IN
DEGREE PROJECT ENVIRONMENTAL ENGINEERING, SECOND CYCLE, 30 CREDITS
STOCKHOLM SWEDEN 2017 ,
Exploring the technological potential for improving energy efficiency of residential space heating in the UK by 2050
SOFIAN DEMDOUM
KTH ROYAL INSTITUTE OF TECHNOLOGY
TRITA TRITA-IM-EX 2017:13
Sofian Demdoum
Master of Science Thesis
STOCKHOLM /2017/
Exploring the technological potential for improving energy efficiency of residential space heating in
the UK by 2050
PRESENTED AT
INDUSTRIAL ECOLOGY
ROYAL INSTITUTE OF TECHNOLOGY
Supervisor:
Oleksii Pasichnyi
Examiner:
Monika Olsson
TRITA-IM-EX 2017:13 Industrial Ecology,
Royal Institute of Technology
www.ima.kth.se
KTH ROYAL INSTITUTE OF TECHNOLOGY
Exploring the technological potential for improving energy efficiency of residential
space heating in the UK by 2050
Submitted by Sofian Demdoum
Supervisor: Oleksii Pasichnyi Examiner: Monika Olsson
A thesis submitted in fulfillment for the degree of Master of Science
in the
Division of Industrial Ecology Department of Sustainable Development,
Environmental Science and Engineering
June 2017
“The roots of education are bitter, but the fruit is sweet.”
Aristotle
Abstract
The UK has pledged to reduce its greenhouse gas emissions by 80 % by 2050, compared to 1990 levels. With the residential sector accounting for roughly a quarter of the UK’s total carbon emissions, and with space heating accounting for roughly two thirds of residential energy con- sumption, addressing heating demand will prove instrumental for the UK to reach its targets.
The large-scale deployment of low-carbon heating technologies, combined with significant im- provements in energy efficiency, is seen as the primary way of reducing both energy demand and carbon emissions.
In this thesis, the technical potential for energy efficiency improvements in the UK’s residen- tial space heating sector is explored. Three areas of improvement are identified: the thermal performance of the building envelope, the efficiency of heating equipment, and the use of smart heating devices. In all three areas, significant potential exists to further reduce energy demand for space heating, and the associated carbon emissions.
To understand the effects of different technology adoption patterns on energy demand for space heating by 2050, a selectively disaggregated bottom-up model of the UK’s building stock is developed. The model projects energy demand for space heating for four different technology adoption scenarios, based on projections of future total heated floor area.
In the ‘Minimal effort’ scenario, which assumes that the least possible amount of effort is made to improve the efficiency of the space heating sector, energy demand by 2050 increases slightly, by 4.3 %. In this scenario, the achieved minor efficiency improvements are not able to offset the increase in the total heated floor area, leading to this small increase in space heating energy demand.
In the ‘Efficiency focus’ scenario, which assumes that significant effort is made to improve the efficiency of the space heating sector, energy demand by 2050 decreases by 34.5 %. Despite these significant efficiency improvements, fossil fuels still make up roughly three quarters of the fuel mix, as no major shift to cleaner energy sources has been achieved. In contrast, in the ‘Re- newables focus’ scenario, which focuses on shifting towards renewable heating options, energy demand decreases slightly less, by only 32 %, but fossil fuels make up only around 46 % of the fuel mix, due to the uptake of low-carbon options such as heat pumps, district heat, and biomass.
The analysis carried out in this thesis shows that the UK’s residential sector consists of many old and inefficient buildings, still heavily relies on fossil fuel-fired heating equipment, and makes nearly no use of smart heating devices to further reduce energy demand for space heating.
Clearly, the technical potential for achieving energy efficiency improvements is significant, in all three of the identified improvement areas. However, achieving the targets set forth by policy- makers will require strong efforts, given the relatively bad current condition of the residential space heating sector. Existing barriers to achieving these improvements should be identified and addressed immediately, to ensure timely efforts are possible.
Keywords: Energy Efficiency; Residential; Space heating; United Kingdom; Bottom-up; 2050
Acknowledgements
I would like to express my deep gratitude to several people who have made this master’s degree, culminating with this thesis, a fantastic experience.
First, I want to ackowledge the work done by the people at InnoEnergy, who make it their life’s work to provide students with this tremendously interesting degree program in the field of energy. Special thanks go to professor Johan Driesen and Mar Martinez at KU Leuven for everything they have done for the students in this master’s program.
During this thesis work, I could benefit from the guidance of PhD candidate Oleksii Pasich- nyi. Our discussions have been an immense help to shape my work and keep the focus of the topic. Thank you for the considerate and engaging feedback I always received.
This thesis was part of an internship at McKinsey Energy Insights. Thank you, Bram, Matt, Magda, and the broader team, to make this internship an amazing learning experience. It has been a true pleasure to be part of this team.
Throughout my studies, I have received the relentless and unconditional support of my parents.
Thank you for always supporting me, for giving me the opportunity to pursue my interests, and for always being there when I needed you.
And lastly, thank you, Marija, for inspiring me, for being there when I needed you, and for always believing in me.
Stockholm, June 2017
This thesis was typeset in L
ATEX using a template provided by Sunil Patel, who himself modified a template
provided by Steven Gunn. The template was published under CC BY-NC-SA 3.0 and has been abridged and
altered by the author.
