Assessment of the Costs,
Performance, and
Characteristics of UK Heat
Networks
Final Report
2
© Crown copyright 2015.
URN 15D/022 – Assessment of the Costs, Performance, and Characteristics of Heat UK
Networks
You may re-use this information (not including logos) free of charge in any format or medium,
under the terms of the Open Government Licence.
To view this licence, visit www.nationalarchives.gov.uk/doc/open-government-licence/
or write to the Information Policy Team, The National Archives, Kew, London TW9 4DU,
or email: [email protected].
Any enquiries regarding this publication should be sent to us at [email protected].
3
Contents
Executive summary .................................................................................................................... 6
Introduction ............................................................................................................................... 6
Data collection .......................................................................................................................... 6
Cost data .................................................................................................................................. 7
Typical benchmark assessment ............................................................................................... 7
Heat sales .............................................................................................................................. 11
Carbon dioxide emissions ...................................................................................................... 11
Conclusions ............................................................................................................................ 11
Data collection ....................................................................................................................... 11
Data benchmarking ................................................................................................................ 11
Heat sales .............................................................................................................................. 12
CO
2
emissions ....................................................................................................................... 12
1. Introduction ....................................................................................................................... 14
Background to this study ........................................................................................................ 14
Purpose of this project ............................................................................................................ 14
Scope of this report ................................................................................................................ 15
Confidentiality ......................................................................................................................... 15
The project team .................................................................................................................... 15
2. Data capture methodology ................................................................................................ 16
Outline approach .................................................................................................................... 16
Data collection ........................................................................................................................ 16
Selecting schemes ................................................................................................................. 16
Statistical background ............................................................................................................ 17
Collecting data ....................................................................................................................... 17
Feasibility studies .................................................................................................................. 18
Online survey ......................................................................................................................... 18
Scope of data collection ......................................................................................................... 19
Heat supply plant ................................................................................................................... 19
Internal distribution ................................................................................................................ 20
Heat network schemes evolvement and history ..................................................................... 21
4
Data checking and verification ................................................................................................ 21
Cost data ................................................................................................................................ 22
Identifying gaming and bias .................................................................................................... 22
3. Description of data captured ............................................................................................. 23
Introduction ............................................................................................................................. 23
Description of heat network schemes by size ......................................................................... 23
Description of heat network schemes by delivery type ........................................................... 24
Typical benchmark assessment ............................................................................................. 24
Use of the typical benchmark values……………………………………………………………….24
Performance typical benchmarks…………………………………………………………………..25
Thermal storage……. ............................................................................................................ 26
Capital cost typical benchmarks ............................................................................................. 28
Operational cost benchmarks ................................................................................................. 32
Heat sales .............................................................................................................................. 33
Carbon dioxide emissions ...................................................................................................... 36
Calculation of CO
2
emissions ................................................................................................ 36
CO
2
emissions from heat sources ......................................................................................... 37
CO
2
emissions from DH schemes.......................................................................................... 38
4. Conclusions ...................................................................................................................... 41
Data collection and availability ............................................................................................... 41
Data benchmarking ................................................................................................................ 41
Heat sales .............................................................................................................................. 41
CO
2
emissions ........................................................................................................................ 41
5
This document has been prepared by AECOM Limited for the sole use of DECC and in
accordance with generally accepted consultancy principles, the budget for fees and the terms of
reference agreed between AECOM Limited and the DECC. Any information provided by third
parties and referred to herein has not been checked or verified by AECOM Limited, unless
otherwise expressly stated in the document. No third party may rely upon this document without
the prior and express written agreement of AECOM Limited.
6
Executive summary
Introduction
There are thought to be over 2,000 heat networks and communal heating schemes of various
sizes in the UK serving 200,000 dwellings and 2,000 commercial and public buildings.
1
The
largest heat network schemes are predominantly found in cities and on university campuses.
There are also a large number of smaller schemes in the domestic sector, often linking
communally heated blocks of flats. This extent of heat networks represents around 2% of the
domestic, public sector, and commercial buildings heat demand. Benefits from the increased
use of heat networks could include energy cost and Carbon Dioxide (CO
2
) emissions reductions
for the UK, through allowing the exploitation of lower CO
2
and higher efficiency forms of
generation. These could include the use of CHP, biomass, heat pumps, waste heat and low
grade heat sources.
The key aim of this project is to provide evidence-based knowledge on costs of heat networks,
as well as on their performance and characteristics to support the evaluation of policy options
involving heat networks. This would address the information gap identified by DECC. This work
has focused on the gathering of robust data and evidence to enhance the understanding of
typical mixed residential and commercial heat network schemes in the UK.
This project was led by AECOM, and supported by Sweett Group for the purposes of cost
benchmarking and verification.
Data collection
A list of data requirements was prepared and agreed with DECC to form the basis of the data
collection. This comprised those variables relating to the performance and costs of heat
networks.
