Thermodynamic efficiency of low-carbon domestic heating

systems: heat pumps and micro-cogeneration

S. Cooper¹*, G. P. Hammond¹ ² and M. C. McManus¹ ²

1. Department of Mechanical Engineering, University of Bath, Bath, UK

2. Institute for Sustainable Energy and the Environment (I-SEE), University of Bath, Bath, UK

* Corresponding author: Department of Mechanical Engineering, University of Bath, Bath, BA2 7AY, UK.

Email: s.cooper2@bath.ac.uk Telephone: 01225 384550

Abstract:

Energy and exergy analysis is employed to compare the relative thermodynamic performance of low carbon

domestic energy systems based on Air Source Heat Pumps and micro Combined Heat and Power (cogeneration)

units. A wide range of current units are modelled under different operating conditions representative of the

United Kingdom to determine the energy and exergy flows from primary energy inputs through to low carbon

heating system and then to end use. The resulting performances are then analysed in order to provide insights

regarding the relative merits of the systems under the different operating constraints that may be experienced

both now and into the future. Although current mid-range systems achieve comparable performance to a

condensing gas boiler, the state-of-art offers considerable improvements. Micro Combined Heat and Power units

and Air Source Heat Pumps have the technical potential to improve the energy performance of dwellings. The

relative performance and potential of the systems is dominated by the electrical characteristics: the grid

electrical generation efficiency, the power-to-heat demand ratio and the availability of electrical export. For total

power-to-heat demands below 1:1.5, Air Source Heat Pumps have greater improvement potential as their energy

efficiency is not constrained. At higher power-to-heat ratios, micro Combined Heat and Power units offer the

potential for higher overall efficiency and this generally occurs irrespective of whether or not the thermal energy

from them is used effectively.

Keywords:

low carbon, heat pump, micro combined heat and power, cogeneration, heating, exergy analysis
1 INTRODUCTION

The carbon emissions and energy performance of the United Kingdom (UK) domestic sector must be improved.

Domestic space heating accounted for 20% of total primary energy consumption of the UK in 2010, contributing

13% of CO2 emissions [1]. It is estimated that a 29% reduction in the CO2e emissions associated with domestic

heating will be required as the UK meets its commitment to a total reduction of 34% relative to 1990 by 2020

[2].

To achieve this will require more efficient heating systems. Although ambitious improvements in building

standards are envisioned, it is estimated that 80% of the UK’s 2050 building stock is currently standing [3] and

even with extensive refurbishment, domestic heat and power demand are likely to remain substantial [4]. Micro

Combined Heat and Power units (mCHP) and Air Source Heat Pumps (ASHP) have both been suggested as

devices with the potential to reduce carbon emissions and energy demand [5]. They are increasingly attracting

attention, with major field trials of both devices conducted in recent years [6–9].

The approach taken in this study has been to model a selection of actual mCHP and ASHP units under a variety

of different assumed conditions and then to analyse the resulting range of energy and exergy performances to

draw general conclusions about the future applicability of these systems. Energy and exergy flows have been

modelled in terms of three stages: (1) primary energy conversion and distribution, (2) mCHP or ASHP unit and

(3) heat distribution in the dwelling (see Figure 1). In comparing the relative system performances, the ratio of

power-heat consumed at the end use stage is shown to be a key determinant of which system has the greater

potential efficiency.
Figure 1: Representative energy flows

2 BACKGROUND

2.1 Combined Heat and Power Units

Combined Heat and Power units combine the generation of power with heat recovery in order to maximise

efficiency. Stirling Engine mCHP (SE-mCHP) units are currently the most popular type of mCHP in the UK

private domestic sector as Internal Combustion Engine based mCHP (ICE-mCHP) devices tend to be larger and

pose more challenges to successful integration into the domestic environment. The higher capital cost of fuel

cell based mCHP has contributed to limited adoption to date but they still attract interest due to their potential

for high electrical efficiency [10]. The two leading fuel cell technologies are Polymer Exchange Membrane Fuel

Cell (PEMFC) and Solid Oxide Fuel Cell (SOFC). Uptake of PEMFC units has been greater (e.g. [7]) but the

higher electrical efficiency and simpler fuel reformer requirements of SOFC units means that they are

potentially the more attractive option in the longer term [11].

2.2 Heat Pump Systems

Heat pumps use a thermodynamic cycle to transfer heat from a source to a sink at higher temperature. In the

context of domestic heating they extract heat from the outside air, ground or water. Most commonly, a vapour

compression cycle is employed with an electrically driven compressor. Performance is measured by the heat

output versus the consumed electricity and expressed as the Coefficient of Performance (COP):

2.1

where C is the COP, Q is the heat delivered and W is the total work input to the heat pump unit (excluding any

component of power which is used for pumping hot water outside the heat pump).