Contents
Abstract ii
Acknowledgements iii
List of Figures vii
List of Tables ix
Abbreviations x
Physical Constants xi
Symbols xii
1 Introduction 1
1.1 Energy challenges in the 21 st century . . . . 1
1.1.1 Energy security and global warming . . . . 1
1.1.2 The energy transition . . . . 2
1.2 Global energy demand and the role of the buildings sector . . . . 3
1.3 Residential buildings and the importance of space heating . . . . 5
1.4 Aim and objectives . . . . 7
1.5 Scope and system boundaries . . . . 8
1.6 Outline of the thesis . . . . 9
2 Literature Review 11 2.1 Energy system models . . . . 11
2.1.1 General characteristics of energy system models . . . . 11
2.1.2 Energy system models for the residential sector . . . . 14
2.2 Scenario analysis . . . . 15
2.3 Energy efficiency improvements in the residential sector . . . . 16
3 Methodology 18 3.1 The overall approach . . . . 18
3.2 The current state of the UK’s residential space heating sector . . . . 19
3.3 Technology solutions for residential space heating . . . . 20
3.4 Modeling future space heating demand . . . . 21
Contents
3.4.1 Model specifications . . . . 22
3.4.2 Estimating future energy demand for space heating . . . . 22
3.4.3 Scenarios for future efficiency improvements . . . . 23
4 The current state of the UK’s residential space heating sector 25 4.1 Energy demand in the UK’s residential sector . . . . 25
4.1.1 The UK’s total final energy demand . . . . 25
4.1.2 Final energy demand in the UK’s residential sector . . . . 25
4.2 Building characteristics of the UK’s residential sector . . . . 26
4.2.1 Evolution of the housing stock . . . . 26
4.2.2 Dwelling characteristics . . . . 28
4.3 Factors influencing a dwelling’s demand for space heating . . . . 29
4.3.1 Outside factors affecting dwellings’ demand for space heating . . . . 30
4.3.2 The impact of residents’ behavior on dwellings’ space heating demand . . 31
4.3.3 The technological state of UK dwellings . . . . 31
4.4 Energy demand for space heating in UK dwellings . . . . 40
5 Technology solutions for residential space heating 43 5.1 Improvement opportunities in non-technological drivers . . . . 43
5.1.1 The type of dwellings . . . . 43
5.1.2 The size of dwellings . . . . 44
5.1.3 Residents’ behavior . . . . 44
5.2 Improvement opportunities in the building envelope . . . . 44
5.2.1 Improvements in existing buildings . . . . 44
5.2.2 Improvements in new buildings . . . . 45
5.3 Improvement opportunities in heating equipment . . . . 47
5.3.1 Improvements in fossil fuel-fired heating equipment . . . . 47
5.3.2 Improvements in electrically driven heating equipment . . . . 48
5.3.3 Improvements in other heating equipment . . . . 49
5.3.4 Improvements through adoption of novel heating technologies . . . . 49
5.4 Improvement opportunities through smart devices . . . . 50
6 Modeling future efficiency improvements 51 6.1 Model specifications . . . . 51
6.1.1 External projections and assumptions . . . . 51
6.1.2 Technological state of the space heating sector . . . . 53
6.2 Defining scenarios for future efficiency improvements . . . . 56
6.2.1 Results of scenario development . . . . 56
6.2.2 The ‘Minimal effort’ scenario . . . . 57
6.2.3 The ‘Limited effort’ scenario . . . . 59
6.2.4 The ‘Efficiency focus’ scenario . . . . 60
6.2.5 The ‘Renewables focus’ scenario . . . . 61
6.3 Simulation results . . . . 63
6.3.1 Evolution of the thermal performance of the building envelope under the four scenarios . . . . 63
6.3.2 Evolution of the performance of heating equipment under the four scenarios 64
6.3.3 Evolution of the contribution of smart devices under the four scenarios . . 66
Contents
6.3.4 Realized efficiency improvements in the four scenarios . . . . 67
6.3.5 Final space heating energy demand in the four scenarios . . . . 68
7 Discussion and conclusion 73 7.1 Discussion . . . . 73
7.1.1 Non-technological drivers . . . . 73
7.1.2 Building envelopes . . . . 74
7.1.3 Heating equipment . . . . 74
7.1.4 Smart heating devices . . . . 75
7.1.5 Implications for the UK’s residential space heating sector . . . . 75
7.2 Conclusion . . . . 78
7.3 Limitations and applicability . . . . 80
7.4 Recommendations for future work . . . . 82
7.4.1 Expanding the work done in this thesis . . . . 82
7.4.2 Broadening the system boundaries . . . . 83
Bibliography 84 A Model specifications and results 90 A.1 Building stock evolution . . . . 90
A.2 Model results . . . . 90
List of Figures
1.1 The relentless rise of atmospheric CO 2 concentrations . . . . 3
1.2 Global final energy consumption per sector . . . . 4
2.1 Overview and grouping of bottom-up and top-down modeling techniques for the domestic sector . . . . 15
3.1 Visualization of the three improvement areas and underlying improvement levers that are considered in this thesis . . . . 20
4.1 Historical evolution of final energy demand in the UK, split per sector . . . . 26
4.2 Historical evolution of energy demand in the UK’s residential sector, split per end-use . . . . 27
4.3 Evolution of the total number of dwellings, total residential floor space, and resulting average dwelling floor space . . . . 29
4.4 Average mean temperature between 1981 and 2010 . . . . 30
4.5 Average annual sunshine duration between 1981 and 2010 . . . . 30
4.6 Evolution of the average indoor temperature in UK dwellings between 1970 and 2012, and evolution of the share of centrally heated homes . . . . 32
4.7 Heat loss parameter of different types of dwellings in the UK in 2012 . . . . 33
4.8 Heat loss parameter of different age categories of UK dwellings in 2012 . . . . 34
4.9 Fuel mix in the UK’s residential space heating sector from 2000 to 2013 . . . . . 36
4.10 Evolution of the number of installed non-condensing and condensing boilers in the UK between 1975 and 2014 . . . . 37
4.11 Energy demand for residenial space heating in the UK from 1970 to 2012 . . . . 41
4.12 Normalized space heating energy consumption per unit area of floor space be- tween 2000 and 2012 . . . . 42
5.1 Whole-building U -values of selected European countries . . . . 46
6.1 Evolution of the number of dwellings in the UK between 2012 and 2050 . . . . . 53
6.2 Evolution of the average heat loss parameter of the building stock in the four scenarios . . . . 63
6.3 Evolution of the average seasonal conversion efficiency of the heating equipment stock in the four scenarios . . . . 65
6.4 Evolution of the average savings achieved through smart thermostats in the four scenarios . . . . 66
6.5 Evolution of the average energy use intensity of the UK’s building stock under the four scenarios . . . . 68
6.6 Evolution of total final energy demand for space heating in the UK under the
four scenarios . . . . 69
List of Figures
6.7 Evolution of final energy demand for space heating in the UK under the four scenarios . . . . 70 6.8 Evolution of the space heating fuel mix in the UK under the four scenarios . . . 71 7.1 Energy use for residential heating in National Grid’s ‘Gone Green’ scenario until
2030 . . . . 81
7.2 Energy use for domestic heating in DECC’s baseline scenario until 2050 . . . . . 81
List of Tables
1.1 Top-3 countries’ shares of proved reserves for fossil fuels . . . . 2 2.1 A comparison of the key features of bottom-up and top-down energy models . . . 13 4.1 Distribution of UK dwellings by age category in 2014 . . . . 28 4.2 Different space heating systems in use in 2012, together with the share of UK
dwellings with such a system installed . . . . 40 5.1 Historical U -values and air permeability standards in the UK’s building regulations 46 6.1 Morphological table for energy efficiency improvements in residential space heat-
ing in the UK . . . . 57
6.2 Description of the four scenarios selected for in-depth analysis . . . . 58
6.3 Evolution of the space heating fuel mix in the UK under the four scenarios . . . 72
A.1 Evolution of the UK’s building stock . . . . 90
A.2 Evolution of the UK’s total residential floor space . . . . 90
A.3 Evolution of final energy demand for space heating in the minimal effort scenario 90
A.4 Evolution of final energy demand for space heating in the limited effort scenario 91
A.5 Evolution of final energy demand for space heating in the efficiency focus scenario 91
A.6 Evolution of final energy demand for space heating in the renewables focus scenario 91
Abbreviations
ASHP Air-Source Heat Pump BAU Business As Usual
COP Coefficient Of Performance EUI Energy Use Intensity GDP Gross Domestic Product GSHP Ground-Source Heat Pump Mtoe Million Tonnes of Oil Equivalent
OECD Organisation for Economic Co-operation and Development ppm Parts Per Million
SPF Seasonal Performance Factor
UK United Kingdom
US United States
Physical Constants
Million tonnes of oil equivalent 1 Mtoe = 41.868 · 10 15 Joule
Petajoule PJ = 10 15 Joule
1
As different types of crude oil have different calorific values, the exact value of one Mtoe is defined only by
convention. Here, the value defined by the International Energy Agency is used.
