulations and rarely concern actual operational performance. Without this necessary focus, designs frequently fail to meet expectations in terms of in-use energy and the associated CO2 emissions. In some cases this may be due to factors such as poor build quality or poor initial design assumptions; however, the behaviour of the building occupants is a critical factor in all cases. Current efforts to reduce this performance gap focus on the assumption that by providing information to occupants about their undesirable behaviour they will act rationally to reduce their energy usage. However, people’s attitudes and behaviours do not always coincide and, although this method may serve to raise awareness, addressing only this one aspect of human cognitive processing may merely result in short term and relatively small energy savings. Dual-process models of cognition posit two modes of cognitive processing and behavioural control. One is conscious and deliberate and targeting interventions towards this system will improve knowledge and attitudes towards energy consumption. The other is unconscious and automatic. Targeting interventions towards this system should encourage energy efficient behaviour in situations where attention is not consciously focussed upon energy-saving. This paper explores the potential of targeting the two different aspects of cognitive processing to influence building users into displaying energy efficient behaviour. It concentrates on incorporating these techniques into the design of building services and their control systems. It concludes that cognitive errors during user interaction with buildings can be responsible for inadvertent energy use, but corrective interventions for these need to be specifically targeted at the cognitive system from which they result. At present the provision of information is used for this purpose and while this may correct some errors it will not remove them all.

Keywords: Occupant behaviour, energy efficiency in buildings, building controls, user- centred design.

1. Introduction Following global concerns over climate change the UK government has committed to an 80% reduction in CO2 emissions by 2050, in relation to a 1990 baseline. As buildings are

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ responsible for as much as 50% of total UK CO2 emissions (BIS, 2010) they represent an area where substantial reductions will need to be made if these targets are to be met. As a key element of the government’s strategy to achieve this target, recent alterations to Part L Building Regulations impose far stricter limits on estimated regulated CO2 emissions. Additionally, planned changes to regulations will require all new residential buildings to be ‘zero carbon’ by 2016 and all new commercial buildings to be ‘zero carbon’ by 2019 (DCLG, 2007). These factors have led to a significant increase in the amount of buildings currently being designed and constructed with the objective of being energy efficient. However, there is growing evidence to suggest that many supposedly energy efficient buildings do not, in practice, meet the intended levels of energy performance. In fact, CO2 emissions of up to three times the design expectations have been documented (Bordass et al., 2004). This is because the current design methodology is not intended to predict actual energy consumption. Instead the main priority is to demonstrate compliance with Part L Building Regulations (Menezes et al., 2011) and as a result designs regularly fail to achieve the anticipated levels of in-use energy consumption. There are various reasons for the gap between predicted and actual energy consumption. These may include poor build quality, such as gaps in insulation or thermal bridging, poor initial design parameters, such as assuming all systems are turned off when not in use, or ‘value engineering’ where inferior systems are substituted during the construction phase without prior consultation with the design team (Bordass et al., 2004). However, in all instances the behaviour of the building occupants is one of the most crucial factors that can lead to huge variations in a building’s energy consumption. The occupants can often compromise the complex systems that energy efficient buildings regularly employ for heating, cooling, ventilation, and lighting, leading to high levels of wasted energy (Demanuele et al., 2010). Unregulated loads, such as I.T. equipment and personal heaters, can also cause large and unanticipated increases in energy consumption. In an office building these unregulated loads can account for as much as 37% of the regulated associated emissions (DCLG, 2009). Traditional efforts to reduce energy consumption in buildings have revolved around the ‘information deficit model’ which assumes that people will act rationally on any information provided to them and modify their behaviour accordingly (Owens & Driffill, 2008). The obvious example of this is with the use of signs urging building users to turn lighting or appliances off when they are not required. There is evidence to suggest that this type of intervention may influence attitudes but often has a negligible effect on behaviour (McKenzie-Mohr, 2000). Owens and Driffill (2008) point to examples from the domestic sector that show, despite constant information campaigns the adoption of energy efficiency measures has been relatively limited and in some instances behaviour has actually become more energy intensive. This phenomenon parallels the situation in classical economics, in which the predicted behaviour of markets is based upon the assumption that individuals will respond rationally to maximise their expected gains; yet in many cases they demonstrably fail to do so (Ariely, 2009). This and other, deficiencies in classical economics have led to the foundation of behavioural economics, incorporating insights from psychology into human deviations from ‘rational’ behaviour (Kahneman, Slovic & Tversky, 1982). These revelations demonstrate that behaviour is often influenced by a myriad of factors with lack of knowledge being only one of these (McKenzie-Mohr, 2000). Consistent with this, dual process models of cognition state that there are two cognitive systems governing judgements of choice; System I which is automatic and responds directly

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ to the external environment with little conscious intervention, and System II which is conscious and reflective (Evans, 2008). Current efforts to reduce occupant energy consumption, such as information posters or email prompts operate on System II. This is because these interventions rely on the user consciously assimilating the information for later use. Although there has been some research into how targeting the automatic system of human cognition can be used to engender energy efficient behaviour (Schultz et al., 2007; Thaler & Sunstein, 2008), there has been little into its potential implementation within buildings and the consequential impacts on energy consumption. This paper will look at some of the ways these two aspects of cognitive processing can be incorporated into the building service design and the control design in buildings. 2. Dual Process Models of Cognitive Processing & Behavioural Control Dual process models of cognition have become increasingly popular over the previous decade. Although they can be criticised (Evans, 2008) as over-simplistic, they provide a useful framework for which to consider various methods of influencing user behaviour. The model makes a distinction between two cognitive systems that govern automatic and deliberative decision-making. Both of these systems are prone to errors of various types. When considered in the context of an occupant’s interaction with a building and its systems, these errors can often cause behaviours that result in inadvertent energy use. It is important to emphasise that although the root cause of the error may be different the resulting behaviour can often be the same. It is therefore imperative to understand what cognitive error is producing the observed behaviour so corrective interventions can be successfully targeted. To consider how errors of the different cognitive systems directly lead to behaviour that increases energy use an example where a building user operates a heating system and a window concurrently will be presented for each situation. For the system to function efficiently the user must first switch the heating system on and subsequently close the window. 2.1 System I

System I is characterised as non-conscious, automatic, heuristic and irrational. It is highly influenced by the individual’s surrounding environment. For instance, the order and position that foodstuffs are presented to customers at a cafeteria has long been known to influence their choices (Thaler & Sunstein, 2008). Heuristics are mental short cuts which people employ unintentionally during the decision making process. Our surrounding environments are complex and to function properly within them requires constant decision making. Analysing all this information before making decisions would be time-intensive so instead people routinely employ rules of thumb to quickly and efficiently make choices. However, this process is prone to biases and mistakes. In terms of interactions with building there are two main types of errors that can cause inadvertent energy usage; action slips and post-completion errors. The theories behind these types of errors were designed to account for human error in operating complex systems as well as everyday errors in mundane situations. Action slips refer to the performance of an action which was unintended (Norman, 1981). They often occur when the intended action and the unintended action have similar initial stages, such as when driving to the shops on a weekend you inexplicably arrive at your place of work (Norman, 2002). For the heating system example, an action capture would occur

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ when the user engages the heating control, but then instead of closing the window, they turn the lights off instead. Post-completion errors occur when the primary goal of an activity is achieved, but a terminal step is forgotten (Byrne, 2008). A common example of this is failing to attach an aforementioned document to an email message, or leaving the filling cap open after refuelling a vehicle. For the heating system example, a post-completion error would occur when the heating control is operated, but the user subsequently fails to close the window. 2.2 System II System II is characterised as conscious, controlled, systematic, and rational. It allows for a more systematic and reflective analysis of information during the decision making process. Errors and subsequent inadvertent energy use of this system can result from two main sources. Firstly, the user may have inappropriate attitudes or intentions regarding energy consumption in buildings and may consequently operate systems with little regard for the amount of energy that they consume. This situation is especially common in non-domestic buildings where the user is not usually financially penalised for wasting energy. For the heating system example, the heating controls are operated, but the user leaves the window open as they do not care that energy is being wasted.

Secondly, if a user possesses a poor mental model of how a system operates then they are likely to use it incorrectly. The poor mental model can be due to inexperience of the user or due to an inadequately designed system. The latter situation arises because designers and users often possess very different conceptual models of a system. Figure 1 shows the flow of information between the designer and the user. Communication between the two actually takes place at the system image, which is created from a combination of the system’s physical structure as well as other pertinent information, such as instruction manuals or labels. The user’s mental model is formed through their interaction with the system image without direct input from the designer. Designers often assume that the users will have exactly the same conceptual model that they do and it is this discrepancy which can lead to the user operating the system incorrectly and inefficiently (Norman, 2002).

Figure 1. Conceptual models of designers and users and how communication flows between them (Norman, 2002). This is particularly relevant to building control systems as incorrect and inefficient use of controls is a major cause of energy consumption in buildings. Controls which are easy to use and intuitive tend to be energy efficient as the users only operate them as and when needed (Bordass et al., 2007). During a post occupancy evaluation for a UK EcoHomes site rated ‘excellent’, it was discovered that 67% of surveyed residents could not properly program their

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ thermostats (Combe et al., 2010). For the heating system example, the heating controls are operated incorrectly as the user does not adequately understand them and the controls fail to give any accurate feedback. However, the window is successfully closed. 3. Corrective interventions

It can be seen from the errors described that although their root cause is different the actual observable behaviour is often the same, in this case the heating is often on while the window is left open. Understanding the cause of the behaviour is critical in designing and targeting effective interventions to eradicate them. Figure 2 shows what corrective interventions can be used to mitigate errors caused from the different systems. These interventions will be expanded on and their use for the heating example will be explored.

System Cause of Error Possible Corrective Interventions

System 1: Affordances, Non- Defaults, conscious, Action not as intended Forcing Functions automatic, Social Norms heuristic, irrational

Education & Inappropriate training System 2: Attitude/Intentions Conscious, controlled, systematic, Affordances, rational Poor Mental Model of Education & training, System Transparent design & feedback

Figure 2. The corrective interventions for the different cognitive systems. 3.1 Affordances Affordances provide intuitive signs to the user of how the object or system should be operated. Norman (2002) describes the affordances of doors; with plates for pushing and handles for pulling, and demonstrates how easily even these basic design features can be implemented incorrectly resulting in frustration for the user. Used properly affordances can indicate to the user the correct, and in this case, the most energy efficient way of operating the system. If there is little or no obvious affordance the system is prone to incorrect use and inadvertent energy consumption. For the heating system example, better affordance on the heating control increases the likely hood that the user will operate the control properly. 3.2 Defaults When confronted with several options for a situation with which they are unfamiliar many people will not make an explicit choice and will instead settle for the default option. Johnson and Goldstein (2003) show how levels of participation in an organ donation program can be drastically affected by whether the box on the requisite form requires the individual to tick it to opt in or tick it to opt out of the program.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ The default settings for building services and their control systems can have a large impact on the amount of energy consumed. During the post-occupancy evaluation of the Heelis building, Swindon, it was discovered that the kitchen extract fans were continually operating on a much higher level than was actually required (Fordham, 2007). Defaults are often utilised in building services to return a system to a particular low-energy setting after a pre-set period of time. For the heating system example, defaults could be used to restore the heating control to its original level when the building is un-occupied, such as during the night or at the weekend. 3.3 Forcing functions Forcing functions constrain actions by making one stage impossible without first completing a prior stage. Requiring keys to be in the ignition before a car can be started is a good example of this (Norman, 2002). Forcing functions can be an effective way of ensuring that users operate the system as the designer intended, but can prove unpopular as they are often inconvenient for the user. For the heating system example, the user would be unable to activate the heating system unless the window is first closed. 3.4 Social norms Social norms are the behavioural expectations that are present within a specific group. Individuals are highly influenced by those around them and will often take behavioural cues from what other people are doing in the same situation. Schultz et al. (2007) carried out an experiment to measure the effect social norms could have on domestic energy consumption. They informed around 300 households in California about how much energy they were using relative to the average of households in their neighbourhood. As a direct result households who were above the average consistently lowered their energy consumption. However, a ‘boomerang’ effect occurred where households that were using much less than the average actually increased their usage towards the average. This effect was reduced by the inclusion of emoticons indicating whether the current energy usage was socially approved or disapproved. For the heating system example, the user has repeated witnessed other users operating the heating controls and then shutting the window. To avoid any social ramifications they too carry out this behaviour in the observed order. 3.5 Education and training This is currently the intervention most often employed to influence building user behaviour. It includes user manuals, staff training, and informative signs and posters. Although it is undeniably important to educate users about energy efficient behaviour it is not in isolation an effective measure. This type of intervention is often used as it is perceived being relatively cheap and easy to implement. Evidence suggests that while it may have an initial effect, users will often revert to their old behaviours after a short period of time (McKenzie-Mohr, 2000). Education and training can be an effective intervention where they are used to improve the user’s mental model of a particular system. If errors are occurring due to the user’s erroneous conception of how the system functions then training can be provided to ensure that they can use it correctly and efficiently. For the heating system example, prior training for the user could allow them to operate the heating control correctly due to a more appropriate mental model of the system.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 3.6 Transparent design and feedback Errors made by a user due to a poor mental model of the system can be greatly reduced by making the design much more transparent and intuitive. Natural mapping between control systems and the results of their operation can be an extremely effective way of achieving this. Norman (2002) highlights examples of how a four ring cooker hob and its control knobs can be simply laid out taking advantage of natural mapping and removing the need for any labelling. Feedback is an important consideration for building services and their controls. It is essentially sending the user information about what effect their action has had. To be at its most effective feedback needs to be salient, close in time, and specific to the conducted action (Byrne, 2008). Without adequate feedback the user is unlikely to form a sufficient mental model of how the system operates further perpetuating errors. An example of effective feedback within building can occasionally be seen in mixed mode ventilated buildings. Users in these buildings can often compromise the carefully controlled ventilation settings by opening and closing windows at inappropriate times. This can inadvertently lead to unacceptable levels of thermal comfort as well as high energy consumption. Mixed mode ventilation systems occasionally use a traffic light LED displays to indicate to the user when it is appropriate to operate the windows without compromising the system. For the heating system example, feedback could be incorporated into the design so that when the user operates the heating system a warning light and sound would result from the window if it was open. 4. Conclusions This paper has discussed how dual process models of cognition may be used as a framework to consider how to influence building user behaviour. Both systems are prone to different types of error and in the context of user interaction with buildings these errors can result in inadvertent energy use. Although the resulting energy use behaviour from these errors can be the same they can have different sources which need to be addressed in different ways. Currently, the vast majority of attempts to influence user behaviour revolve around the ‘information deficit model’ and as a result are primarily targeted at System II cognitive processing. The assumptions of this model; that there is a direct link between attitudes and behaviour and that the user will necessarily act rationally on information that they receive is flawed (McKenzie-Mohr, 2000). Although it is important to educate users about how to properly use control systems it is not by itself an effective comprehensive strategy to achieve discernable and long term energy saving. Instead, the errors stemming from System I cognitive processing need to be appreciated. The interventions for targeting this system, such as affordances, defaults, forcing functions and social norms can provide effective and long term energy efficient behaviour regardless of the knowledge and attitudes of the user. However, there has been little research into this topic and corrective interventions of this type will need careful consideration and testing. Some interventions can produce quite different results from what was initially intended, as with the ‘boomerang effect’ during social norm interventions (Schultz et al., 2007). It is clear that if the UK is to meet its ambitious emissions targets then building designers cannot continue to marginalise the building users during the design process. It is futile to integrate ever more complex energy efficient technology into buildings if the users will be incapable of using them Designers often claim that the problem is not with the design of the systems, but with people operating them incorrectly. However, it is important to emphasise

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ that systems are created with the implicit purpose of allowing users to perform necessary functions; they are not designed as entities themselves. If the majority of users cannot operate the system easily and consistently to achieve their desired purpose then the design has failed.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ References

Ariely, D. 2009. Predictably irrational. HarperCollins. Bordass, W., Cohen, R., & Field, J. 2004. Energy Performance of Non-Domestic Buildings: Closing the Credibility Gap. Building Performance Congress. 2004. Bordass, W., Leaman, A., & Bunn, R. 2007. Controls for End Users: a guide for good design and implementation, Building Controls Industry Association (BCIA), March 2007. Byrne, M. D. 2008. Preventing post-completion errors: How much cue is enough? In Proceedings of the Thirtieth Annual Conference of the Cognitive Science Society, 351-356. Austin, TX: Cognitive Science Society. Combe, N., Harrison, D., Dong, H., Craig, S. & Gill, Z. 2010. Assessing the number of users who are excluded by domestic heating controls. International Journal of Sustainable Engineering 4, 84-92. Demanuele, C., Tweddell, T. & Davies, M. 2010. Bridging the gap between predicted and actual energy performance in schools. World Renewable Energy Congress XI, 25-30 September 2010, Abu Dhabi, UA. Department for Business, Innovation and Skills (BIS). 2010. Low Carbon Construction, Innovation and growth team: Emerging Findings. BIS, 2010. Department for Communities and Local Government (DCLG). 2007. Building a Greener Future: Policy Statement, DCLG, London. Department for Communities and Local Government (DCLG). 2009. Zero carbon for new non- domestic buildings. DCLG, London. Evans, J. 2008. Dual-Processing Accounts of Reasoning, Judgment, and Social Cognition. Annual Review of Psychology 59, 255–278. Fordham, M. 2007. Heelis National Trust’s HQ. Building Magazine 5th November 2007. Online. Accessed: 7th February 2011. Available at: http://www.building.co.uk/heelis-national-trust’s-hq- reviewed-by-max-fordham-engineer/3098715.article. Johnson, E. J., & Goldstein, D. G. 2003. Do defaults save lives? Science, 302, 1338-1339. Kahneman, D., Slovic, P., & Tversky, A. 1982. Judgment under uncertainty: Heuristics and biases. Cambridge: Cambridge University Press. McKenzie-Mohr, D. 2000. Promoting Sustainable Behaviour: An Introduction to Community-Based Social Marketing. Journal of Social Issues, 56, 543-554. Menezes, A.C., Cripps, A., Bouchlaghem, D. & Buswell, R. 2011. Predicted vs. Actual Energy Performance of Non-Domestic Buildings. Third International Conference on Applied Energy, 16-18 May 2011, Perugia, Italy. Norman, D.A. 1981. Categorisation of Action Slips. Psychological Review, 88, 1-15. Norman, D.A. 2002. The Design of Everyday Things. New York: Basic Books. Owens, S. & Driffill, L. 2008. How to change attitudes and behaviour in the context of energy. Energy Policy 36, 4412–441. Schultz, P. W., Nolan, J., Cialdini, R., Goldstein, N., & Griskevicius, V. 2007. The constructive, destructive, and reconstructive power of social norms. Psychological Science 18, 429-434. Thaler, R. & Sunstein, C. 2008. Nudge. London: Penguin.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Dynamic Stall For A Vertical Axis Wind Turbine In A Two-Dimensional Study

R. Nobile1,*, Dr M. Vahdati1, Dr J. Barlow1, Dr A. Mewburn-Crook2

1 University of Reading, Reading, UK 2 Wind Dam Renewables Ltd, Swansea, UK

* Corresponding author [email protected]

ABSTRACT The last few years have shown that Vertical Axis Wind Turbines (VAWTs) are more suitable for urban areas than Horizontal Axis Wind Turbines (HAWTs). To date, very little has been published in this area to assess good performance and lifetime of VAWTs either in open or urban areas. At low tip speed ratios (TSRs<5), VAWTs are subjected to a phenomenon called 'dynamic stall'. This can affect the fatigue life of a VAWT if it is not well understood. The purpose of this paper is to investigate how CFD is able to simulate the dynamic stall for 2-D flow around VAWT blades. During the numerical simulations different turbulence models were used and compared with the data available on the subject. In this numerical analysis the Shear Stress Transport (SST) turbulence model seems to predict the dynamic stall better than the other turbulence models available. The limitations of the study are that the simulations are based on a 2-D case with constant wind and rotational speeds instead of considering a 3-D case and variable wind speeds. This approach was necessary for having a numerical analysis at low computational cost and time. Consequently, in the future it is strongly suggested to develop a more sophisticated model that is a more realistic simulation of a dynamic stall in a three-dimensional VAWT.

Keywords: Vertical Axis Wind Turbine (VAWT), urban area, computational fluid dynamics (CFD), dynamic stall, turbulence model

Nomenclature

N number of rotor blades ...... λ tip speed ratio ...... (ΩR/ U∞) c airfoil/blade chord ...... mm θ azimuth angle ...... deg t thickness of the blade ...... mm α angle of attack ...... deg s span of the blade ...... mm V relative wind speed ...... m/s Rr radius of rotor ...... mm ω rotational speed ...... rad/s Rm radius of central mast ...... mm PIV Particle Image Velocimetry ...... U∞ undisturbed velocity ...... m/s Ω rotation frequency/vorticity ...... rad/s

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ INTRODUCTION In the last few decades, the production of electricity from wind turbines has seen a rapid growth in many countries around the world. The major drivers are the recent need to reduce CO2 emissions into the atmosphere and meet the growing demand for electricity. One promising alternative for the future generation of electricity is the installation and integration of wind turbines in the built environment combined with other alternative sustainable systems [1], [2]. The benefits are mainly generation of electricity on the site where it is needed with reduction in transmission losses and cable costs.

To date, very little research has been carried out on VAWTs on understanding the aerodynamics and flow interaction between blades during the operation of a wind turbine [3]. And, the progress of small wind turbines in the built environment has been mainly focused on Horizontal Axis Wind Turbines (HAWTs) rather than Vertical Axis Wind Turbines (VAWTs). But several studies have shown that VAWTs are more suitable for urban areas than HAWTs [2], [4], [5], [6]. The advantages are mainly: omni-directional without a yaw control, better aesthetics to integrate into buildings, more efficient in turbulent environments and lower sound emissions [7]. In addition there is some research speculating that VAWTs are appropriate for large scale of 10 MW or more, as VAWTs can operate mechanically better than HAWTs [8], [9]. They can withstand high winds due to their aerodynamic stall behaviour [4]. Generally, the aerodynamic analysis of a VAWT is very complicated, as the blades are called on to operate in unsteady flow, pitching relative to the mean flow and cutting the stream tube twice [4]. One important aspect to consider during the operation of a VAWT at low wind speeds is the generation of a phenomenon called dynamic stall. The phenomenon is mainly characterised by the development of vortices that will interact with the airfoil of the blades and have a substantial impact on the design and power generation of the wind turbine [10]. The main purpose of this work is to understand how CFD is able to approach the development of dynamic stall around the blades of a straight-bladed VAWT. In this paper a number of simulations are explored to understand dynamic stall around the rotor of a 3 straight-bladed VAWT. To reduce time and memory costs a 2-D case is explored for all numerical simulations. The Computational Fluid Dynamics (CFD) Software used was ANSYS CFX 12.0. Here, the three turbulence models analysed were the k-ε model, the standard k-ω model and the SST (Shear Stress Transport) model. The numerical simulations were studied for different tip speed ratios and compared with the small amount of experimental data available in literature.

METHODOLOGY In order to understand the physics involved during dynamic stall of a straight-bladed vertical Darrieus wind turbine, a 2-D rotor is proposed and analysed. This 2-D approach is adopted for reducing time and computational costs. The airfoil analysed, for the VAWT, is a NACA 0018 and its characteristics are listed in Table 1. The rotor is composed of 3 blades and a central mast. The solid model of the rotor was generated with ProEngineer 4.0 and imported into ANSYS CFX 12.0.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Table 1: Properties of the rotor

NACA0018 Cord Thickness Span Rotor Number Mast Radius c t s Radius Blades Rm (mm) (mm) (mm) Rr N (mm) (mm) 490 88.2 50 3000 3 75

The mesh, as shown in Figure 1, is mainly composed of three sub-domains: one fixed sub-domain outside the rotor, one dynamic sub-domain around the blades of the rotor and one fixed sub-domain for the remaining part of the rotor. The mesh was generated by adopting a Sweep Method with one element deep and the total number of elements is 4.58x105, as listed in Table 2.

Fig. 1: Mesh for the 2-D rotor of the VAWT

Table 2. Sub-domain properties

Sub-domain Number of elements 1) Fixed sub-domain outside rotor 20*103 2) Dynamic sub-domain around blades 42*104 3) Fixed sub-domain inside rotor 18*103 Total number of elements 4.58*105

All sub-domains were meshed by using only triangle elements, as they are more appropriate for simulations involved fluids [11]. The mesh around the blades of the rotor, which is a wake development region, was refined through the use of the facing sizing and inflation options available in ANSYS 12.0. In this region, the use of prism elements is able to capture boundary layer effects more effectively and efficiently. The three different meshes were linked together with the use of domain interfaces. Symmetrical boundaries were used for the top and bottom parts of the 2-D model with no-slip boundary conditions at the two sides. An opening boundary was chosen for the output and a constant wind speed of 6 m/s, with a turbulence intensity of 5%, was defined for the inlet. This wind

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ speed corresponds to the operational time of many VAWTs in the built environment. The two wall sides were placed at 4c from the diameter of the blades respectively. The outlet and inlet were placed 4c and 10c away respectively. Table 3 gives a summary of the parameters employed in the four different cases where different time steps and tip speed ratios (TSRs) were defined. For all transient simulations a total time was defined to give enough time for the flow to develop around the blades of the rotor. Finally, the residual target, in the convergence criteria, was set to be 10-4.

Table 3. Input data for the four cases analysed

Simulation Tip speed Angular Time for one Angle of Time Step Case ratio speed rotation attack Simulation (rad/s) (s) (deg.) (s) Case 1 λ1=2.3 ω1=4.7 t1=1.33 -30≤α1≤30 0.0037 Case 2 λ2=3.0 ω2=6.0 t2=1.05 -21≤α1≤21 0.0029 Case 3 λ3=4.0 ω3=8.0 t3=0.78 -15≤α1≤15 0.0022 Case 4 λ4=5.0 ω4=10 t4=0.63 -12≤α1≤12 0.0017 The effect of different TSRs are presented and discussed in the conclusions section of this paper.

Dynamic Stall Although VAWTs have several advantages over HAWTs, the aerodynamics around the blades is very complicated [10]. VAWTs, during their operation, are called to work under both static and dynamic stall conditions. For a rotational airfoil, static stall will develop at high tip speed ratios with low angle of attacks for a steady flow, while dynamic stall will take place at low tip speed ratios with high angle of attacks for an unsteady flow. Static and dynamic stall are characterised by large recirculation separated flows. Consequently, the blades are subjected to cyclic forces due to the variation of incidence angle of the blade relative to the wind direction [12]. Although the presence of dynamic stall at low TSRs can have a positive impact on the power generation of a wind turbine, the formation of vortices can generate other problems such as vibrations, noise and reduction of fatigue life of the blades due to unsteady forces [12]. Larsen et al. [13] show that dynamic stall is mainly characterised by flow separations at the suction side of the airfoil. This can be summarised in four crucial stages: 1) Leading edge separation starts, 2) Vortex build-up at the leading edge, 3) Detachment of the vortex from leading edge and build-up of trailing edge vortex, 4) Detachment of trailing edge vortex and breakdown of leading edge vortex. The sequence of these four flow events will generate unsteady lift, drag and pitching moment coefficients with a large range of flow hysteresis dependent on the angle of attack [14]. The expression of the angle of attack α adopted for the simulation, without induction factor, is given by Eq. (1):

 sin    arctan  (1)    cos 

where θ is the azimuth angle and λ the TSR. In this study, as shown in Table 3, four different cases are analysed with different parameters that are highly dependent on the TSRs.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Turbulence Models Wang et al. [15] show that the most popular turbulence models, adopted in the CFD community, are mainly Direct Numerical Simulation (DNS), Large Eddy Simulation (LES) and Reynolds-Averaged Navier-Stokes (RANS). The DNS method today requires a large amount of computing resources and time. The LES method is more appropriate for 3-D simulations. Therefore, the only method adopted for this 2-D numerical study was the RANS method. The three RANS turbulence methods analysed, due to low computational costs, are: the standard k-ω model, the standard k-ε model and the SST model. A more detailed description about the turbulence methods can be found in the book by Wilcox [16].

RESULTS In this section of the paper a number of numerical simulations are analysed for different TSRs. The numerical simulations obtained during the present study are mainly compared with the study carried out by Wang et al. [15], as this previous numerical study showed a good agreement with experimental data. The main differences are summarised and listed in Table 4.

Table 4. Main differences between the two numerical methods

Present Study Wang et al. Study ANSYS CFX FLUENT Rotor with 3 blades and central mast Single pitching blade Free-stream turbulence intensity 5% Free-stream turbulence intensity 0.25% Variable time step with angular Constant time step velocity Quadrilateral elements Triangle elements α=100+150sin(ωt) α=arctan(sinθ/λ-cosθ) No wake interaction Wake interactions Initial input from converged steady state No converged steady state

In this numerical study, a number of cycles of the rotor were calculated at different TSRs until a periodic solution was achieved. The total time was set to be 4s to give enough time for the rotor to reach a steady periodic state in all four cases. In Fig. 2 and Fig. 3, the left side, show graphically the results obtained for this numerical analysis, while the right side compare the CFD results of Wang et al. with experimental data by Lee and Gerontakos [17]. Furthermore, Fig. 2 and 3 show how the lift and the drag coefficients, Cl and Cd, are affected by different angles of attack and TSRs. The curve shapes are in good agreement with the experimental data obtained by Lee and Gerontakos. Here, the final results are more realistic and less fluctuating than the CFD simulation conducted by Wang et al [15]. Also, a strong instability at high angles of attack is observed and is thought to be due to the deep dynamic stall that is typical for low TSRs. However, the few exceptions are the absence of a second peak on the curves due to the presence of a trailing edge vortex and the lack of intersection points between upstroke and downstroke paths that are seen in the experimental graphs on the right.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

SST Model

Angle of attack α CL, λ=2.3 CL, λ=3 CL, λ=4 CL, λ=5

Fig. 2. Lift coefficient CL for the two numerical studies and experimental data

SST Model

Angle of attack α

CD, λ=2.3 CD, λ=3 CD, λ=4 CD, λ=5

Fig. 3. Drag coefficient CD for the two numerical studies and experimental data

However, the development of several peaks, especially for negative angle of attacks and low TSRs can be associated with the development of upstream wakes that will interact with the downstream blades. Also the presence of a central mast will generate several wakes that will affect the flow downstream. The different hysteresis loops in Fig. 2 and 3, show clearly the development of two phenomena called dynamic and static stall. Dynamic stall typically will develop at high angle of attacks for λ<4 and is characterised by its fluctuating nature, while static stall will take place at low angle of attacks for λ ≥4 with smoother curves and less intersection points. One important consideration is that the range of angle of attack is different from the experimental data obtained by Lee and Gerontatos. This is mainly due to the use of Eq. (1) instead of using α=100+150sin(ωt) that is typically adopted for the case of a pitching motion single blade. Finally, the numerical results are analysed by comparing the evolution of the shed vorticity with the experimental data available in the field of dynamic stall. Fig. 4 shows the vorticity field at θ=1200 for the three turbulence methods adopted in this numerical study. The SST method shows a better agreement with the experimental data obtained by Ferreira et al. [18] and Wang et al. [15] than the k- and k-ε methods that are more dissipative. This turbulence method is able to show more accurately the generation of vortices at the leading and trailing edges than the other two turbulence methods. A more detailed explanation will be given in the next section for the SST method, as the results best agreeing with experiments

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Fig. 4. Vorticity field for θ=120o for the three different turbulence methods

SST Results In this section only the SST method is analysed in its several stages during a complete revolution of the rotor when a periodic solution is achieved. Fig. 5 clearly shows the several stages involved through dynamic stall for a TSR of 2.3 at different azimuth angles θ. This specific TSR was chosen because it corresponds to the case with deep stall. Fig.5 shows that for an azimuth angle θ between 1800 and 2400 the flow is almost attached to the blade. Then at an angle of θ=2600 a leading edge vortex will start to develop and expand until 2800. Also, in this stage there is the generation of a trailing edge vortex that will detach at approximately 3000. Afterwards, there is a progressive reattachment of the flow to the blade with leading and trailing edge vortices moving downstream. At θ=3600 there is again the presence of a second small leading and trailing edge vortices that will disappear between 200 and 400. Then the development of a third leading edge vortex will take place at 600 followed by a trailing edge vortex at 800. Finally, the flow will start to become laminar and to reattach to the blade until the same dynamics will start again from the beginning stage. From the final results it can be stated that the SST model shows a good agreement with the four stages related to dynamic stall mentioned above. Furthermore, this 2-D study has the advantages of showing how the two kinds of vortices will move down and eventually affect the physics of the blades that are called to work downstream.

Fig. 5: Vorticity field at different azimuthal angles θ for the SST model at λ=2.3

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ CONCLUSIONS A numerical analysis of the physics involved during dynamic stall of a rotor of a straight- bladed VAWT, composed of 3 blades with profile NACA0018, was conducted at four different TSRs. In this study three RANS turbulence models have been explored for the four cases analysed. The lift and drag coefficients are in good agreement with the study carried by Wang et al, but a number of differences have been found between the two studies. The most relevant are the absence of a second peak on the curves due to the development of the trailing edge vortex and the lack of intersection points at high TSRs. But at low TSRs there is an increase in the number of intersection points especially for negative angles of attacks that can be related to deep dynamic stall. This instability is mainly associated with the development of upstream wakes from upstream blades and mast that interact with the downstream blades. In this numerical study, the analysis has proved the presence of two different phenomena called dynamic and static stall that are highly depended on the TSRs adopted. In here the SST model is examined in terms of vorticity distributions around the blades. In general the method shows more reasonable accuracy with some existing wind tunnel experiments than the k-ε and k-ω turbulence methods that seem to be more dissipative. Also, the method is able to show the main four phases involved during dynamic stall. But one important observation for the SST method is the presence of single-vortices instead of having several small vortices that are typically found around the airfoil for PIV dynamic stall tests. A better improvement can be achieved in the future investigation of a 3-D case where the LES and the DES methods are strongly recommended. The two methods will take into consideration the 3-D nature of the vortices developed during dynamic stall. In general this paper shows the aerodynamics involved at different TSRs and angles of attack for a 2-D rotor with central mast. This is necessary because the development of dynamic stall in VAWTs can have a substantial impact on both the design and power generation of a wind turbine.