The data collection was separated into three complementary data gathering exercises carried
out in parallel with the aim to obtain accurate and completed sets of broken down data on cost,
performance and other characteristics. These consisted of:
An in-depth questionnaire sent to stakeholders with information on a selection of
existing heat network schemes considered to be representative of the main types of
networks in the UK. This was subsequently followed up by further engagement to
look to clarify information provided and fill-in data gaps;
Identification and examination of hypothetical schemes taken from feasibility studies,
deemed representative of current and future heat network developments;
1
The Summary Evidence on District Heating Networks in the UK. DECC, 2013 paper identified 1,765 heat
networks but DECC believe this figure to be nearer to 2,000. The nature of the heat network market means that
there are no robust datasets collating information on all networks.
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/212565/summary_evidence_district_
heating_networks_uk.pdf.
7
An on-line survey addressed to a wide range of stakeholders involved in different
aspects of heat networks with a view to obtain additional and/or complementary
information. The aim of collecting this data was to provide additional information for
benchmarking the existing schemes and feasibility studies, and to further inform
DECC’s evidence base on heat networks’ cost and performance.
A selection of 14 existing heat network schemes considered to be representative of schemes in
the UK were identified by experts from both AECOM and DECC. The selection process included
considerations such as successful implementation and level of performance. This was to ensure
that the data from the schemes would represent a successful and deliverable scheme.
The selection of 14 heat networks limits the extent to which detailed statistical analysis could be
conducted on the received datasets. This was impacted further by only receiving suitable data
from 7 of the 14 schemes.
Cost data
All capital costs are presented as 2013/14 prices. For heat network schemes where capital cost
information is one or more years prior to this, the RICS Building Cost Information Service (BCIS)
General Building Cost Index has been used to provide an uplift.
Typical benchmark assessment
‘Typical’ benchmarks have been created from mean averages of data from across the schemes.
Where data is considered as inconsistent, for example an outlier, or missing, the data from the
scheme for the particular benchmark is not included in the average calculation. The typical
benchmark values can be used to understand the costs and performance of existing heat
networks in the UK. In theory they can be used to construct hypothetical heat networks based
on some simple input parameters, such as annual heat demand from customers.
Typical benchmarks have been produced for performance, capital cost, and operation costs,
and are summarised in Tables 1 – 4 (see following pages 8-10).
The results are shown for bulk supply, and non-bulk supply schemes separately. Bulk heat
networks are defined as those where the main scheme operators deliver heat in bulk to major
distribution points, but who do not have responsibility for final delivery to the end customers.
Non-Bulk schemes are those where the operator or manager of the scheme is responsible for
final delivery to the individual customer (each dwelling or flat). Where relevant, an overall
average is also provided, although this is not suitable for some metrics.
A description of the metrics used in the cost and performance tables is provided in the
Appendix.
8
Table 1: Typical benchmarks describing the performance of heat networks. All heat demand figures are
presented as aggregate customer (at customer connection or bulk supply point) heat demands.
OVERALL
PERFORMANCE
BENCHMARKS
AVERAGE AVERAGE MAX MIN
Number
AVERAGE MAX MIN
Number
% seasonal heat load
factor of network
% na 23% 28% 20% 4 17% 24% 13% 3
Distribution losses
1
% heat
generated
na 6% 11% 3% 4 28% 43% 12% 3
Distribution losses
1
per m
distribution network
kWh / m na 544 850 231 4 767 5942 416 3
Linear heat density MWh
demand per m network
2
MWh / m 7.6 9.7 18.3 5.9 4 4.7 7.8 3.1 3
Thermal store capacity
per MWh heat demand
3
m
3
/MWh
0.015 0.016 0.016 0.016 1 0.014 0.014 1
Length of internal
pipework per dwelling /
flat
4
m 13.3 0 13.3 20.8 9.2 3
Operating temperature -
flow
Degrees C 88 92 95 82 4 84 85 81 3
Operating temperature -
return
Degrees C 62 68 75 55 4 54 60 46 3
Network parasitic
electricity consumption
% heat
demand
na 2.9% 4.0% 1.0% 4 1.9% 2.0% 1.7% 2
1
Difference between heat generation and heat demand served divided by the heat generation
2
Average figure for heat demand served per m of buried pipe in the network
3
Average figures for schemes with thermal storage is provided.
4
Only applicable to non bulk schemes which serve individual dwellings/flats/units
BULK SCHEMES
9
Table 2: Capital cost typical benchmarks normalised to annual heat demand (D – Domestic, ND – Non
Domestic, HIU – Hydraulic Interface Unit). The data presented in BOLD is provided as an aggregate where
available by the heat network scheme through the data collection process. All values presented in italics
are provided by heat network schemes at the disaggregated level where available. Therefore the aggregate,
and sum of disaggregated values are not always equivalent.