Performance specifications and test conditions are standardised for ASHP application to space heating [12] and

potable hot water supply [13]; these definitions are followed in this study. Although Ground Source Heat Pumps

(GSHP, those that employ the ground as their heat source) are generally more efficient, the market potential for

Air Source Heat Pumps (ASHP) is considered larger in the UK due to the additional installation requirements

associated with GSHPs [14].

2.3 Exergy

To consider exergy, consideration must be made of the Second Law of Thermodynamics as well as the First

Law; unlike energy, exergy is destroyed in any process that involves irreversibility:

∑ ∑ ∑ 2.2

where ∑ is the sum of exergy inputs to the system, ∑ is the sum of desired exergy outputs from the

system, ∑ is the sum of any other exergy outputs from the system and is the amount of irreversibility

associated with the process. The exergy efficiency is then defined [15]:

2.3
Considering the case of heat transfer, , across the system boundary at constant temperature ( ), the associated

exergy transfer is given by the maximum work that could be obtained from that heat transfer:

2.4

where is ambient temperature and T is the temperature of the heat sink (e.g. heat emitters or inside air).

Kotas (1980) provides a methodology for calculating the exergy content of chemical enthalpy. It is convenient

to consider the exergy content of the fuel as related to its enthalpy of combustion, H, by a factor :

2.5

Allen & Hammond [17] have collected values of for various fuels. These are adapted for inclusion in Table 1:

Table 1: Exergy-energy ratio of fossil fuels. Adapted from [17]

Fuel

Coal 1.03

Fuel Oil 1.01

Natural Gas 0.94

3 METHODOLOGY

Performance data characterising 10 ASHP units and 11 mCHP units were gathered in order to provide an

overview of the range of units currently available. Although nominal performance data are readily available

([10], [18]) they are generally specific to particular steady-state conditions. Therefore, to analyse the energy

flows at each of the stages in Figure 1 in a way that enabled meaningful comparison between the units, a

dynamic model was constructed. The model includes a component for each stage, calculating energy and

exergy flows at each stage for each time step. Different permutations of building insulation, climate, heat emitter

size and heating control system were selected as representative of the range of conditions that might be expected

between now and 2030 (see Table 3). Once comparable performance results were determined for each unit,

these were analysed for different ratios of final electrical power demand to heat demand, derived by assuming

different domestic electrical power demands.

3.1 Grid electrical supply model

In order to calculate the primary energy and exergy requirements associated with each system, it is necessary to

know how the electricity which either supplies the ASHPs or is displaced by the mCHP electrical production is

generated. Two cases were considered in order to demonstrate the effect of assumptions regarding grid supply.

Firstly, a Combined Cycle Gas Turbine (CCGT) achieving a constant 54% electrical efficiency [19], equivalent

to 50.1% efficiency at point of use (taking transmission and distribution losses as 7.3% of electricity generated,

[20]). Secondly, a dispatch model was constructed by using the electrical generation totals for 2030 from the

―Market Rules‖ (―MR2030‖) scenario developed by the Transition Pathways project (see Table 2, [21]) and

historic generation data [22]. The dispatch model uses a modified merit order approach, similar to [23];

renewable and nuclear generation are prioritised, followed by industrial CHP, CCS plant and then CCGT and

Coal for peaking. It is assumed that there is some overlap in the operation of the thermal plant; that is, CCGT

and coal plant supply a proportion of demand when the CCS plant is operating above 80% capacity. This is in

contrast to a strict merit order assumption where the conventional plant would only operate once the CCS plant

was at maximum capacity. Hammond & Stapleton [24] were followed in taking the input to non-thermal

renewable energy systems and imports to be equal to their electrical output. CCS plant electrical efficiency was

taken from [25]. Transmission and distribution losses were assumed to remain the same.
As this study is comparative, the marginal primary energy requirement for electrical power was calculated each

time step. The amount of electricity generated from renewable and nuclear sources changes less than the amount

generated from fossil fuels when demand is affected by the heating systems and it is these changes which are

taken into account. This has the interesting effect that the average marginal primary energy and exergy

requirements for electricity are higher in the ―MR2030‖ scenario than for the CCGT plant even though CO2

emissions rates are lower. Marginal primary exergy requirements were calculated as the product of the primary

energy requirements and the factors in Table 1.

Table 2: Electricity generation mix (output mix). Data calculated from [21] & [25]

Transition Pathways, Average electrical

Generating plant
“Market Rules” 2030 generation efficiency
type
mix

Gas CCGT 9.3% 43%

Coal CCS 16.5% 27%

Gas with CCS 16.4% 37%

Nuclear 17.1% 34%

Wind 20.2% -

Other renewables 7.3% -

Imports 4.9% 34%

Other (including 64%

14.3%
CHP & coal)

Losses from natural gas pumping, leakage and other operator activities amounted to 2% of the total natural gas

energy input in 2010 with a resultant energy requirement for natural gas supplied to mCHP units of 1.02. Losses

upstream from input to the national transmissions system (e.g. extraction and processing) are not included in this

study.