Symbols
◦ C Temperature Degrees Celsius
E Energy Joule
EU I Space heating energy use intensity kW h/m 2
K Temperature Kelvin
s Smart thermostat savings %
W Heat loss Watt
η Seasonal conversion efficiency %
To my parents.
Chapter 1
Introduction
1.1 Energy challenges in the 21 st century
1.1.1 Energy security and global warming
Mankind in the 21 st century is facing major challenges concerning the way energy is currently being consumed. In their Sustainable Development Goals, the United Nations summarize the challenges as “ensuring access to affordable, reliable, sustainable, and modern energy for all ” (UN, 2016). Holdren (2006) brings these challenges down to two key issues, that stand out above the others due to “their combination of difficulty and danger ”: Energy Security and Global Warming.
The first challenge, Energy Security, is a matter of national security, and has long been acknowledged by many governments to be a major issue of pressing importance. It relates to the fact that, to this day, fossil fuels still make up the lion’s share of the world’s total energy consumption, with a global share of around 81% in 2013 (The World Bank, n.d.). This domi- nance of fossil fuels is ubiquitous. They made up 81.5% of the U.S.’s total energy consumption in 2015, 72.6% of that of the EU in 2013, 88.1% of that of China, and a similar share for many other regions (EIA, 2016a; Eurostat, n.d.b; The World Bank, n.d.).
While fossil fuels remain the primary source for meeting the world’s energy needs, cheaply ex- tractable and reliably deliverable conventional oil and natural gas are dwindling away at an alarming rate. British Petroleum (2016) estimates that the world has proved reserves for an- other 51 years of oil and 53 years of natural gas. Of course, these numbers should be seen in their appropriate context. Proved reserves increase every year as new resources are discovered and become economically viable to extract. No doubt exists, however, that fossil fuels are a finite resource, and that their extraction is increasingly complicated and costly (Holdren, 2006).
This problem is exacerbated by the fact that the remaining resources are unevenly spread across
the world. For oil, natural gas and coal, the three countries with the most reserves hold 44.2 %,
48.6 % and 57.0 % of the world’s proved reserves, respectively, as shown in table 1.1. In con-
trast, more than half of the European Unions energy consumption in 2014 came from imported
sources (Eurostat, n.d.a).
Section 1.1 Energy challenges in the 21 st century
Table 1.1: For oil, natural gas and coal, the three countries with the most reserves hold 44.2 %, 48.6 % and 57.0 % of the world’s proved reserves, respectively.
(Source: Adapted from British Petroleum (2016))
Total of
top-3 countries
Oil Venezuela
17.7%
Saudi Arabia 15.7%
Canada
10.1% 44.2%
Natural gas Iran 18.2%
Russia 17.3%
Qatar
13.1% 48.6%
Coal United States
26.6%
Russia 17.6%
China
12.8% 57%
The increasingly small amount of fossil fuel reserves concentrated in a small number of some- times politically unstable nations poses a serious threat to the security of energy supply in the regions of the world that rely heavily on imports of these fossil fuels.
The second challenge, Global Warming, is a global issue, prone to much controversy and conflicting interests, and comes with a great deal of uncertainty as to the extent and geograph- ical distribution of the effects, and a significant time lag (Oreskes, 2004; Holdren, 2006). This has led to governments or special interest groups downplaying the severity of its effects or even denying it completely (McRight and Dunlap, 2011). Fortunately, global warming is starting to be universally accepted as true and caused by human action (Oreskes, 2004). It refers to the phenomenon whereby the Earth’s average surface temperature is increasing at a, historically speaking, unusually rapid pace primarily due to the release of harmful components in the at- mosphere as people burn fossil fuels (Earth Observatory, n.d.). These harmful components are called ’greenhouse gases’, as they intensify the Earth’s natural greenhouse effect, causing global warming (IPCC, 2007). Anthropogenic greenhouse gas emissions have been artificially raising the concentration of greenhouse gases in the atmosphere at an ever-increasing rate, as shown in figure 1.1, mostly through combustion of fossil fuels, but also from cutting down forests that would otherwise absorb part of the emitted carbon (Nasa, n.d.).
1.1.2 The energy transition
To counter the two previously discusses energy challenges, major efforts are needed to fundamen- tally change the way energy is consumed. The necessary changes are such a radical shift from the modern-day global energy system that we have always known, that it is not far-stretched to speak of an Energy Transition (IEA, 2015; Engie, n.d.; McKinsey & Company, 2016). In its 2016 edition of the World Energy Outlook, the IEA speaks of an “energy landscape in flux ” (IEA, 2016, 32), gathering momentum thanks to the Paris Agreement on climate change of November 2016, and representing a strong and unified signal of determination to accelerate the transition to a cleaner and more efficient energy system.
From this wording it can deduced that, according to the IEA, this energy transition can be
brought down to two key measures: Renewable Energy Sources, to make the global energy
system cleaner, and Energy Efficiency, to make it more efficient.