OVERALL
CAPITAL COSTS
AVERAGE AVERAGE MAX MIN
Number
AVERAGE MAX MIN
Number
Heat Connections £ / MWh na £25 £36 £8 4 £624 £1,179 £129 3
Cost substations ND and D
blocks
£ / MWh £16 £12 £16 £8 2 £19 £39 £5 3
Cost heat meters ND and D
blocks
£ / MWh £4 £3.7 £3.7 £3.7 1 £4.4 £8.0 £0.8 2
Cost HIUs Dwellings
1
£ / MWh £253 0 £253 £461 £46 3
Cost heat meters
Dwellings
1
£ / MWh £170 0 £170 £232 £107 2
Cost internal pipework
connection to HIUs
1
£ / MWh £492 0 £492 £782 £203 2
Cost for connections
prelims
2
£ / MWh
Overhead and profit
connections
2
£ / MWh
SUM OF DISAGGREGATED
CONNECTION COSTS
£ / MWh na £16 £20 £12 0 £939 £1,523 £362 0
Heat Network £ / MWh £150 £150 £239 £80 4 £150 £168 £132 2
Mechanical capital cost –
total
£ / MWh £70 £62 £102 £42 4 £80 £92 £72 3
Civil capital cost – total £ / MWh £74 £79 £136 £25 4 £64 £76 £52 2
Preliminary capital cost –
total
3
£ / MWh £8 £10 £13 £7 2 £3 £3 £3 1
Overhead and profit –
total
£ / MWh £10 £10 £13 £7 2 0
SUM OF DISAGGREGATED
HEAT NETWORK COSTS
£ / MWh £161 £160 £265 £80 £148 £171 £128
Ancillary plant associated
with network (eg
pumping, treatment, etc)
£ / MWh na £68 £68 £68 1 £137 £137 £137 1
Thermal store £ / MWh £14 £17 £17 £17 1 £12 £12 £12 1
1
Only applicable to non bulk schemes which serve individual dwellings/flats/units
2
None reported - 2 schemes indicated these costs are included but can't be disaggregated
3
Prelim costs are believed to be mainly included within mechanical, civil and/or total costs
BULK SCHEME
NON BULK SCHEMES
no data
no data
10
Table 3: Capital cost typical benchmarks normalised to non MWh metrics
Table 4: Operation cost typical benchmarks normalised to annual heat demand and capacity.
OVERALL
OPERATION COSTS
AVERAGE AVERAGE MAX MIN
Number
AVERAGE MAX MIN
Number
Heat network
maintenance cost
£ / MWh £0.6 £0.4 £0.4 £0.4 1 £0.6 £0.9 £0.3 3
Heat network management
cost
£ / MWh
Substation maintenance
cost
£ / MWh
HIUs maintenance cost £ / MWh £9 0 £9 £16 £2 3
HIUs maintenance cost £/MW £820 0 £820 £1,039 £658 3
Heat meter maintenance
cost
£ / MWh £3.4 0 £3.4 £9.0 £0.1 3
Avg annual staff cost for
metering, billing and
revenue collection
£ / MWh £11.1 £2.5 £4.6 £0.4 2 £16.9 £34.8 £0.1 3
Annual business rates £ / MWh £6 £7 £8 £5 2 £6 £8 £2 3
BULK SCHEMES
no data
no data
OVERALL
CAPITAL COSTS
AVERAGE AVERAGE MAX MIN
Number
AVERAGE MAX MIN
Number
Network capital costs per
length (main network-
buried)
£/m £984 £1,242 £1,472 £874 4 £468 £514 £422 2
Network capital costs per
length (internal pipe)
£/m £169 0 £169 £244 £94 2
Substations cost per kW
capacity
£/kW £32 £28 £40 £15 2 £35 £53 £16 3
Domestic HIUs cost per
dwelling
£/dwelling £1,075 0 £1,075 £1,326 £738 3
Heat meter cost per
building (ND and bulk)
£/building £2,878 £1,949 £1,949 £1,949 1 £3,343 £6,136 £551 2
Heat meter cost per
dwelling
£/dwelling £579 0 £579 £668 £491 2
Thermal store cost per m3 £/m3 £962 £1,080 £1,080 £1,080 1 £843 £843 £843 1
BULK SCHEMES
11
Heat sales
Heat prices for heat network schemes are unregulated and calculated using a number of
different methods. The price of heat is often determined by the heat source, which can result in
a wide range of costs and structures. Information on heat sales prices was collected as part of
the data capture exercise and compared with counterfactual prices for the domestic sector
which include gas supply, boiler maintenance and replacement.
Carbon dioxide emissions
The CO
2
savings provided by a heat network scheme can be calculated by comparing the
scheme with a counterfactual heating source. The main counterfactual technology appropriate
to heat network schemes is the condensing gas boiler which is predominant in urban areas.
This report assesses potential CO
2
savings from schemes under a range of assumptions
describing the current and future electricity grid emissions.
Conclusions
Data collection
Data has been collected from a representative sample of 7 district heat network schemes and
used to assess the costs and performance of heat networks. Additional data from feasibility
studies and an online survey has helped validate the data from existing schemes.
The data collection exercise has highlighted the difficulty in obtaining accurate and consistent
data across existing heat network schemes in the UK, and the need to develop a framework in
which the schemes can be better monitored and understood. The fragmented and nascent
nature of the heat network industry in the UK means that data is either not collated, or not easily
accessible, but could be used if available to help understand and support further heat network
development.
This study has also identified the complexity of heat network schemes, with many technical and
financial approaches. Understanding the boundary of the heat network schemes is important -
whilst some heat network schemes are relatively simple with all heat supply, distribution and
customer connections financed and operated by the same organisation, the majority are more
complex with different organisations responsible for different elements of the scheme.