References

[1] J. Knight, “Urban wind power: Breezing into town,” Nature, vol. 430, pp. 12-13, Jul. 2004.

[2] S. Mertens, Wind energy in the built environment: concentrator effects of buildings. TU Delft, 2006.

[3] R. Howell, N. Qin, J. Edwards, and N. Durrani, “Wind tunnel and numerical study of a small vertical axis wind turbine,” Renewable Energy, vol. 35, no. 2, pp. 412- 422, Feb. 2010.

[4] A. Mewburn-Crook, “The Design and development of an augmented vertical wind turbine.” School of Mechanical, Aeronatical and Production Engineering, Apr-1990.

[5] S. Stankovic, N. Campbell, and A. Harries, Urban Wind Energy. Earthscan, 2009.

[6] C. J. S. Ferreira, G. van Bussel, and G. van Kuik, “2D CFD simulation of dynamic stall on a Vertical Axis Wind Turbine: verification and validation with PIV measurements,” presented at the 45th AIAA Aerospace Sciences Meeting and Exhibit, 2007, pp. 1-11.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ [7] C. Hofemann, C. J. Simao Ferreira, G. J. W. Van Bussel, G. A. M. Van Kuik, F. Scarano, and K. R. Dixon, “3D Stereo PIV study of tip vortex evolution on a VAWT,” 2008, pp. 1-8.

[8] G. Marsh and S. Peace, “Tilting at windmills: Utility-scale VAWTs: towards 10MW and beyond?,” Refocus, vol. 6, no. 5, pp. 37-42, 2005.

[9] G. Marsh, “Wind turbines: How big can they get?,” Refocus, vol. 6, no. 2, pp. 22- 28, March.

[10] C. J. Simão Ferreira, A. van Zuijlen, H. Bijl, G. van Bussel, and G. van Kuik, “Simulating dynamic stall in a two-dimensional vertical-axis wind turbine: verification and validation with particle image velocimetry data,” Wind Energy, vol. 13, no. 1, pp. 1- 17, 2010.

[11] “ANSYS Meshing: Application Introduction.” ANSYS, 2009.

[12] N. Fujisawa and S. Shibuya, “Observations of dynamic stall on Darrieus wind turbine blades,” Journal of Wind Engineering and Industrial Aerodynamics, vol. 89, no. 2, pp. 201-214, Feb. 2001.

[13] J. W. Larsen, S. R. K. Nielsen, and S. Krenk, “Dynamic stall model for wind turbine airfoils,” Journal of Fluids and Structures, vol. 23, no. 7, pp. 959-982, Oct. 2007.

[14] J. A. Ekaterinaris and M. F. Platzer, “Computational prediction of airfoil dynamic stall,” Progress in Aerospace Sciences, vol. 33, no. 11-12, pp. 759-846, Apr. 1998.

[15] S. Wang, D. B. Ingham, L. Ma, M. Pourkashanian, and Z. Tao, “Numerical investigations on dynamic stall of low Reynolds number flow around oscillating airfoils,” Computers & Fluids, vol. 39, no. 9, pp. 1529-1541, Oct. 2010.

[16] D. C. Wilcox, Turbulence Modeling for Cfd. DCW industries La Canada, 2006.

[17] T. Lee and P. Gerontakos, “Investigation of flow over an oscillating airfoil,” Journal of Fluid Mechanics, vol. 512, pp. 313-341, 2004.

[18] C. J. Simão Ferreira, A. van Zuijlen, H. Bijl, G. van Bussel, and G. van Kuik, “Simulating dynamic stall in a two-dimensional vertical-axis wind turbine: verification and validation with particle image velocimetry data,” Wind Energy, vol. 13, no. 1, pp. 1- 17, 2010.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Sustainable Procurement – Challenges For Construction Practice.

R.J. Belfitt*1, Prof. Martin Sexton2, Dr. Libby Schweber2, Brian Handcock3.

1 Technologies for Sustainable Built Environments, University of Reading, UK.

2 Innovative Construction Research Centre, University of Reading, UK. 3 Morgan Sindall plc. Rugby, UK.

* Corresponding author: [email protected]

ABSTRACT The impetus for sustainable procurement in construction is growing in response to regulatory, societal and client demands. A strong impetus, however, needs to be translated into effective sustainable procurement strategies and practices. The evidence to date suggests that this translation has been patchy across the sector as a whole and, in many cases, within single construction companies. There are number of barriers embedded in traditional procurement and organisational structures and practices which are hampering the often deep changes required.

The aim of this paper is to draw on and critically reflect upon the general and construction specific sustainable procurement literatures. From this, a number of theoretical and empirical questions will be distilled which will inform the investigation of the sustainable procurement policies and practices within Morgan Sindall plc. and the design of action research interventions to bring about targeted change and innovation to enhance its sustainable procurement performance. Keywords: Sustainable procurement, construction.

1. INTRODUCTION

Sustainability is an issue which is being given increasing attention by both government and industry and pressures are increasing on businesses to improve sustainability performance. Whilst many of the issues are nothing new, sustainability is being used to group together a variety of both new and more established issues. There are a number of differing definitions of sustainable development. One of the most widely used is that of the Brundtland Report (World Commission on Environment and Development 1987) which defines it as “development that meets the needs of the present without compromising the ability of future generations to meet their own needs”. This definition however does not suggest criteria for putting sustainability ideas into practice. Nor does it suggest what principles sustainability contains. The ‘needs’ of the present are likely to differ between varying groups. For example the needs of a construction company, seeking to make a profit from construction of a building are likely to be significantly different to those of the end user, who desires a place to work or live. In the same way, the needs of future generations may not be the same as the needs of people today.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ A widely used concept for describing sustainability is that of the triple bottom line (Norman and MacDonald 2004). This concept suggests that sustainability is made up of three main components of environmental, social and economical sustainability. Still, this does not give substance as to what is contained within each component. In 2005, the UK government released its Sustainable Development Strategy, which included five guiding principles of “living within environmental limits”, “ensuring a strong, healthy and just society”, “achieving a sustainable economy”, “promoting good governance” and “using sound science responsibly”. It then goes on to outline “Sustainable consumption and production”, “climate change and energy”, “natural resource protection and environmental enhancement” and “sustainable communities” as priority areas for immediate action. (Department for Environment Food and Rural Affairs 2005). In 2005 the European Union Council adopted “Guiding principles of sustainable development” in which four “key objectives” and ten “policy guiding principles” for sustainable development are outlined (Commission of the European Communities 2005). Here we see even two governmental bodies taking differing approaches to sustainability. For a business entity, the results would be significantly different again. It seems clear that the criteria of what constitutes sustainability is a much debated subject and it is often the case that what one body considers important sustainability factors may not match the opinions of another.

The construction industry in the UK is responsible for around 8% of GDP (Department for Business Enterprise & Regulatory Reform 2008). The sector is made up of companies from the very small, with a handful of employees, to very large national and multinational corporations employing thousands. Customers of construction projects are both public and private, with the public sector representing about one third of direct construction procurement

Procurement is the acquisition of goods and services. This could include anything from office supplies, to construction materials, to the services of contractors and sub- contractors. Procurement is one way in which companies interact with one another and it could be considered that the actions of a company are only as sustainable as those of its suppliers. There is an increasing amount of literature available on sustainable procurement (Seuring and Müller 2008). However, the work applying these principles to construction is much more limited.

This paper will first outline some general sustainable procurement influences and barriers from the literature on sustainable procurement. It will then look more specifically at the construction industry and how these ideas might be applied within the construction context. Finally, it will look to form some questions that will be used to develop an investigation of sustainable procurement practices within Morgan Sindall plc, a large UK construction firm, and how these practices can be enhanced.

2. SUSTAINABLE PROCUREMENT

The concept of sustainable procurement draws attention to the responsibility of a company for activities outside its own boundaries (Meehan and Bryde 2011). There are a number of influences for increased interest in sustainability and hence sustainable procurement practices. Some main examples are suggested below.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ The first of these drivers is government policy. In 2005, the UK released its Sustainable Development Strategy where it pledged to “lead by example” in how it spends money (Department for Environment Food and Rural Affairs 2005). This was followed in 2007 by the UK’s Sustainable Procurement Action Plan which aimed to make the UK government EU leaders in sustainable procurement practice (Department for Environment Food and Rural Affairs 2007). It is uncertain whether the private sector should follow this example, as their needs may be significantly different to those of private companies. It could however have a significant influence on companies competing for contracts from the public sector. As government buyers look to companies with improved sustainability credentials, so companies might have to look to their supply chains to ensure they are able to meet the new demands. Indeed, in the Sustainable Procurement Action Plan, an intention is set out that government “will include appropriate requirements for suppliers and sub-contractors to provide products and services that comply with agreed mandatory standards and assist in the delivery of departmental sustainable operations targets” (Department for Environment Food and Rural Affairs 2007). In addition, legislation on product specification will directly influence the purchasing decisions that can be made by companies. Legislation, for example the RoHS directive in electronic equipment in the EU, could be introduced for a variety of reasons, such as to improve product safety, remove products that are harmful to the environment or remove products that use unethical production methods. In effect, this reduces the procurement options available to a company, before they can consider their own requirements.

Reputation has been suggested as another key factor that motivates companies to improve the sustainability of their supply chain. It has been suggested that focal companies (those that are seen as governing the supply chain, having the direct contact with customers and designing the product or service) may be held responsible for the impacts of their whole supply chain (Koplin, Seuring et al. 2007) (Seuring and Müller 2008). This is particularly true of companies based upon a strong brand. As such, companies use sustainable procurement as a method of ensuring that their supply chain meet environmental and ethical criteria in order to maintain their reputation.

Customer requirements may also form a key driver to adopt sustainable procurement methods. It has been suggested that consumer awareness is a significant driver in the adoption of more sustainable practices (Dobers and Wolff 2000). In addition, it is thought that corporate buyers are increasingly exerting pressure on their supply chain to improve sustainability performance, especially in the environmental arena (Morton, Green et al. 1996). This pressure may not just be exerted by the immediate customer, but from their customers further up the supply chain (Morton, Green et al. 1996). This need to meet customer requirements for sustainable products and services is likely to force companies to make more sustainable procurement decisions. Pressures from customers may again originate from the customers’ own set of pressures and needs. Thus, the purchasing decisions made by a company do not only reflect its own drivers, but those of others, adding a further set of constraints.

As well as the drivers of the implementation of sustainable procurement including those listed above, there are a number of barriers which discourage companies from sustainable procurement. Three major commonly cited examples of these barriers are increased cost, increased effort and insufficient supply chain communication (Seuring and Müller 2008). Further discussion of barriers and incentives to sustainable procurement will be discussed with more focus on construction later in the paper.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 3. SUSTAINABLE PROCUREMENT IN CONSTRUCTION

3.1Drivers

Recently, attempts to implement sustainable procurement ideas into the construction industry have started to appear. The UK government strategy for sustainable construction lists procurement as one of the ‘means’ to achieve the ‘ends’ of improved sustainability performance (Department for Business Enterprise & Regulatory Reform 2008).

In the development of the strategy, a number of drivers are listed that form the business case for sustainable procurement. These include value for money, reputation issues, market differentiation and regulation and legislation (Department of Trade and Industry n.d.). The influence of these factors on specific construction companies may vary greatly. Suggestion of some reasons for these variations for different factors appears below.

Value for money is suggested as a driver for sustainable procurement as the increased use of whole life costing methods reduces the overall expenditure throughout the life of the building. However, the value of whole life costing is unlikely to be realised by the construction company, or in many cases, the client. The benefit of, for example, reduced running costs of a building may not be seen by the constructor, as they are not the end user. Often, the end user is not the client, adding a further layer of disconnection between the constructor and the user. This could be exaggerated even further if considering end of life costs, as the users at the end of life may not be the same even as the intended users at the construction stage. This may point to a need for significant change to existing business models in order to take a more long-term approach.

Such a change to a long term approach however may not lie within the priorities of construction firms. The issues appear to lie more in the way that clients secure investment to fund construction projects. Changing the finance methods for construction clients is unlikely to be a priority of construction companies. The extent to which this driver influences those who are in the position to make changes in response to it is therefore in need of investigation.

Brand reputation has been discussed in general terms already. It is suggested that focal companies are held responsible for the environmental and social impacts of their whole supply chain, and that this is particularly true of companies with a strong brand, such as clothing companies (Koplin, Seuring et al. 2007; Seuring and Müller 2008). The impact of this will depend largely on the impact of brand image on the construction company. If consumers and clients are keen to use construction companies with a good record of sustainability, it is likely to be in the constructor’s interests to ensure they employ strong criteria throughout their supply chain and maintain a positive reputation. However, this will depend largely on the extent to which construction companies are held to account by clients, investors and shareholders and the extent to which companies are able to measure and publicise their achievements.

Market differentiation may have a large role to play in encouraging companies to adopt sustainable procurement practices. If a company sees an opportunity to market itself as having a completely sustainable supply chain, it may be able to attract business from customers who value this and are looking to do the same. As before, this advantage depends largely on the importance attached by clients to sustainability and their

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ willingness to pay for the perceived benefits. It has been suggested that adopting sustainable construction practices may provide a competitive advantage, but the extent to which this is realised is unknown (Tan, Shen et al. 2011). As the government has pledged to procure more sustainably in future through the sustainable procurement action plan, it would suggest that there is a market for this at least in government funded projects. Client requirements could play a major role here. It is suggested that clients are placing certain requirements into contracts for environmental performance, for example (Sterner 2002).

Government regulation and legislation can also play a key role. If companies are legally required to purchase sustainably, for example, this will force them to look at their supply chain at least to meet the requirements of legislation.

3.2 Barriers

Although there are a number of potential advantages for a company adopting sustainable procurement practices, they are not widely implemented. It has been estimated, for example, that uptake of whole life costing in the private sector lies at around 5% (Department of Trade and Industry n.d.). A number of barriers have been noted in the construction industry when trying to implement other procurement advances which may apply to sustainable procurement. Some of these are discussed here.

In research in which questionnaires were issued to commercial managers within a single contracting company, Wood and Ellis (2005) noted that with respect to partnering relationships in construction, there were instances where, whilst perceptions and even experiences were positive, relationships sometimes remained cost driven, even when there were perceived benefits in a change to the new approach. This experience seems contradictory, in that a methodology with perceived benefits is not taken up or the transition to the new methodology is slow. One reason for this could be ‘inertia’ within the company (Meehan and Bryde 2011). This develops from the institutionalisation of a routine within an organisation. As firms try to maintain a sense of reliability, processes become routine. This results in the implementation of change becoming more difficult, as it relies on an upheaval from the existing routine (Meehan and Bryde 2011).

A further reason could be that the staff involved feel that other pressures placed upon them force them to make decisions that do not align with a sustainable procurement strategy. This could indicate that there is conflict between pressures on staff and they find the greater of these to be driving them towards maintaining the more traditional approach. Such conflicts of incentives may be another barrier to sustainable procurement. This can be despite the fact that many companies are documenting sustainable procurement strategies in their annual reports (eg. Morgan Sindall 2010). It would be interesting to find out to what extent these policies actually influence the procurement decisions within a company. The actual business decisions are likely to be based upon a number of differing pressures, and the way in which such pressures overlap will be key to the final decisions made.

Another barrier could be that the benefits of making sustainable procurement decisions may not be received by those that incur the extra cost. An example is that previously mentioned of whole life costing (Department of Trade and Industry n.d.). Whilst the end user may benefit from, for example, lower energy demand or the final owner may benefit

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ from a design that allows for easy recycling after demolition, the constructor and possibly even the client may not gain any benefit from this. If costs are increased, there may be little motivation in this respect for a company to adopt new practices, moving away from a capital cost based approach.

Further, it has been suggested that, in government procurement for example, one department may incur extra cost that could reduce costs for another department, resulting in an overall cost reduction (Department for Environment Food and Rural Affairs 2007). If a similar principle were applied to a business, one department may be dissuaded from choosing a more sustainable procurement option, despite the overall benefits to the company, in order to keep within departmental budgets. The impact of this will be very much dependent on the organisational structure of the company in question.

Finally, another possible barrier is the motivation of individual staff members to improve sustainability performance. Resistance from employees could work to undermine efforts of the company at a higher level. This resistance could come as a result, for example, of staff feeling that they are being given extra work or having extra pressures put on them. In this case, they may be reluctant to change their practices if they do not see significant reward for their efforts.

These two points indicate that the organisational structure can have significant impacts on the ability and motivation of a company to implement sustainable procurement practices. It would be useful to find out how the organisational structure within companies affects their procurement decisions. This information on the practices within the companies should be useful in identifying structural changes that may be needed in order for new practices to be accepted and successfully implemented.

It is clear that a number of different factors affect procurement decisions, acting as both drivers and barriers to procuring sustainably. It is important for any company to decide what represents sustainable development within its own context. This may well be influenced by what is of value to important stakeholders, the opinions of whom may compel the company to change practices. The investigation of where these influences for sustainability originate, and how these are balanced with other needs within in procurement strategy is an important area for investigation. It may be the case that current procurement and sustainability strategy do not have the same aims, and the ‘hierarchy’ of these influences will be the deciding factor in which strategy actual decisions reflect. As such, understanding not only what drives or restricts sustainable procurement, but relative importance attributed to each factor is very important in order to understand how decisions are actually made.

4. CONCLUSIONS

The discussion in this paper suggests that it will be necessary to understand a number of aspects of the company in order to develop an effective and robust sustainable procurement policy. Whilst the list of issues discussed is by no means exhaustive, it indicates the wide range of factors affecting sustainable procurement. Following the discussion of these presented above, three questions are presented below for use in research within a large construction company, namely Morgan Sindall plc. of sustainable procurement within construction.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

 What factors provide the current motivation for sustainable procurement? Where do these originate?  What organisational culture features in construction companies and how does this affect the decisions they make?  How do strategies of sustainability and procurement interact and are there areas of conflict which prevent the objectives of one strategy being achieved by the need to meet those of the other?

It is believed that by using these questions as a starting point for further research, a better understanding of the practices in and influences on the construction industry can be developed through the research to take place in Morgan Sindall plc. Having acquired this information, it will enable a clearer view of where improvements need to be made within current practices in order to make procurement within the industry more sustainable.

REFERENCES

Commission of the European Communities (2005). Communication from the commission to the council and the European Parliament. Brussels. Department for Business Enterprise & Regulatory Reform (2008). Strategy for Sustainable Construction. London, Department for Business, Enterprise & Regulatory Reform. Department for Environment Food and Rural Affairs (2005). Securing the future: delivering UK sustainable development strategy. Norwich, TSO (The Stationery Office). Department for Environment Food and Rural Affairs (2007). UK Government Sustainable Procurement Action Plan, DEFRA (Department for Environment, Food and Rural Affairs). Department of Trade and Industry. "DTI ‘Strategy for Sustainable Construction'- consultation events: Procurement and whole life costs." Retrieved 4th May, 2011, from http://www.bis.gov.uk/files/file37179.pdf. Dobers, P. and R. Wolff (2000). "Competing with 'soft' issues - from managing the environment to sustainable business strategies." Business Strategy & the Environment (John Wiley & Sons, Inc) 9(3): 143-150. Koplin, J., S. Seuring, et al. (2007). "Incorporating sustainability into supply management in the automotive industry - the case of the Volkswagen AG." Journal of Cleaner Production 15(11-12): 1053-1062. Meehan, J. and D. Bryde (2011). "Sustainable procurement practice." Business Strategy & the Environment (John Wiley & Sons, Inc) 20(2): 94-106. Morgan Sindall (2010). Annual report and accounts 2010. Morton, B., K. Green, et al. (1996). "Purchasing and environmental management: Interactions, policies and opportunities." Business Strategy & the Environment (John Wiley & Sons, Inc) 5(3): 188-197. Norman, W. and C. MacDonald (2004). "Getting to the Bottom of "Triple Bottom Line"." Business Ethics Quarterly 14(2): 243-262. Seuring, S. and M. Müller (2008). "From a literature review to a conceptual framework for sustainable supply chain management." Journal of Cleaner Production 16(15): 1699-1710.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Sterner, E. (2002). "'Green procurement' of buildings: a study of Swedish clients' considerations." Construction Management and Economics 20(1): 21 - 30. Tan, Y., L. Shen, et al. (2011). "Sustainable construction practice and contractors' competitiveness: A preliminary study." Habitat International 35(2): 225-230. Wood, G. D. and R. C. T. Ellis (2005). "Main contractor experiences of partnering relationships on UK construction projects." Construction Management and Economics 23(3): 317 - 325. World Commission on Environment and Development (1987). Our common future. Oxford, Oxford University Press.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

A Consumption And Emissions Model Of An RTG Crane Diesel Generator C. E. Knight1*, V. M. Becerra2, W. Holderbaum3, R. Mayer4 1 Technology for Sustainable Built Environments, University of Reading, United Kingdom 2 School of systems Engineering, University of Reading, United Kingdom 3 School of Systems Engineering, University or Reading, United Kingdom 4CRESS Ltd, Reading, United Kingdom

*Corresponding author: [email protected]

ABSTRACT The UK is increasingly reliant upon the service sector, as over the past few decades the manufacturing capabilities have dropped. This has lead to vast quantities of goods being imported from foreign manufacturers, 95% of these imports is done by sea. The port facilities of the UK operate fleets of Rubber-Tyred Gantry (RTG) cranes that move the containers from ship to lorry or train for land distribution. The cranes that do this are run 22 hours a day for 340 days a year. The need for mobility means that these cranes cannot be connected to an external electricity supply, but must operate islanded power systems. This is achieved through an onboard diesel generator rated at the peak power demand from the crane. The use of these cranes leads to high emission of pollutants into the local environment and vast quantities of fuel use. The current trend of increase of fuel prices around the world means that, port operators are paying more to operate the RTG cranes each year. This paper presents computer model of the diesel generator of a typical RTG crane at the Port of Felixstowe. The model is validated with data provided by Port of Felixstowe. Keywords – Diesel generator; Simulink; synchronous machine;

1. Introduction With the costs of energy increasing, simulation is fast becoming the most suitable way of analysing systems and testing their responses. In fact it is now common practice in many disciplines from engineering to banking [1]. This paper outlines a computer model implemented using Matlab/Simulink [2] that describes the operation of the Diesel Generator of an RTG (Rubber-Tyred Gantry) Crane. The RTG Crane is operated at most major port facilities around the world, the largest capable of lifting 42 tonne loads. The crane is free roaming as such able to move completely independently around the port site, it is used to move shipping containers from storage areas to lorries for transport via road networks. Power systems that operate independently of mains power are known as ‘islanded’ systems [3]. The duty cycle imposed on these machines is very demanding, often rapid cycles of lowering and rising can be followed by periods of idle time. This duty cycle places heavy demands on the Diesel Generator power supply. The supply consumes great quantities of diesel fuel to supply the power to the induction motors used on the crane. The attitudes toward fuel consumption are changing however, with rising costs and a growing awareness of the effects of such consumption [4]. Users are now

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ looking for ways to reduce or remove their dependence on fuel [5-10]. The RTG Crane has great potential for energy storage technologies to reduce consumption given its operation and construction. However their operational demands do not allow for these cranes to be offline for the long periods required to develop an energy storage system and so simulation must be used. To allow testing of the effects of energy storage a computer model will be used. This model will be tuned to respond in the manner of the actual diesel generator this will then allow the implementation of a consumption and emissions model. The consumption and emission data will then allow an analysis of the effects of energy storage systems on the diesel generator. This paper will set out the detail of the Diesel Generator and its computer model which is implemented using Matlab/Simulink’s toolbox SimPowerSystems [11]. This toolbox is created specifically to allow modelling of power electronic systems, it provides models of many 3 phase components such as electrical machines and drives. This will then be validated against real data from the manufacturer, before the consumption and emissions model are described and added to complete the model. 2. Diesel Generator The diesel generator is comprised of 4 major parts:

 Diesel Engine  Synchronous Machine  AVR (Automatic Voltage Regulator)  Speed Regulator

(a) (b) Figure 1. Diesel Generator Components. (a) Engine. (b) Synchronous Machine.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ The two main components are shown in Fig 1 above, being the diesel engine and the electrical machine used by the RTG Crane. The AVR and Speed Regulator control these components to provide the required actions. The diesel generator provides a torque to the electrical machine rotor which causes it to rotate inside the machines electric field. Through the laws of induction a voltage is induced in the stator coils of the electrical machine [12]. This voltage is dependent upon the speed of rotation, which determines the frequency so this must remain constant [13]. The electromagnetic field determines the amplitude of the voltage. From the stator windings this voltage can then be used by a load connected to the electrical machine. There are many performance characteristics for generators; our focus in this case is the power/consumption curve. This curve shown in Figure 2 has been produced from manufacturer data shown in Table 1.

Fuel System Load RPM 1500 100 90

80 Prime Power 25% g/KWh 227 70 60 50 Specific Fuel 50% g/KWh 203 40 Consumption 30 75% g/KWh 198 20 at: 10

0 Consumption (l/hr) Consumption 100% g/KWh 200 0 25 50 75 100 Ouput Power (% Load)

Table 1. Fuel consumption data for engine. Figure 2. Graph of converted consumption values. The data presented above shows that the diesel engine’s fuel performance varies greatly with load, with 100% load the engine is consuming in excess of 90 litres of diesel an hour. In reality the current fuel consumption rate is constantly changing due to the duty cycle and the containers being of different weights. The model will need to simulate the diesel generator performance in real time under known load; the power output of the engine will be monitored and the data from Figure 2 will be scrutinised to gain a current consumption rate. 3. Model The model of the diesel generator was implemented using SimPowerSystems, which is a toolbox within Matlab/Simulink. Standard blocks were used for simulation of the diesel generator, their parameters were then changed to represent the actual machine. The most critical block shown in Figure 3 is the synchronous machine [14], this model uses pu (per unit) values rather than SI values [15]. This meant some parameters like the stator resistance had to be calculated.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Figure 3. The Symbol and Electrical Model of a Synchronous Machine. The settings for the machine required by Matlab were taken directly from manufacturers data, as such assumed to be accurate at the time of construction. The parameters are shown entered in Matlab in Figure 4.

Parameter Units Value Nominal Power VA 500e3 Line-to-Line Voltage Vrms 380 Synchronous Reactance (d-axis) Hz 50 Transient Reactance (d-axis) PU 2.47 Subtransient Reactance (d-axis) PU 0.14 Synchronous Reactance (q-axis) PU 1.48 Subtransient Reactance (q-axis) PU 0.17 Leakage Reactance PU 0.08 Transient Short Circuit Time Constant (d-axis) S 0.185 Subtransient Short Circuit Time Constant (d-axis) S 0.025 Subtransient Open Circuit Time Constant (q-axis) S 0.025 Stator Resistance PU 0.035 Inertia Coeficient Kgm2 18.07 Friction Factor PU 0 Pole Pairs p 2

Figure 4. Electrical Machine values as entered in Matlab. The diesel engine does not have a standard block form in SimPowerSystems, we may model it using standard blocks though. The critical information required from the engine is its power output, from this the consumption and emissions can be obtained. The model of the diesel engine component, Figure 5, is greatly simplified, the Speed Regulator maintains the speed at 1500 rpm (Revolutions Per Minute). An actuator is used to represent the engine itself, it takes the output from the speed regulator and calculates an appropriate a torque response. This is put through a time delay, to simulate the time delay observed in the actual engine.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Figure 5. Diesel Engine model. The output from the diesel engine section is a power quantity in pu, this forms the input to the electrical machine model described above. This power signal is then used to calculate the electrical output of the diesel generator.

Figure 6. Combustion and Emission model.

The combustion and emissions model, Figure 6, implements the consumption rate against load curve of Figure 2. The model takes in the measured power delivered by the engine and outputs a rate of fuel use. A simple gain block takes this combustion rate and calculates the emission rate according to (2).

2.68kg/CO2 per Litre Diesel Consumed (2) As the model is designed to work in real time simulations the output from the model is a rate of consumption not actual consumption. The data output will be stored as a Matlab variable during the simulation. Once complete the data is plotted and integrated to produce the data from which the performance of energy storage can be evaluated. 4. Results The model was tested against manufacturer data, shown in Figure 7, the 3 phase fault data for the electrical machine was chosen as a validation test set.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Figure 7. 3 Phase Fault response. The data shows that under 3 phase fault conditions the current output will rise to 7kA, a rapid decay over 0.25s will see current fall to 800A, after a second the current will reach steady state at 2,900A. A simple Matlab block implements a 3 phase fault, the simulation was set up to run for 11 seconds with the fault occurring after a second. The instantaneous currents and voltages from the electrical machine are monitored and the rms (Root Mean Square) equivalent is found [12]. This value was calculated using a Function block implementing Equations (3) and (4)

√( ) (3)

√( ) (4)

where and are the rms voltage and current respectively, Vabc and Iabc are the instantaneous voltages and currents on each phase. The simulation was run using the ode23tb, Matlab has many ODE (Ordinary Differential Equation) solvers, 23tb uses a Runge-Kutta formula to resolve the equations [2][16]. This solver is used for ‘stiff’ systems, i.e. systems with time constraints of different orders of magnitude. A maximum step size has to be specified in the settings to prevent Matlab increasing step size during steady state periods which causes inaccurate results.

Figure 8. Model response to 3 phase fault.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Figure 8 shows the response of the model to the 3 phase fault after a second. The current reaches peak at 7950A, before settling to steady state at 2920A. The settle time is as expected from the response however the amplitudes in the transient stage is not. This discrepancy is due to the excitation system, which contains the AVR used to control the electrical machine. At its default settings it is set up to regulate much larger machines than the one simulated in this project. The excitation system was updated with settings for a machine of similar size to that used in this application. The results of this, Figure 9, show a much closer correlation to the desired response.

Figure 9. 3 phase fault response, undated excitation system. The transient condition now more closely represents the actual response of the machine, tuning of the excitation system will allow more accurate transient state representation. The results to both show excellent agreement between the model and the validation data in the steady state condition. The duty cycle imposed on the crane is dominated by periods of steady state behaviour, so this allows us to begin basic analysis of the diesel generator with duty cycle data. The initial testing on the combustion model was done with a simple duty cycle, often the container to be moved is below others in a stack, this requires the crane to move others to reach the desired container. The duty cycle to represent this consists of a series of lifts, it has been modelled using a repeating active power demand of variable size. For simplicity loads were selected to demonstrate the combustion models ability to represent this as a combustion rate. Figure 10 shows the output from Matlab for this simulation. The first two peaks show a noticeable step in the falling side of the response, this is due to the engine meeting the power demand during the time it is demanded so rate lowers and then the demand being removed the rate falls to zero more quickly. The final three peaks have too great an angle to show this effect. The time of the power demand was longer than the time constant of the diesel engine ensuring an appropriate response from the model.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Figure 10. Plot showing combustion rates for increasing loads. Figure 10 shows that the model is capable of representing the typical spectrum of combustion rates observed on the actual cranes. 5. Conclusion In this paper a diesel generator system and its function was outlined, a model was then presented using the Matlab/Simulink environment. The conditions for the model were outlined as well as the evaluation techniques. The model was tested using 3 phase fault data from the manufacturer, as such confidence in the models steady state response is high. An outline for a system to calculate the combustion and emission rates for the engine has been explained and implemented on the model. The combustion model was tested with a basic duty cycle taken from realistic crane data and has shown realistic combustion rates can be seen for the correct loading. Future work will include tuning the excitation system to gain accurate transient state analysis and validation of the combustion model against data recorded from the crane during its working duty cycle. 6. References [1] Ortjohann, E.; Mohd, A.; Schmelter, A.; Hamsic, N.; Lingemann, M.; , "Simulation and Implementation of an Expandable Hybrid Power System," Industrial Electronics, 2007. ISIE 2007. IEEE International Symposium on , vol., no., pp.377-382, 4-7 June 2007.

[2] Sybille, G. Brunelle, P. Giroux, P. Casoria, S. Gagnon, R. Kamwa, I. Roussel, R. Champagne, R. Dessaint, L. Lehuy, H.; 2003; “SimPowerSystems: For Use with Simulink”; TransEnergie Technologies Inc; http://www.mathworks.com/help/releases/R13sp2/pdf_doc/physmod/powersys/powersys. pdf

[3] Best, R.J.; Morrow, D.J.; McGowan, D.J.; Crossley, P.A.; , "Synchronous Islanded Operation of a Diesel Generator," Power Systems, IEEE Transactions on , vol.22, no.4, pp.2170-2176, Nov. 2007.

[4] Sandip D. Shah, David R. Cocker III, Kent C. Johnson, John M. Lee, Bonnie L. Soriano, J. Wayne Miller, Emissions of regulated pollutants from in-use diesel back-up generators, “Atmospheric Environment”, Volume 40, Issue 22, July 2006, Pages 4199- 4209, ISSN 1352-2310.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

[5] L. Leclercq, B. Robyns, and J. Grave, Control based on fuzzy logic of a flywheel energy storage system associated with wind and diesel generators, “Mathematics and Computers in Simulation”, 2003, pp.271-280.

[6] J. Kondoh, I. Ishii, H. Yamaguchi, A. Murata, K. Otani, K. Sakuta, N. Higuchi, S. Sekine, M. Kamimoto, Electrical energy storage systems for energy networks, “Energy Conversion and Management”, Volume 41, Issue 17, 1 November 2000, Pages 1863- 1874, ISSN 0196-8904.