3.2 mCHP models

Performance metrics were gathered for 11 different mCHP units. A ―grey box‖ approach was taken, based on

that developed by IEA ECBCS Annex 42 [26] [27], using two lumped thermal capacities. Parameters for two of

the units (one SE-mCHP unit and one ICE-mCHP unit) were taken directly from Annex 42 calibration work,

while parameters for two other ICE-mCHP units, one SE-mCHP and two PEMFC-mCHP units were calculated
based upon experimental data provided in the same report [28]. Steady-state performance was generally directly

available (in this study, all values are converted to be relative to the higher heating value, HHV, i.e. the gross

calorific value of fuel). However, in some cases, the relevant thermal inertias and heat transfer coefficients had

to be inferred from the warm up and cool down characteristics of the unit and (in the case of the SE-mCHP)

from the part-load electrical efficiency. It was not possible to model the variation in electrical efficiency of the

ICE-mCHP units in this way and so an additional performance map was used. Similarly, in the case of fuel cell

mCHP units, a full Annex 42 type model was not developed; instead, a similar approach was taken to model the

thermal inertia of the units but the electrical and thermal efficiencies were interpolated between observed values

at different fuel flows. This was done to enable the consistent use of reported performance data for other units

(two PEMFC-mCHP units, one SE-mCHP unit and one SOFC-mCHP unit [29], [30], [31], [32], [33]).

Although these simplifications reduce confidence in the model’s ability to predict the exact performance of a

unit in specific circumstances, it is reasonable to assume that they will indicate the unit’s performance under

similar circumstances.

The units (apart from the SOFC-mCHP unit) are assumed to operate with a heat-led control methodology. It was

assumed that there is no restriction to the export of electrical power and that this electricity displaces other

demand locally (i.e. without distribution losses). This is not the case in locations where electrical export is not

allowed (e.g. Japan) and may not be the case in locations where tariff structures disincentivise export (e.g. UK).

The effect of this assumption is highlighted in the discussion section. In all results, net electrical power (i.e. net

of inverter losses) is used. The SOFC-mCHP unit is constrained to a very low maximum rate of change of

output. Therefore, rather than operating as a heat-led unit, four operating regimes were designed with different

prioritisations between maximising electrical efficiency and minimising excess heat generation. These four

regimes are treated as separate units. Two are constrained to always operate at the fuel flow corresponding to

peak electrical efficiency or above while the other two can be turned down. One unit of each of these pairs is

also oversized such that the fuel flow corresponding to peak heat demand also corresponds to maximum

efficiency. In each case, excess heat is rejected and only the heat used is included in efficiency calculations.

The heat flows within each mCHP unit were calculated using a one second time step (to ensure modelling

stability given the low ratio between the thermal inertia and the heat transfer coefficient in some cases). Heat

flows outside the mCHP units were calculated using a one minute time step (see [34]).
3.3 ASHP models

Steady state performance data for 10 ASHP units were gathered from reports published by BRE Ltd [35] and the

Swiss Warmepumpen-Testzentrum (WPZ) [36]. Each performance measurement is specific to a given outside

air (heat source) temperature and a flow (heat sink) temperature. Linear interpolation of the isentropic efficiency

of the heat pump between these measured points was used to determine the COP of each ASHP when either the

outside air temperature or the flow temperature differed from the relevant test temperature. A similar approach

to that taken with the mCHP units was used to model the dynamic response of the ASHP units. However, given

the low time constant of the units and the fact that they use modulating drives, their overall energetic

performance is less sensitive to the value of their thermal inertia and so it was sufficient to estimate this from

their size and the dynamic response of other ASHP units where available.

The ASHP units considered cover a range of nominal performances from the mid-range to the state-of-the-art

currently available. Previous studies [37–39] have reviewed technical development and used exergy analysis at

component level to identify that there are various aspects of ASHP design that can be optimised to improve the

COP. An on-going trend towards higher performance is also observed in the units which are available and it is

likely that the current state-of-the-art performance will be commonly achieved by mid-range units by the end of

the decade.

3.4 Dwelling model

Figure 2: Dwelling heat flows simplified model

Given that the aim of this study is to analyse patterns and trends in the energetic performance of the domestic

heating systems under consideration, it was decided that the most appropriate approach was to model a

relatively high total number of permutations (see Table 3) with a dwelling thermal model sufficiently simplified

to enable this. A lumped capacitance model was therefore selected; this includes a single building fabric thermal

inertia with internal and external heat transfer coefficients, an internal air thermal inertia, internal gains, an

effective solar gains area and an air infiltration rate (Figure 2). These parameters were calibrated against the

thermal characteristics of a detailed model of a building created in ESP-r by Dr. N. Kelly and Dr. J. Hong at

ESRU, University of Strathclyde to assist with a similar study (see [40], [41]). The effective building heat loss

coefficient is 169W/K; lower than the average for the entire UK stock (247W/K) but similar to the average heat

loss from flats (167W/K) [42]. A second dwelling with identical inertia and gains but with reduced external heat

loss and air infiltration (equivalent heat loss coefficient of 118W/K) was used to represent improved building

standards. Test Reference Year climate data was obtained from the PROMETHEUS project [43], based on

UKCP09 climate modelling [44]. Two modelled climates were selected for the study: present day Glasgow

(coastal) and central London in the 2030s. Two heat emitter systems were modelled; these were sized with heat

transfer coefficients such that flow temperatures of 45°C and 35°C are required to maintain an indoor

temperature of 20°C in the standard building when the outside air temperature is 0°C. A hot water tank of 200

litre capacity was used in each dwelling with insulation limiting its heat loss (to the inside air) to 2W/K.