Historic investment and payback on infrastructure also need to be considered when assessing
heat networks. A large number of existing schemes have developed over a number of years,
and some operators have been able to adopt existing heat supply and distribution infrastructure
for little or no charge, effectively a grant to the scheme. Some other more recent schemes have
received large capital grants for infrastructure investment. This means that not only is historic
cost information difficult to obtain, more importantly it does not feature in the operator’s
economic model. Whilst this may enable existing schemes to operate in a financially viable
manner, it is not a replicable model for new schemes where all investment costs will need to be
included.
Data benchmarking
The data collected in this study has been used to produce a set of benchmarks or typical
values. Key conclusions are as follows:
Performance:
Distribution losses range widely with averages of 6% and 28% for each type of
network. The losses for the bulk schemes are in line with commonly used
12
assumptions with losses of around 10% or less although these are representative of
the more urban schemes examined in this report, and the losses could be higher in
less dense schemes. The losses for the non-bulk schemes are significant with one
scheme at 43%. Evidence suggests that high losses can be experienced through
internal distribution pipework within buildings, and it will be important that the
specification and design of internal pipework and operating parameters are
controlled to reduce losses on schemes.
Capital costs:
Heat network buried pipe costs are typically around £150 / MWh annual, whilst the
connection costs range significantly from £25 / MWh for bulk schemes to £624 /
MWh for non-bulk schemes. The key area of sensitivity is therefore around the heat
connections and configuration. In particular, individual dwelling connections in the
form of HIUs and heat meters appear to dominate the capital cost of domestic
schemes with individual connections.
Buried heat network pipework costs average around £1000 / m over a heat network,
although this ranges from £422 / m to £1472 / m depending on the size and nature
of the schemes.
Operation costs:
Operation costs associated with the heat network appear to be low, and not
considered significant by organisations. However there will be higher operation
costs associated with other elements of the heat network scheme, such as heat
sources and energy centre buildings.
The highest operation costs appear to be associated with HIU and heat meter
maintenance, and meter reading / billing activities.
Heat sales
This report has assessed the heat sales prices used by the 7 schemes, and compared these
with counterfactual prices for the domestic sector. The analysis suggests that in low heat
demand dwellings, the counterfactual prices can be relatively high once boiler maintenance and
replacement is included, and current heat sales prices from heat networks may be under-
valuing the heat.
This undervaluation may allow economic operation of some current heat networks which have
benefited from historic investment or grant funding, but higher prices may be required for future
networks which incur all capital expenditure. The analysis suggests that higher prices may be
obtained, whilst still offering consumers a discount over counterfactual heating costs.
CO
2
emissions
The CO
2
emissions from heat distributed by heat networks depend on the type of heat source
and the fuel used. Larger heat networks predominantly source heat from gas fired CHP at
present although other sources such as energy from waste are also used.
The analysis demonstrates the sensitivity of the CO
2
calculations to assumptions for grid
electricity emissions, and the need for improved advice and guidance for these calculations. In
particular, the valuation of grid CO
2
at an average or marginal generation factor may not be
suitable. It is recommended that a framework is developed for CO
2
calculations. This should
13
take into account the future electricity grid emissions projections, the power generation
technologies which will be displaced under different load conditions by decentralised
generation, and the operation regime for decentralised technologies. Without this, the lifecycle
CO
2
emissions of heat network schemes cannot be suitably and consistently assessed, making
investment in heat networks and heat supply technologies higher risk. DECC is addressing this
by investigating CHP operation in further detail.
2
2
https://www.gov.uk/government/publications/bespoke-natural-gas-chp-analysis
Introduction
14
1. Introduction
Background to this study
There are thought to be over 2,000 heat networks and communal heating schemes of various
sizes in the UK serving 200,000 dwellings and 2,000 commercial and public buildings.
3
The
largest heat network schemes are predominantly found in cities and on university campuses.
There are also a large number of smaller schemes in the domestic sector, often linking
communally heated blocks of flats. This extent of heat networks represents around 2% of the
domestic, public sector, and commercial buildings heat demand. Benefits from the increased
use of heat networks could include energy cost and Carbon Dioxide (CO
2
) emissions reductions
for the UK, through allowing the exploitation of lower CO
2
and higher efficiency forms of energy
generation including the use of CHP, biomass, heat pumps, waste heat and low grade heat
sources.
DECC’s response to consultation in December 2013 suggested that heat networks will be part
of a future RHI policy review. In order to inform how this may be developed, DECC identified
the need for additional evidence to better understand the costs and performance of heat
networks before policy options can be assessed.
Purpose of this project
The key aim of this project is to provide evidence-based knowledge on costs of heat networks,
as well as on their performance and characteristics to support the evaluation of policy options
involving heat networks. This would address the information gap identified by DECC.
This work has focused on the gathering of robust data and evidence to enhance the
understanding of typical mixed residential and commercial heat networks in the UK. To this end,
this work has collected data on characteristics, performance and cost of several major existing
heat networks in the UK at a high disaggregated level of detail to allow for an analysis and
identification of the main impact variables affecting the performance and financial viability of
heat networks, as well as to evaluate sensitivities. In addition, and to aid the understanding of
and eliminate any potential bias and gaming in the data obtained, several hypothetical schemes
were also studied. This part of the work focused on the gathering and close analysis of a
sample of published feasibility studies produced to develop heat networks’ schemes. Finally, to
enhance the robustness of the findings, additional information was collected via an online
survey to stakeholders.