[7] Ribeiro, P.F.; Johnson, B.K.; Crow, M.L.; Arsoy, A.; Liu, Y.; , "Energy storage systems for advanced power applications," Proceedings of the IEEE , vol.89, no.12, pp.1744-1756, Dec 2001 doi: 10.1109/5.975900

[8] Mellor, P.H.; Schofield, N.; Howe, D.; , "Flywheel and supercapacitor peak power buffer technologies," Electric, Hybrid and Fuel Cell Vehicles (Ref. No. 2000/050), IEE Seminar , vol., no., pp.8/1-8/5, 2000 doi: 10.1049/ic:20000268

[9] Barton, J.P.; Infield, D.G.; , "Energy storage and its use with intermittent renewable energy," Energy Conversion, IEEE Transactions on , vol.19, no.2, pp. 441- 448, June 2004

[10] Hebner, R.; Beno, J.; Walls, A.; , "Flywheel batteries come around again," Spectrum, IEEE , vol.39, no.4, pp.46-51, Apr 2002.

[11] Sybille, G.; Brunelle, P.; Hoang Le-Huy; Dessaint, L.A.; Al-Haddad, K.; , "Theory and applications of power system blockset, a MATLAB/Simulink-based simulation tool for power systems," Power Engineering Society Winter Meeting, 2000. IEEE , vol.1, no., pp.774-779 vol.1, 2000.

[12] Irwin, J.D; Nelms, R.M; 2005; “Basic Engineering Circuit Analysis”; 816 p; John Wiley and Sons; USA; ISBN 9780470128695.

[13]Hughes, A; 1993; “Electric Motors and Drives: Fundamentals, Types and Applications”; 352 p; Newnes: ISBN 0750617411.

[14] Wilde, T; 2005; “Electrical Machines, Drives and Power Systems”; 960 p; Pearson Education; ISBN 0131969188.

[15] Mohan, N. Undeland, T. Robbins, W; 2002; “Power Electronics: Converters, Applications, and Design”; Wiley; ISBN 0471226932.

[16] Stroud, K.A. Booth, D.J.; 2007; “Engineering Mathematics”; Palgrave Macmillan; ISBN 9781403942463.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Challenges In Intelligent Management Of Power And Cooling Towards Sustainable Data Centre

S. Luong1*, K. Liu2, James Robey3 1Technologies for Sustainable Built Environments, University of Reading, United Kingdom 2Informatics Research Centre, Henley Business School, University of Reading, United Kingdom 3Capgemini UK, Woking, Surrey

*Corresponding author: [email protected]

ABSTRACT Power consumption in data centres has been rapidly increasing and based on current trends this will inevitably cause major challenges for both data centre designers and operators. More than a decade ago, a standard server cabinet housed computers using single core processors drawing a low (5kW) amount of power. As each cabinet contained one or two computers the energy required to provide conditioned air was not a major concern for data centre operators. In contrast, in 2011, a server cabinet of the similar dimension can house up to 128 blade servers making them more densely populated therefore drawing more power and increasing heat dissipation per square meter. Due to the increase in heat dissipation the data centre cooling systems are at risk of being unable to adequately provide enough cooling. Both increased density in computer hardware and cooling demands are having a huge impact on power consumption. This paper examines the challenges data centres are facing regarding power and cooling management. It also makes references to the environmental requirements and specifications for data centres recommended for organisations based on best practices. We also look at one of the current practices for building a sustainable state-of-the-art data centre and finally present a recommended approach to reducing power consumption in cooling systems.

Keywords: Data centres, sustainability, power, cooling, intelligent management

1. INTRODUCTION

Data centres around the world are encountering power, cooling, space and environmental issues whilst supporting the growth of an organisation. Due to the demands of more processing capability and storage space data centres are on the verge of running out of floor space and consuming more electrical power than it can be supplied with. As IT equipment become more compact, they draw more electrical power per sq. feet. Data centres consume large amounts of electricity and it has been estimated that the total magnitude is believe to be between 1.5 per cent and 3 per cent of total electricity generated. In addition, the cost of powering and cooling IT equipment for three years is equivalent to “1.5 times the cost of purchasing server hardware” (Brill, 2007). According to a report written by Sheihing (2009) the cost of powering data centres in the United

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ States cost $4.5 billion and this figure is predicted to grow 12% per year. And, according to the Code of Conduct on Data Centres Energy Efficiency (European Commission, 2008) data centres in Western Europe had consumed 56 TWh of electricity in the year 2007. If data centre power consumption follows the predicted unsustainable trend then there would be a serious power shortage as energy suppliers will not be able to cope with such demands.

The management of power and cooling towards a sustainable data centre is a combination best practices, hardware and software. Best practices include following standards, guidelines and using the best approaches. Hardware involves the selection of the IT equipment to install in the data centre and software comprises of the technologies used for the management of power and cooling. There needs to be a balance between the three categories in order to achieve energy efficient and sustainable data centre. For example, a data centre professional might design a new data centre to include the best practices but must also consider there could be potential for the data centre to use heat reclamation technology that is further enhanced by a cheap software solution.

In this paper, we will discuss the challenges of power and cooling in section 2, examine the environmental requirements and specifications in section 3 by exploring how the requirements have changed previously to help data centres reduce power consumption and better manage their cooling systems. Section 4 describes how a state-of-the-art data centre utilise best practices and standards to construct the leading sustainable data centre. Sections 5 present a recommended approach to the intelligent management of power and cooling, and finally conclude the paper in section 6.

2. THE CHALLENGE OF POWER & COOLING

There has been a dramatic increase in the number of organisations adopting environmental sustainable policies in order to be environmentally friendly and ultimately reduce their operation costs. One of the key factors that are always taken into consideration when an organisation implements their strategy is reducing IT operation costs. Data centres rely on power availability and transmission capabilities, which are generally being affected by increasing demand for more computing power. In United States, the energy used by data centres had more than doubled between the year 2000 and 2006 as illustrated in Figure 1 below.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Figure 1 Past and Projected Electricity Use by Data Centres in the U.S. (U.S. Environmental Protection Agency, 2007)

Figure 1 also shows predicted scenarios so that comparisons can be made on the future of energy use. Based on current historical trends, in the year 2000, data centre energy usage was approximately 30 billion kWh/year and in 2011 is predicted to peak over 120 billion kWh/year. Clearly the Historical Trend scenario is highly unsustainable due to the cost of maintaining data centre operation and the lack of power stations to provide enough electricity. Organisations would want to try to achieve the best practice or state of the art scenario, which is to significantly reduce their energy consumption.

There are many challenges a data centre operator will constantly encounter due to the ever changing technology. Some of the common challenges include new system deployments or upgrades, scalability and life cycle costs. To elaborate on challenge of deploying new systems or upgrading them, in the past, when demand was not high it was very easy to commission new servers and add more CRAC units inside the data centre. The energy cost for powering and cooling servers was not a major issue for organisations. However, today’s IT equipment is considerably more compact and delivers greater performance than a decade ago causing the power density to increase dramatically. Because technology is always advancing a data centre should be designed to be able to adapt and scale to accommodate these changes. But realistically, it is difficult to determine the numbers of years a data centre should be designed for knowing that in future high performance computing will always need more energy.

The major challenge for data centre operators is how to cool densely packed IT equipment. A decade ago, a server rack’s power design was approximately 5kW. This is expected to rise to 37kW by the year 2014. The more power each rack consumes, the more heat is generated. Essentially, including investments in advanced cooling solutions there has to be a trade-off between maximising the floor space with many low-

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ performance servers or have fewer and more powerful servers so that less floor space is used. The technical challenge here is how to deliver cooling to or how to remove heat from the IT equipment. We will consider the two solutions and the technical challenge involved in air and water cooling.

Air cooling involves cooling the entire facility and server platform with air. This normally involves a direct expansion (DX) chiller, air-side economiser and fans all of which increases the energy consumption in a data centre. However, some of this equipment such as using an air-side economiser to exhaust hot air out of a data centre maybe costly and difficult to retrofit. They often do not have a good return on investment if the initial data centre design did not include this type of equipment. On the other hand, if a data centre was designed to maximise the equipment’s usage they can be very efficient. The other limitation that affects air cooling and air flow to the server racks is the practical constraints in the existing design of a data centre. For example, if a data centre has a raised floor plenum height of 24 inches with under floor cabling system causing obstruction these will obviously set an upper boundary on the air cooling. It is difficult to determine the upper boundary in data centres due to the lack of real-time airflow values but instead data centre operators determine this based on their operational experience. As for cooling the actual server racks, the manufacturers typically provide airflow data to inform data centre operators how much cooling the server rack needs. Ultimately, the ability to optimise the data centres capacity to provide cool air and reduce power consumption is down to how it is configured, i.e. using hot aisle/cold aisle configuration (Wang, 2006), following best practices on placement and proximity of perforated tiles, and using blanking plates and cut-outs to prevent hot and cold air mixing. To further optimise the air cooling capacity containment doors and roofs are used for either the hot or cold aisle (U.S. Department of Energy, 2011). But all of these enhancements come at an additional cost and complication.

The alternative, water cooling, aims to bring cooled liquid closer to the server racks. Water is a significantly better thermal conductor than air (Engineering Toolbox, 2011) and so it should be more effective at heat removal. There are several variants of liquid cooling ranging from chip/rack level cooling to liquid-cooled door that is positioned behind the racks. Chip and rack level liquid cooling are clearly the best solutions for removing heat especially from the CPU. But this type of solution is highly complex and expensive which would not be considered a mainstream solution but possibly for situations where many high performance server racks have to be densely packed together. Regardless of which liquid cooled variant is used there still needs to be some airflow to cool the other components on the server platform and for the actual heat exchange between server rack and the liquid cooling system. Liquid cooling solutions can be simply retrofitted into data centres with the benefit of a raised floor plenum but ultimately there is the potential risk of the fluids evaporating or leaking onto the IT equipment causing

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ serious physical damage. As with all liquid cooling solutions there is the additional cost of installing a leak detection system. There are many discussions (Ellsworth, et al., 2008; Sharma, et al., 2009) around the efficiency of liquid cooling as at the basic level in air cooling there is a heat exchange using liquid in the chillers. But with water cooling we are bringing that heat exchange closer to the racks to better cool the servers. But, as mentioned previously, there still needs to be some airflow to cool the entire server platform.

3. ENVIRONMENTAL REQUIREMENTS & SPECIFICATIONS

Before 2002, the thermal designs for IT equipment was rather complex due to the lack of standards and environmental specification. Equipment vendors manufactured their products based on their own specifications so data centres generally had to accommodate varying environmental requirements, which made cooling and heat removal more challenging. Post 2002, data centres continue to accommodate products from different vendors. A technical committee formed by the American Society of Heating, Refrigeration and Air Conditioning Engineers (ASHRAE, 2002) published “Thermal Guidelines for Data Processing Environments” that maintains an Equipment Environment Specifications, which is the recommended operating envelope and now a widely accepted standard for manufacturing IT equipment for data centres. This ensured that participating vendors manufacture their products within the specified design envelope therefore alleviating some of the cooling challenges data centres were facing. Furthermore, in order to help data centres reduce their energy consumption the Equipment Environmental Specifications had been revised as shown in table 1.

2004 Version 2008 Version Low End Temperature 20oC (68oF) 18oC (64.4oF) High End Temperature 25oC (77oF) 27oC (80.6oF) Low End Moisture 40% RH 5.5oC DP (41.9oF) High End Moisture 55% RH 60% RH & 15oC DP (59oF)

Table 1 Comparison of 2004 and 2008 recommended operating envelope

In 2004, the temperature and humidity boundaries for IT equipment was initialised allowing manufacturers to test their products within that design envelop to ensure the equipment operates reliably within those boundaries as high operating temperatures may cause hardware failure or reduced reliability. In 2008, the recommend operating envelope was revised to help reduce power consumption. Although, this does not ensure optimum energy efficiency it does offer a window of opportunity to save energy in contrast to the

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 2004 version. This can be achieved by lowering the air optimiser fan speeds or the DX (direct expansion) chilling units to allow the ambient temperature in the data centre to be raised.

Improvements made to IT equipment such as variable fan speed offers data centres the flexibility to operate at the recommended temperatures. The temperature of the components on the IT equipment is affected by the ambient temperature and the component temperature tracks closely to the ambient temperature. For example, based on a scenario that the equipment uses a constant fan speed at max power an inlet ambient temperature of 17oC would make the component temperature to be 40oC and an inlet ambient temperature of 38oC would make the component temperature to be 60oC. Considering in this scenario that 60oC is the allowable temperature range for reliable operation a variable fan speed can operate at a slow fan rate at lower temperatures to save energy. Between the inlet ambient temperatures of 16oC to 25oC the fan rate is on low and the component temperature remains at 60oC but beyond 25oC the fan rate increases to maintain a constant component temperature of 60 oC therefore not affecting the reliability of the components. Overall, a variable fan speed would have saved more energy during the 16oC to 25oC operation temperature in comparison to a constant fan speed.

As with the temperature, raising the boundary limits for relative humidity could affect the performance and reliability of the components (ASHRAE, 2008). High relative humidity causes conductive anodic filament failure to printed circuit boards. In addition, high relative humidity and common atmospheric contaminants causes hydroscopic corrosion. Low relative humidity causes electrostatic discharge (ESD) and very high voltages can build up in very dry environments. In severe instances, ESD can damage sensitive electronics causing hardware failures.

Following the standards and guidelines had initially helped data centres tackle some of the cooling challenges but as power density increased over the years the recommended operation temperature guidelines was revised to help reduce power consumption. As IT vendors continue to improve their manufacturing process and produce more durable equipment it is expected that the Equipment Environmental Specification is likely to be revised again so that data centres can further improve their energy efficiency and have more flexibility to how they accommodate IT equipment.

4. THE STATE-OF-THE-ART COOLING AND POWER REDUCTION

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ The efficiency of a data centre is measured by the Power Usage Effectiveness (PUE) metric (The Green Grid, 2007). PUE is a ratio of the total power consumed by a data centre over the power consumed by the IT equipment. The theoretical minimum PUE value is 1.0 which means for every watt of IT power, no additional watt is consumed for cooling or distributing power to the IT equipment. The report to the U.S. Congress (U.S. Environmental Protection Agency, 2007) listed four categories of data centre efficiency as shown in table 2 below.

Scenario Current Trends Improved Operations Best Practices State-of-the-Art PUE 1.9 1.7 1.5 1.4

Table 2 Data Centre Efficiency Targets (U.S. Environmental Protection Agency, 2007)

In July 2010, the industry average data centre PUE measurement was 2.5 (Miller, 2010). During 2010, IT firm Capgemini constructed a new highly resilient (Uptime Institute, Tier 3 certified) data centre called Merlin (Capgemini, 2010), which achieved a PUE of 1.11 largely driven by its use of fresh air cooling within a modular approach. Other factors contributing to the industry leading PUE and sustainability credentials included its location which enabled the use of fresh-air cooling, and its use of renewable and reusable components. Merlin can accommodate up to 12 modular data halls within a 3,000m2 brown-field warehouse. Each module has the capability of housing 104 racks.

Figure 4 illustrates the floor plan of one of the data centre module. The centre design had followed the approach of a front-to-back configuration to ensure the separation of hot and cold air. Rather than using an under floor plenum to supply cold air and letting hot air naturally rise to the ceiling, the module is on a single floor with the cold air supply contained and fed directly to the front of the cabinets while the hot air is ducted and circumvented directly away to the Air optimiser or exhausted out of the module. Essentially, both cold air and hot air are contained and ducted directly to and from the IT equipment to maximise the module’s potential in delivering cool air and removing heat from the source. Air temperature, humidity and velocity sensors are located in each cold aisle and are connected to a building management system (BMS). The cold aisles have doors with louvers which the BMS sets to control how much cold air should be supplied to each of the four cold aisles. Ultimately energy is saved through the use of 12 variable speed fans which are adjusted by the BMS to meet the server cooling demands. The climate controlled Module positioned at left side of the module is controlled by the BMS which supplies air at the right temperature. Because this is a modular data centre, if required, another Module could be attached to the right side to supply cold air.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Figure 4 Merlin data centre module floor plan (Capgemini, 2010)

Merlin’s use of free fresh air cooling carefully channelled through the IT equipment is critical to it achieving its low PUE.

5. RECOMMENDED COOLING METHODOLOGIES

As mentioned in the introduction, achieving a sustainable data centre is a combination of best practices, hardware and software. Data centre professionals design the infrastructure based on best practices, guidelines and standards but ultimately for the purpose of distributing power to the IT equipment and to provide an environment specifically for cooling high performance computers. Clearly, there has been major advancements in all three categories: best practices are starting to include the use of free cooling to supply cold air or photovoltaic panels for generating electricity on-site, variable speed fans on server components or water cooling heat sinks pre-built onto the processors for quick and easily installation, and intelligent BMS software for climate control.

In future work, we will examine the optimal density of server racks and how it affects the whole cooling process, energy consumption, total cost of ownership and data centre design in general. This paper discussed the challenges of cooling a data centre and the standards or practices involved which raises the research question how close should the server racks be installed together in both air and water cooling solutions; are organisations willing to upgrade to a more powerful server if they knew this would help tackle the challenge of power and cooling in data centres thus reducing their total cost of ownership.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ We will also research the use intelligent agents to improve power distribution and cold air delivery to the IT equipment. There is a lot of research into the use of intelligent agents or multi-agent systems for building control with the emphasis of minimising energy consumption and provide a comfortable environment for human users (Duangsuwan & Liu, 2009; Wang et al., 2010; Wang et al., 2010). However, there appears to be a gap in this field where intelligent agents are not applied to industrial buildings that do not accommodate human users but only IT equipment. There is a lot of potential where building control in data centres could be optimised or made intelligent and giving it the ability to learn to improve power consumption and cooling systems. The following steps will require a survey as to why data centres has not explored the use of intelligent agents and how intelligent agents could be a possible solution to help tackle the on-going challenge of managing power and cooling.

6. CONCLUSION

Data centres are increasingly becoming challenging to maintain due to the increasing density of IT equipment, the difficulties in heat dissipation and heat removal from the source. Organisations are usually constrained by their existing data centre facilities and therefore are looking for cost effective solutions to ensure their IT services can be maintained at a lower cost using less energy and minimising the amount of carbon emissions produced. Designing a sustainable data centre is a combination of best practices, hardware and software.

The business challenge across the data centre industry is trying to keep energy expenses and carbon emissions low whilst maintaining a reliable service and delivering increasing demands for computing power. As technology continues to decrease in size, increased performance will be more densely packed to each server rack requiring more energy and generating more heat. Both air and water cooling solutions have technical challenges and limitations. When trying to improve the cooling system in an existing data centre, there is the risk of the equipment not performing to expectations due to incompatibility with the data centre design. The existing infrastructure and the initial design considerations (e.g. the choice of height for the under floor plenum) will have already determined the upper boundary (in terms of heat dissipation) and lifespan of the data centre. It then becomes a question of how long before IT equipment reaches the data centre’s upper boundary when it can no longer deliver sufficient cooling and therefore forcing an organisation to consider building an additional data centre.

The environmental specifications for data centres were first defined in 2004 when data centre designers and IT vendors voluntarily participated in using the standard to help

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ reduce the challenge of power and cooling. In 2008, the specifications were revised. IT hardware vendors are now manufacturing more durable hardware which allows data centre operators to raise the ambient temperature and humidity levels inside its data centres enabling the reduction of power consumption and carbon emissions. It is expected the specifications will be revised again in the near future as IT vendors manufacture more durable hardware able to handle hotter environments.

The state-of-the-art data centre constructed by Capgemini has achieved a world leading PUE of 1.11 by innovatively designing a modular data centre combined with best practice air handling enabling free fresh air cooling. Data centres continue to improve through the various ways of utilising best practices and approaches, and being selective about the IT equipment deployed. Future work will involve researching the optimal density of server racks and the use of intelligent agents for building control in data centres.

REFERENCES

ASHRAE, 2004, Thermal Guidelines for Data Processing Environments, ASHRAE Publication.

ASHRAE, 2008, 2008 ASHRAE Environmental Guidelines for Datacom Equipment – Expanding the Recommended Environmental Envelope.

Brill, K. G., 2007, Data Center Energy Efficiency and Productivity, White Paper, The Uptime Institute, Inc., The Uptime Institute Symposium 2007: The Invisible Crisis in The Data Center: How IT Performance is Driving the Economic Meltdown of Moore’s Law.

Capgemini 2010, A Closer Look at Merlin – Technical specifications for the world’s most sustainable data centre, Capgemini UK.

Duangsuwan, J. & Liu, K., 2009, Normative Multi-Agent System for Intelligent Building Control, 2009 Pacific-Asia Conference on Knowledge Engineering and Software Engineering, IEEE, 978-0-7659-3916-4/09.

Ellsworth, M. J., et al., 2008, The Evolution of Water Cooling for IBM Large Server Systems: Back to the Future, 11th Intersociety Conference on Thermal and Thermomechanical Phenomena in Electonic Systems, 2008. ITHERM 2008, 978-1- 4244-1701-8.

European Commission, 2008, Code of Conduct on Data Centres Energy Efficiency Version 1.0.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Miller, R., 2010, How A Good PUE Can Save 10 Megawatts, Data Center Knowledge, 13th September 2010, [online] Available at: www.data centerknowledge.com, [accessed: 29/5/2011].

Scheihing, P., 2009, DOE Data Center Energy Efficiency Program, U.S. Department of Energy, Energy Efficiency and Renewable Energy.

Sharma, R., et al., 2009, Water efficiency management in datacenters: Metrics and Methodology, IEEE International Symposium on Sustainable Systems and Technology 2009, ISSST ’09, 978-1-4244-4324-6.

The Green Grid, 2007, The Green Grid Data Center Power Efficiency Metrics: PUE and DCIE, Technical Committee White Paper.

U.S. Environmental Protection Agency, 2007, Report to Congress on Server and Data Center Energy Efficiency Public Law 109-431, ENERGY STAR Program, August 2, 2007.

Wang, D., 2006, Cooling Challenges and Best Practices for High Density Data and Telecommunication Centers, Proceedings of HDP’06, IEEE, 1-4244-0489-4/06.

Wang, Z., et al., 2010, Multi-Agent Control System with Intelligent Optimization for Smart and Energy-Efficient Buildings, IECON 2010 – 36th Annual Conference on IEEE Industrial Electronics Society, IEEE, 978-1-4244-5226-2/10.

Wang, Z., et al., 2010, Multi-Agent Intelligent Controller Design for Smart and Sustainable Buildings, Systems Conference, 2010 4th Annual IEEE, IEEE, 978-1- 4244-5883-7/10.

Engineering Toolbox, 2011, Thermal Conductivity of some common Materials and Gases, The Engineering ToolBox, [online] Available at: www.engineeringtoolbox.com, [accessed: 29/5/2011].

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

A Methodology To Quantify The Environmental Impacts Of The Microsoft Windows Operating Systems

Daniel R. Williams*, Yinshan Tang

Technologies for Sustainable Engineering University of Reading, Physics Building, Reading, RG6 6AF, United Kingdom

*Corresponding author: [email protected]

ABSTRACT The Operating System (OS) that manages computer hardware has large energy saving potentials; however no energy or cost quantification model exists. Microsoft Windows is estimated to be installed on more than 1 billion computers worldwide with a market share of over 89% (Market Share, 2011). Each computer uses electrical power within the home and organisations and thus is a significant power consumer and indirect carbon emitter. This research presents a model that quantifies the energy consumed by a PC when using different versions of Microsoft Windows. The model focuses upon the benefits of using Windows 7 and factors in advanced power management and power saving features. The model is setup so that any OS type can be analysed and thus forms a methodology framework from which to build further. The model uses detailed scientific assumptions and empirical data from various studies. The model has already been successfully implemented as a driver to reduce power consumption in a case study. Using the model on an UK company estate of 50,000 PCs has revealed that over 22,000 tonnes of CO2e could be saved by simply turning on the power management features of Windows 7 and using efficient hardware.

Keywords: Environment, green IT, software, carbon, operating system.

1. INTRODUCTION Information and Communication Technology (ICT) activities produce approximately 2-

3% of the world’s manmade CO2e (Gartner, 2007) and thus is a prime target for energy reduction and improved process efficiency. Most computer systems are powered by an Operating System (OS) which manages energy, hardware and provides the foundation on which to run applications. Microsoft Windows is the most populous OS in the world (Market Share, 2011) and thus can be considered as presenting a prime energy saving opportunity.

Microsoft released the Windows 7 OS with two key power saving technologies; improved power management and more efficient process and component management (Microsoft,

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 2009b). Microsoft has attempted to improve the efficiency of software processes and the energy management of hardware components. Combined, these improvements allow the computer to process information more efficiently thus providing the ability to save energy.

Power management techniques are not a new concept however Windows 7 is the first Microsoft OS to come pre-set with power management turned on. This simple and subtle change creates a default energy saving state without the necessity for user intervention. Studies have found that computers within an organisation and at home often have power management features turned off (Blackburn and Collins, 2009). A key feature of Windows 7 is the enforceability of power management within an enterprise scenario, as granular power management has been included within the ‘group policy’ management tool. Therefore any computer running within an enterprise can be centrally setup to follow power management rules and thus can be enforced.

Computer hardware and OS configurations can be highly flexible and thus energy consumption varies accordingly. Hardware components are designed with energy management features which can be controlled by the OS. Developments in hardware and software design mean that advanced savings can be achieved when the latest hardware and OS are combined. Therefore providing a scenario-independent quantity of the amount of energy that can be saved by using these features is not possible.

A model (Microsoft, 2009a) developed by for the Microsoft Corporation attempted to estimate the energy savings that could be realised by using the Windows Vista OS compared to older versions of the Windows OS. The model does not include specific power management variables and assumes a best practice scenario. This model however provides a good foundation to build a scientific, detailed and academic methodology and model.

The aim of this research is to create a methodology which quantifies the energy use and savings that can be realised when using the Windows OS and its various power saving features. Four versions of the Windows OS will be included to provide an estimate of the

energy and thus environmental impacts (CO2e) that each creates focusing upon the innovative features of Windows 7. Hardware variations will also be factored into the methodology to take account of the latest hardware design and power saving features. The cost implications of energy use will also be included within the model to highlight potential savings that could be realised.

A Windows OS energy modelling tool will be developed to use the developed methodology providing a graphical visualisation interface to analyse PCs within an enterprise scenario in order to optimise energy use and reduce environmental impacts via OS implementations. The model scenario will be based upon upgrading a PC estate with previous versions of the Windows OS to Windows 7 over a 3 year time period. The modelling tool will be aimed at two audience levels, technical experts and a general business manager. Therefore a goal of the model is to split the model variables and

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ assumptions into two tiers of detail. The first tier to offer a basic level of input which relies upon model assumptions, the second tier allowing scenario specific details and assumptions to be used in order to refine results further.

2. METHODOLOGY & MODEL In order to structure the methodology and thus the overall model, the model structure used by Microsoft (2009a) was analysed and used as a foundation. This foundation was built upon with detailed and scientific model variables to determine the power usage of hardware when using a Windows OS and its power saving features.

In order to appeal to the two different target audiences, two tiers of information were determined. The first tier contains the most basic inputs needed to complete the model and relies on the second tier for assumptions and data. The second tier contains empirical data and assumptions that can be utilised or changed if certain details are known to the scenario.

The following sections list models tiers and the variables included within each. Detailed calculations and data are not included within this report due to its technical complexity and overall size.

2.1. Tier 1 Tier 1 contains variables that must be entered according the scenario being analysed. If only Tier 1 information is entered, the model can use Tier 2 assumptions to calculate a model result.

2.1.1. PCs Installed This model variable allows information about the current number of Desktop, Laptop and Thin Client PCs in use to be specified. PCs in this case are defined as those machines that are currently running a Microsoft Windows OS that will be included within the scenario analysis. The difference between the power consumption of an average Desktop and Laptop is considerable due to the hardware technologies employed. Generally a Laptop has lower power consumption than a Desktop PC (Energy Star, 2010). Thin Clients can have very low power consumption and are often used in business situations where only simple tasks are being completed.

2.1.2. Monitor Information Information about the number and type of monitors currently in use was factored into the model as energy usage by a CRT monitor can dwarf that of an equivalent size LCD monitor. The monitor count includes those used by Desktop PCs and monitors used to extend a Laptop, however inbuilt laptop monitors are not included in this section.

2.1.3. OS Profile A PC estate OS profile is required to calculate current and future energy use based upon the differences between the various versions of Windows. The OS profile includes the percentage of PCs currently using Windows Other, Windows XP, Windows Vista and

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Windows 7. To account for hardware differences, the percentage of OS version per Laptop, Thin Client and Desktop PC type also needs to be defined.

2.1.4. Current Power Management Power management tools exist in each version of the Windows OS at varying levels of complexity and enforceability. Many third party advanced power management software applications are also available for the Windows OS. Therefore this model section takes into account by PC type and OS version the number of PCs that may already have Power Management in use. An opportunity to detail the type of power management is available in tier 2.

2.2. Tier 2 Tier 2 offers model variables that can be used with secondary data or with scenario specific primary data.

2.2.1. Site Power Consumption & Cost Businesses are often subject to site or business specific electricity prices thus to calculate the cost of using PC equipment, the cost paid for electricity (pence/KWh) is required. This information is not always immediately known and therefore a method was sourced to calculate based upon the size and power consumption of the business being analysed. The UK Government’s Department of Energy and Climate Change (DECC, 2010) publish an average ‘pence per KWh’ for non-industrial companies. Using the analysed scenarios business size and power consumption information one can determine the average price paid.

The model analyses power costs over three years and accordingly a variable to account for the expected growth in power costs is included. Ofgem (2010) suggest energy prices could increase 25% over the next 10 years compounded this could increase per year energy prices by 2.26%.

2.2.2. Estate Growth Rate PC estates can grow and shrink according to business need and change. Thus this model section allows an average growth or decrease rate in the number of PCs to be input. If growth rates are not understood, a sensible historical growth rate could be used. ICT equipment for some companies will be a major growth area therefore this information will allow the model to factor in future additions to the PC base. From discussions with Microsoft UK, a sensible growth figure of 6% pa was determined.

2.2.3. Work Schedule Use time is an important factor in order to calculate how long PC equipment is being used for. This model section allows the average working hours for the scenario being analysed to be input. Within the enterprise scenario, commonly a set working day and week is defined as 8 hours a day, 5 days a week. Having the ability to change the work schedule was viewed as important as, for example, a service based company working 24 hours, 7 days a week would receive incorrect results using the standard working day.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 2.2.4. User Profile PCs energy consumption and use time can vary according to the types of process being performed by the user. To account for this variation a section to scale power consumption and PC use time based upon a user profile was included. This section utilises PC user profile research conducted by Alinean Inc (2008). In summary, five types of user were identified along with a power scalar and active and inactive PC use time; high performance workers (140%), knowledge workers (100%), structured task workers (95%) and data entry workers (85%) (Alinean Inc, 2008). An organisational analysis is required in order to define the percentage of each worker type. Once this information is known, an average hardware power scalar and use time is calculated. Using an analysis by Alienan Inc (2009) resulted in an average power scalar of 96% and a PC active use of 7.3 hours per day.

2.2.5. PC Use Profiles & Power Management The aim of this section is to model the number of hours each piece of equipment (Desktop, Laptop, Thin Client, Monitor) is in an active, idle or sleep state. This information will be used in later sections to calculate the power consumption at the various activity state levels.

A PC Use Profile was calculated for each OS version with or without power management. Within each PC Use Profile the hours in active, idle and sleep states were calculated. Active hours were taken directly from the average PC active use hours as calculated by the user profile section. For the case that no power management is used, sleep hours were set to 0 with idle hours being the remainder of the day’s hours after active hours has been accounted for. Where power management was used, idle state hours were calculated by specifying the power management setup of each OS which resulted in an amount of time saved per component (see next paragraph). Sleep hours were set as the remainder of the hours in a day when active and idle state hours were taken into account.

To calculate the reduction in idle time that power management features can provide a section to specify the power management features, per OS, was created. The main power management features of Windows 7 were used as variables for this section. Therefore, this section includes the time after entering idle that the Monitor ‘dims’ (2 min - Windows 7 only) and ‘powers off’(10 min), the time that sleep mode is activated (30 mins) and how enforceable (%) the power management regime is. Specifying how enforceable power management is per OS was found to be an intangible variable due to its complexity. Windows 7 was set to 100% enforceable representing the state of the art situation. Vista and non-Vista PCs were set to 80% and 50% enforceable respectively. A variable to account for the number of periods that the idle sequence was interrupted was also included (4 periods). Laptops include both a monitor and computer in one and thus for laptops the proportion of power that is attributed to the monitor was specified; this figure was found to be an average of 30% (Derived from a study completed by Mahesri & Vardhan, 2005). From this information the amount of time saved per PC, monitor and Laptop was calculated.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Other power management features of Windows 7 not included here, such as hard disk management, are accounted for in the ‘Windows 7 Efficiency Savings’ section (see later) due to their variability on differing hardware setups. This section does not take into account faulty hardware which can stop sleep mode from occurring as realised in a report by Blackburn and Collins (2009).

2.2.6. Manual Power-off The PC use profile section does not take into account PCs that are manually powered off. The manually powered-off PC hours were set to an average of 36% of all users for standard working days and 72% for weekends and holidays following a study by Roberson et al. (2004). This value could also be assessed from a basic survey of current users to refine results.

2.2.7. Power Consumption To calculate the power consumption at each PC hardware activity state, the power consumption of each type of PC and monitor at each state is required. Information on the power consumption (watts) of hardware at active, idle (inactive), sleep and power-off state is available for most recent and new hardware from Energy Star (2010) or from the hardware manufacturer. In the case that multiple hardware models are used for one PC or monitor section, then a weighted average can be used. The power scalar determined in the User Profile section was applied to each components power use.