Domestic hot water demand was based on national averages [45]. Two heating control systems were used: (i)

aiming for 55°C flow temperature using an on-off thermostat with a temperature dead-band of 2°C (although the

actual flow temperature depends upon the conditions) and (ii) heat generation proportional to the difference in

temperature between the inside air temperature and a control set point (20°C), flow temperature dependent upon

the thermal characteristics of the heat emitters system and the ASHP / mCHP unit. These permutations are

summarised in Table 3:
Table 3: Simulation permutations

Grid electrical Heating unit Control Climate Building Other

supply electrical
demand

 3 SE-MCHP  Fixed
 Transition temperature  London,  Current insulation  2230kWh/yr
Pathways,  3 ICE-mCHP 2030s
modelled  Improved insulation  4460kWh/yr
(on-off
“Market thermostat
Rules”  4 PEMFC- and  8920kWh/yr
scenario for MCHP
controlled)  Glasgow,
UK (2030)  Variable 1970’s  Standard heat emitters
 1 SOFC-mCHP temperature
 CCGT (in four  Improved heat emitters
based on
configurations)
proportional
 10 ASHPs controller

The exergy transfers associated with the heat flows to the heat emitters and to the air inside the dwelling were

calculated using equation 2.4 and the flow, return, inside air and outside air temperatures.

In order to consider the efficiencies of the whole system (from primary energy inputs to final use of energy) in

terms of the ratio between thermal and electrical demand, three different annual electrical demands (i.e.

appliances and lighting, excluding heating) were assumed based on the UK annual average (4460kWh/yr [1]),

twice the UK annual average and then half of it.

4 RESULTS & DISCUSSION

4.1 mCHP unit performances

Figure 3: Variation in mCHP electrical efficiency with thermal efficiency.

The calculated mCHP electrical and thermal efficiencies are plotted in Figure 3. A large field trial of PEMFC

units in Japan [7] reported higher performances that those modelled (average efficiencies of 28.8% electrical

and 37.8% thermal). A lack of available detail regarding the units used and the conditions in the trial prevents

direct comparison but the results are clearly significant and are approximated as an ellipse on the graph for

completeness. The ICE-mCHP units tend to show quite consistent electrical efficiency whereas the smaller SE-

mCHP units suffer greater electrical performance degradation with the higher cycling associated with less

consistent heat demands [46]. One of the PEMFC units (―mCHP7‖) does not show the same range of thermal

efficiency variation. This is because the unit is undersized for most domestic heating duties and so operates

more consistently; however, the use of an auxiliary heater is not taken into account and would significantly

reduce the system electrical efficiency. Although intended to be indicative only, the modelling supports the

suggestion that the SOFC-mCHP unit should be run continuously, rejecting heat as necessary. The slow ramp

rate of the unit makes it difficult to fully utilise the 20% thermal recovery that is theoretically possible.
The exergy efficiencies of the mCHP units are plotted against the power fraction they supply, Fp, in figure 4,

where:

4.1

where P is the electrical power generated by the mCHP unit.

Figure 4: Variation in mCHP exergy efficiency with power fraction.

It can be seen that although the total energy efficiency of most of the mCHP units being considered is high, the

relatively low exergy value (c. 8%) of heat delivered at this temperature means that the exergy efficiency of

units is low unless their power-heat output ratio is high. The proximity of the 100% energy efficiency plot at

lower power fractions constrains any significant improvements in exergy efficiency such that it must involve an

increased electrical efficiency. Despite achieving a lower total energy efficiency, the exergy efficiency of the

SOFC-mCHP units is therefore much higher.

4.2 ASHP unit performance

The annual average COP of each ASHP under the various conditions is plotted in Figure 5.

Figure 5: Variation in ASHP annual average COPs with average (heat-demand-weighted) emitter

temperatures.

There are clear groupings of results with each of the 16 permuations of building construction, emitter size,

climate and control system resulting in a different mean emitter temperature. Within each of these permuations

there is a range of about 35% between the performance of the most efficient ASHP (usually ―ASHP6‖) and the

least efficient ASHP (usually ―ASHP2‖). As expected, for each ASHP unit, higher mean emitter temperatures

are associated with a lower average COP. However, there are separate trends for the two types of building

(illustrated for ―ASHP1‖); the average COP in the well insulated building is typically similar to that in the less-

well insulated building despite a lower average flow temperature. This is probably due to the better insulated

house exhibiting a shorter heating season; far less heat is required but it is required when the outside

temperature is colder. Notably, using a control methodology with constant flow temperature can cause an

increase in power consumption of more than 30%, a similar effect to that observed by Kelly & Cockroft [47]
who calculated a comparable average COP (c. 2.7) using a detailed model without ―weather-compensation‖

control.