3
The Summary Evidence on District Heating Networks in the UK. DECC, 2013 paper identified 1,765 heat
networks but DECC believe this figure to be nearer to 2,000. The nature of the heat network market means that
there are no robust datasets collating information on all networks.
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/212565/summary_evidence_district_
heating_networks_uk.pdf.
15
Scope of this report
The primary output from this study is a dataset of infrastructure cost and performance
characteristics for a range of heat networks. This report provides an overview of the data
collection undertaken and details of the cost and performance data received.
The analysis and outputs presented in this report examine only the heat network infrastructure,
which takes heat from a thermal source, and delivers it to customers. The analysis does not
include the thermal sources and associated buildings and infrastructure, neither does it include
the customer’s own heating systems.
Additional data has been collected as part of this work relating to complete heat network
schemes including the heat source, and operation costs. This additional data is not presented
in this study, but is referred to in the methodology sections. This data has been collected for
future analysis by DECC.
Confidentiality
The data collected by AECOM for this work includes a large amount of commercially sensitive
information. All descriptions and results in this report are anonymous and individual schemes
are not identifiable. For this reason, the schemes are characterised by ranges, and all financial
information is normalised based on annual heat sales and other metrics.
The project team
This project was led by AECOM, and supported by Sweett Group for the purposes of cost
benchmarking and verification.
Data capture methodology
16
2. Data capture methodology
Outline approach
A list of data requirements was prepared and agreed with DECC to form the basis of the data
collection. This comprised those variables relating to the performance and costs of heat
networks.
The data collection was separated into three complementary data gathering exercises carried
out in parallel with the aim to obtain accurate and completed sets of broken down data on cost,
performance and other characteristics. These consisted of:
An in-depth questionnaire sent to stakeholders with information on a selection of
existing heat network schemes considered to be representative of the main types of
networks in the UK. This was subsequently followed up by further engagement to
look to clarify information provided and fill-in data gaps;
Identification and examination of hypothetical schemes taken from feasibility studies,
deemed representative of current and future heat network developments;
An on-line survey addressed to a wide range of stakeholders involved in different
aspects of heat networks with a view to obtain additional and/or complementary
information. The aim of collecting this data was to provide additional information for
benchmarking the existing schemes and feasibility studies, and to further inform
DECC’s evidence base on heat networks’ cost and performance.
The following sections describe each of these data gathering exercises in further detail.
Data collection
Selecting schemes
A selection of 14 existing heat network schemes considered to be representative of schemes in
the UK were identified via a process which included a pre-selection of potentially relevant
schemes extracted from AECOM’s heat networks database. These schemes were extracted
according to:
Sector: to identify Mixed Residential and Commercial schemes;
Age: to concentrate on schemes recently completed but with at least one year in
operation;
Size: to select large or medium/large schemes (of over 500 dwellings/units
connected with a preference for over 1,000 connections);
Management / Ownership / Operation: to identify well established organisations who
would be more likely to be able to provide complete, good quality data on the
different aspects of the schemes (performance, characteristics and costs).
17
A final selection of schemes was made by experts from both AECOM and DECC, which
included other considerations such as successful implementation and level of performance. This
was to ensure that the data from the schemes would represent a successful and deliverable
scheme.
In general, the existing schemes selected were based around gas fired CHP, due to the
prevalence of these types of schemes at this scale. Energy from waste schemes were also
identified and included in the initial data capture process. For the purposes of this report, only
the heat network’s characteristics are of interest and information collected on the energy
sources is not included in the analysis.
Statistical background
The selection of 14 heat networks limits the extent to which detailed statistical analysis could be
conducted on the received datasets. This was impacted further by only receiving suitable data
from 7 of the 14 schemes. However the study team and steering group believe that the
schemes from which data was obtained represent a suitable sample for the analysis of future
UK heat networks.
Research by Databuild for DECC in 2013 identified 1,765 heat networks in the UK
4
. However
out of these 1,765 networks, only 75 are classified as large (over 500 homes or 10 non-
domestic buildings), and these are the scale of network of interest to this study. Of the large
networks, around half are identified as sourcing heat from CHP (20 networks have uncertain
sources of heat), and two-thirds use gas as the main fuel source. In addition, around 70% of
the large networks were developed before 1990.
The Databuild research therefore demonstrates that although there are a large number of heat
networks in the UK, only a small number of these are at a significant scale, and of these, a
majority source their heat from CHP and / or natural gas. The age profile suggests that most of
the large networks will also have less accessible data on initial capital costs, and so be less
useful for cost data collection.
In light of the Databuild research, the targeted sample of 14 networks is considered to be a
good representative sample of heat networks systems envisaged applicable for large-scale
deployment.
It is important to note that this study has identified the lack of collection and coordination of data
on schemes by operators and developers. In general, there is little historic information
available, and other data is fragmented between a range of stakeholders. Even for recent
schemes where it would be expected that sufficient data would be available, incomplete
datasets were received and the data collection process was not simple. The state of the
industry therefore means that large samples of consistent data which may be used for statistical
analysis are not easily accessible and/or available, and this study provides the most
comprehensive dataset available to date.