2.2.8. Hardware Migration Upgrades to new OS are commonly timed to occur with a hardware refresh. New hardware components often use less energy due to an increasing focus upon energy saving and from technical advancements. Therefore a model variable to account for the decrease in power per upgraded PC was included. Alinean Inc (2008) found that when upgrading from a Pentium D to a Core/ Core Duo and Energy Star rated system, a decrease of 5.2%, 28.6% and 33.0% was realised in active, idle and sleep modes. An old to new hardware specific analysis is required in order to understand exact power consumption differences between new and old.

2.2.9. Windows 7 Efficiency Savings Windows 7 includes improved software process efficiencies which are aimed to reduce energy consumption and increase PC usability. A model variable was included to account for this saving. The amount of power reduction that Windows 7 has over previous systems will depend on the hardware being used and the PC use profile. Therefore this value cannot be quantified precisely without hardware specific power measurement. In the case that measurement is not available values measured by Hacker (2010) and Lahoty & Nataraja (2009) can be utilised. It was found that Windows 7, when used on the same hardware and using default hardware setups saved power using its advanced process management over Windows XP. According to Hacker (2010) when the computer was in active mode Windows 7 saved an average of 10.60% and when inactive saved 10.50%.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 2.3. Results From the variables described in previous sections, a calculation of the amount of power being used per component is undertaken by the model. The following sections provide an overview of the calculations.

2.3.1. Component Growth The quantity of a particular component every year is calculated using the component quantity and ‘estate growth rate’ value. A migration rate value of 33% per year was used to change from the current OS version to Windows 7. This calculation also determines the number of components using and not using power management using values from the ‘power management profile’.

2.3.2. Activity State Power Consumption For each activity state (Active, Idle, Sleep and Power Off) the total power consumed per year for each component is calculated. This calculation utilises the Work Schedule, PC use profile and power consumption variables. Within each of the power mode sections (apart from Active mode) the calculation is completed twice to differentiate between components with or without power management. Power Off mode’s hours are calculated from the ‘Manual Power-off’ data. This is subtracted from the ‘Inactive power mode’s’ ‘number of hours’ value.

2.3.3. Weekend Power Consumption The number of hours a component is left turned on or off over the weekend is calculated using the ‘manually powered-off’ values. An average power consumption value per weekend is then calculated utilising the results from the activity state power consumption calculation. This section is calculated twice to account for components with or without power management activated.

2.3.4. Total Power Consumption across all Modes Total power consumption per component is calculated by totalling the power consumption values per activity state and weekend power consumption. Finally total power consumption per year for all of a component type is determined and includes differences in power management.

2.3.5. Costs The cost of the power used per component and for all components using the power cost and power cost growth variables is completed to highlight cost per component.

2.3.6. Carbon Equivalents Translating electricity use into an environmental impact was achieved by relating the total electricity used per section to a UK average electricity carbon emission factor (Table 1).

Using a CO2 equivalent (CO2e) value means that other greenhouse gases produced by electricity production and consumption (such as nitrogen oxide and methane) are taken into account showing the overall impact to the environment. To provide context and

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ tangibility to those that are not technically or scientifically minded, some interesting and useful carbon equivalents are provided. Each figure (Table 1) has been calculated from a scientifically recognised source and is specified for the UK where appropriate.

Table 4 - Carbon Equivalent Calculations

Equivalent Type Value Source & Notes DEFRA (2010). Average electricity UK average = 0.544 kg This value differs by country e.g. carbon emissions CO2e per kWatt hour US average = 0.718 kg CO2e/kWh

Average CO2e per petrol car = 0.334 kg CO2e / mile and average CO2e per diesel car = 0.319 kg Average car carbon UK average = 2839.89 CO2e / mile (DEFRA 2010). emissions kg CO e per year 2 Average distance travelled by car per person in the UK = 9241 miles (DfT 2009). As calculated for an average UK Average home carbon UK average = 5500.00 house hold from The Energy Saving emissions kg CO e per year 2 Trust (2010) This value can vary by location and Average kg’s of carbon woodland type. 4050.09 kg/acre is accumulated per year 4050.90 kg CO2e per an average from a variety of contained in one acre of year scientific sources, the main being forest Nowak & Crane (2002) and Baral & Guha (2004).

CO2e per Mile travelled UK Average = 0.319 kg Average CO2e per diesel car by an average diesel car kg CO2e / mile (DEFRA 2010). Assumes a trip of 5600 kilometres using a conversion factor of 0.135 CO2e per Air Trip to UK Average = 894.53 kg CO2e per pkm (passenger New York, USA kg CO2e kilometre), plus a 9% uplift to take into account non direct routes (DEFRA 2010). 2.3.7. Online Tool The model results have been adapted into an interactive online tool (Figure 1). The tool is live on the Microsoft Environment website (www.microsoft.com/uk/environment).

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Figure 7 - Tier 1 of the model has been developed into an interactive online tool

3. DISCUSSION This research demonstrates a methodology and a two tier model structure that enables the calculation of energy used within a Windows OS environment. Key features included within the methodology are power saving technologies, enforceability levels, hardware technologies and OS process efficiencies. The methodology highlights how the Windows OS environment can be used to reduce power consumption and thus begin to reduce environmental impacts whilst not compromising productivity or the PC experience.

The presented model and methodology allows a fully customisable Windows 7 scenario to be modelled according to the power management needs of a user. Implementing granular power management variables allows the model to truly reflect reality unlike previous tools created.

Unless every variable is fully customised, results will be calculated from assumptions and secondary data and thus can only provide an estimate of power consumption. A key model assumption which quantifies savings from Windows 7 via process efficiencies cannot be relied upon as very robust as this value will depend upon hardware used and computer tasks being performed. Hacker (2010) did realise Windows 7 specific savings through a limited hardware study and thus this variable will be a prime focus for future research.

This model does not take into account the potential environmental impacts or costs of replacing hardware and destroying older hardware. This needs to be taken into account and lifecycles of hardware should always be maximised so as to extend lifetime and reduce embedded carbon costs. Windows 7 was designed to run on older hardware and

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ thus it should not always be assumed that a new hardware upgrade is needed (this scenario can be modelled using this methodology).

The model and methodology has been adopted by Microsoft UK Environment and is being used on a number of case studies. Using the model on an UK company estate of

50,000 PCs has revealed that over 22 thousand tonnes of CO2e could be reduced by simply turning on the power management features of Windows 7 and using efficient hardware. These case studies will help refine and improve the models secondary data store and overall applicability by comparing and reviewing model and actual results.

4. REFERENCES Alinean Inc (2008) latest estimates based on research observations over the past 15 years of TCO / ROI assessments. Technical Report, Alinean Inc, Orlando, USA. Baral, A. and Guha, G. (2004) Trees for carbon sequestration or fossil fuel substitution: the issue of cost vs. carbon benefit. Biomass and Bioenergy, 27(1): 41-55. Blackburn, M. and Collins. G. (2009) Why Power Schemes are not enough. White Paper, 1E. London, UK. [Online]. Available: http://www.1e.com/contenthub/index.aspx#tabs-2 DECC (Department of Energy and Climate Change) (2010) Prices of fuels purchased by non-domestic consumers in the United Kingdom excluding/including CCL (QEP 3.4.1 and 3.4.2). [Online]. Available: http://www.decc.gov.uk/en/content/cms/statistics/source/source.aspx DEFRA (2010) 2010 Guidelines to Defra / DECC’s GHG Conversion Factors for Company Reporting (version 1.1). Technical Document, Defra, London, UK. DfT (Department for Transport) (2009) Table NTS0904: Annual mileage of 4-wheeled cars: Great Britain, 2002 to 2009. Technical Document, Department for Transport, London, UK. DUKES (Digest of UK Energy Statistics) (2010) Chapter 5 Electricity. Technical Report. DECC, London, UK. Energy Star (2010) Desktop & Integrated Computer and Laptop Product List. Technical Report, U.S. Environmental Protection Agency, Washington, USA. [Online]. Available: http://www.energystar.gov Gartner (2007) Gartner Estimates ICT Industry Accounts for 2 Percent of Global CO2 Emissions. [Online]. Available: http://www.gartner.com/it/page.jsp?id=503867 Hacker, A. (2010) Maximizing the Impact of Effective Power Management with Windows 7. Technical report, Mindteck, Enola, PA, USA. [Online]. Available: http://www.mindteck.com/whitepapers/Enabling-Green-Computing.pdf Lahoty, P. and Nataraja, R. (2009) Green Computing with Windows 7. Technical report, Mindteck, Enola, PA, USA. [Online]. Available: http://www.mindteck.com/whitepapers/Green_Computing-with_Windows- 7Ver_1.0.pdf Mahesri, A. and Vardhan, V.(2004) Power consumption breakdown on a modern notebook. Power-Aware Computer Systems, 4th International Workshop (PACS), Portland, OR, USA.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Microsoft (2009a) Desktop and Server Energy Models. Technical report, Microsoft Corporation, Redmond, USA. Microsoft (2009b) Windows 7 Power Management: Power Management Improvements in Windows 7. Technical report, Microsoft Corporation, Redmond, USA. Nowak, D. and Crane, D. (2002) Carbon storage and sequestration by urban trees in the USA. Environmental Pollution, 116 : 381–389. Ofgem (2010) Project Discovery. Technical Report, Ofgem, February 2010, London UK. Roberson, J., Webber, C., McWhinney, M., Brown, R., Pinckard, M. and Busch, J. (2004) After-hours Power Status of Office Equipment and Inventory of Miscellaneous Plug-Load Equipment, 2004. Technical Report LBNL-53729-Revised, Lawrence Berkeley National Laboratory. The Energy Saving Trust (2010) Surprising Statistics! [Online]. Available: http://www.energysavingtrust.org.uk/Climate-Change

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Multifunctional, Adaptable Facades

Bridget Ogwezi1*, Dr Richard Bonser1, Dr Geoff Cook1, Jonathan Sakula2

1 School of Construction Management and Engineering, University of Reading 2 Buro Happold, London, UK

*Corresponding author: [email protected]

ABSTRACT The current state of the planet, in terms of energy use and climate change means that by all means possible, we must move to a more sustainable way of living. Buildings account up to 40% of energy consumed, hence there must be a determined effort to reduce their energy consumption in maintaining a suitable internal environment. The building façade acts as the medium through which the internal environment can be controlled through interaction with the external environment. The building façade can define the internal living space by controlling the flow of energy through it. In order to reduce the energy used in achieving design criteria in terms of lighting, thermal properties, acoustics, air quality and security, the façade needs to be able adapt to the changing external environment and the complex internal needs of the users. This study details the need for multifunctional, adaptable facades and what they achieve. Finally, the paper discusses the possibility of multifunctional, adaptable skins being developed in the light of current research.

Keywords: Environmental control, adaptable facades, artificial neural networks, building skin, computer controlled facades

1. INTRODUCTION: CREATING AND DEFINING AN ENVIRONMENT

In the simplest form, by building walls that sit on a foundation and are topped by a roof, another environment is carved out from the existing open and unregulated one. The elements of the external environment can be moderated by this outer layer, the façade of the structure. Rain does not penetrate into the inner area unless there is an opening, the same applies to wind and sunlight penetration is limited to transparent parts or openings in the façade. This principle of a façade applies if you are erecting a tent or building a skyscraper.

In any building, the façade should not be regarded as just an outer face or a means of delineating an area; it is in fact in a key component to the design functionality of the

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ structure. Even though facades, in several cases, do not bear structural loads from the rest of the building, precise design of every facet of it is crucial to the successful use and lifespan of the entire structure.

Defining the internal environment to detailed specification, however, requires a measure of complexity of design and construction. Careful consideration must be given to

a) What will the building be used for? Is there a possibility this can change? b) Where is it located in relation to climate and the existing micro-climate? c) Who are the intended users of this building? d) How are the façade elements and design requirements to be combined successfully?

In the book, Intelligent Skins, the author states that “The façade is the single greatest potential controller of its (the building’s) internal environment.” (Wiggington and Harris, 2002)

For several years in building construction history, due to the extensive use of mechanical ventilation systems, there was little the need to design building façades that would have such impact. In terms of internal environmental control, making the façade air tight would have been a high priority, and whatever the external conditions, by using HVAC systems and artificial lighting, the interior was made comfortable. (Barkhume, 2007)

Even before the issue of climate change and global warming became a major driver in the move to reduce energy use in buildings, engineers were looking for ways to make buildings more efficient and more cost effective in the long run, and move towards facades that respond to their environment and the users to produce a structure that is sustainable for its entire duration.

Global warming is widely believed to be due to anthropogenic causes: the burning of

fossil fuels as an energy source and other industrial processes which release CO2 and other green house gases into the atmosphere.

Buildings are responsible for up to 40% of all energy consumed worldwide (WBCSD, 2008), and legislation is now in place to ensure that figure is reduced. It is with the façade that a new artificial environment is created and provides a controlled link between the interior and exterior environments.

With sophisticated design, the façade can specifically define that environment according to user specifications.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

2. MULTIFUNCTIONALITY

Expected performance criteria for a façade can be broadly split: (Skelly 2000)

(1)Aesthetics Prevent condensation and mould growth

Maintain a good appearance Control humidity

Communicate status Keep out wind

Communication information Keep out rain and snow

Allow views Keep out insects, vermin

Control solar radiation

(2)Structure/ structural integrity Utilisation and control of daylight penetration Be a self supporting and stable Keep out unwanted sound Transfer loads Control glare Prevent biological damage from plants fungi Keep out unwanted odours and pollutants Be durable Provide visual privacy Control sound emission Provide facility to blackout windows Control vibrations Allow controlled natural ventilation

Limit infiltration (3)Access, maintenance and buildability Prevent uncontrolled fabric heat loss or Support maintenance and repair gain

Address buildability issues, handling, storage, etc (5)Security and safety Integrate with other building systems Keep out intruders Provide safe operable openings Provide fire resistance Address manufacturing issues Provide security in extreme events: seismic and blast loads (4)Control and protect the internal environment

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ A façade that can successfully carry out several of these functions can be described as multifunctional.

As mentioned, the façade, more than any other component of the building, can determine the amount of energy required to run it. Thus the inclusion of micro power generation in the form of photovoltaic panels in facades adds another category to façade functionalities; that of energy harvesting, as a means of reducing the carbon footprint and imported energy for running the building.

3. THE NEED FOR ADAPTABILITY

An adaptable façade is one that responds statically or dynamically to differing conditions. The conditions might be changes in weather during the year, air quality in the building, the position of the sun, use of internal space etc. Under each set of conditions, the façade or elements of the façade changes and therefore changes what it ‘delivers’ to the building users.

To reach prescribed levels of efficiency and functionality and deliver suitable internal conditions, the façade needs to respond to the changes in demand placed on it from both internal and external factors.

In order to continually deliver an acceptable internal environment, the façade must be able to change in some way, in response to certain stimuli.

Successfully controlling the internal environment by interacting with the external environment brings about the greatest need for an adaptable façade.

The primary function of a building is to create an environment conducive for the users to carry out intended activities within. The conduciveness, in terms of human comfort, can be described in the following categories:

1. Visual comfort: good views, sufficient lighting 2. Thermal comfort: temperature, internal air movement at acceptable levels 3. Sound regulation: loud, distracting sounds from neighbours, traffic reduced to minimum 4. Good air quality: low levels of air toxicity (Skelly 2000) 5. Securitas: is the building safe in the event of a bomb blast or earthquake?

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Building users want to feel the same temperature, not be affected by glare or loud noises during all periods of occupation but the external environment that produces these loads is highly variable, it changes over minutes, hours, weeks, months, and years and may further be complicated by global warming.

Internally, the environment is made variable by the increasing sophistication of building design and contents. These contents introduce further heating loads which can be controlled by the environmental systems incorporated into the building. The energy required to keep the building within suitable comfort levels can be greatly reduced if the façade can ‘damp’ the effects of the external climate loads.

A major challenge in designing an interior environment to suit the users is the users themselves. User perception of comfort is highly variable and it is not possible to create a space in which all the people would be completely satisfied with the ambient thermal conditions. Differences in age, background, clothing and gender produce differences in what an individual terms as comfortable. Studies show that the lack of control over their personal environment can lead to dissatisfaction within the building, no matter how advanced the building management system is.

Studies also suggest that in office buildings, natural or mixed mode ventilation would be preferred by users over air conditioning](Arens 2007), however, by opening a window, heat can be lost making it an uncomfortable temperature for another building user.

Problems also arise in technological development when designers work without adequately factoring in the users and their response to the technology. As with any change presented to individuals, if the terms of the change are not acceptable then it is very likely to be criticised and dismissed. This is the socio-technical dilemma; where building users find ‘work-arounds’ to avoid being constrained by new systems introduced.

Another dimension in ensuring façade performance, adaptability and user comfort are the ‘conflicts’ that arise within façade design; satisfying one area of demand leads to a compromise in another. For instance: how do you ensure daylighting is maximised while controlling solar glare or solar again? Fresh air needs to be circulated with minimal input from the HVAC system; how do you allow for this natural ventilation and also control the amount of noise that gets in? Well designed façade systems and materials make it possible for these conflicts to be addressed, and adaptable systems provide these solutions with added efficiency while attempting to reduce energy use and develop a more sustainable building landscape.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 3.1 Traditional Façade design

Traditionally, in desiring to control the interior environment, facades have been designed under the concept of passive architecture, where the building structure itself copes with internal and external variabilities. For example, in building design, buildings that are ‘open plan’ with large spaces and a relatively small number of columns can easily be adapted to change of use, by adding internal partitions.

Each façade of a building experiences differences in their exposure to natural elements; the facades face different directions, and thus experience different environmental conditions. Wind can drive rain in very particular directions, and the façade that is downwind will receive less rain and wind at that particular than the façade facing the rain. By recognising the difference in what each façade faces, in terms of climate and microclimate, then by adjusting factors such as the glazing ratio, using reflective glass or adjusting the air tightness, the façade as a acts a controlled moderator of the exterior variabilities.

An example of traditional design is the South American mud masonry house: Mud, or adobe, is used in hot climates as a temperature moderator. The principles of thermal mass apply here, where the adobe absorbs heat during the hot daytime, preventing the interior from becoming as hot as the exterior. As the temperature cools down in the evening the bricks contract and give off this heat, some of it is released into the dwelling, keeping it warm at night and the rest of it is discharged to the atmosphere. This effect can be felt by standing close to a concrete or stone or masonry building in the evening after a hot day; the facade is quite warm even though the external temperature is relatively low.

3.2 Adaptable Facades

Adaptable facades are able to respond to changes in internal and external demands in both static and dynamic ways.

3.2.1 Static adaptable facades

By realising the potential of materials and smart materials, a façade can change its physical properties and react in a predetermined way to internal and external conditions. A very old example is the Inuit igloo. The dwelling, made of snow and ice, uses the fact that snow is an excellent insulator to keep the interior of the dwelling at up to a

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ temperature of 16C from body heat only. (Holihan et al 2003) This façade keeps the amount of external energy needed to ensure human comfort to an absolute minimum. The façade readily adapts itself to the higher internal temperature: when the ice on the inside melts, it runs and fills the cracks and crevices that may have occurred during construction. The melted ice solidifies and makes the façade air tight.

The trend towards highly glazed curtain wall systems in buildings has been a driver in the development of certain materials very useful in climate control. Double glazing and double skin facades, even with gas filled cavities would not adequately meet internal comfort demands in terms of reducing glare and solar heat gains (IDCOP 2006).Switchable glazing technology: electrochromic and photochromic glass: These glass nano-coatings change the light transmittance in glass to moderate solar glare, solar gain and provide privacy in response to an electric current and natural light respectively.

Even though these technologies are well established, they have their limitations. Photochromic glass, for instance, is autonomous, but as it reacts to light and not heat, a significant amount of heat through solar gain, would already be in the building before blocks it out.

Suspended particle devices: particles are suspended in solution between the plates of glazing, the particles are normally in random motion, colliding and thus reducing transparency, when the energy is applied, these particles become aligned and the glass is transparent. (DesignBuilder 2010)

In general, switchable glazing has not yet adequately controlled solar glare, because even though they can reduce light transmission by up to 90%.(DesignBuilder 2010), solar glare is up too 1000 times more bright than the light needed for reading.

Switchable Thermal Insulation is an adaptable system using the high thermal conductivity properties of hydrogen. A steel panel, welded around an evacuated fibre board can have its insulation properties range from 0.002w/m.k to 0.1w/m.k by releasing small amounts of hydrogen when it is heated. Thus, the panel can switch between ‘insulator’ in cloudy winter days to retain heat and summer days to prevent the interior space getting too hot. In its ‘conductor’ state, it allows useful solar heat gains in winter sunshine. (IEA 2001)

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 3.2.2 Dynamic Facades

Dynamically adaptable facades are those with moveable parts. These parts that can be controlled on different levels; as a ‘whole building’ approach, with a centralised building management system, it could be controlled on a floor by floor level, or by actuators on individual façade elements which may or may not allow user interference.

Computer controlled facades use Building Management Systems to control the interior environment through the flow of information from sensors located in the building skin. These sensors can detect light levels, heat levels, wind speeds, air quality, and through actuators that are designed along with the façade elements, effect changes in the building skin. While some say that smart materials will replace computer controlled façade systems, advances in smart technology in the automotive industry, for instance, seem to suggest that there is still a lot more room for advancement in computer controlled facades.

Shading devices can be linked to a central control system that tracks the path of the sun and moves the shading elements to continually provide internal glare protection. Automated louvres can respond to levels of daylight and the sun angle and the slats will be oriented accordingly to reduce glare and still admit daylight. These systems are particularly useful when considering that the sun angles differ, not just during a day, but in different seasons of the year. Dynamic shading has been shown to reduce energy consumption of up to 30% in a UAE office setting. (Hammad and Abu-Hijleh 2010)

Natural ventilation is also an aspect of dynamic façade control. Openings designed into the façade can be opened or closed on response to a sensor that determines air quality within the building, or heat readings.

The dynamic façade also extends to sophisticated energy harvesting in which solar collectors on the façade track the path of the sun to optimise the radiation falling on it.

This is not to say that a dynamic façade must be controlled by a computer system; the so-called ‘Swindow’ (IEA ECBCS 2009), a window that self adjusts its opening angle in relation to pressure differences on its faces due to air movement.

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 4. POSSIBLE DEVELOPMENTS IN MULTIFUNCTIONAL, ADAPTABLE FACADES

Research in biomimetics investigates how lessons can be learned from living organisms and be adapted to solve man made issues. As suggested by Schittich (2003), by considering a building façade as a ‘skin’ that can respond as it would on a living organism, lessons can be learned and systems developed to make a building façade multifunctional and adaptable that uses minimal energy and is sustainable.

Wigginton and Harris(2002) suggest that that by developing the skin concept with Artificial Neural Networks, important advances can be made in multifunctional, adaptable façade development. Artificial neural networks are computational models that mimic the nervous systems of animals (Gershenson). The nervous system is what links the skin to the brain, and by this analogy, the building façade to the building management system. ANN has the ability to learn adaptively; that is, perform operations based on training or experience. In its program, The Human Body by the BBC (2011) it was shown that the human skin, the link between the external physical environment and the internal one can adapt over time to respond in ways that it would not normally respond. Though ANN’s were first developed in 1943, they are only currently being funded for further research. ANN’s have been used to ‘train’ an adaptive façade to provide optimum light transmittance into a building.(Skavara 2009).

REFERENCES

Arens, E (2007) Assessment of Indoor Environments, Proceedings of Roomvent2007. pp 8-12

Axel Ritter speaking at the Australian Royal Melbourne Institute of Technology’s Green Building and Design Conference, held on 9-10 September 2009 in Melbourne, Australia

Barkhume, A (2007). Innovative Building Skins:Double Glass Wall Façade [accessed online at http://njit.academia.edu/AllenBarkhume/Papers/95093/Innovative_Building_Skins_Dou ble_Glass_Wall_Ventilated_Facade 27/05/2011]

BBC (2011) Documentary. Inside the Human Body Ep2/4

DesignBuilder webpage [accessed 18/04/2011]

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Gershenson, C. Artificial Neural Networks for Beginners [accessed 28/05/2011 http:/arvix.org/ftp/cs/papers/0308/0308031.pdf

Hammad, F. and Abu-Hiljeh, B. (2010). The energy saving potential of using dynamic external louvres in an office building. Energy and Buildings Vol 42 Issue 10 pp.8

Holihan et al (2003) How warm is an Igloo?, Cornel University Library. BEE453 - 2003 Student papers. [accessed 18/04/2011)URI: http;/hdl.handle.net/1813/125]

IEA, Energy Conservation in Buildings and Community Systems Programme (2001), Proc. Int Conf and Workshop. Pp17

IEA, Energy Conservation in Buildings and Community Systems(2009) Annex 44, Integrating Environmentally Responsive Elements in Buildings, Expert Guide - Part 2 [accessed online http:/www.ecbcs.org/docs/Annexe_44_Expert_Guide_RBE.pdf 28/05/2011]

IDCOP:Innovation in Design, Construction & Operation of Buildings for People (2006) Building Envelope Technologies. Report 2. pp 2-3

Schittich, C. ed. (2003) In Detail: Building Skins. Shell, Skin, Materials. Base,l Boston, Berlin. Birkhauser

Skavara, M.E., (2009)Learning Emergence: Adaptive Cellular Automata Façade Trained by Artificial Neural Networks.Dissertation: Msc Adpative Architecture and Computation. Bartlett School of Graduate Studies. UCL, London

Skelly, M. (200) 'IBC automated window project'. [accessed online at http://www.cwct.co.uk/ibcwindow/adaptive/performance.html 20/04/2011]

Skelly, M. (200) 'IBC automated window project'. [accessed online at Occupant comfort http://www.cwct.co.uk/ibcwindow/adaptive/comfort.html 20/04/2011]

Wigginton, M and Harris, J. (2002). Intelligent Skins. Oxford: Elseiver Science pp. 13,19

World Business Council for Sustainable Development( 2008), Energy Efficiency in Builkdings; Transforming the market [accessed 7/04/2004 at http://www.wbcsd.org/DocRoot/rVDgBRKvPngUrqivMHNM/91719_EEBReport_WE B.pdf 25/05/2011]

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Heat Demand Analysis Of Residential Development In London Connected To District Heating Scheme

R. Burzynski1*, M. Crane2, R. Yao3 and V.M. Becerra4

1 Technologies for Sustainable Built Environments, University of Reading, UK 2 SSE Utility Solutions, Thatcham, UK 3 School of Construction Management and Engineering, University of Reading, UK 4 School of Systems Engineering, University of Reading, UK

* Corresponding author: [email protected]

ABSTRACT

To impose CO2 emissions reductions the UK Building Regulations require developers of new residential buildings to calculate expected CO2 emissions arising from their energy consumption using methodologies such as SAP 2005 or, more recently SAP 2009. However, these calculations are rarely verified with real energy consumption and related CO2 emissions. This paper presents results of heat demand analysis based on individual flat weekly heat demand data, collected from recently built residential development of several hundred of flats connected to district heating network. The paper presents a methodology for separating out the domestic hot water use (DHW) and space heating demand (SH) and compares site measured DHW to the DHW demand calculated using SAP 2005 and 2009 methodologies. The analysis shows also variance in consumption of the heat between both size of flats and tenure (privately owned or housing association).

Keywords: Domestic hot water, space heating, energy consumption, heat demand.

1. INTRODUCTION

There are number of publications regarding modelling and estimating domestic hot water consumption in residential properties. The main methodologies, have been described by the authors in the earlier paper[1]. However, there are few publications regarding measured DHW and space heating consumption and the accuracy of estimates made with Standard Assessment Procedure 2005 (SAP 2005)[2] and SAP 2009[3]. This situation lead the authors of this paper to undertake post occupancy analysis of the heat consumption for a new residential development located in London. This paper presents part of the results of this research, which is related to domestic hot water consumption. The authors also wanted to verify whether there is any clear correlation between type of tenure and the heat consumption of a property.

Section two of this paper presents details of methodology used to separate space heating and hot water consumption out of overall heat consumption. Section three shows results of the analysis of selected data mainly average hot water consumption for different

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ types of flats, tenures and the level of its variation. Section four presents conclusions regarding the methodology used in the analysis and short discussion of the results.

2. METHODOLOGY

It is generally acknowledged that consumption of hot water in residential buildings vary widely. The recent EST study [4] confirmed that it is mainly dependent on the number of occupants living in given property. However, this information is usually unknown during the design phase. Therefore, there is a need to use factors known for designers that could help to normalise hot water consumption used by different properties. These factors are the number of bedrooms and the floor area of a property.

The difficulty with the analysis was that the space heating (SH) and domestic hot water consumption (DHW) data are jointly measured by one heat meter. Therefore, there was a need to develop a methodology to separate these two components. This methodology is based on the assumption that the heat consumption during the summer time does not include a space heating component, therefore is solely DHW which can be extrapolated to give an estimate of annual DHW consumption. By deducting this consumption from the total heat consumption, it is possible to calculate heat consumption used for space heating. Further parts of this section describe details of the methodology used to analyse available data.

Data set Scottish and Southern Energy provided a set of weekly readings from heat meters installed in more than 150 social housing flats and 300 privately owned flats located in south east London. The automatic heat and electricity metering system reads data from the meters daily and logs the information into a central database. The metering system is not based on counting kWh pulses but it logs the actual heat or electricity meter readings. Therefore, the data is assumed to have high level of accuracy. The same data is used for billing the residents and there have been few data quality issues. For each flat the following information was available: flat number, floor level, number of bedrooms, floor area, type of ownership, hand over date and weekly meter reading starting usually from the hand over date. The first residents moved into the flats in summer 2009 and the last flats were not occupied until summer 2010.

Flats The flats were built between 2007 and 2010 to the Building Regulations Part L 2006 standard. All properties are connected to district heating scheme (DH). A plate heat exchanger installed in each flat provides instantaneous DHW. Space heating system is equipped with a programmable thermostat and radiators are fitted with TRVs valves. This configuration of DH, the controls, DHW and SH (radiator) temperatures are very similar to a combi boiler system. The only significant difference is that the heat exchanger for domestic hot water is sized to deliver 55kW of hot water, which is nearly twice the output of a combi boiler.

Selecting suitable data set and period of analysis In this analysis, the statistics are based on a full year of data obtained from flats occupied between 05/04/2010 and 03/04/2011. Flats with occupancy period of less than one year were excluded from the analysis as were flats which were occupied after

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 07/03/09. This four weeks period was treated as settlement period during which both heat and electricity consumption might significantly vary from normal consumption. However, the date of beginning of occupation was not known, therefore it had to be determined by analysing available heat data with the following criteria. 1. Occupation date had to be later than property hand over date. 2. The minimum heat consumption had to be not less than 10 kWh per week. The minimum consumption had been determined based on average weekly consumption of hot water for single bed flat with one occupant[5].

There were some issues with reliability of provided data. For example readings for some weeks were not available, some reading were inconsistent with previous or next reading, in some periods heat demand was zero suggesting that the flats were vacant or the residents were away for extended period of time. In many instances, despite the fact that the flats had been handed over, the readings did not show any heat or showed only small electricity consumption, which might suggest that the flats were not occupied yet. Such data were either corrected using interpolation, readings were omitted in the analysis or the flats were completely excluded from further analysis.

Dividing overall heat consumption into hot water and space heating components As there was no separate meter for measuring space heating and domestic hot water consumption it was necessary to split these two parameters using methodology described below.

To find the period during which all heat consumption can be treated as hot water only consumption a heating degree-day (HDD) factor was utilised. Period with low number of HDD should indicate the fact that the building and flats do not need space heating any more and the existing heat consumption is driven by hot water demand. However, before the HDD can be calculated it is necessary to define a base temperature. A base temperature is a temperature of external air at which given building or flat does not need any auxiliary heating. As this temperature was in practice unknown an analysis of building heat load against HDD for different base temperatures was conducted. For this purpose weekly HDD for different base temperatures has been obtained from Degree Days.net [6] website. This website provides HDD data calculated based on mean, minimum and maximum daily temperatures obtained from Weather Underground [7] website using an enhanced version of a set of equations developed by the UK Meteorological Office. More details about this methodology and equations can be found on the UK Meteorological Office website [8].

Based on SAP 2009 [3] it was assumed that the monthly DHW consumption has varied. Based on the measured, average hot water consumption during selected 9 weeks and the variation in daily consumption presented in Table 5 the data has been extrapolated to other months of a year.

Table 5 Monthly variation factor of average daily hot water consumption (volume) [3] Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Annual 1.1 1.06 1.02 0.98 0.94 0.9 0.9 0.94 0.98 1.02 1.06 1.1 1

Flats with gaps in heat readings or zero heat consumption for more than two weeks out of the nine were excluded from the analysis as the error in estimation of hot water

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ consumption would be too large. These gaps are likely to arise from residents going on summer holiday and hence are not representative of the rest of the year.

In the final step, for each flat, the estimated annual DHW consumption was deducted from the total heat consumption of given flat, giving annual space heating consumption.

3. RESULTS

Selected sample of flats for analysis Out of 450 flats 222 had sufficient number of reliable records to be included in the analysis with many flats excluded due to occupancy period less than one year. Out of 222 flats 100 are owned by private owners and 122 by housing association. Some details of the analysed sample are presented in Table 6. Table 6 Number and floor area of flats selected for analysis of hot water demand. No. of No. of Flat size [m2] Ownership bedrooms flats Min Average Max 1 29 41.6 42.8 43.4 Private flats 2 53 63.6 65.8 81.5 3 18 67.6 76.2 89.5 1 51 41.6 43.0 45.5 Housing Association flats 2 50 63.6 66.7 81.2 3 21 80.3 83.3 90.0 1 80 41.6 42.9 45.5 All flats 2 103 63.6 66.2 81.5 3 39 67.6 80.0 90.0 Private flats - 100 41.6 61.0 89.5 Housing Association flats - 122 41.6 59.6 90.0

Hot water demand analysis base period The HDD base temperatures used in SAP 2005 for individual dwellings varied between 9.4º C for middle level flats and 11.3º C for flat over the unheated car park. However, the analysis of the whole building heat consumption against HDD shows that the highest level of correlation is for base temperature of 14.5º C. The results of further analysis show that for the base temperature of 14.5º C the number of HDD per week did not exceed 3 for period between 28/06/2010 and 29/08/2010. Therefore, this period was selected as base period and the heat consumption within it was treated as domestic hot water consumption only. Consequently, the selected period provided a sample of 9 weeks of data.