Section 1.2 Global energy demand and the role of the buildings sector
Figure 1.1: The recent and relentless rise of atmospheric CO
2concentrations. In 2013, carbon dioxide levels surpassed the symbolic 400 ppm milestone for the first time in recorded history.
(Source: Nasa (n.d.))
The first measure, Renewable Energy Sources, refers to the use of energy sources that are not of a depletable nature. Rather, they are being renewed through the natural functioning of the Earth, hence the name ’renewable’. Today, the most prevalent renewable energy sources include hydro, wind, solar, geothermal and biomass. Renewable energy sources counter both challenges effectively. First, they emit little to no greenhouse gases when producing energy, thus mitigating climate change 1 . Second, they are present everywhere on Earth in at least some useful forms, enhancing energy security for countries otherwise relying heavily on imported fossil fuels.
The second measure, Energy Efficiency, refers to more efficient utilization of primary en- ergy. Generally, increasing energy efficiency is one of the most cost-efficient ways of reducing the environmental burden of energy use and strengthening energy security (IEA, 2016; Holdren, 2006). It lowers the amount of primary energy needed for delivering the same useful energy, thus reducing total energy consumption and related greenhouse gas emissions, if combustion of fossil fuels was used for delivering the desired energy. Almost all of the Nationally Determined Contributions (NDCs) submitted to the COP21 2 address energy efficiency (UNFCCC, n.d.).
1.2 Global energy demand and the role of the buildings sector
In 2014, the world’s total final energy demand amounted to 9425 Mtoe. Final energy demand can be disaggregated into different end-uses, as shown in figure 1.2. Perhaps surprisingly, the
1
Hydro, wind and solar energy sources emit no greenhouse gases when producing energy, but are responsible for some emissions during manufacturing, installation, maintenance, dismantling and decommissioning. Geothermal power plants are associated with some greenhouse gas emissions, depending on the type of system used, but these are much smaller than for fossil fuel-based plants. Biomass is more controversial in that respect. If sustainably sourced, its environmental footprint is low. On the other hand, unsustainably sourced biomass can have a significant environmental footprint, on par with that of fossil fuels.
2
COP21 refers to the 2015 United Nations Climate Change Conference, held in Paris in November, 2015.
Section 1.2 Global energy demand and the role of the buildings sector
buildings sector is the largest energy-consuming sector, accounting for 35 % of the world’s final energy consumption in 2010 and a similar share of carbon emissions (IEA, 2013).
Figure 1.2: Buildings, transport and industry each account for roughly one-third of global final energy consumption, with the buildings sector coming out on top with a share of 35 %.
Note: “Other sectors” includes agriculture, forestry, fishing and other non-specified.
(Source: Adapted from IEA (2013))
Usually, within the buildings sector, a distinction is made between residential 3 and commercial buildings, as they are quite different from each other in terms of energy consumption patterns and underlying drivers.
The residential sub-sector encompasses all houses, apartments and other types of dwellings where people live. The commercial sub-sector encompasses all buildings whose main purpose is to allow for activities related to trade, services, public administration, etc.
In the residential sub-sector, energy consumption is closely linked to factors such as population, number of households, floor space, building envelope age and characteristics, income and income growth, consumer behavior, climate, etc. (Swan and Ugursal, 2009).
In the commercial sub-sector, energy consumption is closely linked to factors such as level of economic activity, floor space, climate, etc.
The residential sub-sector is by far the largest of the two, accounting for roughly three quar- ters of total final energy consumption in buildings. Even though energy consumption in the commercial sub-sector is growing much faster than in the residential sub-sector, the latter is expected to remain dominant until 2050 (IEA, 2013).
Despite the importance of the residential sector in the world’s total energy consumption, the dynamics behind its energy demand are largely still an undefined energy sink. This in contrary
3
In this thesis, the words residential sector, domestic sector, and households are used interchangeably to
indicate the energy-consuming sector consisting of living quarters for private households. End-uses of energy
commonly associated with this sector include space heating, water heating, lighting, cooking, space cooling,
and running a variety of other household appliances and consumer electronics. The residential sector excludes
institutional living quarters.
Section 1.3 Residential buildings and the importance of space heating
to energy consumption of the other major sectors, industry, transport, and the electric power sector. Swan and Ugursal (2009) impute this to four key reasons:
• The residential sector consists of a broad variety of structural sizes, geometries, and ther- mal envelope materials
• Resident behavior can vary widely and can impact a given dwelling’s energy consumption by as much as 100 %
• Due to privacy issues, collection of energy data at individual household level is limited
• Detailed sub-metering of the end-uses in a household has prohibitive costs
1.3 Residential buildings and the importance of space heating
Within the residential sub-sector, the IEA (2013) distinguishes between six ’energy services’ - services that provide utility to a dwelling’s residents and which in the process consume energy.
These six services are space heating, space cooling, cooking, water heating, lighting, and res- idential appliances and electronics. All energy consumed in residential buildings can more or less be attributed to one of these six services.
Dwellings make use of various energy sources for delivering services. In OECD countries, non- OECD Europe and Eurasia, electricity and natural gas are the dominant fuel types used. In other parts of the world, biomass and waste remain the most widely used energy sources.
While the used fuel types vary in different parts of the world, one thing is shared: conventional fuels continue to be the dominant energy source in residential buildings. In 2010, residential space and water heating needs were still largely met through combustion of fossil fuel- and traditional biomass-based equipment, with a global share of 85 % (IEA, 2013).
Globally, space heating and cooking dominate residential energy consumption, where the former predominates in OECD 4 countries and the latter predominates in non-OECD countries (IEA, 2013).
Energy demand for space heating, the largest service in terms of energy consumption, is influ- enced through a variety of drivers.
One of the key drivers is the climatic conditions in which the dwelling needs to provide heat- ing services. Other key drivers include behavior of the residents in terms of setting the room temperature, which is largely determined by their wealth, as confirmed in a study by Bentzen and Engsted (2001); the dwelling’s building envelope characteristics in terms of insulation, air tightness, and architectural design; the size of the dwelling in terms of floor space that requires heating; and the efficiency of the equipment used in terms of energy conversion efficiency and distribution channels (e.g. adequately sealed and insulated ducts and pipes).
Given the importance of space heating in the context of residential buildings’ energy consump- tion, efforts in reducing said service’s energy demand will prove instrumental in achieving the
4
OECD stands for the Organisation for Economic Co-operation and Development, and currently includes 35
(mostly developed) countries.