The exergy delivered as heat to the heat emitters from each ASHP is plotted against the electrical power

(exergy) consumed by each ASHP in Figure 6. In contrast to the energy efficiency, the exergy efficiency of each

ASHP (i.e. the gradient of a trend line for that unit) does not vary greatly with ambient air temperature (typical

variation of +/- 3% across working range). For example, the exergy efficiency of ―ASHP1‖ averages about 27%

for the conditions modelled.

Figure 6: Exergy delivered against power consumed for ASHPs under varying conditions

4.3 Overall system

The variation of the energy and exergy efficiencies of the whole energy system (i.e. primary energy inputs

through to final energy use) with the power fraction can be considered by reference to Figure 7 (energy

efficiencies plotted for TD = 20ºC). The power fraction (x-axis) referred to is the fraction of electrical demand

(appliances and lighting, Z), compared to total energy demand in the dwelling except in the case of the SOFC-

mCHP units which generate more electricity than is demanded. In these cases, it refers to the ratio of electricity

generated (P) compared to electricity generated and heat used (equation 4.1).
The data associated with the mCHP units is similar to that in figure 4 but includes losses in the distribution of

natural gas and the exergy loss as heat is transferred from the heat emitter system to the air in the room (at lower

temperature). The SOFC-mCHP unit achieves the highest exergy efficiency but is limited to high power

fractions. The dashed line shows the effect on the SOFC-mCHP whole system exergy efficiency of adding an

auxiliary boiler to reduce the power fraction; it is likely to remain higher than the alternatives until the power

fraction is below about 65%.

Figure 7: Exergy and energy efficiency of ASHP and mCHP based whole systems against power fraction

The exergy efficiency of the ASHP systems is highly dependent on the efficiency of grid electrical generation.

This is shown clearly on the graph by the two distinct groupings of ASHP system performances relating to the

two electrical supply scenarios considered. For comparison, plots are provided for a 90% efficient condensing
gas boiler with each of electrical supply alternatives; if electricity is supplied by the ―MR2030‖ mix, the system

performances are generally similar but with electricity supplied by the CCGT, the ASHP systems show higher

efficiency in almost all cases.

In the cases with power fractions between about 35% and 60% (i.e. heat to power ratios of 2:1 to 0.7:1), both

types of unit offer comparable performance (for the ―MR2030‖ mix), with operating conditions and constraints

determining the highest exergy efficiency available. At the higher end of this range of power fractions, the

PEMFC-mCHP units from Tokyo Gas field trial would represent the highest performance but they are not

plotted on this graph as discussed earlier.

At lower power fractions (i.e. heat to power ratios above 2:1), most of the ASHP systems have the potential to

achieve greater than 100% energy efficiency and it is impossible for a conventional mCHP system to exceed

this. For the higher performing systems with electricity supplied by CCGT, 100% primary energy efficiency can

occur with a power fraction as high as 40% (i.e. heat to power ratio of 1.5:1). Conversely, if unrestrained

electrical export is allowed, the high electrical efficiency achieved by using the SOFC-mCHP unit means that

energy is saved even if the thermal energy is unused. Under these conditions, the assumed reduction in

electrical distribution losses achieved by the SOFC-mCHP unit is more significant than the heat recovered or the

electricity used by the more efficient ASHPs. The SOFC-mCHP system efficiency would therefore be higher

than that of an ASHP system regardless of the average COP achieved. For high performing systems, the overall

energy performance is largely dictated by the relative grid electrical efficiency and restrictions on power export,

not on the provision of heat.

The largest exergy losses are due to inefficient generation of electricity and to the low exergy value of the heat

flow. Because of the low exergy content of the thermal energy, relatively small changes in the exergy efficiency

of the systems have the potential for a larger impact on their energy performance; this sensitivity partly explains

the wide range of system performances reported in the literature. Although outside the scope of this analysis, it

should also be noted that savings in the electrical demand of appliances will have a proportionally larger effect

on the exergy and energy required earlier in the system than savings at other points. This analysis considers the

energetic and exergetic performance of the units; for a holistic assessment of their relative merits, consideration

should also be given to other factors such as their whole-life environmental impacts and the economic aspects of

their adoption.
Technical constraints such as local grid infrastructure may limit power export. It is feasible that a clustering

approach (i.e. several ASHPs with some fuel cell mCHP units) could be employed to maximise overall

efficiency in situations involving multiple low power to heat ratio demands. It has been suggested that this

would also minimise daily variation in grid power demand [48] and it is likely that this effect would also be

seasonal.