Collecting data
Representatives of these schemes were approached with an introduction to the research and to
explain the requirements for information and data collection procedures. A questionnaire was
then sent to each of these representatives requesting detailed information on the schemes.
Following subsequent discussions with some of the stakeholders about the level of detail being
requested, a shortened summary questionnaire was prepared and issued which required fewer
resources to complete.
4
See footnote 3.
Data capture methodology
18
The data capture process was conducted through a combination of e-mailed responses to the
questionnaires, follow up phone calls, and data gathering meetings held between AECOM and
the stakeholders. A total of eight questionnaires were returned, and following an assessment of
the data quality and completeness, seven were deemed suitable for inclusion in the analysis.
One scheme excluded could not provide essential information due to commercial reasons and
therefore this scheme was excluded from the analysis.
Following subsequent data evaluation for completeness and quality, additional one to one
contacts were made with participants for the following purposes:
To clarify information, confirm validity of suspect data and correct inaccurate data
(e.g. wrong units or incorrectly entered numbers). This data cleansing exercise
enabled further clarification of certain results obtained, and enhanced the overall
level of data quality obtained.
To supply missing data. This enhanced the overall data quantity obtained.
Feasibility studies
In parallel to the data from existing schemes, a number of feasibility studies were identified from
different sources and detailed data similar to that requested for the existing schemes was
extracted from seven of them. One scheme was then excluded from the seven due to an
absence of information relating to size of heat demand preventing the economic model from
being constructed.
The use of data from feasibility studies has the following benefits:
By using recent studies, the costs are current and allows for benchmarking and
quality checking of the existing schemes where cost information may be missing or
out-of-date;
Feasibility studies are generally ‘complete’ and include all cost elements to allow a
full economic evaluation to be constructed. In existing schemes, the operational
structures can mean that not all costs are available or easily attributable (see later);
The feasibility studies identified were by parties deemed independent of the
schemes. Therefore the scope for bias and gaming should be removed. This allows
the data extracted to be used to cross-check the data from existing schemes;
The costs in feasibility studies are generally hypothetical and do not reflect costs
associated with specific difficulties (for example installation problems) of real
schemes. They are therefore more appropriate for developing economic models of
generic schemes.
Whilst the data from feasibility studies offer a number of benefits, the data is generally of a more
aggregated nature due to lack of detailed information for costing, and therefore is less useful for
detailed analysis.
Online survey
An online survey was conducted in addition to the data collection from existing schemes and
feasibility studies. This was used to collect information covering a wide range of cost and
performance parameters from heat network schemes, the operational requirements, and
individual items of equipment. The survey was publicised and distributed amongst members of
19
the CHPA
5
, UKDEA and the urban energy Vanguards Network. In addition, the survey was sent
directly to additional stakeholders identified and other AECOM contacts by targeted email.
The aim of the survey was not to collect complete datasets describing operational heat network
schemes, but to obtain information on different components and elements of heat network
schemes to allow comparisons and benchmarking to be conducted.
Ten responses were received from participants ranging from energy services companies and
scheme operators, to individual component manufacturers. Information on 14 heat network
schemes was provided with 4 of these believed to be schemes located outside the UK. The
information obtained on schemes was generally not complete, but useful data was extracted to
complement benchmarking and the existing scheme and feasibility studies datasets (e.g.
information on heat demands served and energy centres characteristics and cost).
Scope of data collection
A heat network scheme comprises a number of elements. These can principally be broken
down into:
Heat generation plant
Distribution network
Customer connections and customer distribution
Simplistically, the sale of heat therefore includes the cost of generating the heat, the cost of
distribution the heat, and the cost of connecting to the customer. After the point of connection,
the customer is responsible for the remaining costs.
In reality, the structure of schemes can be more complex with potentially multiple heat sources,
and multiple levels of customer connection. In addition, there can be a number of organisations
involved with the ownership and operation of the scheme.
The following sections describe some of these complexities alongside Figure 1 (see following
page). These are important to consider when interpreting data collected from schemes.
Heat supply plant
This report aims to examine heat network components and therefore is not directly concerned
with the characteristics of the heat supply plant. However it is important to understand the
boundaries of the network in relation to the heat supply plant to ensure that the relevant heat
network costs are included.
In a simple scheme, all heat supply plant may be in the form of an energy centre and may be
operated by a single organisation (the scheme operator) in combination with the heat network.
The energy centre which contains the heat supply plant may also contain heat network ancillary
components such as pressurisation equipment, water treatment plant and a thermal store. With
common ownership, it should be possible to obtain the costs and performance of these heat
network components separately to the heat supply plant.
In a more complex scheme where there may be multiple organisations and / or energy centres,
the heat network components may be dispersed. For example one organisation may be
responsible for the heat network pipework, whilst another may be responsible for the ancillary
heat network components. It is therefore important to understand what is included in the data
provided by respondents.
5
The CHPA has recently renamed itself as the Association for Decentralised Energy (ADE).
Data capture methodology
20
In some cases, secondary heat sources are distributed in customer buildings. The
characteristics of a baseload heat network with distributed boilers may be different to a peak
load network and again needs to be considered. For example, the heat network substations
could be much lower capacity than in a peak load network per a MWh of heat delivered. Some
of the schemes in this report have this configuration for some or all customers.