Section 1.3 Residential buildings and the importance of space heating
often ambitious targets concerning energy consumption in the buildings sector set forth by for example the European Union (European Commission, 2010). Various levers exist through which space heating energy consumption can be reduced. An often cited improvement lever is chang- ing residents’ energy-related behavior, for example by lowering the desired room temperature or ensuring rooms are only heated when needed (Swan and Ugursal, 2009). To what extent progress will be made in this area is uncertain. One thing is sure, as people get wealthier, they become more demanding in terms of comfort and more negligent in terms of their behavior, which leads to believe the future holds little good (Fazeli et al., 2016).
A second, more promising improvement lever will be the focus of this thesis. Here, the fo- cus will be on looking at the potential for energy efficiency improvements in residential space heating achieved through adoption of superior technologies. Cockroft and Kelly (2006) have identified the large-scale deployment of renewable technologies, in combination with improved energy efficiency as a means of curbing rising energy demand and carbon emissions.
As mentioned at the end of section 1.1, energy efficiency is seen as one of the most cost-efficient ways to reduce energy consumption. The focus of this analysis will be on the United Kingdom, which is done for three reasons. First, the UK is responsible for a large share of space heating energy consumption in the EU, accounting for 13 % of the EU28’s 5 total space heating energy consumption in 2013, behind only Germany and France. This is partially due to the sheer size of the UK in terms of population (almost 64 million in 2013, or 12.7 % of the EU28), but also due to the UK’s climate, which is among the colder ones in the EU, necessitating significant space heating. Second, as a member of the European Union 6 the UK is bound to stringent targets in order to contribute to a more competitive, secure and sustainable energy system in the EU, and to meet its long-term target for greenhouse gas reductions by 2050 (European Commission, n.d.). Third, for the UK, all national statistics, press reports, and other information sources are available in English, which facilitates data collection.
A prerequisite for understanding the technological potential for efficiency improvements in res- idential space heating in the UK is the development of a solid and up-to-date starting point, which covers all technological aspects relevant to the space heating sector in sufficient detail.
Building on top of that, the relevant technology options must be identified that could lead to efficiency improvements, taking into account both current and future technologies.
Up-to-date models that focus on residential space heating in the UK, keep a holistic view on all technological aspects that affect space heating energy demand, go to the required level of detail, and incorporate both current technologies and technologies expected to take up significantly in the future, are not sufficiently available.
Johnston et al. (2005) have developed an extensive bottom-up, physically-based building stock model of the UK’s residential sector, looking at the building envelope and at heating equipment.
However, their model dates back to the early 2000s, and significant changes have occurred since then. Bell and Lowe (2000) have proven the feasibility of modernizing the UK’s existing building stock, but stick to well established (early 1980s) technology. Dowson et al. (2012) have reviewed the thermal performance of the UK’s building stock, but don’t look at other factors affecting
5
The EU28 encompasses all 28 member states of the European Union, with Croatia in 2013 being the latest addition.
6
At the time of writing, the UK is still a member of the EU. How the Brexit will affect the UK’s climate
policies remains speculative, but it can be expected that its climate targets will be in line with those of the EU.
Section 1.4 Aim and objectives
energy demand for space heating. No academic literature was found that quantified the impact of smart heating devices on energy demand for space heating in the UK.
Various technical reports on the future of the UK’s heating sector can be found, such as De- taEE (2012); DECC (2013); Element Energy (2015). However, most of these reports focus on the heating supply and the shift to low-carbon options, but don’t cover the full range of po- tential efficiency improvements, or at least don’t explicitly quantify their impact. Furthermore, these technical reports often do not fully disclose the used methodology and data points.
1.4 Aim and objectives
This thesis intends to contribute to the existing academic knowledge by exploring the techno- logical potential for energy efficiency improvements in residential space heating in the United Kingdom through adoption of improved technological options.
This is achieved on the one hand by looking at the current state of the energy performance of the UK’s space heating sector, in terms of technological options that are in place, and in terms of their energy performance. And on the other hand, by identifying superior technologies and quantifying their potential for improvement that could be achieved if people adopted them.
Together, these two parts provide a view on the technological potential for efficiency improve- ments in the UK’s space heating sector.
In addition, this thesis aims to provide insights into the effects on future space heating energy demand of adopting these superior technologies, under three simple yet distinct scenarios. The evolution will be modeled until the year 2050, as this is a reference year that is commonly used in energy and climate planning (Johnston et al., 2005; Cockroft and Kelly, 2006; McKenna et al., 2013).
The main contribution of the obtained results is to present a holistic assessment of the cur- rent state of the UK’s residential space heating sector, to identify technological options for efficiency improvement and assess their improvement potential, and to give an indication of the effects of current and accelerated technology adoption patterns on space heating energy demand in the UK by 2050.
This could then give an idea of how much additional effort will be needed in the years to come in order to achieve the goals and targets set forth by policy makers. The first part of the thesis, where the potential for energy efficiency improvements is analyzed, then provides a view on the
’tool-set’, so to speak, that can be leveraged in order to reach these goals and targets. The second part, where the resulting space heating energy demand are investigated, then provides a view on the effects of certain adoption patterns.
The principal research question of the thesis is:
What is the technological potential for energy savings through efficiency improvements in residential space heating in the United Kingdom by the year 2050?
Three sub-questions can be identified, that together provide an answer to this principal research
question:
Section 1.5 Scope and system boundaries
• Which aspects related to residential space heating energy consumption in the UK present the highest technological potential for efficiency improvements, and what is the current average performance of UK dwellings for these as- pects?
• For these most promising aspects, what are the particular technological op- tions that have the highest technological potential to increase space heating efficiency, and how does their average performance compare to the technolog- ical options currently in use?
• How will energy demand in the UK’s domestic space heating sector evolve un- der different scenarios of the adoption of the identified technological options?
1.5 Scope and system boundaries
It is important to stress that the goal of this thesis is not to take a system perspective on the UK’s energy system, and from there model reductions in primary energy demand through adoption of superior technologies. Instead, the focus lies only on the delivered end-use energy that is consumed on the premises of final customers. This delivered end-use energy depends on several factors, one of them being the performance of the technologies used for providing heat to and maintaining that heat in a dwelling. Improvements in the latter are the focus of this thesis.
The energy system boundaries observed in the thesis are the residential, end-user premises where final energy is consumed to deliver heat to dwellings. The focus of this thesis is on understanding the technological potential for reducing this final energy delivered to end-user premises.