Domestic hot water consumption The results of the analysis show that in general the DHW consumption calculated per square meter of flat floor area decreases with the flat size. However, this decrease is not very consistent. A comparison of the results of the analysis of measured DHW consumption with results of SAP 2005 modelling is presented in Table 7. The DHW data represents heat measured at the input to DHW plate heat exchanger and therefore does include heat loses from the heat exchanger and DHW pipework to taps. It does not

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ include loses related to, district heating distribution pipework to heat exchanger. Therefore, the measured values have been compared with corresponding values from cell 39 for SAP 2005 and cell 45 of SAP 2009 worksheets.

Table 7 DHW consumption categorised according to type of ownership and flat size.

DHW Site Average DHW SAP No of Average Annual DHW SAP 2005 Difference Difference 2009 per Ownership Bed- Flat Size Consumption per per Bedroom in Average in Average Bedroom rooms [m2] Bedroom [kWh/m2/year] [%] [%] [kWh/m2/year] [kWh/m2/year] 1 42.8 25 32 29 27 9 Private flats 2 65.8 18 26 49 21 22 3 76.2 17 25 46 20 16 Housing 1 43.0 16 32 100 27 69 Association 2 66.7 13 26 105 21 67 flats 3 83.3 19 24 26 19 -2 All flats 1 42.9 19 32 67 27 41 2 66.2 15 26 72 21 40 3 80.0 18 24 34 19 6

For example for privately owned flats the average annual DHW consumption decreases from 25 kWh/m2/year for 1-bedroom flat to 18 kWh/m2/year and 17 kWh/m2/year for 2 and 3-bedroom flats respectively. For housing association owned flats the average annual DHW consumption for 1-bedroom flat was 16 kWh/m2/year which decreased for 2 bedroom flat to 13 kWh/m2/year and increased to 19 kWh/m2/year for 3 bedroom flats.

Comparison of SAP 2005 modelling results of DHW consumption with site measurements leads to rather significant overestimation of the demand. For example, SAP 2005 estimated that for the 2-bedroom flat of 66.22 m2 the hot water consumption should be 26 kWh/m2/year, whereas the average measured consumption for such flat was only 15 kWh/m2/year (average for all 2-bedrrom flats). This means that SAP 2005 overestimated the consumption by 72 %. For 3-bedroom flat this overestimation factor was lower – 34 %, which is still significant. The relative position of site average, SAP 2005 and SAP 2009 for average size of 1-bedroom flat can also be observed on Figure 2 and Figure 3. In these figures, the value of the average consumption should be read from the horizontal axis rather then the vertical one.

However, if the new SAP 2009 were used for estimation of DHW demand the result would be much closer to the on site measured consumption. In this case the difference between measured DHW consumption and calculated using SAP 2009 for 1 and 2- bedroom flat would be reduced to 41% and 42 % respectively and for 3-bedroom flat to 6%. Therefore, this analysis indicates that the new SAP methodology is more accurate at estimating DHW demand than SAP 2005.

The analysis of the DHW data revealed also very wide range of hot water consumption of individual flats. The consumption can be as low as 2-3 kWh/m2/year and as high as 71 kWh/m2/year, almost irrespective of the flat size. More detailed analysis of low consumption flats showed that these flats had frequent and often regular periods of residents’ absence revealed by weeks of significantly lower heat consumption. Low

100

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ average electricity consumption suggests also that a single resident most probably occupied these flats. On the other hand a closer look at the flat with the highest average heat consumption per square meter shows both high heat and electricity consumption, which suggest that this 1-bedroom flat might be more densely occupied than other flats. Comparison of the minimum, average and maximum DHW consumption categorised by property ownership number of bedrooms is presented in Figure 1 and Table 7.

Extremems of DHW Consumption for kWh/m2/year Individual Flats 80 71 68 70 61 60 50 Min 40 30 Average 19 18 20 15 Max 10 3 2 2 0 1 bed 2 bed 3 bed

Figure 1 Comparison of minimum, maximum and average of DHW consumption for different sizes of flats.

Table 8 Comparison of minimum, maximum and average of DHW consumption for different sizes of flats and types of ownership. DHW Site Annual Consumption per Variation in DHW Site Annual No of Bed- 2 Ownership Bedroom [kWh/m /year] Consumption [%] rooms Min Average Max - + 1 8 25 56 68 126 Private flats 2 2 18 61 91 245 3 2 17 68 89 299 1 3 16 71 81 343 Housing 2 3 13 39 79 203 Association flats 3 3 19 51 83 165 1 3 19 71 84 270 All flats 2 2 15 61 90 299 3 2 18 68 89 275

Analysis of the distribution of the average DHW shows that for all three types of flats the most frequent value is shifted towards lower consumption levels. This can be clearly seen on the Figure 2 presenting DHW distribution for 1-bedroom flats.

101

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 1 Bed Flats 18 18 16 Housing Association 16 14 Private Flats 14 12 Site Average 12 10 SAP 2005 10 8 SAP 2009 8 6 6 4 4 2 2 0 0 <3 <6 <9 <12<15<18<21<24<27<30<33<36<39<42<45<48<51<54<57<60<63<66<69<72 kWh/m2/y

Figure 2 Distribution of occurrences for estimated DHW consumption for 1-bedroom flats. Moreover, Figure 2 vreveals also that the DHW consumption of housing association owned flats is located more among lower consumption ranges rather than higher where private flats are more dominant. This is true also for 2-bedroom flats. However, 3- bedroom flats have slightly different distribution of frequency of occurrence. In this case housing association owned flats lay more on the left side of Figure 2 if compared to privately owned flats. It is also worth noticing that for this type of flat the SAP 2009 estimation of DHW is almost the same as the consumption measured on the site. This suggests that SAP 2009 DHW estimation for larger flats (3-bedroom with average floor size of 80m2)

3 Bed Flats 7 8 6 Housing Association 7 5 Private Flats 6 Site Average 5 4 SAP 2005 SAP 2009 4 3 3 2 2 1 1 0 0 <3 <6 <9 <12<15<18<21<24<27<30<33<36<39<42<45<48<51<54<57<60<63<66<69 kWh/m2/y

Figure 3 Distribution of occurrences for estimated DHW consumption for 2-bedroom flats. The above observations are quantified in Table 9, which reveals that the average consumption of DHW in privately owned flats is higher by 35% and by 28% in case of 1-bedroom and 2-bedroom flats respectively. However, in case of 3-bedroom flats the consumption in housing association flats is higher than in private flats by 12 %.

Table 9 Domestic Hot Water Consumption Comparison No. of Housing Private Difference bedrooms Association kWh/m2 kWh/m2 %

1 25 16 55 2 18 13 38 3 17 19 -11

102

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

4. CONCLUSIONS

Developing methodology for assessing annual DHW and space heating consumption through separation of the two out of overall heat consumption revealed several uncertainties and issues. For this new development, the major issue was careful assessment of each flat’s electricity and heat consumption based on which the start of occupancy period had to be recognised and sufficient number of reliable records ensured from the summer period. In this methodology significant uncertainty factor lays in the assumption that for selected flats the DHW consumption pattern is representative for whole year and can be used to extrapolate available data to other moths of the year. However, not having data from separate DHW meter made it impossible to verify this assumption. Taking the above into account it can be concluded that this methodology is useful for estimates of split between hot water and space heating consumption. However, it might be questionable in case of using it to validate or verify results of modelling with SAP and other models.

The results of the analysis suggest that for the analysed flats the annual DHW consumption seems to be overestimated by SAP 2005. However, the improved formula for calculation of hot water consumption introduced in SAP 2009 significantly reduced this situation. As a result, in some cases the results of SAP 2009 DHW estimation were very close to the measured one.

Finally, the results also show some significant differences in hot water consumption between flats occupied by social housing and of privately owned flats residents. The DHW consumption was higher in private 1 and 2-bedroom flats by 55% and 38% respectively. Only in 3-bed flats the average consumption was higher than in private flats, however only by 11 %.

5. REFERENCES

[1] Robert Burzynski, et al., "Review of domestic hot water demand calculation methodologies and their suitability for estimation of the demand for Zero Carbon houses.," presented at the 1st TSBE Engineering Conference, Reading, UK, 2010. [2] DECC, "The Government’s Standard Assessment Procedure for Energy Rating of Dwellings (SAP 2005)," DECC, Ed., Version 9.81/9.83 ed. Garston: BRE, 2009. [3] DECC, "The Government’s Standard Assessment Procedure for Energy Rating of Dwellings," DECC, Ed., Version 9.90 ed. Garston: BRE, 2010. [4] EST, "Measurement of Domestic Hot Water Consumption in Dwellings," DEFRA 2008. [5] R. Burzynski, "All Houses Meter Readings. Zero Carbon Homes - Chalvey.," Scottisch and Southern Energy, Slough, 2011. [6] BizEE_Software. (2011, 28/04/2011). Heating & Cooling Degree Days - Free Worldwide Data Calculation. Available: www.degreedays.net [7] Weather_Underground. (2011, 28/04/2011). Weather Underground. Available: www.wunderground.com

103

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ [8] UK_Met_Office. (2011, 28/04/2011). UKCP09 gridded observation data sets — frequently asked questions. Available: www.metoffice.gov.uk/climatechange/science/monitoring/ukcp09/faq.html#faq

104

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Considering Occupants’ Comfort In Sustainable Building Refurbishment Projects

M.Agha-Hossein1,*, A. Elmualim2, M. Williams2, A. Kluth3

1Halcrow Group Ltd/TSBE, University of Reading, UK 2University of Reading, UK 3Halcrow Group Ltd, London, UK

* Corresponding author. [email protected]

ABSTRACT Research shows that, as a result of poor energy efficiency, a significant amount of the UK’s total energy consumption is wasted. Taking the principles of user-centred design into account as well as changing building occupants’ behaviour could help to prevent this energy loss and considerably cut carbon emissions per year. This paper presents an overview of the role of building occupants on energy performance of non-domestic buildings. It further introduces Halcrow’s current research project on how to improve the energy performance of their recently refurbished and occupied Headquarters in London, while increasing the comfort, satisfaction and productivity of their employees. An employee benchmark survey was conducted at the pre-occupancy stage. The purpose of this survey was to identify the employees’ level of satisfaction with their current workplace and, also, to indicate employees’ motivation and energy awareness level. The result indicates that the majority of the respondents are neither satisfied nor dissatisfied with their current workplace. With regard to employees’ sustainability awareness, most of the respondents said that they were not fully aware of Halcrow’s sustainability targets. This paper provides the results of this survey in detail. Keywords: Energy awareness, occupants’ comfort, pre-occupancy survey, non- domestic buildings and refurbishment

INTRODUCTION About 25% of carbon emissions in the UK (100 million tonnes of CO2 per year) are generated by non-domestic buildings (Hogain, 2010). According to Carbon Trust (2005), every day about £7 million, which is about 21% of the UK’s total energy costs, is wasted in UK industry due to poor energy efficiency. In comparison, it is estimated that changing buildings users’ behaviour could save this money for the companies and cut carbon emissions by 22 million tonnes per year (Carbon Trust, n.d. cited in Opus Energy, 2010). Clements-Croome (2003) explains that healthy buildings are very likely to increase productivity and save energy at the same time. Since comfort and productivity are very closely related (Morrow, 1995), care must be taken to balance energy savings with occupants’ comfort. Research shows that if building occupants are not comfortable with their work environment they will adapt their environment to meet their needs; even if this increases energy usage (Hadi & Halfhide, 2011). It is argued

105

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ that green technologies such as efficient lighting and advanced ventilating systems will enhance interior environmental quality and therefore be beneficial to human well being and productivity (Browning et al, 1995). However, technologies which are designed to improve the energy efficiency of a building must engage with the users, or the building will under-perform and energy savings will be limited (Gray, 2010).

Halcrow Group Ltd has recently refurbished a leased 1930’s, 5-storey office building in Hammersmith, London. This building is now occupied by about 450 people who have moved from Halcrow’s previous offices (Vineyard House (VH) and Shortlands) adjacent to the site. As staff satisfaction has a central place in social sustainability, Halcrow wishes to investigate innovative interventions currently available to reduce their energy consumption in their new HQ, while increasing their employee’s satisfaction and well-being. To do this, a survey was carried out at the pre-occupancy stage to understand employees’ needs and expectations about their work environment. The findings from this questionnaire were considered at the design stage, where applicable, of the new building. The results also used as a benchmark for evaluating the new building’s performance. This paper demonstrates some of the findings from the questionnaire considering VH building only.

1.1 Vineyard House (VH)

VH is also a 1930s, 5-storey building in Hammersmith. The building was poorly ventilated by two air-conditioning units on the roof which pumped air around the building. The building benefited from natural lighting, and fluorescent (‘energy savers’) artificial lighting was used to supplement the natural lighting. Three boilers served radiators through out the building, which were used for heating. However, as the building was poorly insulated, portable heaters were used during winter. The offices were mostly open-plan, but there were a few cellular offices as well. There was no quiet thinking area in the building and the only socialising area was the canteen on the ground floor which was too small to accommodate the number of the employees in VH and unwelcoming. METHODOLOGY

An employee survey was conducted to collect data regarding employee satisfaction, needs and expectations. This benchmark survey was used as a tool to enable the employees to confidentially express how they felt about their work environment. The first part of the questionnaire included items concerning demographic factors such as age, sex and employment status. Also, in this part, employees were asked to specify their modes of transportation to work and their willingness to work at home. In the second part, employees were asked to indicate their levels of satisfaction with their workplace physical environment, use of interior space, indoor facilities and current policies. For these questions, 5-point response scales were used, where: 1= Strongly Agree, 2= Agree, 3= Neither Agree nor Disagree, 4=Disagree and 5 = Strongly Disagree. In the latter part, employees were asked to state whether they were aware of Halcrow’s sustainability targets and whether they felt personally responsible for contributing to Halcrow’s sustainability objectives. This survey was sent via internet to all employees in Vineyard House (VH) and Shortlands buildings and stayed open for two weeks. Initially, 197 completed surveys were returned, (representing 34% response

106

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ rate). Having excluded data from ineligible participants and questionnaires with missing data on more than 70% of items on “satisfaction with workplace” questions, the final sample consisted of 189, 162 from VH and 27 from Shortlands.

RESULTS

This part of the paper illustrates the findings from the pre–occupancy employee survey, for VH employees only.

3.1 Demographic Questions

Over two-thirds of the 162 VH respondents were male, 32.7% (53) were female. Most respondents, 41% (66), were between 26 and 35 years old, 5.6% (9) were 25 years old and under, 23.6% (38) were between 36 and 45, 16% (26) were between 46 and 55 and 13.7% (22) were over 55 years old. In terms of employment status, the majority of the respondents, 95.1% (154), were full-time. More than 70% respondents (115) indicated that they were based in VH 75%-100% of the time.

3.2 Mode(s) of Transport

About 98% (159) respondents answered this question. Their responses are shown in Table 1.

Table 1: Mode(s) of transport you normally use to commute to your office

Frequency Percent Bus/Train/Underground/Foot/Bicycle 64 40 Car (in combination with other modes) 27 17 Bicycle (in combination with other modes) 32 20 Foot (in combination with other modes) 79 49 Bicycle (only) 5 3 Foot (only) 11 7 Car (only) 7 4

3.3 Using the Canteen in VH

About half the respondents (81) specified that they rarely used the canteen facilities in VH.

3.4 Office Set Up:’ Where You Sit in the Office’

The majority of the respondents, 76.4% (123) worked in an open plan office with more than 5 people, 17.4% (28) worked in a multi-occupant cellular office (5 people or less), 5.6% (9) had their own single cellular office and only 0.6% (1) of the respondents hot- desked in an open plan office.

107

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 3.5 Employees’ Satisfaction with Their Workplace

3.5.1 Physical Environment

Approximately 86% (139) of participants answered all the questions (8) in this category; the results are illustrated in Figure 1 overleaf.

Figure 8: I feel satisfied with the physical environment of my workplace

3.5.2 Use of Interior Space

Approximately 84% (136) of the participants answered all the questions (13) in this category; the results are illustrated in Figure 2.

108

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Figure 9: I feel satisfied with the use of interior space in my workplace

3.5.3 Indoor Facilities

Approximately 92% (149) of the participants answered all the questions (5) in this category; the results are shown in Figure 3.

Figure 10: I feel satisfied about the indoor facilities in my workplace

109

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 3.5.4 The Policies

Approximately 98% of participants (159) answered this question and 93% of them were satisfied with “Flexible Working Hours” policy and 68% were satisfied with “Working from Home” policy.

Table 2 shows the mean employees’ satisfaction level in each category. The 4 items together showed a satisfactory level of reliability (Cronbach’s alpha = 0.89), so overall satisfaction scores could be derived; overall satisfaction mean scores are also shown in Table 2.

Table 2: Mean Scores for Employee's Satisfaction

Physical Use of Indoor The Overall Employees’ Environment Interior Facilities Policies Satisfaction Level at Space VH

Frequency 139 136 149 159 111

Mean 3.45 3.18 3.20 2.02 2.98

Analysis was conducted to assess whether overall satisfaction differed between office set-ups (e.g. open plan, cellular offices, etc) (see Figure 5 overleaf). Overall satisfaction of respondents in single cellular offices (Mean=2.61, Std. Error of Mean=0.14) was higher than that of those in open-plan offices (Mean= 2.98, Std. Error of Mean=0.047). This difference was significant, t (92) = -2.104, p=0.038.

Figure 11: VH Employees' Satisfaction in different office set-ups

110

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

3.6 The Positive Effect of Office Environment on Employees’ Productivity, Well-being and Enjoyment

Figure 6 indicates the degree to which respondents felt that their current workplace had a positive effect on their productivity, well-being and enjoyment at work. The overall mean score (Cronbach’s alpha = 0.902) was 3.07.

Figure 12: The Positive Effect of VH Environment on Employees' Productivity, Well-being and Enjoyment

An analysis was performed to assess the correlation between employees’ overall satisfaction in VH and the positive effect of VH environment they perceived. A significant positive correlation (r = 0.602, p < 0.001) was found, indicating that satisfaction was positively associated with perceived positive effect

3.7 Employees’ Awareness of, and Attitudes towards, Halcrow’s Sustainability Targets

Figure 7 shows the level of employees’ awareness of Halcrow’s sustainability targets and whether they felt personally responsible for contributing to these targets.

111

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Figure 13: Employees' Awareness of and Attitudes towards Halcrow's Sustainability Targets

DISCUSSION AND CONCLUSIONS

A sustainable workplace has a positive impact on the environment while adopts practices to improve its employees’ satisfaction. To achieve a sustainable workplace behavioural intervention is perhaps going to be the key and that physical work environment might help drive certain behaviours. In this study, an employee survey was used as a tool to improve workplace sustainability by engaging the building’s occupants.

The sample of respondents was broadly representative of the employees at VH, as indicated by responses to the demographic questions and the question regarding office set-up.

The overall employees’ satisfaction score indicated that the respondents, on average, were neither satisfied nor dissatisfied with their workplace. Considering the four categories of employees’ satisfaction separately, the respondents were not satisfied with their workplace indoor temperature, indoor air quality and opportunity for personal control of the immediate environment. Inappropriate thermal conditions affect dexterity and increase physiological stress (Zhang, Barrett, 2010); therefore, it was important to consider these issues in the building’s refurbishment. Regarding the use of interior space, most of the respondents were not satisfied with the availability of contemplation areas and also with the auditory privacy of the individual workspace. Providing spaces for different work-styles, in the new building, was considered at the design stage. The data presented in Table 1 (Modes of transport) indicate that the majority of the employees used sustainable modes of transport to commute to work.

About half the respondents indicated that they rarely used the canteen facilities in VH.

112

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ As is shown in Figure 5, and as confirmed by t-test, respondents who worked in single cellular offices were more satisfied than those who worked in open plan offices. However, it should be borne in mind that the sample size for single cellular offices was small compared to that for the open plan offices.

Figure 7 indicated that the majority of the respondents did not know about Halcrow’s sustainability targets and most reported that they did not know how to contribute towards Halcrow’s sustainability objectives. This could be because the channels of communication were ineffective

This survey and the post-occupancy survey will indicate whether the refurbishment is meeting expectations that the new building is more sustainable than the old HQ building, and guide thinking on the need for new technical, behavioural and policy changes that are necessary to maximise sustainability performance within the budget and other practical constraints that have been identified.

References Browning, W. Roman, J., 1995, Green and the Bottom Line: Increasing Productivity through energy efficient design

Carbon Trust, 2005, Management Guide: Creating an awareness campaign, Queen’s Printer and Controller of HMSO

Carbon Trust, n.d., Cited in Opus Energy 2010, Rising employee awareness, [online], Available from: http://electricityadvice.opusenergy.com/module/page-254/employee-awareness.cfm (Accessed 13/12/2010)

Clements-Croome, D., 2003, Environmental Quality and the Productive Workplace, [online], Available from: http://www.cibse.org/pdfs/2acroome.pdf (Accessed 03/04/2011)

Gray, I., Technology Strategy Board: Driving Innovation, (11 August, 2010), Press release, [online], Available from: http://www.innovateuk.org/_assets/pdf/press-releases/11_aug_10_ucd.pdf (Accessed 13/12/2010) Hadi, M. & Halfhide, C., 2011, Green Buildings: Understanding the role of end user behaviour, In: D, Barlett (Ed.), Going Green: The Psychology of Sustainability in the Workplace, Leicester: The British Psychological Society, pp. 31-35.

Morrow, W., 1995, Personal Environments and Productivity in the Intelligent Building, Intelligent Building Institute Intellibuild 95, Atlanta, Georgia

Ni Hogain, S., n.d., A New Energy Strategy: The second report of the Zero Carbon Britain project, Zero Carbon Britain 2030, CAT Publication, M. Kemp, Ed. 2010, pp.4. Zhang, Y., Barrett, P., 2010, Findings from a post-occupancy evaluation in the UK primary schools sector", Facilities, Vol. 28, pp.641 - 656

113

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

A Review Of Currently Available Standards And Software Tools For Assessing Life Cycle Greenhouse Gas Emissions From Buildings.

H. J. Darby1*, A.A. Elmualim2, F. Kelly3 1Technologies for Sustainable Built Environments, University of Reading, UK 2School of Construction Management and Engineering, University of Reading, UK 3Peter Brett Associates LLP, Reading, UK

*Corresponding author: [email protected]

ABSTRACT Buildings are estimated to be responsible for around 50% of greenhouse gas (GHG) emissions in the UK. There is potential to reduce GHG emissions through both the operational and embodied forms. To date, attention has been focused on operational reductions, although in terms of meeting the UK carbon reduction targets it may be more important to consider early embodied reductions. This paper reviews currently available standards for assessing life cycle GHG emissions from buildings, evaluates a sample of available software tools for calculating these emissions and investigates the present values of future emissions. The study demonstrates the lack of a standardised approach to the calculation of embodied emissions and of reliable data on emission factors for building materials and processes. It indicates that embodied emissions can be a major consideration and that weighting of future emissions appears to be an important factor. Reduction factors accounting for the period emissions are present in the atmosphere during a 100-year assessment period had only a modest effect but weighting to allow for future decarbonisation of the energy supply had a large effect.

Keywords: Greenhouse gas emission; building; embodied; operational, carbon.

1. INTRODUCTION

Global temperatures are expected to rise by between 1.1 and 6.4oC this century, depending, to a large extent, on the quantity of greenhouse gases (GHG) we emit to the atmosphere from now onwards. This warming is expected to have very negative effects on many peoples and ecosystems and, therefore, minimising our greenhouse gas emissions is a priority.

Buildings are estimated to be responsible for around 50% of GHG emissions in the UK (BIS, 2010). Potential reductions involve both operational emissions, produced during use, and embodied emissions, produced during manufacture of materials and components, and during construction, refurbishments and demolition.

To date, the UK government has focused attention on reducing the, apparently, larger operational element. For example, the Building Research Establishments Environmental Assessment Method for buildings (BREEAM) allows the best environmental rating to be achieved without any consideration of embodied emissions

114

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ (BRE, 2009a) and the UK Building Regulations have a phased approach to increasing operatonal energy efficiency, resulting in reduced operational carbon emissions (Oc), in the period up to 2020. This focus, together with the lack of reliable, readily available and usable data on embodied emissions, has hampered the consideration of embodied carbon emissions (Ec) during building design.

Currently, there is lack of a consistent and accepted approach to the calculation of Ec, the relationship and interaction between the Ec and Oc is not well understood, and there is considerable uncertainty and variability in the available information on emission factors for building materials and processes.

In terms of meeting the UK carbon reduction targets of 34% by 2020 and 80% by 2050 (measured against the 1990 baseline), it may be equally, if not more, important to consider early embodied carbon reductions in buildings, rather than just future operational reductions. The UK Low Carbon Transition Plan (DECC, 2009a) includes measures for decarbonisation of the electricity supply over the period 2010 to 2050 (DECC, 2009b). This together with more efficient lighting and M&E equipment installed in future refits is likely to significantly reduce future operational emissions. These issues lend further weight to the argument for considering early embodied reductions. Methods of evaluating the present value of future emissions, may allow more realistic comparisons to be made between the embodied element, occurring largely at construction stage, and the operational element, occurring over the lifetime of the building.

The process of determining Ec, although simple in principle, is not straightforward in reality due to uncertainty and lack of information on emission factors for all the materials and processes involved. There is considerable variability between the currently available databases for emissions from materials alone and there are many software tools and online ‘carbon calculators’ available purporting to provide whole life assessments. In many cases it is unclear exactly where the assessment boundaries have been drawn and which data sources have been used.

This paper reviews the currently available standards relating to the assessment of life cycle greenhouse gas emissions from buildings, and describes a case study of a book storage building in the UK, where Ec has been investigated using three available software tools together with the author’s own evaluation. Methods of assessing the present values of future emissions, based on the period emissions are present in the atmosphere, and on future decarbonisation of the energy supply are investigated. This work is part of a research project aimed at producing information on embodied emissions for different types of building, components and forms of construction, in a simplified form, which can be readily used by building designers to optimise building design in terms of minimising overall GHG emissions.

2. LIFE CYCLE GHG EMISSIONS AND STANDARDS

The internationally recognised standard for life cycle assessment (LCA) in environmental management is ISO 14040 (BSI, 2006a), which encompasses all environmental impacts. Life cycle greenhouse gas emissions, or carbon footprints, are a

115

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ sub-set of this more comprehensive LCA.

In this paper the carbon footprint of a building is defined as the total GHG emissions caused by the building project, from extraction of raw materials, design and production, use of the building and through to disposal at the end of life, as illustrated in Figure 1.

In simple terms a building project is itself a product comprising of processes using a collection of sub-products, each with their own processes.

Figure 1: Life cycle of a product (UNEP, 2010)

Currently available standards for assessing GHG emissions, in the UK include:

 The Greenhouse Gas Protocol (WRI and WBCSD, 2004) developed by the World Resources Institute and the World Business Council for Sustainable Development in 1998 and revised in 2004

 ISO 14064 (2006) Greenhouse Gases (BSI, 2006b, 2006c, 2006d), gives guidance on quantification and reporting of GHGs at the organisational and project levels, and on validation and verification of assertions.

 PAS 2050 (BSI, 2008) uses the ISO 14040 methods and clarifies the assessment of GHG emissions for goods and services. All GHGs contributing more than 1% of the total footprint are required to be included. It also considers a method for taking account of how emissions at different times during a product life cycle can have different impacts on climate, and for the inherent uncertainties in life cycle assessments. It appears to be aimed more towards smaller scale ‘products’, but apparently the principles could be applied to full building projects. Although not considered as a British, European or International Standard, it is categorised as a “publically available standard”, which is essentially a step in the procedure of standardisation, short-cutting the full process to meet the urgent need for a consistent assessment method. It is understood that the intention is for it to be developed into ISO 14067 at some future date.

 CEN Technical Committee TC 350 “Sustainability of construction works” (BSI, 2011a) is developing a harmonised approach to the measurement of embodied and operational environmental impacts of construction products and whole buildings, across the entire life cycle. At this stage the following parts of Standard EN 15643: Sustainability of construction works have been produced:

116

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ - Part 1: General framework (BSI, 2010). - Part 2: Framework for the assessment of environmental performance (BSI, 2011b).

Again these encompass a wide range of environmental impacts, and it is not clear at this stage when the full suite of standards will become available and if there will be any specifically dealing with GHG emissions. However it does provide the diagram reproduced in Figure 2, showing the information required at the different life cycle stages, which are needed to carry out a full assessment.

The system boundary in EN 15643 does not include reuse, recovery and recycling potential, although this can be included as an optional stage. This appears to be contrary to the general approach of the other standards , which generally only permit the deletion of a life cycle phase if it does not significantly change the overall conclusions of the study. End of life treatment of certain common building materials, such as steel and timber, can substantially affect the overall GHG emissions.

Figure 2: The information modules applied in the assessment of environmental

performance of a building from its life cycle stages in BS EN 15643-2

Key issues in determining building carbon footprints is the uncertainty and variability in the available information on emission factors for building materials and processes. This

117

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ results from a lack of transparency and of a consistent and accepted method for the calculation of embodied emissions. Common differences are in the different system boundaries used and whether the emissions considered are from all GHGs (commonly reported as carbon dioxide equivalent or CO2e) or from just carbon dioxide (CO2).

The development of these standards will no doubt, improve the consistency and reliability of emissions reporting, although there is still scope for variability in the data, life cycle boundaries selected, and assumptions made

3. CASE STUDY

3.1 Description Of The Building

The case study building is a book storage facility, completed in October 2010, with a total internal floor area of 11,578m2, designed to house around eight million volumes, in carefully regulated internal temperature and humidity conditions. It consists of both warehouse and office type accommodation. It has a steel framed structure on mass concrete foundations, in-situ concrete floor and precast concrete insulated sandwich cladding panels to the external walls. The 12m-high main storage chamber has insulated metal deck roof cladding and the lower 7m high section has a composite concrete and metal deck roof to support the building plant and equipment. The building is well insulated, with high thermal mass provided by the external wall panels, and is well sealed, achieving an on-site air test result of 1.6m3/h/m2.

3.2 Estimate Of Emissions

Three commonly available software tools were used to estimate emissions. These are described in the Sections.3.2.1 to 3.2.3 below.

Each of these tools have their individual limitations and therefore a separate assessment of Ec was made, based on a detailed analysis of ‘as constructed’ information to determine quantities, types of building materials, and processes used. Carbon emission factors from a variety of sources were then used to calculate Ec. (Anderson et al., 2009), (Concrete Centre, 2010); (EA, 2010), (DECC, 2011), (Granta Design, 2010), (Hammond and Jones, 2008), (Hutchins, 2010), (TTL, 2009), (WSA, 2010).

Only one of the tools estimated Oc and therefore an estimate, based on a predicted 2 annual ‘Building Emission Rate’ of 20kgCO2/m , provided by the building designers, was used as a comparison. This gives a predicted annual Oc of 231tCO2 for the whole building.

3.2.1 Envest 2

Envest 2 (BRE, 2009b) is a web based tool for estimating the lifecycle environmental impacts and costs of office buildings. It is normally available on a subscription basis, although it was provided on a free trial basis for the purposes of this case study. It was developed by the UK Building Research Establishment and uses their own extensive research data on materials, operational use, impacts and weightings.

All environmental impacts are measured using a single points scale called ecopoints. This includes an impact called ‘Climate change’, which is an estimation of the GHGs

118

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ emissions and includes Oc and Ec, the latter including both the building fabric and services.

Input is in a simplified form based on the building overall dimensions and configuration and on a choice of building elements and material specifications. Material quantities are not required for the input.

3.2.2 CES Eco Selector

CES Eco Selector (Granta Design, 2010) is a general software tool used for product design or assessment. It contains an extensive materials database, which includes engineering properties, ecologiical impact and price data, and allows product design to be optimized in terms of material, ecological impact and cost. The software is produced by Granta Design Limited, who specialise in materials information technology, in collaboration with Cambridge University, and is available via purchased licence.

The integral Eco Audit tool estimates energy use and CO2 emissions for each of the material, manufacture, use and disposal life cycle phases.

All material quantities need to be calculated prior to input to the software.

3.2.3 Environment Agency Carbon Calculator

The Environment Agency Carbon Calculator (EA, 2010) is an Excel spreadsheet designed to calculate the Ec of construction projects. It estimates CO2 emissions from materials and associated transportation, personal travel, and site energy use and waste management, and is available to download, free of charge, from the Environment Agency website.

Information on material emissions is primarily obtained from the ‘Inventory of Carbon and Energy’ (ICE) database (Hammond and Jones, 2008), and all quantities need to be calculated prior to input to the software.

3.3 Assumptions

The following assumptions were made for the case study: 60 year building life; three M&E refits at 15, 30 and 45 years and a general building refurbishment at 30 years.

119

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ End of life treatment has been assumed to be based a hierarchy of: recycled where possible (e.g., steel); downcycled to a usable but lower grade material (e.g., concrete); incinerate for energy generation (e.g., timber); landfill.

- Transport emissions were based on actual travel distances from suppliers to site, where this was known. Distances could be determined for the majority of the large quantity materials, but estimates were made where this information was not available. - Emissions from construction and demolition works on site were based on the actual construction period and an estimate of the period required for demolition.