Figure 1: Schematic showing a comparison of a ‘simple’ and ‘complex’ scheme. This can be used to
illustrate both the physical nature of the scheme, but also the organisational arrangements.
Internal distribution
Internal distribution pipework is required to distribute heat from the primary heat network’s
interface to areas of the building. In a simple heat network, this may be the customer’s own
heating system which is maintained and operated by the customer. However in multi-tenanted
buildings the internal distribution pipework may be operated by another party and connect to
individual customers via secondary interface units.
Whilst the internal distribution pipework forms part of the overall heat distribution system, it is
not generally considered as part of the heat network. In most modern heat network schemes,
the heat network delivers heat to a substation within each building, after which internal pipework
distributes heat to individual customers (such as flats) with potentially individual heat interface
units. The network operator would typically be responsible for the network up to the building
CUSTOMER BUILDING
HEAT
SUPPLY
NEW NETWORK
CUSTOMER
HEATING
SYSTEM
A: SIMPLE SCHEME
INTERFACE
MULTI-CUSTOMER BUILDING
NEW NETWORK
EXISTING NETWORK
CUSTOMER
1
CUSTOMER
2
HEAT
SUPPLY 1
HEAT
SUPPLY 2
HEAT
SUPPLY 3
B: COMPLEX SCHEME
INTERFACE
21
and the heat substation, and a third party (such as a social landlord or a management
company) would be responsible for the internal pipework.
The data in this report presents heat networks as two types based on the scope of the scheme
covered by the data provided:
Bulk heat supply schemes. These provide heat and sell in bulk to multiple customers,
such as a block of flats. Another body is responsible for distribution and sales to
individual customers.
Non-bulk schemes. These include distribution and sales to individual customers.
It is possible that some schemes included in this report classified as ‘Bulk’ also include
extensive secondary internal pipework for distribution to individual customers, and are therefore
effectively ‘Non Bulk’. However they are treated as Bulk for this report due to the data provided
only covering the Bulk elements of the scheme.
Heat network schemes evolvement and history
Due to the complex nature of heat network schemes and the need for large scale investment,
heat networks have often developed in a phased process. This means that investment may
have taken place over a number of years by a number of parties. It is also possible that existing
individual schemes have amalgamated including the connection of individual communally
heated blocks.
The result of this is that an operational heat network scheme may include components which
were developed for another purpose, and which have been effectively paid off during their
lifetime. One example is a scheme which has adopted and extended an existing heat network.
This means that the economic models for some existing heat network schemes may be very
different to new schemes. For example, some existing schemes may be financially viable due
to historic investment and adoption of infrastructure, whilst the same scheme as a new build
without these historical benefits may be unviable. It is therefore important to understand that
the economic characteristics of heat networks when used in an economic model may not be
representative of existing schemes performance. It can be argued that some existing heat
network schemes in the UK have benefited from some form of grant, whether through direct
financial grants, adoption of existing infrastructure, cross subsidies from energy generation, or
historic investment pay-off, and that this could be seen to distort the realistic economic
performance of heat network schemes in the UK in absence of support mechanisms.
In this study, all capital costs of components which form part of the heat network are included
where possible. Despite their costs not featuring in particular schemes, they are a required
component and as such should be accounted for to understand the characteristics of heat
networks.
Data checking and verification
AECOM and Sweett Group have conducted a data checking process on the datasets received,
to ensure that the resulting financial models are correct. This data checking process included:
Basic checking of data formats and units to ensure that questionnaires have been
correctly interpreted and completed.
Follow up questions with data providers where information is missing or identified as
potentially erroneous.
Simple checks on data to check for consistency within schemes. For example:
Data capture methodology
22
Checking that the total heat sales plus heat losses on the network equal the
total heat production.
Assessing efficiencies from fuel consumption and heat and electricity
outputs.
Checking for consistency in data between heat network schemes, including using
information from the feasibility studies and on-line survey responses. This was
partially conducted through a process of benchmarking, some outputs of which are
described in section 3.
Cross referencing performance data provided by schemes with existing AECOM
experience of heat network schemes and best practice performance.
Cross checking capital cost information with data from a range of other schemes
within the Sweett Group experience.
This process identified a number of potential errors or areas of uncertainty which have been
resolved. However, it is important to note that whilst the authors have conducted the data
checking process, this study is fundamentally reliant on the quality of information provided by
external stakeholders.
Cost data
All capital costs are presented as 2013/14 prices. For heat network schemes where capital cost
information is one or more years prior to this, the RICS Building Cost Information Service (BCIS)
General Building Cost Index has been used to provide an uplift to the capital cost data.
Identifying gaming and bias
Due to the use of the information from this study to inform future policy analysis, there is a risk
of gaming and bias from the data providers. This risk is most likely from commercial
organisations rather than public sector heat network scheme operators.
The data checking and verification process has been used to identify where this may be a
factor, and in the datasets received, no obvious instances of gaming or bias were identified
(although this does not rule out the potential). It is important to recognise that due to the
relatively limited number of schemes data has been received from, and the range of schemes, it
is not possible to conduct robust statistical analysis.