5 CONCLUDING REMARKS

Energy and exergy analysis has been conducted for a range of mCHP and ASHP devices, modelled as discrete

units but within the context of the whole energy system. The current mid-range ASHP and SE-mCHP units have

broadly comparable performance to a condensing boiler with grid supplied electricity for normal power to heat

demand ratios. However, appropriately installed, state-of-art ASHP units and SOFC-mCHP units have the

potential to achieve 40 – 50% primary energy savings.

Efficient SOFC-mCHP units can achieve higher efficiency than new CCGT units (once grid losses are

accounted for). Although energy savings are clearly possible, economic and environmental considerations must

be fully taken into account to provide a holistic appraisal [49].

The relative performance and potential of the systems is dominated by the electrical characteristics: the grid

electrical generation efficiency, the power-to-heat demand ratio and the availability of electrical export. For total

power-to-heat demands below 1:1.5, ASHPs have greater improvement potential as their energy efficiency is

not constrained. At higher power-to-heat ratios, mCHP units do offer the potential for higher overall efficiency

and this generally occurs irrespective of whether or not the thermal energy is used effectively. In practice, it is

likely that a combination of the two systems would provide the best performance.

6 ACKNOWLEDGEMENTS

This work was principally supported by the Engineering and Physical Sciences Research Council (EPSRC)

[Grant number EP/G031681/1] as part of the SUPERGEN Highly Distributed Energy Futures (HiDEF)

consortium. The authors gratefully acknowledge the interchange made possible under this programme. The

assistance of Nick Kelly and Jun Hong of ESRU, University of Strathclyde and of Ian Richardson and Murray

Thompson of CREST, Loughborough University in calibrating the building models is gratefully acknowledged.

The 'market rules' transition pathway was developed by a consortium of nine university partners funded via the
strategic partnership between e.on UK (the electricity generator) and the EPSRC to study the role of electricity

within the context of 'Transition Pathways to a Low Carbon Economy' [under Grant EP/F022832/1]. This is co-

led by the second author (GPH), as is the renewed consortium project entitled 'Realising Transition Pathways'

with funding provided from 2012 by the EPSRC alone [under Grant EP/K005316/1]. Finally, the authors are

grateful to two anonymous reviewers for their valuable comments and suggestions. The authors' names are listed

alphabetically.
7 ABBREVIATIONS AND NOMENCLATURE

ASHP Air Source Heat Pump

COP Coefficient of Performance

ICE Internal Combustion Engine

mCHP Micro Combined Heat and Power

PEMFC Proton Exchange Membrane Fuel Cell

SE Stirling Engine

SOFC Solid Oxide Fuel Cell

A Total non-renewable primary energy input

C Coefficient of Performance

E Exergy

FP Fraction of energy at end use delivered as electrical power

H Enthalpy of combustion (higher heating value)

P Power output from mCHP unit

Q Heat flow

TD Room temperature

TE Heat emitter temperature

T0 Ambient temperature

W Work input to heat pump

Z End use electrical demand

 Energy efficiency

 Ratio of chemical exergy to enthalpy of combustion

 Rational (Exergy) efficiency

8 FIGURES

Figure 1: Representative energy flows

Figure 2: Dwelling heat flows simplified model

Figure 3: Variation in mCHP electrical efficiency with thermal efficiency.

Figure 4: Variation in mCHP exergy efficiency with power fraction.

Figure 5: Variation in ASHP annual average COPs with average (heat-demand-weighted) emitter temperatures.

Figure 6: Exergy delivered against power consumed for ASHPs under varying conditions

Figure 7: Exergy and energy efficiency of ASHP and mCHP based whole systems against power fraction

9 REFERENCES

[1] J. Palmer and I. Cooper, Great Britain’s housing energy fact file. London: DECC, 2011.

[2] DECC, The UK Low Carbon Transition Plan. London: TSO, 2009.

[3] B. Boardman, Home Truths: A low-carbon strategy to reduce UK housing emissions by 80% by 2050.

Oxford: Environmental Change Institute, University of Oxford, 2007.

[4] P. F. G. Banfill and A. D. Peacock, ―Energy-efficient new housing - the UK reaches for sustainability,‖

Building Research & Information, vol. 35, no. 4, pp. 426-436, Jul. 2007.
[5] J. Cockroft and N. J. Kelly, ―A comparative assessment of future heat and power sources for the UK

domestic sector,‖ Energy Conversion and Management, vol. 47, no. 15–16, pp. 2349-2360, Sep. 2006.

[6] Carbon Trust, ―Micro-CHP Accelerator Interim report,‖ The Carbon Trust, London, 2007.

[7] Y. Yamada and K. Nishizaki, ―Next generation model of the world’s first residential PEMFC

cogeneration system goes on sale,‖ in Proceedings of the 24th World Gas Conference, Buenos Aires, 5 -

9 October 2009, 2009, pp. 1-7.

[8] M. Miara, ―Two large field-tests of new heat pumps in Germany,‖ in 9th IEA Heat Pump Conference,

Zurich, 20 - 22 May 2008, 2008.