3.4 Results From The Case Study

A summary of the estimated emissions from the three software tools and the author evaluation is given in Table 1 and in Figure 3.

CES Eco Selector EA Carbon Calculator Envest 2 Author

tCO2 tCO2 tCO2e tCO2

Material 5111 117% 4423 101% na 4373 100%

Manufacture 366 90% 465 115% na 405 100%

Transport 161 32% 379 74% na 511 100%

Sub-total 5639 107% 5267 100% 4358 82% 5289 100%

Use (Oc) na na 23637 171% 13838 100%

End of life -1819 641% na na -284 100%

Total 27995 149% 18843 100%

(na=not available) Table 1: Estimated emissions

The results show that the Sub-total Ec figures (material + manufacture + transport), varied by +7% to -18%compared with the author. This is effectively the ‘cradle to site’ Ec. If the end of life values are included (‘cradle to grave’ Ec) the variation is up to - 24%. End of life figures were only available for the CES and Author cases. Figures for Oc were only available from the Envest 2 and Author figures and the variation is +71%.

120

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Figure 3: Estimated emissions

4. PRESENT VALUE OF FUTURE EMISSIONS

Previous studies (Darby, 2010) have shown that lifetime Oc is generally larger than Ec. However, Oc occurs over the lifetime of a building (generally 60 to 100years), whereas the vast majority of Ec occurs at the start of a building’s life. In terms of the timeframe in which carbon reductions need to be made, it is possible that carbon savings made at the start of a building’s life could be more valuable than predicted savings in the future. The effect of future decarbonisation of energy supply could have a profound effect on future emissions, as could more efficient lighting and M&E equipment installed in future refits. It is argued (Stern, 2006) that the cost of measures to mitigate climate change increase for every year the measures are delayed. For these reasons, two methods of weighting future emissions are investigated, using the data from the author’s estimate of emissions for the case study building, which may allow more realistic comparisons to be made between EC and OC.

The first is based on PAS 2050, which accounts for the reduced period emissions are present in the atmosphere during a 100-year assessment period and the weighting factor is given by Equation (1). This is a simplified version of the approach outlined by the IPCC (Solomon et al., 2007).

100 X 100 i Weighting Factor  i1 i (1) 100

Where i is the year in which emissions occur and X is the proportion of total emissions occurring in that year i.

The second approach is based on the UK Government Markal-Med model scenarios for decarbonisation of the electricity supply over the period 2010 to 2050 (DECC, 2009b). Mean values of the scenarios considered have been used to give reduction factors to be applied to each of the annual Oc and the future Ec. These factors apply to future electricity supply but for the purposes of this study they are applied to energy supply as a whole

Table 2 shows the relationship between Ec and Oc for the different scenarios considered in the study.

121

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ The range of Ec as a percentage of the total emissions was found to be 27% to 57%.

unweighted PAS 2050 PAS 2050 + Markal Cradle to site Cradle to grave Cradle to site Cradle to grave Cradle to site Cradle to grave Ec 27% 32% 31% 32% 57% 57% Oc 73% 68% 69% 68% 43% 43% Total 100% 100% 100% 100% 100% 100% Table 2: Case study building: proportions of Ec and Oc

Figure 4 shows CO2 emissions for the case study building lifetime phases. Applying the PAS 2050 reduction factors has a relatively small impact, whereas applying the PAS 2050+ Markal reduction factors has a large effect.

Figure 4: Llife cycle CO2 emissions with weighted future emissions

The emission profiles over the building lifetime are clearly demonstrated in Figure 5. This indicates that Ec is equivalent to approximately 23 and 25 years of Oc respectively in the unweighted and PAS 2050 cases and more than 60 years in the PAS 2050+Markal case.

Figure 5: Ec and Oc as a percentage of whole life emissions

122

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

5. DISCUSSION AND CONCLUSIONS

The estimates of Ec (cradle to site) using the three software tools and author evaluation appear relatively close, but this result needs to be treated with care. If examined more closely there are differences in what is being measured. The end of life stage is only assessed in the CES and Author estimates and, if included, the variation is 24%. The results are for CO2 in the case of CES, EA and Author and CO2e (the Kyoto full basket of GHGs) in the case of Envest 2. More data are available for CO2 and it is estimated (Hammond and Jones, 2008) that, for most building materials, this represents the vast majority of GHGs (e.g., around 95%). However, for plastics other GHGs are more significant. It is anticipated that as materials data evolves and becomes more comprehensive, factors including all GHGs will become available.

Greater differences were observed between the estimates for different materials. For example, steel sections 91%, concrete 56%, timber 1200% and blockwork 184%, expressed as a percentage of the minimum. These essentially reflect the difference in emission factors used and can have an overwhelming effect on an assessment, depending on the quantities of the materials involved, and can lead to incorrect choices of materials and construction systems if assessments are made at the design stage.

The case study indicates that Ec can be an important consideration in lifecycle CO2 emissions from buildings. A review of other previous case studies (Darby, 2010) indicated that Ec varied between 79% and 108% of Oc for warehouse buildings and was as low as 2% for some other buildings. In this study the range was found to be between 19% and 133%, demonstrating that different assumptions and boundary conditions can produce widely differing results and confirming the need for standardisation of Ec data and assessment methods. Assessment of lifecycle carbon emissions should not be considered a precise process due to the many uncertainties and assumptions that have to be made. Nevertheless, it can be a useful tool in assessing where emission reductions can be made most effectively.

Oc in this case study is based on estimates and predicted values. The building was only completed in October 2010 and, therefore, there is currently no data available to assess their accuracy, although these will be available for future updates of the study.

Weighting of future emissions appears to be an important factor to consider. The PAS 2050 reduction factors, to reflect the period emissions are present in the atmosphere during a 100-year assessment period, had only a modest effect, possibly due to the relatively short building life. Longer life buildings would increase this effect but would conversely tend to reduce the Ec proportion of the whole life emissions. In contrast, weighting due to decarbonisation of the energy supply has a large effect, although a more detailed analysis of whether the rate of decarbonisation of the electricity supply will be reflected in the overall energy supply is required before the method used in this study can be validated. No attempt has been made to include discounting based on the cost of measures to mitigate climate change increasing with time but is considered that it would be useful to investigate this issue further.

It is considered that the results of this case study make a useful contribution to the overall bank of data on Ec in buildings.

REFERENCES

Anderson, J., Shiers, D., Steele, K. (2009), The Green Guide to Specification, Fourth Edition, IHS BRE Press ISBN 978-1-84806-071-5.

123

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ BIS (Department for Business, Innovation and Skills) (2010), Low Carbon Construction, Innovation and growth team: Emerging Findings (report), BIS, London. BRE (Building Research Establishment) (2009a), BRE Environmental Assessment Method (BREEAM), www.breeam.org (consulted November 2009). BRE (Building Research Establishment) (2009b), Envest: Environmental Impact and Whole Life Costs for Buildings (software), BRE, Watford. BSI (British Standards Institution) (2006a), BS EN ISO 14040:2006: Environmental management – Life cycle assessment – Principles and framework, BSI, London. BSI (British Standards Institution) (2006b), BS ISO 14064:2006: Greenhouse gases - Part 1: Specification with guidance at the organization level for quantification and reporting of greenhouse gas emissions and removals, BSI, London BSI (British Standards Institution) (2006c), BS ISO 14064:2006: Greenhouse gases - Part 2: Specification with guidance at the project level for quantification, monitoring and reporting of greenhouse gas emission reductions or removal enhancements, BSI, London. BSI (British Standards Institution) (2006d), BS ISO 14064:2006: Greenhouse gases - Part 3: Specification with guidance for the validation and verification of greenhouse gas assertions, BSI, London. BSI (British Standards Institution) (2008), Publicly Available Specification PAS 2050:2008: Specification for the assessment of the life cycle greenhouse gas emissions of goods and services, BSI, London. BSI (British Standards Institution) (2010), BS EN 15643-1:2010 Sustainability of construction works - Sustainability assessment of buildings - Part 1: General framework, BSI, London. BSI (British Standards Institution) (2011a), BS EN 15643-2:2011 Sustainability of construction works - Sustainability assessment of buildings - Part 2: Framework for the assessment of environmental performance, BSI, London. BSI (British Standards Institution) (2011b), www.bsigroup.com/Standards-and- Publications /Committee-Members/Construction-committee-members-area/M350- standards/?id=158921, (consulted April 2011). Concrete Centre (2010), Concrete Structures 10, The Concrete Centre, Camberley, Surrey. Darby, H. (2010) The Carbon Life Cycle of Buildings: A Review of the Current UK Carbon Emissions Reduction Strategy for Buildings”, TSBE Conf 2010, University of Reading. DECC (Department of Energy and Climate Change) (2009a), The UK Low Carbon Transition Plan, DECC, London. DECC (Department of Energy and Climate Change) (2009b), Analytical Annex: The UK Low Carbon Transition Plan, DECC, London. DECC (Department of Energy and Climate Change) (2011), UK emissions statistics, http://www.decc.gov.uk/en/content/cms/statistics/climate_change/gg_emissions/uk_emi ssions/2009_final/2009_final.aspx, (consulted April 2011) EA (Environment Agency) (2010), Carbon Calculator, version 3.1.2 (software), Environment Agency. Granta Design (2010), CES Selector 2010 (software), Granta Design, Cambridge.

124

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Hammond, G. and Jones, C. (2008), Inventory of Carbon & Energy (ICE) Version 1.6a, University of Bath. Hutchins (2010), UK Building Blackbook, The Capital Cost and Embodied CO2 Guide, Volume Two: Major Works, Franklin and Andrews, London. Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K.B., Tignor, M., Miller H.L. (eds.) (2007), Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, 2007, Cambridge University Press, Cambridge. Stern, N. (2006), Stern Review on the Economics of Climate Change, www.hm- treasury.gov.uk/stern_review_report.htm, (consulted 20/05/10). TTL (TRADA Technology Limited) (2009), Construction Briefings: Timber carbon footprints, TTL, High Wycombe. UNEP (United Nations Environment Programme) (2011), www.unep.fr/scp/lifecycle/ (consulted April 2011) WRI (World Resources Institute) and WBCSD (World Business Council for Sustainable Development) (2004), The Greenhouse Gas Protocol, A Corporate Accounting and Reporting Standard, revised edition, WRI&WBCSD, WSA (World Steel Association) (2010), LCI Data for Steel Products (report), WSA, Brussels, Belgium.

125

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Sustainable Planning In The Era Of The Localism Bill

T.McGinley*1, K. Nakata2 J. Robey3,

1 Technology for Sustainable Built Environments, University of Reading, United Kingdom 2 Informatics Research Centre, University of Reading, United Kingdom 3Capgemini UK, Forge End, Woking, Surrey, GU21 6DB

* Corresponding author: [email protected]

ABSTRACT The Localism Bill is currently being read in the House of Lords. The Bill presents a series of policy reforms that are being promoted by the government to empower citizens to participate in local decisions under the umbrella term of the ‘Big Society’ (BS). This paper addresses concerns outlined in the Environmental Audit Committee (EAC) report on Sustainable Development in the Localism Bill. We focus our analysis on the policy of neighbourhood planning which aims to give people greater ownership of plans and policies that affect their local area. The concerns of the EAC report are addressed by compiling a list of requirements for ‘sustainable’ neighbourhood planning (SNP). In addition to the requirements we developed a set of factors to represent the capabilities of two cutting edge community engagement platforms (The Place Station and Simpl.co). By mapping the capabilities of the CEPs to the requirements identified for SNP, we highlight a number of gaps which require further research.

Keywords: Community engagement, community engagement platforms, participation, sustainable development

1. INTRODUCTION

Enabling community engagement in planning decisions through ‘neighbourhood planning’ forms a key policy of the ‘Big Society’ (BS) policy reforms. The aim of neighbourhood planning is ‘to give people greater ownership of plans and policies that affect their local area’ (CLG, 2011). There has been some criticism of neighbourhood planning’s consideration of the sustainable development agenda (EAC, 2011; Porritt, 2011). This paper will examine the neighbourhood planning concerns and use these to produce a set of the requirements for sustainable neighbourhood planning (SNP). The Confederation of British Industry called in August 2010 for the Localism Bill to have been given royal assent by the end of February 2011 (CBI, 2010). However at the time of writing this paper the bill is being prepared for second reading in the House of Lords. The CBI recommendations highlighted the damage that could be done to the renewable energy economy if planning reform was delayed or diluted. Due to this delay this report concentrates on the studies and commentary based on the initial readings of the Localism Bill. SNP will require community engagement in order to be effective. One method to engage communities is with community engagement platforms (CEPs). This paper assesses if the concerns about neighbourhood planning can be alleviated with the capabilities of state of the art CEPs. The identification of the gaps between the

126

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ requirements of SNP and the functional capabilities of the CEPs will indicate areas of future work to develop or employ new tools to bridge the gaps. The gaps may also identify an opportunity for policy reconsideration. The aim of this paper is to identify the gaps between the requirements of ‘neighbourhood planning’ and the capabilities of community engagement platforms using the following objectives:

 identify the technical and non technical requirements of sustainable planning in the era of the Localism Bill;  identify the technical capability gaps in current community engagement platforms;  propose solutions to the capability gaps in community engagement.

2. BACKGROUND

The idea of power decentralisation and enhanced citizen participation is not new. Campbell and Marshall (2000) and Foley and Martin (2000) describe approaches by previous UK governments to enhance public engagement. Smith (2010) suggests a cyclical tide of government policy in the UK that alternate between emphasising the role of government and the responsibility of the individual. The proposed Localism Bill sits within this cycle of increasing the responsibility of the individual; however it also represents the biggest reform of the planning process in the UK for a generation (Ellis, 2011). The government blames the UK's drop in the World Economic Forum’s global competitiveness index from 4th in 2000 to 12th in 2009 in part on what it calls an ‘adversarial planning regime’ (BIS, 2011). One of the primary functions of planning in the UK is to ensure sustainable development (CLG, 2011b). Ellis (2011) is concerned by the government’s justification for planning reform is that ‘sustainable development’ must not get in the way of growth. A recent poll reported that 72% of people are confused as to the exact meaning of the governments ‘Big Society’ (YouGov, 2011). However, although politically the exact meaning may still be evolving, the issues it touches upon including community engagement and citizen participation are well documented in academic literature, however for simplicity this paper focuses on the work of Arnstein (1969) and the ‘Ladder of Citizen Participation’.

The rungs of Arnstein’s ladder are primarily divided, in ascending order, into nonparticipation, tokenism and citizen power. If we suppose that the participatory intention of the Localism Bill will be delivered as presented in the launch documentation for the Big Society, the reforms would address, citizen control, the top rung of Arnstein’s citizen participation. Arnstein’s paper presents concerns that citizen control; supports separatism, fragments public services, is open to exploitation, incompatible with the merit systems and professionalism, more costly and less efficient and requires financial resources to accompany control. This paper will use the concerns provided by Arnstein to provide a context to our discussion of sustainable community planning requirements and capabilities.

4. METHODOLOGY

The approach taken in this paper is to identify the requirements of SNP, assess the capability of two CEPs and propose solutions to the gaps identified between the requirements and capability factors. The requirements were extracted from a set of documents that expressed sustainable development concerns about the neighbourhood

127

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ planning. To achieve this we extracted and analysed an initial set of requirements from the literature looking for duplications, merging and filtering out the less relevant requirements to arrive at a consolidated set of requirements. The capability factors were derived from focus group sessions involving the practitioners of community engagement platform development (Simpl, 2011b; ATU, 2011b) to identify the capability factors to include in the assessment and then assign values according to the platform’s performance for that capability. It was then possible to map the SNP requirements and CEP capabilities on to a graph and identify the gaps. Once the gaps had been identified, solutions to fill the gaps were proposed.

5. REQUIREMENTS

The consolidated set of requirements for SNP has been identified in the sections below. To make the following sections easier to follow the requirements are indexed and prefixed with the letter ‘R’. R1 and R2 define the requirement for neighbourhood planning authorities (NPAs) which in order to define and neighbourhood development plans (NDPs). R3 outlines the main concern of the EAC report, and the remaining requirements deal primarily with the community engagement concerns of Arnstein (1969).

5.1 Define the NPA and avoid unfair influence (R1)

R1 addresses the need to define the NPA ‘qualifying body’ (HOC, 2011; Porritt, 2011) and the need to avoid unfair influence in the process including businesses and individuals (Arnstein, 1969). There is concern that excessive businesses influence in the planning process will lead to business growth being the only agenda without consideration of sustainable development (Ellis, 2011) this is in part due to the lack of accountability of the NPA. Porritt (2011) is concerned that pro growth business objectives in planning and sustainable development are ‘entirely incompatible’. Plans will only be brought into force if they achieve a majority in a community referendum. The referendum alone will not address Porritt’s concerns that individuals should have the right to resist environmentally or socially insensitive plans. Consideration needs to be given in the following requirement to avoid unfair influence and the other pitfalls of ‘citizen control’ in Arnstein’s ladder by reaching underrepresented sections of the community in the development of the plan.

5.2 Define the Neighbourhood Development Plan (R2)

Following the definition of the neighbourhood planning authority (NPA), it is proposed in the Localism Bill that the NPA should produce ‘neighbourhood development plans’ (NDPs) (HOP, 2011). The Environmental Audit Committee (EAC) outlined concerns that poorly managed NDPs could have a detrimental impact on the sustainable development of the communities (HOC, 2011). This concern requires a capacity for community engagement and an ability to collaboratively capture the spatial requirements and aspirations of the community.

5.3 Monitor the cumulative impact (R3)

The planning reforms packaged in the Localism Bill represent the biggest changes to planning in the UK in a generation (Smith, 2011). National targets have been replaced

128

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ with NDPs. Although these will have to conform to the National Planning Framework there is a concern that only 22 out of the 354 English councils have adequate climate change policies in place to meet their 2020 40% emissions reduction targets (Porritt, 2011). Porritt’s concerns highlight the risk to sustainable development at a local level. In addition, the EAC is concerned that although the act of creating local plans at a local level could be a responsive and engaging policy it will result in possibly thousands of disparate plans. This will make the process of simulating changes in planning policy at a national level a complicated and expensive matter of joining up multiple non-standard plans (EAC, 2011). Therefore a technical requirement of SNP would be an ability to monitor the cumulative effect of the plans.

5.4 Asset Transfer Facility (R4)

An asset transfer facility (for example transferring ownership of a library to a community) would provide support for, both the community or NGO the assets are being transferred to, and the department the assets are being transferred from. This may include a technical requirement to enable the transferees to develop their idea about the future use of an asset and an opportunity for the transferor to highlight its suitable assets.

5.5 Enable shared services (R5)

Arnstein (ibid) reports that citizen control can support separatism and fragment public services. This will be of particular concern to local and central government, in a similar way to R3 with repercussions on R7 (reducing cost and improving efficiency). However this requirement highlights a capability of SNP to link and share services from within the same platform that captures the requirements of individuals for neighbourhood planning.

5.6 Rating system and user profiling (R6)

Arnstein (ibid) informs us that citizen control is inconsistent with meritocracies and professionalism. It is therefore a requirement of sustainable neighbourhood planning that the skills, capabilities, interests and concerns of the individuals are recognised.

5.7 Reduce Cost and Improve Efficiency (R7)

The Big Society has been criticised by the Head of the NGO, Community Service Volunteers, as a cover for cuts in services that are being carried out in order to save money (BBC, 2011). Therefore, Arnstein’s (1969) observation that community control can in the long run be ‘more costly and less efficient’ should be of concern to the government when this is coupled to the complexity of R3 (monitoring the cumulative impact of the plans).

5.8 Requires financial resources and support (R8)

Due to ongoing reductions in government funding, SNP needs to be able to identify and attract alternative forms of funding. A possible area of funding may originate from businesses responding to their corporate social responsibility requirements.

129

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 6. CAPABILITY FACTORS

Capability factors are used in this study to represent a community engagement factors platforms capacity perform a function. The factors were developed in discussion with experts from the The Place Station and Simpl.co platforms (ATU, 2011b; Simpl, 2011b). We have assumed that the platform is online and that users have access to the Internet. The factors are listed in table 1 below.

7. REQUIREMENT TO CAPABILITY MAPPING

Having developed the consolidated list of requirements for sustainable neighbourhood planning we mapped the capability factors set out in Table 1 to these requirements. The results can be seen in Table 2 below. We will also track developments of the Localism Bill to see if these requirements are satisfied by changes in the proposed legislation.

Table 1. List of Capability factors including rating system Factor Scale C1 Accessibility - measures the known digital barriers to 1 = lots of barriers, engagement. 6 = no barriers C2 Location - describes the platform’s capability to 1= by country, express the granularity of an idea’s location. 6 = space within a building. C3 Idea development - depicts the capacity of the 1= simple comment system, platform to collaboratively develop an idea. 6= manipulate an object oriented idea. C4 User Centred - demonstrates the capability of the 1 = login, platform to adapt its interface, information and rules 6 = the platform adapts holistically to the user based on the profile of the user requirements C5 Mobility - shows The extent to which the platform is 1 = works on mobile devices, compatible with mobile devices 6 = cross mobile browser support with native apps C6 Communication - describes the ability to utilise 1 = no social media strategy, social media to enable a community engagement idea 6 = scrape all conversations about an idea C7 Scalability - site specific, or can any idea be 1 = site specific, supported, anywhere? 6 = global and extendable C8 Openness - the extent to which the platform is open 1= proprietary, to developers and users. 6 = open source with API C9 Support - the capacity of the system to identify 1= lists all available support support for an idea. 6= personalised support based on user profile

Table 2. Mapping of the compatibility factors (Table 1) to the consolidated requirements Compatibility match R1 Defining the organisation that develops the plan relates to the platforms accessibility (C1) factor, which should include polarisation avoidance. R2 The requirement to define the neighbourhood plan, maps to the Idea development (C3), location (C2) and mobility (C5) factors R3 Monitoring the cumulative impact could be solved technically by linking the data from the NDPs and NDOs in real time to provide the users with real time feedback on the effect of their actions. This would rely on the openness (C8) and location (C2) capability of the platform R4 An Asset transfer facility requirement maps to the Idea development (C3) factor. The platform can assist by attracting and connecting relevant support from business and alternative forms of funding. This is an existing capability of The Place Station and Simpl.co. In future work we will add the support factor to encapsulate more detail about the functionality of the platforms

130

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ R5 The Enabling shared services requirement maps to the communication (C6) and scalability (C7) factor. Provision of a shared services platform for the different services. Simpl.co is an example of a shared social innovation platform for local government. R6 Rating system and user profiling maps to the user centred (C4) factor R7 Reducing cost and improving efficiency maps to the openness (C8) factor, if an open source community development team can be motivated to work on the platform. R8 Requires financial resources and support maps to the support (C9) factor.

8. CAPABILITY ASSESSMENT OF COMMUNITY ENGAGEMENT PLATFORMS

The platforms that are analysed in this study are the Asset Transfer Unit’s The Place Station (referred to hereafter as P1) and FutureGov’s Simpl.co platform (referred to hereafter as P2). The aim of P1 is to ‘introduce owners of land and buildings to social and community entrepreneurs with ideas for transforming their local area’ (ATU, 2010). A new version of P1 is currently being tested that has been designed to respond to the requirements of neighbourhood planning. Simpl.co (P2) is a platform designed to allow communities and social innovators to engage with local authorities in the UK and US. The platform is an example of a shared social innovation platform for local government. In contrast to the P1 platform, P2 is not designed to be a neighbourhood planning platform. However Simpl has been included to provide an example of a state of the art community engagement platform. P2 is not a spatial platform and therefore has a lower location factor than P1; however it is a sophisticated community idea generation tool. Capability P1 P2 simpl.co (P2) The Place Station (P1) Accessibility 5 4 Location 4 1 Idea 3 3 Accessibility Development 6 User Centred 4 4 5 Support Location Mobility 4 5 4 3 Communication 4 4 Scalability 6 6 2 Idea Openness Openness 4 3 1 development 0 Support 5 4

Scalability User Centred

Communicatio Mobility n

Figure 1 The Place Station (P1) and Simpl (P2) Capability Assessment

9. DISCUSSION

The aim of this paper is to identify the gaps that exist between the requirements of sustainable neighbourhood planning and the capabilities of community engagement platforms. From the findings presented in Figure 1 we identify and discuss the capability gaps. Firstly, In order to define the neighbourhood planning authorities and avoid unfair influence (R1) we considered the capability for accessibility (C1) and mobility (C5). In terms of reaching underrepresented sections of the community, mobility may go some way towards addressing the issues of engaging teenagers; it is unlikely that other hard to reach groups, such as the elderly, will be impacted by the mobility capability. The location capability factor (C2) could provide the capability to

131

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ capture the spatial requirements of the community within the geographical context of the plan. P1 was originally designed for the transfer of built assets to community groups and is currently considering the requirements of neighbourhood planning; therefore there is a score bias towards P1 in the location factor. P2 is not a location-based platform and therefore scores low. Demsoc (2011) reported that it is possible for NPAs to overlap. Where this happens it will further frustrate R3 (monitor cumulative effect). P1 has a lower mobility factor than P2; this is because P1 is spatial platform with a similar type of complex content as that required by SNP. It is therefore currently easier to produce a non-spatial community engagement platform on a mobile device. However technologies including Augmented Reality and Location Based Services currently being considered by P1 (ATU, 2011b) as well as mobile game technology, could reduce the gap between R1, R2 and C1 and provide fertile ground for future work.

Sustainable neighbourhood planning requires a capability to monitor the cumulative impact (R3) of neighbourhood planning on sustainable development. Both platforms also scored equally for communication. This capability is closely linked to the openness of the platform but responds exclusively to the platform’s approach to social media. This paper identified the capabilities of openness (C2). P2 scored highly due to an innovative strategy to provide the user data under a creative commons licence (Simpl, 2011b) whereas P1 was investigating licensing an application programming interface (API) for the platform (ATU, 2011b). R3 requires the NPAs of R2 to be produced in a way that is intuitive and engaging for the community whilst being open to automated analysis in order to ease the burden on monitoring the cumulative impact. It may be possible to alleviate some of the requirements of R3 with user profiling (C6) by directing individuals to other relevant services. The gaps in C6 could be addressed with a platform that adapts holistically to the user requirements. One route of analysis would be incorporating attention profiling mark-up language (APML) tools to passively respond to the users requirements.

The lowest scoring capability and therefore the biggest gap in capability currently exists in idea development (C3). C3 enables the communities to develop their built environment aspirations for their community. Both community engagement platforms used a comment system to develop the ideas. P1 also used a rating system responding to R6, whereas P2 opted out of using idea ratings due to a concern that this may unfairly reduce the visibility of an idea, based on a principle that ‘all ideas have value and deserve equal presence in the market place’ (Simpl, 2011b). The strategy of P2 provided a workable solution to R1 in that it reduces the ability of individuals to unfairly influence the neighbourhood development plan. It may be possible to reduce the C3 capability gap by utilising game technology to provide richer collaborative idea development.

Both platforms score highly for their support (C9) capability. P1 offers a phone number with dedicated support for asset transfers to community groups, and both platforms provide dedicated email and social network support to incubate the ideas and connect users with the platforms partners. Future research may include an analysis of the capability of the platforms to allow the users to dynamically build their own support networks using an object-oriented approach to ideas and projects. We are interested to explore the support capability that can be achieved by integrating a corporate social responsibility (CSR) framework to act to entice corporate aid or collaboration and therefore close the capability gap for R8.

132

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

10. CONCLUSION

This paper proposed a method for assessing the suitability of community engagement platforms (CEPs) to the requirements of neighbourhood planning. Future work should conduct a broader study of the available CEPs. The method presented here to analyse the requirements of a community engagement scenario and study can be generalised for many community engagement scenarios that consider using CEPs to address their requirements. A limitation of the research is that the potential users of the platform were not involved in the development of the capability factors and therefore they might not be representative of the users needs. In the coming months we will document the use of new and emerging CEPs in real neighbourhood planning case studies and focus on reducing:

 the support (C9) capability gap, by utilising CSR in response to R9;  the idea generation (C3) gap to enable the development of neighbourhood plans (R2) that include the capability to monitor cumulative impact (R3) ;  the shared services requirement (R5) with particular interest on the effect this can have on lowering cost and improving efficiency (R7) by reducing the gap in openness (C2) and improving user profiling (C4);  R1 (avoid unfair influence) can be addressed by improving accessibility (C1), mobility (C5) and building a stronger user profiling (C4) capability.

Whilst some of the gaps that exist between the requirements of sustainable neighbourhood planning and the CEPs can be filled by improvements in the CEPs, It is not possible to address all the SNP requirement gaps with CEP capabilities alone. Therefore some gaps such as the concern of Ellis (2011) that development and growth will be prioritised over sustainable development, must be explicitly addressed in the legislation.

REFERENCES

Arnstein, S. (1969), A ladder of Citizen Participation, Journal of American Planning Association, vol 35 iss 4 pp 216-224 ATU, (2011), The Place Station, Asset Transfer Unit, February 9th 2011. [Accessed April 18th 2011] URL: http://www.atu.org.uk/Support/theplacestation ATU (2011b), Interview with Sarah Eustace, Kings Cross Hub 5th May 2011. Bassoc, (2011) Steve Wyler to lead Locality - the new organisation created by the basic/DTA merger. Bassoc, 6th January 2011. [Accessed 15th March 2010] URL: http://www.bassac.org.uk/node/992 BBC, (2011) Cuts ‘destroying big society’ concept says CSV head, BBC News Politics, 7th February 2011, [accessed 20th February 2011]. URL: http://www.bbc.co.uk/news/uk-politics-12378974 BIS, (2011), The Plan for growth, Department for Business Innovation & Skills, March 2011 BSN, (2011) What is Big Society? Big Society Network, Accessed 25th February 2011. URL - http://thebigsociety.co.uk/what-is-big-society/ Campbell, H. Marshall, R. (2000), Public Involvement and Planning: Looking beyond the One to the Many. International Planning Studies, vol. 5, iss 3, pp 321-344 CBI, (2010), News Release: CBI warns uncertainty on planning undermining energy investment: business group sets out actions needed to deliver £150bn in low-carbon investment, Confederation of British Industry. URL:http://www.cbi.org.uk/ndbs/press.nsf/0363c1f07c6ca12a8025671c00381cc7/cc7ac0b73334ef ee8025777700379c67 CLG, (2011) Localism Bill: Neighbourhood plans and community right to build: Impact assessment. Department for Communities and Local Government, DCLG Publications, January CLG, (2011b), The planning system, Communities and Local Governments, URL:http://www.communities.gov.uk/planningandbuilding/planningsystem/ Demsoc (2011), White paper on neighbourhood planning, based on case study in Lewes. Democratic Society, 2011 (awaiting publication)

133

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Ellis, H (2011), Planning reform – ‘the continuous revolution’, Town and Country Planning, May 2011, Vol. 80, Issue 5, pp. 216-17 Foley, P. and Martin, S. (2000) A new deal for the community? Public participation in regeneration and local service delivery, Policy & Politics, vol 28 iss 4 pp 479-91 HOC, (2011) Sustainability of Planning reforms, Sustainable Development in the Localism Bill - Environmental Audit Committee. [Accessed on 22nd March 2011] URL: http://www.publications.parliament.uk/pa/cm201011/cmselect/cmenvaud/799/79904.htm HOP, (2011) Localism Bill Schedule 9: Neighbourhood planning, Part 1, Neighbourhood Development Orders. [Accessed on 2nd May 2011] URL: http://www.publications.parliament.uk/pa/cm201011/cmbills/126/11126.284-290.html#j878s Porritt, J. (2011), The Greenest Government Ever: One Year On, Friends of the Earth, A report to the Friends of the Earth. May 2011. Simpl (2011), The Place Station, Simpl: social innovation market place. [Accessed 15 April 2011] URL: http://www.simpl.co/The-Place-Station Simpl (2011b), Interview with Carrie Bishop, Kings Cross Hub 5th May 2011. Smith, M, J. (2011), The intellectual roots of the Big Society, The Big Society Challenge, Keystone Development Trust Publications, 2011 YouGov, (2011), YouGov / The Sun Survey Results, YouGov. February 2011. URL: http://today.yougov.co.uk/sites/today.yougov.co.uk/files/YG-Archives-Pol-Sun-BigSociety- 150211.pdf

134

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Review Of Factors Affecting Uncontrolled Ventilation In Food Supermarkets

S. Sawaf1*, H. Awbi1, J. Barlow2, J. Broadbent3 and B. Gregson3

1Technologies for Sustainable Built Environments Centre, University of Reading, UK 2Department of Meteorology, University of Reading, UK 3Johnson Construction, Gatehead Business Park, Delph, UK

* Corresponding author: [email protected]

ABSTRACT Mechanical ventilation systems account for a substantial proportion of the total energy consumption of retail buildings. Air infiltration through openings in the building fabric can increase the energy demands as well as reducing the level of comfort on the shop floor for staff and customers. Store location and orientation could make it more prone to higher wind penetration. Entrances in supermarkets represent a common case of uncontrolled ventilation which is determined by door sizes, positions and opening frequency, etc. that influence the amount of airflow going through these doors. Draught is caused by wind pressure and buoyancy force and very often these acting together and to reduce their impact, building designer often specifies wind lobbies, which often too costly and not always very effective. This paper will discuss the above factors with respect to large supermarkets to establish a best practice guide that could be followed to minimize the ingress of draught into retail building and reduce energy use.

Keywords: HVAC, infiltration, wind lobby, building location and building orientation.

1. INTRODUCTION

Energy consumption in food supermarkets is around 3.5% of the total UK energy consumption (Ge and Tassou, 2011). Energy consumption in retail supermarkets varies with the size and nature of the services they provide. Heating, Ventilation and Air- Conditioning (HVAC) systems contribute to a considerable amount of the total energy consumption and it is estimated that around 15%-20% of a supermarket’s energy is consumed in HVAC system and this figure varies according to the size of the store and services it provides.