While a holistic perspective would require looking at both the supply and demand side, this is not the focus of this thesis. Firstly, because technological options for efficiency improvements on the supply side are of a completely different nature than those on the demand side. Secondly, because end-users cannot influence the supply side directly, but are only able to influence the final energy demand they require to heat their house, regardless of the impact on primary energy consumption.
The aim of this thesis is to look at energy system improvements from the end-user perspective, and within the end-user’s sphere of influence. As a result, some of the measures in this thesis that are portrayed as ‘improvements’, might actually have a negative impact on the UK’s energy system as a whole.
It should be noted that this does not represent a flaw in the methodology or a too narrow scope of this thesis. On the contrary, it allows for exposing the opposing dynamics inherently present when investigating both the supply and demand side of an energy system. It is then up to policymakers to keep a system perspective, and identify these opposing dynamics by looking at both the supply and demand side. Policies and incentives can then be used to align the different driving forces on both sides, ensuring a pathway towards an optimal energy system is followed.
This thesis can contribute to this complex issue by providing insights in the technological op-
tions end-users have concerning space heating, and what their effects would be on final energy
Section 1.6 Outline of the thesis
demand.
Within the energy system boundary of end-user premises, only energy required for space heating is considered. One caveat here is that technological measures might have an impact on more than space heating alone. For example, a better dwelling insulation not only reduces the need for space heating, but also lowers energy demand for space cooling 7 .
For this thesis, this caveat is not really an issue, as only energy used for space heating is inves- tigated. Some measures might affect other end-uses, but as they fall out of the scope of this thesis, this does not impact the final results.
This would be different, however, for studies that look at all residential end-uses. There, the effects of technological measures on all end-uses must be taken into account.
Energy demand for space heating depends on various drivers. Seligman et al. (1977) iden- tify outside physical conditions, technological state of the building, and residents’ behavior as the primary determinants of a dwelling’s space heating energy demand.
In this thesis, only changes in the technological state of buildings is investigated. While it can be expected that changes to outside conditions and residents’ behavior are likely to occur, tak- ing these into account would obscure the insights that are obtained from looking at changes in technological state of buildings only.
Thus, in summary, the scope of this thesis is limited to analyzing technological options on the demand side of the UK’s domestic space heating sector, and in exploring their impact on final energy demand.
To get a view on the full picture, additional studies are needed that investigate the economics of these technological options, that look at customer preferences to understand actual technology adoption patterns, and that carry out a similar analysis for the supply side. Combined, these analyses would provide the holistic perspective necessary for making optimal decisions on this complex topic. The limitations of this study and recommendations for further work are covered more extensively in chapter 7.
1.6 Outline of the thesis
This thesis is structured as follows. Chapter 1 provided an introduction that sketches the energy transition that the world is currently experiencing, and places residential space heating in that context. In chapter 2, a literature study is conducted where the academic literature is investigated concerning bottom-up energy models, past works on residential space heating efficiency improvements in the UK, and scenario analysis. In chapter 3, the methodology is presented with which the analysis will be carried out in order to obtain an answer to the principal research question. In chapter 4, the current state of the UK’s space heating sector is covered. Chapter 5 provides an overview of the technology options that could lead to efficiency improvements. Chapter 6 brings the findings of the previous two chapters together to develop
7
Energy demand for space cooling in the UK is negligibly small, and therefore this example is less relevant
here. However, for other countries subject to different climate conditions, better dwelling insulation will reduce
the need for both space heating and cooling.
Section 1.6 Outline of the thesis
a model for calculating future demand for space heating based on adoption of the identified
technology options. In chapter 7, the findings of the thesis are discussed and reflected upon,
and the limitations of the study and recommendations for future work are presented.
Chapter 2
Literature Review
This chapter provides an overview of the relevant academic literature that was consulted in order to serve as reference material for this thesis.
In section 2.1, energy system models are covered, starting with a broad overview and then nar- rowing down to energy models for the residential sector.
In section 2.2, scenario development and analysis methods are covered.
In section 2.3, previous academic results on space heating in the UK, particularly on technolog- ical improvements and their impact on future energy demand are covered.
2.1 Energy system models
This section provides an overview of the relevant academic literature concerning energy system models, mainly focusing on those that cover the residential sector.
2.1.1 General characteristics of energy system models
Models for analyzing energy systems became widely available in the 1970s. These models served multiple purposes, such as acquiring a better understanding of optimal supply system design for a given projection in energy demand, being able to take into account interactions between energy and environment, or general energy system planning (Bhattacharyya and Timilsina, 2010). The subsequent proliferation of energy system models has exacerbated the need for systematic and comprehensive reviews of the field. Categorization of energy models can be done to a variety of ways, depending on the desired scope of comparison.
Hoffman and Wood (1976) categorized models based on the underlying modeling technique, and identified linear programming-based methods, input-output approaches, econometric meth- ods, process models, system dynamics-based methods, and game theory approacehs.
van Beeck (1999) distinguished between nine ways of classifying energy models 1 , based on the
1
The work of van Beeck (1999) is based on previous works by Hourcade et al. (1996) and Grubb et al. (1993),
to which the interested reader is referred for more information concerning the classification of energy models.
Section 2.1 Energy system models
purpose of the model (general and specific), the model structure (internal and external assump- tions), the analytical approach (top-down versus bottom-up), the underlying methodology, the mathematical approach that is used, the geographical coverage (project, local, national, re- gional, or global), the sectoral coverage, the time horizon (short, medium, or long term), and lastly the data requirements.
Pandey (2002), on the other hand, categorized energy models by making use of a set of key at- tributes, and identified the paradigm (top-down/simulation versus bottom-up optimization/ac- counting), the space (local, regional, national, or global), the sector (energy and/or macro- economy), and the time (short, medium, or long term) as being the most important ones.
Nakata (2004) made the classification based on the modeling approach (top-down versus bottom- up), the underlying methodology (partial or general equilibrium, or hybrid), the modeling tech- nology (optimization, econometric, or accounting), and the spatial dimension (national, regional, or global).
Bhattacharyya and Timilsina (2010), in their review of energy system models, consider the modeling approach (or paradigm), the sectoral coverage, the time horizon, and the spatial focus as criteria to classify the considered models.
Clearly, energy models are classified using a broad variety of criteria, based on the specific contrasting elements that the classifiers want to highlight. However, only very few models - if any - will fit exclusively into one classification category, which makes comparing classification reviews an ambiguous task.
One recurring theme is present in almost all energy model classifications, though. Pandey (2002), Nakata (2004), Bhattacharyya and Timilsina (2010), and van Beeck (1999) all distin- guish between either a top-down or a bottom-up paradigm. This distinction is deemed to be a fundamental modeling choice, and is further elaborated in the next paragraphs.