[9] Energy Saving Trust, Getting warmer: a field trial of heat pumps. London: Energy Saving Trust, 2010.

[10] I. Staffell, ―Fuel Cells for domestic heat and power: Are they worth it?,‖ University of Birmingham,

Thesis (PhD), 2009.

[11] A. Hawkes, I. Staffell, D. J. L. Brett, and N. P. Brandon, ―Fuel cells for micro-combined heat and power

generation,‖ Energy & Environmental Science, vol. 2, no. 7, p. 729, 2009.

[12] BS EN14511-4:2007, Air conditioners, liquid chilling packages and heat pumps with electrically driven

compressors for space heating and cooling - Part 4: Requirements. London: BSI, 2007.

[13] BS EN255-3:1997, Air conditioners , liquid chilling packages and heat pumps with electrically driven

compressors - Heating mode Part 3 . Testing and requirements for sanitary hot water units. London:

BSI, 1997.

[14] T. Berntsson, ―Heat sources — technology , economy and environment,‖ International Journal of

Refrigeration, vol. 25, pp. 428-438, 2002.

[15] T. J. Kotas, Y. R. Mayhew, and R. C. Raichura, ―Nomenclature for exergy analysis,‖ Proceedings of the

Institution of Mechanical Engineers, Part A: Journal of Power and Energy, vol. 209, no. 41, pp. 275-

280, Jun. 1995.

[16] T. J. Kotas, ―Exergy concepts for thermal plant. First of two papers on exergy techniques in thermal

plant analysis,‖ International Journal of Heat and Fluid Flow, vol. 2, no. 3, pp. 105-114, Sep. 1980.

[17] S. R. Allen and G. P. Hammond, ―Thermodynamic and carbon analyses of micro-generators for UK

households,‖ Energy, vol. 35, no. 5, pp. 2223-2234, May 2010.

[18] B. Thomas, ―Benchmark testing of Micro-CHP units,‖ Applied Thermal Engineering, vol. 28, no. 16,

pp. 2049-2054, Nov. 2008.

[19] Siemens AG, ―SGT5-8000H datasheet,‖ Siemens Energy Inc, Erlangen, 2010.

[20] DECC, Digest of United Kingdom Energy Statistics 2011. London: TSO, 2011.

[21] T. J. Foxon, G. P. Hammond, and P. J. G. Pearson, ―Developing transition pathways for a low carbon

electricity system in the UK,‖ in Technological Forecasting and Social Change, 2008, vol. 77, no. 8, pp.

1203-1213.

[22] Elexon, ―Balancing Mechanism Reports,‖ 2011. [Online]. Available: http://www.bmreports.com/.

[Accessed: 10-Jun-2012].

[23] M. Barnacle, E. Robertson, S. Galloway, J. Barton, and G. Ault, ―Modelling generation and

infrastructure requirements for transition pathways,‖ Energy Policy, pp. 1-16, May 2012.

[24] G. P. Hammond and A. J. Stapleton, ―Exergy analysis of the United Kingdom energy system,‖

Proceedings of the Institution of Mechanical Engineers, Part A: Journal of Power and Energy, vol. 215,

no. 2, pp. 141-162, Jan. 2001.

[25] N. A. Odeh and T. T. Cockerill, ―Life cycle GHG assessment of fossil fuel power plants with carbon

capture and storage,‖ Energy Policy, vol. 36, no. 1, pp. 367-380, Jan. 2008.

[26] N. J. Kelly, J. A. Clarke, A. Ferguson, and G. M. Burt, ―Developing and testing a generic micro-

combined heat and power model for simulations of dwellings and highly distributed power systems,‖

Proceedings of the Institution of Mechanical Engineers, Part A: Journal of Power and Energy, vol. 222,

no. 7, pp. 685-695, Nov. 2008.

[27] I. Beausoleil-Morrison, A. Weber, F. Marechal, B. Griffith, A. Ferguson, and N. Kelly, Specifications

for Modelling Fuel Cell and Combustion-Based Residential Cogeneration Devices within Whole-

Building Simulation Programs. A Report of Subtask B of FC+COGEN-SIM The Simulation of Building-

Integrated Fuel Cell and Other Cogeneration Systems. A. Annex 42 of the International Energy Agency

Energy Conservation in Buildings and Community Systems Programme, 2007, p. 111.

[28] I. Beausoleil-Morrison, Experimental Investigation of Residential Cogeneration Devices and

Calibration of Annex 42 Models. A Report of Subtask B of FC+COGEN-SIM The Simulation of

Building-Integrated Fuel Cell and Other Cogeneration Sys- tems. Annex 42 of the International Energy.

Annex 42 of the International Energy Agency Energy Conservation in Buildings and Community

Systems Programme, 2007, p. 255.

[29] R. Payne, J. Love, and M. Kah, ―Generating Electricity at 60% Electrical Efficiency from 1 - 2 kWe

SOFC Products,‖ ECS Transactions, vol. 25, no. 2, pp. 231-239, 2009.