Natural ventilation strategies in buildings are designed to save energy and remove the need for HVAC systems. However, in mechanically ventilated buildings uncontrolled ventilation affects the operation of the HVAC system. The infiltration will create imbalance between supply air and the extract and could cause discomfort for people inside the building (Tassou et al., 2011). As Building Regulations requirements on infiltration levels have become more stringent, door openings have become the main cause of infiltration in buildings. Different strategies are employed to limit air infiltration through door openings in supermarkets, such as using a wind lobby at the entrance door with an additional air curtain installed at the top of the internal entrance/ exit doors. However, air curtains are highly energy intensive (Tesco, 2008).

135

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Wind flow around buildings is affected by the surface roughness, urban development and other obstructions. These factors can have an impact on the air infiltration into a building. Supermarkets in sheltered areas have lower wind pressure acting on them than those exposed to the wind (Plate and Kiefer, 2001). Building orientation to the wind direction also influences the airflow and pressure distribution around them which affects the air infiltration through openings. Internal layout can also affect wind velocity and comfort inside the store (Kindangen, 1997).

Entrances in supermarkets are the primary cause of draught in the stores. Large food stores have larger openings to provide easy access and egress to customers with trolleys and baskets. Door size is specified in accordance to the store size; some supermarkets chains have their own regulations that they have to meet and these regulations should match the building regulations i.e. a store with a sales floor of 2000m2 requires minimum opening width of 2.4m where a store with a sales floor of 6000m2 requires minimum opening width of 2.8m. To limit draughts, supermarkets can install draught lobbies at the main entrances which have to conform to the relevant building regulation and standards.

The aim of this paper is to review factors influencing uncontrolled infiltration, and present results of preliminary onsite measurements.

2. VENTILATION IN SUPERMARKETS

2.1. Mechanical Ventilation

The primary role of a mechanical ventilation system in a building is to provide the required indoor environment for goods and users. Food supermarkets sell different types of commodities, some of which are kept on the shelf, others are refrigerated. The ventilation system needs to provide adequate ventilation for the goods as well as satisfying the customers and staff ventilation requirements. In supermarkets the recommended internal temperature is between 19°C and 21°C (CIBSE, 2005). Meeting the ventilation demands in such an environment needs an energy intensive system.

HVAC systems in food supermarkets vary according to the size of the store and the services provided. Keeping the temperature within the recommended range in such an environment puts an extra demand on the mechanical system (Tassou et al., 2011). The nature of supermarkets, with refrigerated food storage, puts an extra demand on HVAC systems, especially in winter when heating is also required. The spillage of cold air from display cabinets will create imbalances and requires the system to work harder to maintain the required temperature (Lindberg et al. 2007).

2.2. Air Infiltration

Uncontrolled ventilation through openings not only increases the demand for heating but it also affects the comfort of the people inside the building. Minimizing draughts will conserve energy as well as increase the level of comfort (Cóstola et al., 2010). A lot of work has been done to minimize the rate of infiltration in buildings. Part L of the Building Regulations requires all buildings to control the infiltration by performing tests to assess the rate of infiltration under controlled conditions (Part L, 2010). Some large

136

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ supermarket chains have exceeded the regulations by asking their contractors to achieve better air tightness rates than required (Tesco, 2008). Achieving better air-tightness will ensure the building has a lower Building Emission Rate (BER) and enable it to comply with Part L of the Building Regulations more easily.

Infiltration rates through doors and openings are affected by the weather and the number of people entering and exiting the store. This affects the balance between the supply and extract air (Vatistas et al., 2007). To limit the infiltration through openings in buildings, designers use wind lobbies in front of large openings (Tesco, 2008). Part M of the Building Regulations encourages the use of wind lobbies at the entrances to buildings, providing they provide adequate access to the buildings (Part M, 2010).

3. STORE LOCATION AND ORIENTATION

3.1. Store Location

Food supermarket chains are always looking to maximise their profit by choosing prime locations to open new stores. The selection of locations for new-build stores is made by identifying newly developed urban areas and also by assessing the potential for further development of the same area with the aim of having the store in the centre of the development to ensure current and future profits are maintained (Harvey, 2007). Store location can have an impact on the rate of uncontrolled ventilation penetrating the store which may increase or decrease the rate of uncontrolled ventilation (Britter and Hanna, 2003). Towns and cities are known to have higher temperatures than the surrounding landscapes due to the choice of building materials of building materials (i.e. concrete) impacts on the thermal environment as such materials store up heat; and higher temperature gradients affect the flow and dispersion in urban areas (Cermak, 1996). The local area around the store can significantly affect the wind flow around it i.e. the presence of a car park around a store allows faster wind flow adjacent to buildings. Street width and urban roughness also affect the surroundings, i.e. the more closely packed the buildings, the lower the wind speed (Chang and Meroney, 2003).

3.2. Store Orientation

The orientation of the building, and the layout of openings within it, can affect the rate of infiltration. An opening on the facade will have an impact on the wind pressure coefficient acting on that facade and this will affect the airflow going through the opening (Cóstola et al., 2010). Wind behaves differently with different building shapes and forms. The flow patterns can be determined by the layout and shape of the building and hence determine the airflow in and around the building. There are methods of assessing wind effects on buildings; the most common being Computational Fluid Dynamics (CFD) (Defraeye and Carmeliet, 2010).

Optimum orientation will minimise the effects of wind load on the building. It is common practice to have the building’s major axes in the direction of the predominant wind (Kasperski, 2007); i.e. designing entrance doors on the strong axes might make the building more prone to draughts. Isolated rectangular shaped buildings with flat roofs have high induced suction pressure at the leeward side.

137

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 4. ENTRANCES

4.1. Entrance & Exit Doors

Large openings in buildings, which consist of windows and entrance and exit doors, are the main cause of draughts in buildings. In the case of retail stores, doors are the main cause of draughts. Different types of doors are installed in retail stores, but automatic sliding doors are normal practice in food retail stores. These doors provide easy access and exit for customers with trolleys and shopping baskets. However, automatic doors have sensors for opening and closing, which cannot distinguish between someone passing by the door and someone intending to enter or exit the store. This results in unnecessary opening and closing of the doors.

Furthermore, the time required between opening and closing of doors in food supermarkets is longer compared to other retail shops, as customers are usually pushing trolleys and therefore walking more slowly. The extra time taken for the doors to close will allow a significant amount of air to escape. Automatic sliding doors take some time to fully open and allow full access for customers, it is common practice to have a door opening speed of 1m/s, and for stores with shopping areas of less than 3000m2 to have doors with a width of 2.4m (Tesco, 2008).

4.2. Wind Lobbies

Wind lobbies in buildings are installed to act as a buffer zone to minimize draughts. The size of a wind lobby is specified by the building designer, and must also meet the specifications of Part M of the Building Regulations (Part M, 2010). There are different designs in common use, some of which are embedded inside the building and some built outside the store like a porch, as shown in Figure 1. Door position is based on the individual store requirements.

Figure 14 Left, wind lobby embedded in the store; Right, wind lobby outside the store.

The main design feature of the lobby is to allow unrestricted flow of people through the doors. Sliding doors need more than one second to fully open and the same for closing (Tesco, 2008), and this can be used to estimate the rate of air infiltration through these doors (Vatistas et al., 2007). With an almost continuous flow of people going through the doors together with the time taken for the doors to open and close, the wind lobby can become ineffective.

138

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 5. CASE STUDY

A simple case study was conducted in a food supermarket which is considered to be one of the most energy efficient. This involved a survey of the effectiveness of the wind lobby in modifying internal microclimate on a randomly chosen overcast winter day. The survey was carried out on 25th January 2011 in the afternoon at Cheetham Hill store in Manchester; this is a superstore with a sales floor of 5000m2 and has a car park. The store is located in an urban area but it does not have any building attached to it. Three measurement locations were identified; just outside the store (Figure 2, Right); in the middle of the wind lobby; inside the store adjacent to the main customer entrance/ exit doors (Figure 2, Left). Air curtain have been installed above internal doors.

Testo -435 model- Handheld instruments were used to collect the air temperature, humidity and wind velocity data in the store and outside. The measurements were taken sequentially, temperature at first, wind velocity and finally humidity. When taking the readings, the instruments were held at 1.6m height for all measurements.

Table 10 Temperature, wind velocity and humidity results from case study.

Measured parameter Outside Store Wind Lobby Inside Store (accuracy) Temperature (± 0.2°C) 7.6°C 8.9°C – 9.3°C 16.7°C – 17°C Wind Velocity (± 0.03 m/s) 0.05 – 0.25 m/s 0.02 – 0.2 m/s 0.00 – 0.04 m/s Humidity (± 0.2%) 85% 80% 73.7%

Table 1 shows how the temperature changed in the wind lobby and inside the store, due to air exchange between the outside, the wind lobby and the store. Due to the size of the store, people were flowing in and out of the store on a regular basis, which led to the doors being open for substantial lengths of time. Figure 2 provides an insight into how the doors are fully open and no customers were going through them. The pictures were taken on the day at different times and the doors stayed open even when there were no customers visibly going through the doors. The issue of the draught going through the doors was raised by members of staff and they reported that the draught reduced their productivity.

Figure 15 Left, view from inside store; Right, looking from outside store

139

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ An estimate has been made of the time during which the doors were kept open whilst customers entered and exited the store. A store with a sales floor area of 1000 m2 is designed to have maximum occupancy of 250 customers (Tesco, 2011). If the time taken for each customer to enter is 4 seconds and the same amount of time to exit, the total will be 8 seconds. Assuming all the customers leave the store within an hour of entering it, the period of time when the doors are open will be over 33 minutes within the 60 minute period.

The presence of the wind lobby had limited effect on the wind velocity at the time of the survey; the reading of wind velocity in the middle of the lobby was close to the measurement taken outside. The wind velocity inside the store was lower than outside; however; curtain fans were operational at the time of survey to limit draughts.

Based on the limited measurements in the study, the level of draught does not seem to be an issue when compared to international standards (British Standards, 2005). However, from a subjective view and after careful consideration it can be concluded that a further study will be required to substantiate the data.

6. FUTURE RESEARCH

The Author’s future research is to find solutions that combat air infiltration through doors possibly totally re-design the entrance of a store to counteract the wind flow. In order to evaluate the potential benefits, extensive data will need to be collected on current performance of wind lobbies to enable comparisons to be drawn. Further work on wind lobbies will include Computational Fluid Dynamic (CFD) modelling and analyses of different designs to determine the most suitable design.

Detailed monitoring of energy consumption and temperature distribution in a food retail store is planned. The intention is to obtain and analyse detailed records of energy use distribution over a substantial period of time and then to explore how further reduction opportunities can be identified and evaluated through use of this data.

7. CONCLUSION

Mechanical ventilation systems are installed in stores to provide the required indoor environment for goods and users. Air infiltration affects the operation of the HVAC system by creating imbalance between the supply and extract, which leads to discomfort on the shop floor for customers and staff, and also affects energy consumption. The main source of air infiltration in supermarket is large openings, represented by the customer entrance/exit doors as well as goods delivery doors.

Entrances in supermarkets are the main cause of draughts. Sliding doors in supermarket are common, as they provide an easy of access for customers with trolleys and baskets; however, these doors are prone to faults which lead to longer opening times. Wind lobbies in supermarkets are installed to act as a buffer zone to limit the draughts. Although all wind lobby designs have to comply with Part M of the Building Regulations, no formal guidance are available to advice on the specifications of the lobby in retail stores.

The case study investigating the effect of the wind lobby on a store has revealed that the wind lobby has a little influence on air infiltration into the store. It also showed that

140

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ with the number of people entering and exiting the store, the doors stayed open for a large part of the time. Leaving the doors open might lead to a reduction in the level of comfort and productivity on the shop floor.

REFERENCES

British Standard Institute (2005), BS ISO 7730: Ergonomics of the thermal environment – Analytical determination and interpretation of thermal comfort using calculation of the PMV and PPD indices and local thermal comfort criteria, 46p., London: British standards Institute. Britter, R. E., Hanna, S. R., (2003), Flow and dispersion in urban areas, Annual Review of Fluid Mechanics, vol. 35, pp 469-496. Cermak, J.E., (1996), Thermal effects on flow and dispersion over urban areas: capabilities for prediction by physical modelling, Atmospheric Environment, vol. 30, no. 3, pp. 393-401. Chang, C., Meroney, R.N., (2003), The effect of surroundings with different separation distances on surface pressures on low-rise buildings, Journal of Wind Engineering & Industrial Aerodynamics, vol. 91, no. 8, pp. 1039-1050. Chartered Institute of Building Services Engineers (CIBSE) (2005) Heating, Ventilation, Air conditioning and Refrigiration: CIBSE Guide B, 1-6 p., CIBSE, London, ISBN 1 90328758 8. Cóstola, D., Blocken, B., Ohba, M., Hensen, J.L.M., (2010), Uncertainty in airflow rate calculations due to the use of surface-averaged pressure coefficients, Energy & Buildings, vol. 42, no. 6, pp. 881-888. Defraeye, T., Carmeliet, J., (2010), A methodology to assess the influence of local wind conditions and building orientation on the convective heat transfer at building surfaces, Environmental Modelling & Software, vol. 25, no. 12, pp. 1813-1824. Ge, Y.T., Tassou, S.A., (2011), Performance evaluation and optimal design of supermarket refrigeration systems with supermarket model “SuperSim”, Part I: Model description and validation, Intrnational Journal of Refrigeration, vol. 34, no. 2, pp. 527- 539. Harvey, S., (2007), Site analysis and identification for a proposed supermarket, (available at http://research.omegasoft.co.uk/publications/Site%20Analysis%20and%20Identification %20for%20a%20Proposed%20Supermarket.pdf) accessed 29/05/2011. Kasperski, M., (2007), Design wind loads for a low-rise building taking into account directional effects, Journal of Wind Engineering & Industrial Aerodynamics, vol. 95, no. 9-11, pp. 1125-1144. Kindangen, J., Krauss, G., Depecker, P., (1997), Effects of roof shapes on wind-induced air motion inside buildings, Buildings & Environment, vol. 32, no. 1, pp. 1-11. Lindberg, U., Axell, M., Fahlén, P., Fransson, N., (2007), Appropriate indoor climate for environmentally sustainable Supermarkets – Measurements and Questionnaires, Proceedings CLIMA 2007 conference, pp. 1369 – 1377. Office of the Deputy Prime Minister (2010) The Building Regulations 2000; Approved Document L2A, Conservation of Fuel and Power, 21 p., London: ISBN 978 1 85946 326 0. Office of the Deputy Prime Minister (2010) The Building Regulations 2000; Approved Document M, Access to and use of buildings, 30 p., London: ISBN 10 1 85946 211 1.

141

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Plate, E. J., Kiefer, H. (2001), Wind loads in urban areas, Journal of Wind. Engineering & Industrial Aerodynamics, vol. 89, no. 14-15, pp. 1233-1256. Tassou, S.A., Ge, Y., Hadwaey, A., Marriott, D. (2011), Energy consumption and conservation in food retailing, Applied Thermal Engineering, vol. 31, no. 2-3, pp. 147- 156. Tesco (2011), Fire risk assessment, Crewe Temporary Store , Tesco, pp. 1. Tesco (2008), Specifications for External Lobbies, Tesco, pp.5. Vatistas, G. H., Chen, D., Chen, T., Lin, S. (2007), Prediction of infiltration rates through an automatic door, Applied Thermal Engineering, vol. 27, no. 2-3, pp.545-550.

142

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Bats And Breathable Roofing Membranes: Do The Mechanics Of A Membrane Affect Mechanical Stability?

S. Waring1*, R.H.C. Bonser1, K. Haysom2

1School of Construction Management and Engineering, University of Reading, Reading, UK 2Bat Conservation Trust, 5th Floor, Quadrant House, 250 Kennington Lane, Vauxhall UK

*Corresponding author:[email protected]

ABSTRACT Biodiversity is an important part of sustainability within the built environment, and as the construction industry becomes more aware of ecological concerns, it is playing a greater role in their strategies. Bats constitute an important component of urban biodiversity and several species are now highly dependent on buildings. Along with their roosting habits this makes them particularly vulnerable to environmental changes. As many buildings considered suitable as bat roosts age, their roofs need replacing and traditional roofing felts are frequently being replaced with breathable roofing membranes (BRMs), designed to aid the energy efficiency of a building and reduce condensation in roof voids. Anecdotal evidence suggests that BRMs could pose an entanglement threat to bats, whilst bat claws may alter the structural properties of such membranes. At present the standard industry tests carried out, comprise a tear resistance and nail tear test. Both of these explore the strength of a membrane on a macro scale, and do not take into account the structural stability when the materials are worried on a micro scale. To explore the mechanics behind bat-BRM interactions, we have carried out scissor cutting and trouser tear tests on a wide range of membranes available on the UK market. This is to generate data on the structural properties of membranes on a finer scale. Data has also been collated on the size and curvature of bat claws, which will allow comparison of claw dimensions in regards to membrane thickness. This paper presents the results of these tests, from which we have been able to determine baseline characteristics of BRMs available in the UK and relates this to how BRMs may be affected by interactions with bat claws. This will allow quantification of the wear and tear experienced within a bat roost, by testing at a finer level than the agreement certificates, to take into account the shape and sharpness of claws found on UK bat species known to roost in buildings. We aim to provide a general overview of the mechanical stability of a wide range of BRMs and what this could mean for use of BRMs within bat roosts. The data presented will also indicate the important areas for future research into bat interactions with BRMs.

Keywords: Bats, breathable membranes, roofing, conservation

143

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Potential Carbon Savings Through Hot-Fill Appliances: Field Test Data Validation

D.Saker1*, S.Millward2, M.Vahdati3, E.Essah4, and R.Mayer5

1Technologies for Sustainable Built Environment, University of Reading, UK 2SSE Energy Efficiency, Vastern Rd, UK 3, 4School of Construction Management and Engineering, University of Reading, UK 5Sciotech Projects, University of Reading, UK *Corresponding author: [email protected]

ABSTRACT Improving energy efficiency in the home represents a highly cost effective method of lowering the national carbon emissions and therefore helping to meet government targets. Forthcoming mechanisms such as the Renewable Heat Incentive and the Green Deal help to achieve this by incentivising the installation of domestic insulation and low carbon micro-generation. The use of such micro-generation for heating domestic water may be further benefited by the use of hot-fill appliances, since using pre-heated water as opposed to electricity can be much less carbon intensive. This paper highlights current on-going work in the investigation of hot-fill washing appliances, for example the monitoring of 15 dwellings where such appliances are installed. The relevant importance of the hot water supply is considered together with the concept of a hot water ring main which links sources and sinks.

Key words: Hot-fill appliances, hot water ring main, hot water supply, data logging

144

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Building Services Systems: Heating And Air Conditioning Design Approaches A. Barcellos1*, Dr. E. Essah2, Prof. Hazim Abwi2, C. Woods3

1Technologies for Sustainable Built Environments Centre, University of Reading, United Kingdom 2School of Construction Management and Engineering, University of Reading, United Kingdom 3 Wates Construction, Basingstoke, United Kingdom

* Corresponding author: [email protected]

ABSTRACT In the UK, it has been shown that 46 % of all energy demand arises from buildings and their design systems. The idea of designing a building for temperatures varying from - 15°C to 35°C even if these range of temperatures might happen only once per year has become a problem for the industry. Overdesign leads to inefficiency and a waste of energy. The design of a building depends on many aspects, such as climatic conditions, environment, and many others. Therefore, to overestimate demands and not include the most appropriate applications throughout the design process of building service system has become an issue to the sustainability of buildings. This paper will look at an overview of the technical literature in the context of heating and air conditioning systems design. Moreover, it would provide a step by step approach of the design processes that should be taken into consideration. Finally, it would analyse two different building types, educational and office buildings, to distinguish between the different parameters and characteristics needed for the appropriate heating and air conditioning applications. Hence, this paper aims to provide a summary, with a more realistic and accurate prediction of building service system, of how to maximise energy efficiency of such systems. Keywords: Heating and air conditioning, building system, energy

1. INTRODUCTION Heating and air conditioning systems are designed in order to maintain a desired ambient condition in a space. In order to define these conditions, it is necessary to possess all the needs and functionality of the building and ensure that the correct applications are considered. To select the most suitable system a range of criteria, such as temperature, humidity, operation cost, air movement, and life cycle assessment need to be considered to

145

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ investigate how they shall affect each other through the design process. Heating and air conditioning loads normally contribute in the data process of quantitative evaluation since it narrows the possible system choices to a minimum range of availabilities (Faber & Kell’s, 2008). This is due to the specification of different building types and the impacts that these building services will have on them. Heating and air conditioning capacity requirements will influence the building design and the approach that would be suitable for implementation. Each building design may require a different control zone to maintain their energy demand. It will also determine the services systems that can be optimised in this determinant zoning (ASHRAE 1995). It is also important to mention that even though a specific heating and air conditioning system might be more efficient than other type, it can only be utilised if it matches a few requirements, such as the maintenance of the desired building environment and the system responds to the building without causing discomfort to people. This paper will show the stages necessary to select and design an appropriate heating and air conditioning systems. Moreover, it will provide a technical literature survey of the most important design calculations and other requirements necessary throughout the process of building services selection. Furthermore, it will also differentiate between the design concepts of two building types and the characteristics that are needed to approach the correct system and also distinguish between its main characteristics.

2. SUSTAINABLE BUILDINGS In order to design the appropriate building systems in its most efficient way, one needs to consider sustainability issues. There is a range of strategies and elements that can be put into practise to maximise the benefits of designing heating and air conditioning systems. 2.1 Sustainable strategies for heating and air conditioning There is a range of applications in the design process of heating and air conditioning systems that can include aspects of sustainability. Among some of these (Faber & Kell’s, 2008):

o Reduction of CO2 emissions and energy demand: application of renewable technologies and low carbon policies as well as the utilisation of efficient equipment o Adaption to climate change: reduction in avoidable heat gains and the possibility to adapt design systems to the future changes that might be faced due to the impact of climate needs o Sustainable materials: provide a better life cycle assessment of the materials for the building systems o Reduction of pollution: avoid discomfort in the indoor air

146

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 3. DESIGN PROCESS The process of building systems design requires a series of phases, which needs to take into account many design stages and requirements. In order to design the most appropriate heating and air conditioning systems, this process has to be taken prior to the design itself. It would define the characteristics of the systems that are needed in the process of design and their functionality. The efficiency of building services systems starts its design process by adopting some basic efficiency rules, which consists of: the avoidance of oversized systems, selection of correct pumps and fans, reduction of pressure drops that will lead to pump and fan power, and the recovery and use of waste sources of heating and cooling among others (CIBSE, 2006). The design process is considered under the following stages: 3.1 Phased Design Stages Before designing heating and air conditioning systems, a building program has to be created to define the most appropriate inputs in the design. The program would include a range of data, such as the functionality of the building, its geometric data, materials, fabric, and site location among many others. Besides, in this design stage a range of data with regards to heating and cooling degree hours, monthly sunlight radiation, monthly wind speed and direction, and peak and average temperature would need to be considered. This will provide a better overview of the natural conditions faced in the location where the building is about to be built. Furthermore, the phased design stages have to consider the needs in the design process for every space within a building. This means that an evaluation of indoor and outdoor air quality, ventilation, occupancy time, humidity, and the life cycle of the building would be required. It is essential to know the function of the building rooms in order to select the most suitable heating and air conditioning system. This is necessary because some areas of the building might need heating or ventilation for example. The systems need to perform its demand without affecting other areas of the building (ASHRAE, 1993). 3.2 Schematic Design This is the design stage where comparisons of the range of pre-selected building systems are made. The comparison of requirements, such as the use of electrical or fuel systems, the size of piping and ducts for the application of ventilation requirements, the cost of operation, its suitability with the building function, lighting, and air quality among several others will be able to be distinguished and analysed. 3.3 Preliminary Design The need to consider the local or national energy codes with regards to the building envelope and other systems, such as heating, ventilation and air conditioning (HVAC) or lightning is taken in the preliminary design process. It would ensure that all the heating and air conditioning systems requirements are in accordance to building codes. The analysis of the heat loss and heat gain, and ventilation control are defined in order

147

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ to design the appropriate supply systems and the equipments needed. It would also decide on systems sizes that match with loading requirements. 3.4 Post Design Phase This is the stage where testing for the selected systems would be made and analysed. It should feature requisites of equipment efficiency; systems control approaches, air quality and airflow volume and temperature (ASHRAE, 1993). 3.5 Design Calculations Amongst all the design processes, calculations to select and evaluate building services heating and air condition systems are important to minimise errors in the performance of the building. Some of the most important ones concern the calculation of heat loss and heat gain calculation and also the building’s fabric (CIBSE B1, 2006). Heat loss calculation is analysed from the different temperature between the outside and inside of a building. It consists of two main heat loss elements; air movement (infiltration/ventilation) losses and the fabric losses. To minimise its effect the need of an airtight building that will benefit from internal and also from solar gains can be used in conjunction with thermal modelling to optimise the size of the heating systems’ size (CIBSE B1, 2006). Heat gain calculations need to consider the solar gains, which also depends on the orientation of buildings. Its calculation is based on the peak heat gains from ventilation, conduction and solar radiation gains. Calculation for fabric losses considers the thermal resistance of each layer of the building construction and its internal and external surface resistance, which contribute to the U-values of the materials. 3.6 Building Envelope and the Design process Heating and air conditioning systems need to be designed in conjunction with others building systems. Other systems will have an enormous effect on the performance, function, size, operating efficiency, and maintenance of heating and air conditioning systems. The building envelope can affect the HVAC system in many different ways, such as:

 Roof, floor, walls: the selection of these items can have an impact on the buildings heat loss and gain since it might affect the buildings’ infiltration, solar radiation or even its thermal transmission.

 Envelope construction: it might cause humidification and dehumidification effects due to the transfer of moisture within the building spaces and the outdoors.

 Heat gain and heat loss: it might effect in the capacity of energy conversion of equipment, such as chillers and boilers and also on the size and the requirements of distribution systems such as ducts and pipes.

 Thermal mass: the choice of a strong envelope can affect the operating cycle of the heating and cooling systems since its causes thermal mass effect.

148

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 4. BULDING TYPES DESIGN CONCEPTS Buildings of many different types may be designed to be heated and air conditioned using a wide range of service systems and equipment. The choice in the implementation of a determinant system might be due to the functions needed or the purpose designed for that specific building type. Below, two different building types are distinguished according to their main characteristics and air conditioning needs. It focuses on the design requirements necessary in order to select the appropriate building service system. 4.1 Office buildings Office buildings are normally occupied from 8am to 6pm. Its operational characteristics together with its requirements need to be considered in order to select the most functional systems for the plant. Occupancy rates play a major role in the selection of air distribution system. Offices with computing centres that might have to be operating on a 24h basis for example, will need to consider backup systems for these areas in case of failure in the central HVAC systems. Office Buildings can be characterised for having both peripheral and interior zone spaces. Peripheral zones are the ones, which consist of variable loads due to the changing of sun position and weather conditions. During winter time, these zones require heating. Meanwhile, over intermediate seasons, it might require heating in one side of the room and cooling on the other one. On the other hand, Interior zoning demands a regular rate of cooling during all seasons since their thermal loading is generated mainly from lights, people and office equipment (ASHRAE, 1995). It will require that the interior air conditioning system allows flexibility to match all load situations. Therefore, the choice of variable air volumes (VAV) is the most required for this operation. The system should estimate if the amount of air movement and outdoor air can be matched at the required temperature without causing any overcooling (Keith Shepherd, 1999). Lighting load in office building is of a high output to the heat load of the building. Computer centres and other electric equipment can provide loading as high as 50 to 110 W/ m2 (ASHRAE, 1995). From that, it is extremely important to familiarise with the electrical equipment that will be operated within the building in order to size the appropriate air handling unit for the implementation of the air conditioning system. At the zoning where electric load is higher than 65 W/ m2, heat ought to be removed from the room by the use of air exhausts or water tubing (ASHRAE, 1995). Normally, office buildings use the traditional dual duct, induction, or fan coil systems. When the induction or fan coil design is settled at the perimeter, it also uses an all air systems for the interior areas. For buildings that do not possess an economiser zone, the installation of a bypass multizone on each floor with a heating coil in each exterior zone duct is used (ASHRAE, 2000). This allows building floors with different levels of occupancy to be operated with high levels of energy saving due to its low fan power. Moreover, energy cost can also be reduced if the number of equipment room is reduced meaning it could be turned off in unoccupied areas and high pressure ductwork would not be required.

149

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Total office building space needs are normally occupied by 8 – 10% of its gross area by the building electromechanics. The height that fan rooms require range from 3 to 5.5m since it depends on the distribution system selected. On office floors, induction or fan coil units need 3 to 5% of the floor area. Thus, pipes, ducts and other equipments will need 3 to 5% of each floor’s gross area. All these elements need to be taken into consideration during the heating and air conditioning design system (CIBSE B2, 2006). For smaller office buildings however, it might be a better choice to select a perimeter radiation system with conventional, single duct, low velocity air conditioning systems supplying air from multizone units. For a medium sized building, air source heat pumps may be selected. Meanwhile, in larger ones, internal source heat pumps systems are possible with almost all range of air conditioning systems. 4.2 Educational Buildings To decide on the appropriate heating and air conditioning systems for educational buildings, a range of inputs needs to be considered. First of all, the need of understanding the functions of the school has to be analysed. Some organisations might differ in their educational methods and have night classes for example. Educational buildings have consistent dense occupancy when it is occupied. However, auditoriums, sporting facilities and other areas might vary its occupancy and therefore vary its amounts of ventilation and temperature required. Due to that, educational service systems need to be as flexible as possible within these rooms. Larger external areas such as auditoriums or gymnasiums should have their own HVAC control as well as the administrative office since it normally stays open till longer periods of the day. Libraries should also consider the implementation of air conditioning as a method to prevent books from damages. Computer rooms will need additional cooling tools due to its high heat load that comes from the equipment (ASHRAE, 1995). The choice for heating and air conditioning systems for educational buildings will depend on many things, such as if the educational building is new or existing. Besides, the type of HVAC to be considered also needs to account for seasonal periods of the year. During hot and dry climates, evaporative cooling might be the chosen type of cooling since the usage of air conditioning might not be priority for some educational buildings. However, the use of air conditioning or dehumidification tools in hot and humid environments is advised since it would prevent the formation of mould. Since, educational building are normally not occupied during the night or over weekends and most of the time during the holiday periods, systems which incorporate night setback, hot water temperature reset and boiler optimisation are usually considered in the systems design. Setbacks tools can regulate systems to heat specific areas. After, it cools these areas down, once the spaces heat up due to the occupancy thermal loading (CIBSE B2, 2006). It saves energy by rearranging the heating and cooling environment temperature for the duration of the unoccupied period of the building. Furthermore, the use of appropriate ventilation is a major importance when considering that it would control the air quality of the rooms and avoid odours or diseases to be spread around. Ambient air temperature should range at about 24°C [ASHRAE, 1995].

150

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ 5. CONCLUSION It has been shown that the design process throughout all the stages of a building is of great importance in the achievement of the most appropriate systems design. Its complexity and wide variety of systems options can make it hard to select the most appropriated one for a specific building. By following a general design concept and allow building service engineers to have an earlier input in the design process, the advantages seen will be of high magnitude. The design process showed in this report is a general concept of how heating and air conditioning systems should be evaluated. There are others stages or methodologies that are possible to be used. The one in this report tough is believed to be the most appropriate since it considers every step of the designing process. Building types are relative challenging to define as only having one specific building service systems as a standard. Since, many building types differs from each other according to its own characteristics a general concept of what have been used in the past and that is believed to be correct and more appropriate have been shown in this report. In conclusion, there is a huge gap in what is certain and uncertain in the design of building heating and air conditioning systems. The importance of developing modelling softwares with similar measures and output results would benefit the industry in a creating a standard approach for building types and help building service engineers in their design process.

REFERENCES Air Conditioning Systems Design Manual (ASHRAE), 1993; American Society of Heating, Refrigeration and Air Conditioning Engineers; Harold G. Lorsch, principal investigator ASHRAE HANDBOOK 1995, Heating, Ventilation, and Air - Conditioning Applications, SI edition ASHRAE Handbook, 2000; HVAC Systems and Equipment, SI edition CIBSE Guide, 2006; Energy efficiency in buildings CIBSE Guide B1, 2006; Heating CIBSE Guide B2, 2006; Ventilation and air conditioning Faber & Kell’s, 2008; Heating and Air-Conditioning of Buildings, 10th edition Keith Shepherd, 1999; VAV air conditioning systems

151

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

The Use Of Sustainable Travel Planning Strategies Within Remote Cities

M. H. Ismail1*

1 Technologies for Sustainable Built Environments Centre, University of Reading, United Kingdom and Hereford Futures Limited, Hereford, United Kingdom

* Corresponding author. [email protected]

ABSTRACT The paper considers sustainable travel strategies for remote cities that form a regional centre for a wider area. The strategies aim to minimise private vehicle traffic within the city centre arising from both city residents and commuters in from outside regions, but without adversely impacting on the total inflow of people to the city for business, leisure or educational purposes so as not to affect the city’s economic viability.

The primary case study within the paper is Hereford, United Kingdom – an ancient Norman city within rural Herefordshire. Significant research has previously been conducted as to the transport problems within the city and such research is summarised and built on in the current paper by proposing potential solutions to the problems.

The paper concludes that sustainable travel strategies in such cities are best aligned in zones, with key strategies for the inner zones being walking and cycling and key strategies for the outer zones being “park and walk” schemes to the inner walking/cycling zones.