According to van Beeck (1999), top-down models are often associated with the economic ap- proach, while bottom-up models are often associated with the engineering approach. The former have no explicit representation of technologies, and instead treat them as a black box. This makes it difficult to incorporate detailed projections of technologies into these models. The lat- ter use highly disaggregated data to quantify energy end-uses and describe technological options in detail.
Bhattacharyya and Timilsina (2010) largely agree with these criteria. They associate top- down models with an econometric approach, while bottom-up models are more inclined towards optimization or accounting methods. The former are able to incorporate a varying suite of technology options, but this is usually rather limited and new technologies are then hard to include later on in the modeling period. The latter allow for incorporating an extensive suite of technology options, which are usually pre-defined but which can be added later on in the modeling period.
Hourcade et al. (1996) further subdivides bottom-up models into prescriptive and descriptive
models. Prescriptive models attempt to provide an estimate for the technological potential, and
look at the effects of adopting the most efficient existing technologies only. Descriptive models,
on the other hand, attempt to provide an estimate of the technology mix that would follow from
realistic decisions, that are based on market barriers, human preferences, attitudes towards risk,
etc. As a consequence, prescriptive models tend to be more optimistic than descriptive models,
Section 2.1 Energy system models
and are more suitable for exploration purposes.
The purpose of van Beeck (1999) was to provide guidance in choosing the most suitable en- ergy models for a given modeling task. On this topic, the following insights are of primary importance to the work done in this thesis.
Top-down models are only suitable if historical relationships and patterns of development among the underlying key drivers and variables hold constant for the duration of the projection period, meaning there can be no discontinuity in historical relationships. This makes top-down models more suitable for short-term predictions.
Bottom-up models, on the contrary, allow for modeling the discontinuities in trends derived from past behavior, through explicit modeling of the underlying technologies. When modeling energy demand, these models are inherently capable of differentiating between energy consumed for the different end-uses in a detailed and disaggregated way, based on what the desired energy services are.
The conclusion that follows from the review by Bhattacharyya and Timilsina (2010) is to a large extent aligned with van Beeck (1999). They identify flexibility and the ability to describe technological options in detail as the primary advantages of bottom-up models. Their main disadvantage is the fact that they do not allow for incorporating prince-induced effects.
Top-down models’ main advantages include the fact that they are less data-intensive, and allow for incorporating price-induced effects. Their main disadvantage is their lack of detailed tech- nology descriptions, which also makes them less suitable for projections in the longer term.
The key features of bottom-up and top-down energy models are summarized in table 2.1.
Table 2.1: A comparison of the key features of bottom-up and top-down energy models, as identified by van Beeck (1999) and Bhattacharyya and Timilsina (2010).
Top-Down Models Bottom-Up Models
Use an economic approach Use an engineering approach Give pessimistic estimates on ”best” perfor-
mance
Give optimistic estimates on ”best” perfor- mance
Can not explicitly represent technologies, and do not allow for addition of new technologies
Allow for detailed description of technologies, and allow for addition of new technologies Reflect available technologies adopted by the
market, based on observed market behavior
Reflect technical potential, independent of ob- served market behavior
Use aggregated data for predicting purposes Use disaggregated data for exploring purposes Disregard the technically most efficient tech-
nologies available, thus underestimate potential for efficiency improvements
Disregard market thresholds (hidden costs and other constraints), thus overestimate the poten- tial for efficiency improvements
Determine energy demand through aggregate economic indices (GNP, price elasticities), but vary in addressing energy supply
Represent supply technologies in detail using disaggregated data, but vary in addressing en- ergy consumption
Endogenize behavioral relationships Asses costs of technological options directly Assume there are no discontinuities in historical
trends
Assumes interactions between energy sector
and other sectors is negligible
Section 2.1 Energy system models
2.1.2 Energy system models for the residential sector
An extensive review of modeling techniques tailored to the intricacies of the residential sector is provided by Swan and Ugursal (2009) and Kavgic et al. (2010). As in the reviews covered in the previous section, they identify two fundamental classes of modeling methods used for modeling end-use energy consumption in the residential sector, top-down and bottom-up modeling.
Here, the distinction between these two modeling techniques is made based on the hierarchical position of the input data, as compared to the domestic sector as a whole. Top-down models take external estimates for total residential energy demand, together with other key variables, as given, and use them to attribute this energy demand to different characteristics of the total housing stock. Bottom-up models, in contrast, develop a set of representative houses, and cal- culate the energy demand of this set of houses, to extrapolate the results to represent a whole region or nation. A schematic of the different energy models and sub-models identified by Swan and Ugursal (2009) is given in figure 2.1.
In the top-down approach, the residential sector is modeled as an ‘energy sink’, meaning en- ergy is consumed but one is not concerned with individual end-uses. The key advantages of top-down models are the need for data only at the aggregate sector level, the simplicity of the approach, and the reliance on historical energy values from the domestic sector, which to some degree provide inertia to the model. According to Kavgic et al. (2010), they are often used for investigating inter-relationships between the energy sector and the economy as a whole.
Well-calibrated top-down models are able to provide satisfying predictions for minor deviations from the status quo. One important drawback of top-down models is the inherent incapability to take into account any kind of discontinuous improvements in technology, due to its reliance on historical data.
The bottom-up approach, on the other hand, bases its estimates for energy consumption on an extrapolation, to regional or national levels, of the estimated energy consumption of a rep- resentative set of dwellings. Models in this category make use of input data from a hierarchal level that is lower than the level of the housing stock as a whole. Within this approach, two distinct methodologies are further identified by both Swan and Ugursal (2009) and Kavgic et al.
(2010), the statistical and the engineering or building physics based method.
Statistical methods rely on historical data and make use of regression techniques to attribute a dwelling’s energy consumption to specific end-uses.
Engineering methods explicitly model the energy consumption of end-uses based on the use of specific equipment types and systems and their underlying physical and thermodynamic charac- teristics. Models following this approach have the highest degree of flexibility and capability in terms of modeling new technologies for which no historical consumption data exists. To do this, however, occupant behavior must be assumed, which is often difficult to estimate, as outlined by both Swan and Ugursal (2009) and Kavgic et al. (2010).