[30] K. Sommer, ―Best Integration of a Fuel Cell – based Micro- CHP for Residential Buildings A

contribution to IEA / ECBCS Annex54,‖ in 4th Expert Meeting for Annex 54 Seconda Università di

Napoli / Università di Sannio Naples, Italy September 20-23, 2011, 2011.

[31] E. Thorsteinson, B. Strathearn, G. Mackenzie, and G. Amow, ―Performance Testing of a 1kWe PEM

Fuel Cell Cogeneration system,‖ in Proceedings of 2nd International Conference in Microgeneration

and Related Technologies in Buildings: Microgen 2, Glasgow, 4 - 6th April 2011, 2011.

[32] G. Magri, C. Di Perna, and G. Serenelli, ―Analysis of electric and thermal seasonal performances of a

residential microCHP unit,‖ Applied Thermal Engineering, vol. 36, pp. 193-201, Apr. 2012.

[33] V. Dorer and A. Weber, ―Energy and CO2 emissions performance assessment of residential micro-

cogeneration systems with dynamic whole-building simulation programs,‖ Energy Conversion and

Management, vol. 50, no. 3, pp. 648-657, Mar. 2009.

[34] A. Hawkes and M. Leach, ―Impacts of temporal precision in optimisation modelling of micro-Combined

Heat and Power,‖ Energy, vol. 30, no. 10, pp. 1759-1779, Jul. 2005.
[35] BRE, Ecodan PUHZ-W90VHA air to water heat pump tests. Garston: Building Research Establishment

Ltd, 2007.

[36] Warmepumpen-Testzentrum, Test results of air to water heat pumps based on EN 14511. Buchs,

Switzerland: Institut fur Energiesysteme, Interstaatliche Hochschule für Technik, 2010.

[37] A. Hepbasli and O. Akdemir, ―Energy and exergy analysis of a ground source (geothermal) heat pump

system,‖ Energy Conversion and Management, vol. 45, no. 5, pp. 737-753, Mar. 2004.

[38] Y. Bi, X. Wang, Y. Liu, H. Zhang, and L. Chen, ―Comprehensive exergy analysis of a ground-source

heat pump system for both building heating and cooling modes,‖ Applied Energy, vol. 86, no. 12, pp.

2560-2565, Dec. 2009.

[39] K. J. Chua, S. K. Chou, and W. M. Yang, ―Advances in heat pump systems: A review,‖ Applied Energy,

vol. 87, no. 12, pp. 3611-3624, Dec. 2010.

[40] J. Hong, C. Johnstone, N. J. Kelly, M. C. McManus, and C. N. Jardine, ―Identifying characteristic

building types for use in the modelling of highly distributed power systems performance.‖ 2010.

[41] I. Richardson, M. J. Thomson, and D. G. Infield, ―A high-resolution domestic building occupancy

model for energy demand simulations,‖ Energy & Buildings, vol. 40, no. 8, pp. 1560-1566, 2008.

[42] J. I. Utley and L. D. Shorrock, ―Domestic energy fact file 2008,‖ Building Research Establishment Ltd,

Garston, 2008.

[43] M. Eames, T. Kershaw, and D. Coley, ―On the creation of future probabilistic design weather years

from UKCP09,‖ Building Services Engineering Research and Technology, Oct. 2010.

[44] P. D. Jones, C. G. Kilsby, C. Harpham, V. Glenis, and A. Burton, UK Climate Projections science

report: Projections of future daily climate for the UK from the Weather Generator. University of

Newcastle, UK, 2009.

[45] Energy Saving Trust, Measurement of Domestic Hot Water Consumption in Dwellings. London: Energy

Saving Trust, 2008.

[46] D. Beyer and N. J. Kelly, ―Modelling the behaviour of domestic micro-cogeneration under different

operating regimes and with variable thermal buffering.,‖ in In: Micro-Cogen 2008, 1st International

Conference on Micro-Cogeneration Technologies and Applications, Ottawa, Canada. 29 April - 1 May

2008, 2008.

[47] N. J. Kelly and J. Cockroft, ―Analysis of retrofit air source heat pump performance: Results from

detailed simulations and comparison to field trial data,‖ Energy & Buildings, vol. 43, no. 1, pp. 239-245,

Jan. 2011.

[48] A. Hawkes, D. J. L. Brett, and N. P. Brandon, ―Role of fuel cell based micro-cogeneration in low carbon

heating,‖ Proceedings of the Institution of Mechanical Engineers, Part A: Journal of Power and Energy,

vol. 225, no. 2, pp. 198-207, Mar. 2011.

[49] S. R. Allen, G. P. Hammond, H. A. Harajli, C. I. Jones, M. C. McManus, and A. B. Winnett, ―Integrated

appraisal of micro-generators: methods and applications,‖ Proceedings of the ICE - Energy, vol. 161,

no. 2, pp. 73-86, Jan. 2008.