Keywords: Sustainable travel, soft measures, remote cities

1. INTRODUCTION Cities located within a rural area are often the primary source of employment, higher education, retail and other facilities for a wide local area, which may result in a net inflow of daily visitors due to:

 residents of the city remaining within the city for work, shopping and other needs; and

 residents of the rural area outside the city coming into the city on a daily basis for work, shopping and other needs

Should the primary means of transport be private vehicle, this may result in significant congestion within the city (which has consequent detrimental economic and social impacts) as well as other adverse environmental, social and economic impacts such as

152

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ increased carbon emissions; air pollution; poor public health; and reduced incentive to invest in the area.

Therefore, careful travel planning is essential to minimise private vehicle usage without dissuading people from coming to the city, as it is important (particularly in the current economic climate) to maintain and increase the economic prosperity of the region. The focus in this paper is on sustainable travel solutions for commuters to work, as these form a significant proportion of peak time journeys.

2. CASE STUDY: HEREFORD, UK 2.1. Background

The main case study within the paper is the ancient cathedral city of Hereford, United Kingdom, located within the predominantly rural county of Herefordshire. Hereford has a population of 55,700 whilst the other principal towns within Herefordshire (Leominster, Ross-on-Wye, Ledbury, Bromyard and Kington) have much smaller populations, ranging from 3,200 to 11,100 (Herefordshire Council, 2009). The nearest large city to Hereford is Worcester, which is 21 miles away, whilst the major cities of Cardiff, Birmingham and Bristol are 58, 61 and 65 miles away respectively. These geographical circumstances have led to Hereford becoming the county’s centre for employment, administration, health, education facilities and shopping, resulting in significant pressure on its urban highways and its historic city centre, in which it retains an 11th Century cathedral and other historic buildings.

2.2. Geographical Factors

An additional geographical complication for Hereford city is that the River Wye divides the city between the North and the South. There is only one principal road bridge, at which the A49 crosses the river. This crossing point suffers from significant congestion which causes significant delays during peak commuter travel time (Mouchel report, 2010). There are also two further pedestrian bridges, one of which carries an important cycle route (the Great Western Way).

However, an important positive factor is Hereford’s pleasant and compact city centre, with a pedestrian-only main shopping street (High Town) and numerous historic buildings, which make it a very “walkable” environment.

2.3. New Development

Hereford is currently undergoing a significant redevelopment at the Edgar Street Grid, a 40ha site to the north of the city centre, which will entail new residential, retail, office and leisure facilities. Therefore, it is important that any travel planning options take into account the impact of the new development, in terms of additional residents to the area and additional employment opportunities within the city centre.

153

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ In total, Hereford plans to increase the number of households by 8,500 by 2026 (including those at the Edgar Street Grid) (Mouchel report, 2010).

2.4. Journeys to Work

76% of Hereford residents work within Hereford and 65% of residents’ journeys to work are less than 5km. Therefore, there appears to be significant scope for encouraging more sustainable means of travel to work, given the short distance of typical work journeys. 57% of residents take their private vehicle to work (with an additional 7% being car passengers); 18% walk to work and 8% cycle. Only 6% use public transport to travel to work, all of which consists of bus use (Census data, 2001).

The figures below (Mouchel report, 2010) show that the choice of transport mode to work is strongly affected by the area of the city in which the resident lives. The majority of car journeys to work arise from the outskirts of the city whilst the majority of journeys on foot arise from within the city centre. This may indicate that a key employment zone is within the city centre:

Fig 1: Car journeys to work Fig 2: Journeys to work on foot

2.5. Hereford Travel Objectives

From the above background, it can be seen that the key objectives for Hereford are to:

154

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/  maintain and encourage the relatively high levels of walking and cycling to work within the city centre  reduce private vehicle use from the city outskirts into the city centre  protect the city’s historic core against increased vehicle use  reduce the significant congestion on the A49 crossing point over the Wye river However, due to the current economic climate, and the reduced public sector funding now available for transport issues, low-cost approaches will be favoured to strategies which involve significant capital expenditure or long-term operation and maintenance costs.

3. METHODOLOGY The focus of the research is on “soft measures”. Such measures do not involve investment in new transport infrastructure or technologies but instead focus on changing people’s behaviour so that they make better use of currently available resources (Cairns, 2004). Following a review of available soft measures, a city-wide sustainable travel plan is devised to suit Hereford and other similar cities, which is set out in the Discussion.

4. RESULTS A summary of available soft measures, which can potentially be used to encourage a behavioural change in choice of transport mode, are set out below and discussed in section 5.

Table 1: Available Soft Measures Strategy Details Potential Financial Implications on the Local Authority Pedestrian- Converting the city centre to  Cost of converting roads into only zones pedestrian-only zones would pedestrian-only zones prevent private vehicle use in the  Additional car parking on the inner zones of the city. outskirts of the pedestrian-only zone Reduced car Reducing availability of city centre  Reduced car parking income parking car parking could encourage private  Additional car parking on the vehicle drivers to use different outskirts of the city centre modes of transport due to the inconvenience of locating a parking space. Car parking Limiting city centre parking to a  Reduced car parking income duration maximum duration (up to 3 hours)  Additional car parking on the could minimise use by commuters outskirts of the city centre without deterring other day visitors such as shoppers.

155

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Car parking Increasing city centre car parking  Reduced car parking income charges charges could minimise use by  Additional car parking on the commuters but could also deter outskirts of the city centre other day visitors.  Reduced income from day visitors, and potentially reduced business investment Park and Ride Car parking on the outskirts of the  Additional car parking facilities city centre and public transport and re-routing of public transport connections into the city centre. vehicles (primarily bus) Reduced Local authority subsidies of public  Cost of subsidising public public transport charges to offer reduced transport transport costs rates or season ticket discounts. Public Greater awareness of available  Promotion/marketing cost transport public transport options could promotion and increase public transport use. marketing Walking / Campaigns such as “Walk/Cycle to  Promotion/marketing cost cycling Work Week” can positively promotion and encourage people to switch to marketing walking or cycling. Bicycle Availability of good quality and  Cost of bicycle stands parking secure bicycle parking can (approximately £35 to £100 per facilities encourage cycle use. stand) Car share Providing car share websites or  Cost of maintaining website / schemes other databases allows people databases living in the same area and travelling to the same area to share their car use.

5. DISCUSSION 5.1 Reduction of Private Car Use

It is self-evident that converting the city centre into a pedestrian-only zone will prevent private vehicle use within the area. However, such a measure may be unduly restrictive and unworkable for residents and outside visitors. An alternative therefore may be to enforce driving restrictions within the city centre for specific time periods, rather than a full conversion of the city centre into a pedestrian-only zone.

Parking restrictions within sustainable travel plans have been reported as effective in reducing commuter car use by an average of 24% or more, whilst the reduction in commuter car use was only 10% or more without parking restrictions4. From Table 1, the three strategies relating to parking restrictions (reducing car park availability, reducing parking duration, and increasing parking charges) may all be effective in

156

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ reducing private vehicle use. However, only reduction of parking duration specifically targets commuter driving, whilst the other two methods impact on day visitors as well as commuters.

5.2 Sustainable Travel Alternatives

A potential sustainable alternative to private vehicle use is public transport use. It has been reported that offering public transport discounts can be highly effective in encouraging reduced car use (Cairns, 2004). Hereford has a railway station within its city centre but no other stations in the outskirts of the city, thus preventing commute to work by rail for Hereford residents. Therefore, the primary existing public transport option for Hereford is bus.

As mentioned above, only 6% of residents commute to work by bus, which seems a relatively low proportion. This may, potentially, be explained by the inability (due to space restrictions) to provide bus priority measures (such as exclusive bus lanes) within the historic city centre core. Therefore, bus users may be subject to the same congestion as faced by private vehicle users, but with the added inconvenience of public transport use (such as waiting at bus stops for the bus to arrive).

Given the compact nature and pleasant environment of Hereford city centre, it is therefore considered that sustainable travel options which favour walking and cycling within the city centre, rather than increasing bus use over current levels, are the optimum choices.

5.3 Hereford City Proposal

Figures 1 and 2 above show that private car use and walking are more popular respectively in different parts of the city. It is therefore useful to consider the city centre as “Zone 1” and the outer regions as “Zone 2” with different travel planning strategies for each zone.

Figure 3: Map of Hereford with travel Zone 2 planning zones5

Zone 1

157

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Zone 1 (circled in green - from the Wye Bridge to Hereford railway station) is 0.9 miles in length. The Herefordshire City Council mini-map (Herefordshire Council, undated) states the area within the red square in Zone 1 is within 10 minutes walk from High Town (the main city centre shopping street). From the above discussion, it is considered that the key approach for Zone 1 is to implement some form of parking or driving restriction within Zone 1 and couple this with measures to encourage increased walking and cycling within the Zone.

Whilst parking restrictions may encourage reduced private car use, driving restrictions would guarantee reduced private car use, and may therefore be the more effective option. However, it is not desirable to adversely impact on day visitor numbers to the city. Instead, the key target is commuters into the city for work to tackle peak time congestion. Therefore, a potential solution is to enforce driving prohibitions in the city centre only during peak commuter travel time. A proposed time frame could be, for example, 7.30am to 9.30am and 4.30pm to 6.30pm. The driving restrictions would only be within the red square in Zone 1, so as not to exceed a reasonable walking distance for most people. Naturally, there would be exceptions for disabled drivers, emergency vehicles and buses.

Enforcement of the driving restrictions could involve the placement of CCTV cameras at key road junctions within the city which would record vehicle use. Residents of Zone 1 who need to commute outside of Zone 1 during peak hours may apply for a special permit to be displayed on their vehicle to avoid receiving any penalty should they be recorded on the CCTV cameras. As it is only relatively low proportion of Hereford residents who commute outside of Hereford for work (and not all of them live within the city centre) it is considered that their vehicle use would not detract from the overall advantages of creating the peak time driving restriction zone within the city centre.

Outside of peak time hours, shoppers, tourists and other day visitors would be permitted to use their private vehicles. Whilst this may be disadvantageous from an environmental viewpoint, the potential adverse impact from an economic viewpoint of deterring such visitors may counter-balance this.

This system would therefore reduce road traffic congestion (a significant concern within Hereford); create a more pleasant walking atmosphere; and reduce carbon emissions. In order to encourage residents to accept and appreciate the change (rather than feeling that it has been imposed upon them) promotion and marketing campaigns ought to be used in the lead-up to the change, to promote the positive outcomes of the peak-time pedestrianisation.

Due to the driving restrictions to be implemented within Zone 1, drivers into the city centre from Zone 2 (as well as from outside Hereford) will need parking facilities on the outskirts of Zone 1. It is proposed that such facilities be located just before the A49 road bridge crossing the river Wye, which is currently the main congestion

158

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ hotspot in the city area. This, it is anticipated, will lead to drivers from Zone 2 who wish to reach the city centre driving up to the parking facility and then walking or cycling into the city (using the pedestrian footbridges) rather than attempt to drive across the river on the A49 road bridge.

Therefore, the only drivers who will continue to use the A49 river crossing will be those who do not intend to drive into the city centre, but will instead continue past the city centre heading either to the north or south of the city.

Reducing vehicle numbers on the river crossing in this way would, it is anticipated, result in reduced congestion, so as to allow drivers who wish to bypass the city centre and reach the north or south of the city to move more easily. This would therefore alleviate the economic and social adverse impacts caused by congestion, as well as to reduce carbon emissions by the reduced private vehicle users.

Additional strategies suitable for Zone 2 include strategies to encourage cycling within the city centre. For example, efficient, smart-card operated bicycle hire facilities could be made available at the river crossing car park, to enable drivers to hire a bicycle daily to continue their commute to work once parked. At the city centre, secure bicycle parking would also need to be provided at key locations to assist such additional cyclists.

Further, the setting up and promotion of car share schemes could be valuable in encouraging reduced private vehicle use from Zone 2 (and beyond) to the river crossing facility.

6. CONCLUSIONS Whilst it is a straightforward matter to set out a list of potential travel planning “soft measures” which may encourage sustainable transport use, a more complex issue is selecting which measures are suitable for particular circumstances. In the case of rural cities, such as Hereford, the importance of the city as a centre for a much wider local area cannot be underestimated, and so any selection of measures must take into account and balance the economic impacts, as well as the environmental and social considerations. As such, it is considered that the package of measures proposed above can be applied effectively to rural cities in the same or similar circumstances as Hereford, and can achieve the appropriate balance between environmental, economic and social aims.

The focus in this paper has been on reducing congestion and carbon emissions from private vehicle traffic caused by commuters to Hereford city centre for work. The proposals do not resolve other travel issues which may affect Hereford, primarily the concern that the A49 currently takes “through traffic” directly into the city centre. Current proposals include the development of a new bypass road to allow outside traffic passing through Hereford to bypass the city centre. However, these issues are beyond the scope of this paper.

159

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ The views expressed in this paper are mine alone, as an independent researcher, and do not represent the views of Hereford Futures Ltd or Herefordshire County Council.

REFERENCES 1. Cairns, S., Sloman, L., Newson, C., Anable, J., Kirkbride, A., Goodwin, P. (2004), Smarter Choices – Changing the Way We Travel, Department for Transport 2. Census Data, 2001 3. Herefordshire Council, Mini-Map, 4. Herefordshire Council, The Population of Herefordshire, November 2009 http://www.herefordshire.gov.uk/docs/Hfd_Mini_map_07.pdf 5. Mouchel, Delivering a Sustainable Transport System Stage 2 Study for the West Midlands Region, Growth Point Connectivity Stage 1, April 2010

160

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Methods Used For Sustainable Management Of Time, Cost And Quality Throughout The London 2012 Olympic And Paralympic Programme - Interim Findings Using A Scoping Study

J. M. Grossman1*, M. Radosavljevic2, S. Peace3, L. Warne3 1 Technologies for Sustainable Built Environments, University of Reading, United Kingdom 2 School of Construction Management and Engineering, University of Reading, United Kingdom 3 Chartered Institute of Building, Englemere, Kings Ride, Ascot, Berkshire

* Corresponding author: [email protected]

ABSTRACT The London 2012 Olympic and Paralympic Games Programme (the Games) is currently live, providing an opportunity to obtain first-hand knowledge of large-scale programme management. The programme has promised to deliver the most sustainable games ever. This provides a unique opportunity to capture lessons learned on delivering a programme of this size, on time and within budget, whilst ensuring the sustainable targets are met. Where feasible, these lessons can then be taken forward to all future programmes, including future Olympic Games. This paper aims to comment on the processes used across the programme to manage time, cost and quality. This is based on interim findings from the analysis of information already available and semi-structured interviews. This will integrate within the development of a new Chartered Institute of Building (CIOB) Code of Practice for Programme Management which aims to improve the understanding of this practice within the industry. Keywords: Time, cost, quality, sustainability, programmes

1. INTRODUCTION The construction sector rarely witnesses a programme of such a scale with such high expectations. It has promised to be the most sustainable games ever, taking in to account all aspects from construction to waste and resource management (London 2012, 2009a). The main aim of this paper is to gauge the lessons learned from the management of this programme with regards to meeting the strict targets to the required time scale, budget and quality finish. As research is still ongoing, the paper meets this aim by focusing on the analysis of interim results from a scoping study and examining existing information. The focus of the study is on the methods used for the management of time, cost and quality and how sustainability affected these decisions. The results will then be used as comparable data for future research that will investigate other large programmes for the CIOB CoP. The methods used to research this data will have to be consistent, reliable and justified providing the baseline of the investigation for comparison with future programmes. The Olympic Delivery Authority (ODA) is the public body responsible for developing and building the new venues and infrastructure for the Games and their use after 2012.

161

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ They will play a vital role in meeting the ambitious aim in providing ‘the most sustainable games ever’. The ODA are working together with the consortium CLM in order to deliver a successful Olympic Games. CLM are made up of CH2M Hill, Laing O’Rourke, and Mace and were appointed the ODA’s delivery partner, putting into practice a relatively new concept. The ODA and CLM want to improve methods in achieving higher levels of sustainability and therefore how cities will host future Olympic Games. These include the employment of local labour, regeneration of Stratford and building structures with their future as important as their primary use (London 2012, 2009a). Therefore it is an ideal programme to investigate lessons learned with regards to the management of time, cost and quality especially as the programme is still live. It allows in-depth research into each aspect from the perspective of employees working within the programme and lessons can be taken forward to programmes of all sizes. As mentioned, London 2012 has stringent sustainability targets to meet before its immoveable deadline. Not only will this benefit the local area after the games, it also goes further in contributing to the long term government aim to reduce greenhouse gas emissions by 80% by the year 2050 (DECC, 2008). It is benefiting the industry by providing the workforce with skills to meet such sustainable ambitions and giving them an awareness and resonance with why this is important. To take these lessons into the wider industry research is vital, as sustainability is becoming increasingly significant to the relative success of a construction programme (OGC, 2009a). Considering sustainability ensures a programme will provide for the present without compromising the needs of the future. Sustainable management synthesises sustainability into every management decision within a programme be it environmental, social or economic. It is essential that delivery of a programme is within the time specified, under budget and to a high quality whilst improving the sustainability of operations and helping to reduce carbon emissions (OGC, 2009b). Programme management is a relatively new approach within the construction industry. A programme can be defined as a collection of smaller projects and programme management can be defined as an effective way to manage cost, time and quality across the projects (Lycett et al, 2004). Understanding of the challenges that come with its implementation and practice remain vague. During implementation there can be a lack of awareness and during practice there can be a lack of commitment (Shehu and Akintoye, 2009a). 2. METHOD 2.1 Selection of Method A number of methodologies were considered in the approach to this research. The key concern was to ensure that the data collected would provide reliable information to be able to perform a comparative analysis with other programmes for the CoP. In order to provide a clear basis for the comparative analysis with other programmes, there has to be consistency across all methods. Due to the nature of the project there was a lot of existing information. This information may be hard to generalise and there is no guarantee that any particular method explained in the documentation was exactly how a task was performed (Raferty et al, 1997). Survey was considered as one alternative approach but it was ruled out due to time constraints and fears that a sample from the first study would not equivalently represent the programme against samples from other

162

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ programmes. Surveyed individuals who are sent questionnaires can spend time thinking through the answers, revisiting the questions and thus additionally biasing results. Questionnaire surveys encourage a reductionist approach that will inevitably distance the interviewee from the question. Another alternative, a semi-structured interview, brings forward its own issues; these are the accuracy of their answers and the comparison of their answers with the truth. It is often suggested the interviewee is focused more on self-presentation and persuasion rather than the facts. It is for this reason that collecting data from different sources is essential in confirming the validity of the results (Dainty, n.d.). The main criteria related to the trustworthiness of individual sets of qualitative research are credulity, dependability, transferability and confirmability (Raftery et al, 1997). Therefore, also considering the time available, semi-structured interviews were considered the best empirical method; these can then be contrasted with the available published information. The interviewees should not know the questions beforehand in an attempt to gain an honest response with no pre-determined views. The questions asked are identical so the results can be used as comparable data across the whole programme and down the supply chain. Moving down the supply chain, tier 1 suppliers have the contracting relationship with the ODA; tier 2 contractors have the contracting relationship with tier 1 contractors and so on. 2.2 Questionnaires and Available Information There is a great deal of information available on the programme ranging from quarterly programme reviews to in-depth project briefs. The interview questionnaire asks questions based on the methods used to manage time, cost and quality, working relationships, health and safety and defect management. By covering each of these sections, each question contributes towards the management of time, cost and quality and how sustainability affected decisions. An example of this includes previous working relationships and how this was incorporated into the management of each project. Interviews took place at programme level and are continuing across different tiers of four individual projects. A perspective across different tiers aims to give realistic views of how successful the intended methods were across the programme. Interviews are structured on the questionnaire; however the overall structure ultimately depends on response of the interviewee. Keeping to a closed interview does not allow for further detail that might be essential to the programme. Analysis of documentation available on the Games showed that the programme was heavily based around a document known as the Programme Baseline Report, which was published in 2007 after the Olympics were awarded to London. The original Baseline Report focuses on all aspects of the programme including the fundamental scope, schedule of works, cost and budget and risk for the entire ODA programme. The Baseline Report was updated in 2010 to take account of changes that have occurred and create a structure that should see the Games through to completion and legacy. The majority of the other documents help in describing how the programme should progress to achieve successful and timely completion. 2.3 Interviews Interviews were planned for approximately 6 professional profiles across different tiers within the Olympic Stadium, the Olympic Village, the Main Press Centre (MPC) and the International Broadcast Centre (IBC) and Structures, Bridges and Highways (SBH).

163

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Figure 16 shows the diverse selection of positions chosen for interviewing. On completion of each interview, detailed notes have been written up ready for interpretation and cross-analysis.

 Project Manager from each T1

 Equivalents from significant Sub Contractor (Mechanical/Electrical/Building)

 Key suppliers

 At least 2 Designers (1 Architectural/1 Engineering)

 Foreman

 Programme Level Manager

Figure 16 - Positions to interview Interviewing across the programme gives varying insight into what actually happened with regards to time, cost and quality across different tiers. The Olympic Stadium was selected because it was completed ahead of schedule and under budget (London 2012, 2011). The Olympic Village and MPC/IBC were selected in order to assess how the programme adapted to include these projects after they could no longer be privately funded as was originally planned (London 2012, 2009b). Unfortunately time and availability constraints have often meant that the profiles shown in figure 1 have been adapted where necessary, however, interviews were conducted as close to the selected positions as possible. At the time of writing, 16 interviews have taken place with various professionals throughout the programme within 3 different projects across different tiers. These include the Olympic Stadium, the Olympic Village and the MPC and IBC. Recordings were taken throughout each interview following the approval from each interviewee. The method has so far been successful in accomplishing the aim set forth by this paper. The interviews have produced a range of answers across different projects and across the different tiers. It must be noted there are further interviews to take place across the Olympic Village and SBH (structures, bridges and highways) which will then contribute to later analysis. 3. ANALYSIS The analysis is based on a discussion of the answers from the interviews and available literature from which a list of conclusions will be based on. 3.1 Time, cost and quality as a whole Taking the management of time, cost and quality as a whole, there are a host of methods that ensure high standards are maintained and could be made compulsory in future programmes. 3.1.1 Benchmarking Benchmarking is widely used as a key element to achieve success across all tiers. The benchmark for environmental management across the site was based on the number of incidents per hours worked on the Olympic Park construction site. This was known as the Significant Environment Incident Frequency Rate (SEIFR). It has been suggested

164

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ that benchmarking is used to improve efficiency and allow for company comparison and improvement (OGC, 2008). This view is agreed across the programme as answers from the interviews explain that benchmarking contributes to minimal defects and ensures that the requirements are clear. If possible, creating a full scale mock-up will reduce quality issues. A big issue that causes defects, delays and additions to cost and time is simply the misunderstanding of what is actually required.

3.1.2 Integration It is agreed across the programme that CLM have effectively come together as unit and the appointment of a delivery partner has been successful. Working effectively with the ODA has been a key element to the success of the programme so far. Previous studies have shown that if projects are set up individually and procured separately, as the Olympic venues were, each project may not have to rely on others as much as in other construction programmes reducing knowledge sharing (Lycett et al, 2004). However, project integration within a programme can prove a key factor in communication (OGC, 2009b). Resource sharing between projects increases knowledge sharing and therefore can reduce time and cost and increase quality (Shehu and Akintoye, 2009b). Due to spatial constraints of being on a large programme, project integration was affected, and interviews demonstrate there was a lack of knowledge sharing between projects at construction team level. Integration across the programme certainly increased as the programme progressed through various initiatives including awards and visits to other venues. 3.1.3 Incentives Interview answers also explain that incentives increase productivity. These include contractor financial rewards for meeting milestones (NAO, 2008) or the Olympic badge scheme. Each badge design is based on individual projects and supply is very limited. With the reward of a badge comes a great deal of respect from people within and outside the programme. 3.1.4 Relationships Managing working relationships can also contribute to the relative success of a programme (Shehu and Akintoye, 2009a). Some companies were selected following their previous work on Olympic and Paralympic Games (London 2012, 2006). Discussion at programme level suggested that a combination of experienced and new up and coming professionals/companies that are looking to prove their worth within the industry should be achieved in order to ensure acceptable levels of leniency in managing time, cost and quality. There are fears complacency can creep in if companies or individual professionals have worked together many times previously. Creating the ideal working relationship based on mutual trust, communication and openness will increase the potential for success (OGC, 2009c). 3.2 Time Preparation for the Games is based on a 2-4-1 timeline for overall structure and internal and external communication. In essence, it took two years to plan, four years to build and will take one year to commission. The timing for this programme is unprecedented

165

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ as the end date in immoveable. An example of failure to deal with this difficulty is the recent Commonwealth Games in India. There was a great deal of controversy in the build up to the games as venues were not ready, the Athletics Village was not complete and athletes decided to pull out owing to fears their safety was at risk (BBC, 2010). Following interviews at programme level, there seems to be no doubt that not only will all venues be completed on time, but they will be ready to hand over to LOCOG (The London Organising Committee of the Olympic and Paralympic Games) a year in advance as planned ready for testing and safety inspections. If possible, starting a project in advance of the agreed programme date is also advised. Risks can be managed with respect to the time constraint. Regular milestones must be met and regular meetings held to ensure projects are on schedule or investigate as to why they might not be. Due to the various layers of management any early fears that communication across the tiers would take time were eliminated as regular meetings were organised between programme management and contractors to enhance communication between tiers. 3.3 Cost Cost across the whole programme was very closely scrutinised at every opportunity. Every alteration had to be justified and change could only occur if it was approved by the relevant internal management authority. Managers were given the ability to approve change up to specific values however if the change required was valued higher it would have to go through a change board. Change board required approval from all management minimising any unnecessary cost. Contingency within the programme was only called upon if necessary but potential risks to the programme were covered by this contingency (London 2012, 2009). Even though the Olympic Village suffered from funding issues which then required public sector intervention, the contingency could still account for necessary risks (Webb, 2009). Alongside health and safety, maintaining cost within the budget takes a leading role in the success of the programme since money can only be spent when necessary to improve quality or complete a task to the required standard. From the financial perspective, the programme was broken down into individual projects and measured against the available budget. This differs from other construction programmes where cost in many cases is based around high levels of resource sharing (Shehu and Akintoye, 2009b). Costs across the programme are typically reviewed weekly to minimise unnecessary expenditure and keep the projects within the specified budget. A technique known as ‘Earned Value’ combines cost and time models into a single model that allows monitoring of time and cost in the same breath. It aims to encourage change based on opportunity or availability of resources however the importance and relevance. There is a general understanding that increasing resources early at extra cost will reduce problems at a later date (OGC, n.d.). 3.4 Quality 3.4.1 Creating a Suitable Working Environment Interviewees agree that the culture surrounding the Games is one that involves everybody in the programme and not just the management. The Olympic Park and surrounding areas are dotted with poster campaigns creating an inclusive environment that constantly encourages the workers that they are part of “The Greatest Show on

166

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Earth” and pride is on the line. A programme of such global recognition is not a standard representation of a construction programme but the method used to create a suitable working environment can be appropriately modified and reapplied in future programmes. Quality was encouraged through immense engagement and increased mutual respect between all tiers based on showing and not telling (i.e. leading by example). Simple methods such as this were appreciated by the workforce. There are monitoring procedures across the programme to investigate quality and allow for discussion of any particular problems. These include quality meetings, random daily checks and quality inspection forms. It is understood that within individual projects, quality management often requires additional resources due to integration and reliance on others. Tasks can rely on the completion of previous tasks and, similar to the domino effect, if a task is not completed to an acceptable level, it can have a knock-on effect that can compromise quality. Successful projects should integrate with respect to procurement, understanding exact requirements and risk assessment (NAO/OGC, n.d.) whilst a specific focus on quality should commence from the beginning. 3.4.2 Sustainability Within As previously explained, one of the main outcomes of the Games is the delivery of a sustainable programme. There is an overarching sustainability strategy, which has specific targets cascaded into all the contracts. Such a promise has been reinforced by a high quality waste management system, the regeneration plan of Stratford and that a substantial proportion of structures will be disassembled and re-used almost immediately after the games for legacy (London 2012, n.d.). As programmes are structured around their required outcomes, it is important quality is not negotiated for sustainability (Shehu and Akintoye, 2009b). This gives valid explanation for regular quality and sustainability meetings across the programme and within each project. 3.5 Future Work Future work will focus on completion of the interviews, further scrutiny of obtained results and further analysis of the available literature. This will include workshops that will investigate gaps in the research findings and identification of any lessons learned from the London Olympic Programme that can be transferred to future construction programmes. 4. CONCLUSIONS Answers from the interviews conducted with reference to the relevant literature have lead to the following interim conclusions:

 Benchmarking from the beginning can ensure high quality and minimal defects. Through increased benchmarking the project can prove cost effective in the long run as long as risk is managed effectively.

 Incentives and rewards can increase productivity consistently throughout the programme. Manage trust and working relationships with relative bonuses for hitting targets and completing milestones.

 Aim to employ a combination of trusted and experienced companies/professionals with new up and coming ones that are looking to prove

167

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ their worth within the industry. Complacency should instead give space to building trustworthy relationships between companies and individuals.

 Creating an inclusive environment within a programme can increase pride and as a result raise respect among the participating companies and workforce. In doing so, this can encourage productivity and respect to quality. It is imperative that management leads by example as opposed to dictating requirements so that the workforce understands their importance for successful completion of a project and programme.

 Review cost on a weekly basis to ensure a project is making suitable progress without compromising the budget especially where change may have occurred. Ensure project integration through regular meetings with sub-contractors to discuss relevant and/or pressing issues.

 Quality should be treated as an individual and key factor from the beginning so that it does not suffer at a later date when such considerations might seem too costly and thus neglected. Quality should not only come as a consequence of completion under budget and on time, but also meeting the scope to a high standard. REFERENCES BBC Sport, 2010. Commonwealth Games delays in Delhi cause 'concern'. [Online] Available at: [Accessed 25 April 2011] Dainty, A. Advanced Research Methods in the Built Environment. Methodological pluralism in construction management research. Chapter 1 DECC, 2008. Climate Change Act 2008. [Online] Available at: [Accessed 19 April 2011] London 2012, 2006. 2012 Delivery Partner announced. [Online] Available at [Accessed 15 May 2011] London 2012, 2009a. Towards a One Planet 2012. [Online] London 2012. Available at [Accessed 01 May 2011] London 2012, 2009b. Funding decision announced for Olympic Village and media centres. [Online] Available at: < http://www.london2012.com/news/2009/01/funding- decision-announced-for-olympic-village-and-media-centres.php> [Accessed 03 May 2011] London 2012, 2011. Construction complete on London 2012 Olympic Stadium. [Online] Available at: < http://www.london2012.com/news/2011/03/construction-complete-on- london-2012-olympic-stadium.php> [Accessed 05 May 2011] London 2012, n.d. Legacy after the Games. [Online] Available at: [Accessed 05 May 2011]

168

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/ Lycett, M., Rassau, A., Danson, J., 2004. International Journal of Project Management. Programme management: a critical review, 22 pp. 289–299 NAO, 2008. Preparations for the London 2012 Olympic and Paralympic Games: Progress Report June 2008. Executive Summary, pp. 5 NAO/OGC n.d. Common Causes of Project Failure. [Online] Available at: [Accessed 29 April 2011] OGC, 2008. Benchmarking. [Online] Available at [Accessed 05 May 2011] OGC, 2009a. News - About the OGC. [Online] Available at [Accessed 05/05/2011] OGC, 2009b. Overview of Managing Successful Programmes (MSP). [Online] Available at: [Accessed 05 May 2011] OGC, 2009c. Managing relationships. [Online] Available at: [Accessed 05 May 2011] OGC, n.d. The productive time improvement journey. [Online] Available at: [Accessed 05 May 2011] Raftery, J., McGeorge, D., Walters, M., 1996. Construction Management Economics. Breaking up methodological monopolies: a multi- paradigm approach to construction management research, 15, 291 - 297 Shehu, Z., and Akintoye, A., 2009a. International Journal of Project Management. Major challenges to the successful implementation and practice of programme management in the construction environment: A critical analysis, 28 pp. 26-39 Shehu, Z., and Akintoye, A., 2009b. International Journal of Project Management. Construction programme management theory and practice: Contextual and pragmatic approach, 27 pp. 703–716 Webb, T., 2009. Olympic 2012 village seeks £225m bailout from European Investment Bank. The Guardian, 10 April 2009. [Online] Available at: [Accessed 05 May 2011]

169

Abstracts of Conference Papers: TSBE EngD Conference, TSBE Centre, University of Reading, Whiteknights, RG6 6AF, 5th July 2011. http://www.reading.ac.uk/tsbe/

Modelling Building Semantics: Providing Feedback and Sustainability

H. H. Shah1*, H. Grzybek2, I. Wiafe2, S. R. Gulliver2, K. Nakata2

1Technologies for Sustainable Built Environments (TSBE) Centre, University of Reading, UK

2Informatics Research Centre (IRC), University of Reading, UK

* Corresponding author: [email protected]

ABSTRACT Even minor changes in user activity can bring about significant energy savings within built space. Many building performance assessment methods have been developed, however these often disregard the impact of user behaviour (i.e. the social, cultural and organizational aspects of the building). Building users currently have limited means of determining how sustainable they are, in context of the specific building structure and/or when compared to other users performing similar activities, it is therefore easy for users to dismiss their energy use. To support sustainability, buildings must be able to monitor energy use, identify areas of potential change in the context of user activity and provide contextually relevant information to facilitate persuasion management. If the building is able to present users with detailed information about specific user activity that is wasteful, this should provide considerable motivation to implement positive change.

This paper proposes using a dynamic and temporal semantic model, to populate information within a model of persuasion, to manage user change. By semantically mapping a building, and linking this to persuasion management we suggest that: i) building energy use can be monitored and analysed over time; ii) persuasive management can be facilitated to move user activity towards sustainability.

Keywords: Modelling, energy use, user behaviour, building performance, persuasive modelling

170