Commonly used input information to construct bottom-up models includes building charac-
teristics, equipment and appliances adoption, climatic conditions, resident behavior, etc. This
high level of detail is both a strength and a weakness of bottom-up modeling. On the one hand,
a highly detailed model gives the ability to model distinct technological options and allows for
determining each end-use’s energy consumption. On the other hand, the need for such highly
Section 2.2 Scenario analysis
disaggregated data can be prohibitive in some cases, and also makes the calculation logic and simulation techniques more complex and less transparent. Kavgic et al. (2010) are particularly critical on this issue when reviewing bottom-up models, highlighting the dangers of a severe lack of representative and accurate data when using bottom-up models in actual policy planning.
Figure 2.1: An overview and grouping of common bottom-up and top-down modeling tech- niques for the domestic sector.
(Source: Swan and Ugursal (2009))
The conclusion of Swan and Ugursal (2009) on the attributes and applicability of the identified top-down and bottom-up approaches can be summarized as follows.
• Top-down techniques are most suited for analyzing the supply-side, based on longer-term projections of the energy demand that account for historic response
• Bottom-up statistical approaches are most suited for determining the contribution of each end-use in the total energy demand, and are capable of including behavioral aspects based on billing data and sample surveys
• Bottom-up engineering techniques are most suited for explicitly calculating the energy consumption of each end-use based on a detailed description of a set of representative houses, and these techniques are capable of accounting for the impact of new technologies
2.2 Scenario analysis
This section provides an overview of the relevant academic literature that was consulted in order
to better understand the techniques employed in scenario analysis.
Section 2.3 Energy efficiency improvements in the residential sector
Bradfield et al. (2005) provide insights into the origins and evolution of scenario techniques used in long-range business planning, and classify the methodologies into three main schools of techniques: Intuitive-Logics models, La Prospective models, and Probabilistic Modified Trend models.
An extensive review of the scenario planning literature is provided by Amer et al. (2013), where scenarios are defined as a set of alternative futures that result from some specific combination of trends and policies, which should incorporate and put an emphasis on the facets of the world that are of importance to the forecast.
As studies focusing on the future help to see the present in a different perspective, they should incorporate options that lie beyond the conventional ‘comfort zone’ that is the status quo. It should be noted that scenarios are not meant to predict the future. Instead, they explore various plausible future situations.
According to Amer et al. (2013), most researchers agree that sticking to three scenarios is the optimal approach. It is recommended by Schwab et al. (2003) to develop a trend extrapo- lation scenario, and a best- and worst-case scenario.
In terms of validating the scenarios, Amer et al. (2013) mention internal consistency and plau- sibility as the crucial aspects to consider.
According to Zivkovic et al. (2016), scenarios can be used to i) develop more robust path- ways in the context of uncertain future conditions ii) analyze effects of projected pathways in the future. For the former, external scenarios must be developed, while for the latter, internal scenarios are most suitable.
In the van Notten et al. (2003) scenario typology, a further distinction is made between chain and snapshot scenarios, depending on the “role of time in the scenario”. Here, chain scenar- ios represent a sequence of events that result in a particular future state. On the other hand, snapshot scenarios immediately describe an end state, and don’t focus on the intermediate de- velopments necessary to reach that state, which can be developed in a later stage.
Pereverza et al. (2017) have developed a morphological approach to internal scenario devel- opment and selection in the context of participatory strategic planning for sustainable heating in cities.
This novel approach is developed to deal with some of the limitations present in traditional methods for scenario development, and relies on the creation of a complete scenario space (in this case for heating systems), and a reduction of this space through a cross-consistency analysis.
2.3 Energy efficiency improvements in the residential sector
Johnston et al. (2005) have explored the technical feasibility of achieving CO 2 emission reduc-
tions in excess of 60 % within the UK housing stock by 2050. The study was conducted in light
of the release of the Energy White Paper by the UK government in 2003, where it stated that
one of its energy policy goals is to lead the UK towards a path to achieving a 60 % reduction in
Section 2.3 Energy efficiency improvements in the residential sector
CO 2 emissions by the year 2050 (DTI, 2003).
The issue is investigated by making use of a selectively disaggregated, physically based bottom- up energy and CO 2 emission model of the UK housing stock.
A key feature of the adopted selectively disaggregated approach is the use of only two so-called
’notional’ dwelling types, representing either pre- or post-1996 2 building structures, respectively.
The authors defend this significant simplification on two accounts. First, data and trend pro- jections of a variety of key building stock evolution metrics, such as insulation ownership, use of lights and appliances, or stock replacement cycles, are available only at the overall housing stock level. Second, the impact on energy use and CO 2 emission is small, if compared to build- ing envelope performance and equipment efficiencies. Therefore, as the authors state, in the long term, it is the average performance of building envelopes and domestic equipment across the whole building stock that is of primary importance, rather than the geometric or thermal singularities of various individual dwelling types.
The overall conclusion of this study by Johnston et al. (2005) is that by 2050, it is technically possible to achieve the energy demand and accompanying carbon emission reductions that are necessary to mitigate the effects of climate change, using currently available technology. These reductions appear to be feasible, despite a significant increase in the number of households and the thermal comfort these households will desire. Achieving these energy demand and carbon emissions reductions will be technically demanding, though, requiring the adoption of techno- logical measures at a substantially faster pace than what could be expected on the basis of historical and current trends.
The largest carbon emission reductions are projected to be achieved through a strategic shift away from the direct use of natural has in end-user premises, but rather a move towards elec- trically driven heat pumps, supported by a low-carbon electric power system.
A very carefully documented retrofitting attempt of four representative dwellings in the UK was done by Bell and Lowe (2000). In this case study, measures were taken to reduce the rate of air leakage, and to improve insulation of walls, doors, and windows, which combined led to reduced heating energy requirements of 35 % on average. It was mentioned that, by using well-proven early 1980s technologies, reductions of around 50 % could be achieved at modest cost.
According to Roberts (2008), insulating the walls and roofs of older dwellings without insulation could result in a 50 to 80 % reduction in heat loss through these building elements.
Of course, the actual achieved savings depend on the original performance of the dwellings, meaning that it is more meaningful to look at values for heat loss parameters achieved through retrofits, instead of savings potentials.
In the case study performed by Bell and Lowe (2000), average heat loss coefficients dropped from 225 to 145 W/K, an improvement of approximately 35 %. Converting this to the achieved aver- age heat loss parameter necessitates knowledge on the average floor space of the four dwellings, which is not given. However, it is mentioned that floor areas of the dwellings ranged between 75 and 95 m 2 , which means that the achieved average heat loss parameter would be in the range of 1.70 W/(m 2 K).
2