In addition to the checking and verification process, the disaggregated level of data requested
and inability of data providers to see the impact of data inputs when providing data, means that
the potential for gaming is limited. AECOM were actively involved in the collation of data with
the operators of some schemes, including having access to the operators’ datasets and models,
and through this process observed no instances of where gaming or bias was obviously
introduced or where the opportunity had arisen for introduction.
23
3. Description of data captured
Introduction
This section provides an overview of the data captured, and characterisation of the heat
networks identified. Outputs from the benchmarking exercise are shown which were used to
cross check data and assist with gap analysis.
The data presented in this section follows the comprehensive data collection and checking
process, which included the surveys and substantial follow up including interviews and
meetings. As previously described in this report, there is a clear lack of data collection and
coordination within the industry, and the analysis presented here is considered to be the most
comprehensive dataset describing heat networks in the UK currently available.
Description of heat network schemes by size
Table 5 below provides a summary of the number of schemes for which data has been
obtained. They are characterised by size in terms of peak thermal supply capacity, annual heat
sales, and length of the heat network.
Size of scheme
Existing heat network schemes
(number)
By peak supply capacity
Less than 10 MW
3
10 to less than 25 MW
2
25 to less than 50 MW
2
50 MW or greater
0
By annual heat sales
Less than 5,000 MWh
0
5,000 to less than 10,000 MWh
2
10,000 to less than 25,000 MWh
1
25,000 to less than 50,000 MWh
3
50 MWh or greater
0
By length of heat network
Less than 1,000m
0
1,000 to less than 2,000 m
3
2,000 to less than 5,000 m
2
5,000 to less than 10,000 m
2
10,000 m or greater
1
Table 5: Summary of the number of schemes from which data has been collected. Information is shown for
schemes where relevant data was obtained to allow classification.
Description of data captured
24
Description of heat network schemes by delivery type
For the purpose of this study, and to identify any potential specific traits and differences, heat
networks were differentiated between Bulk and Non Bulk.
Bulk heat networks are defined as those where the main scheme operators deliver heat in a
bulk basis to a major distribution point connections but who do not have responsibility for final
delivery to the end customers. Non Bulk schemes are those where the operator or manager of
the scheme is responsible for final delivery to the individual customer (each dwelling or flat).
In identifying differences between Bulk and Non Bulk schemes, a possible rationale for such
differences is presented. It is possible that differences in results between the Bulk and Non
Bulk schemes are an artefact of the low sample size rather than representing actual differences.
Typical benchmark assessment
The following sections provide ‘typical’ benchmark outputs from analysis of the data collected.
Due to the significant amount of data collected from across the schemes, it is possible for
benchmarks to be formed in a range of ways, using different normalisation factors, and levels of
aggregation. The following sections provide some of the main benchmarks identified, but
further analysis of the data could be conducted for more in-depth analysis.
The ‘typical’ benchmarks are created from mean averages of data from across the schemes.
Where data is considered as inconsistent, for example an outlier, or missing, the data from the
scheme for the particular benchmark is not included in the average calculation. This means that
data from different schemes may be used to form the different benchmarks.
Benchmark information is shown for the existing heat networks from which data was collected.
As described in the methodology, data has also been collected from feasibility studies and from
an online survey. This additional data was helpful to cross check and verify the information
collected from the existing heat networks. However, the data from the feasibility studies and
online survey data was not sufficiently disaggregated or complete to be directly included within
the typical benchmark analysis, and this data is not presented in this report.
It should be noted that all of the following analysis is based on a small dataset as previously
explained which may explain some of the differences identified.
Use of the typical benchmark values
The typical benchmark values presented in the following sections can be used to understand
the costs and performance of existing heat networks in the UK. In theory they can be used to
construct hypothetical heat networks based on some simple input parameters, such as annual
heat demand from customers.
When using these values, it is important to consider the following:
The values are based on a small sample of schemes and whilst efforts have been
made to check the data and ensure its robustness, the statistical validity cannot be
determined.
In some cases where data was not available or deemed unreliable, the typical
benchmark values are based on a smaller sub set of the 7 schemes. This means
that if the typical benchmark values are taken for two or more metrics for use in
analysis, they may not be consistent as derived from different existing schemes.
25
The following information is presented for each typical benchmark:
Description / name
Units
Mean average across all existing schemes
Mean average across Bulk supply and Non-Bulk supply schemes.
Range of values split into Bulk supply and Non-Bulk supply.
Number of schemes from which the values are calculated
The typical benchmarks are normalised using a number of metrics as shown by the units. For
consistency, a complete set of cost typical benchmarks are provided normalised against annual
customer (at individual customers or bulk supply point) heat demand (£ / MWh) to allow the
analysis of network costs based on heat demand.
A description of the metrics used in the cost and performance tables is provided in Appendix A.
Performance typical benchmarks
The outputs from benchmarking performance metrics are shown in Table 6 (see page 26).
Particular points are noted:
The seasonal heat load factor is generally higher for bulk heat networks than non-
bulk heat networks. This is potentially due to the former providing base-load heat
with the peak capacity figures provided including some diversity, whilst for the non-
bulk schemes, the peak demands are based on customer connections and therefore
include less or no diversity.
Distribution losses range widely with averages of 6% and 28% respectively for the
bulk and non-bulk heat networks. The losses for the bulk schemes are in line with
the common understanding that losses for heat networks are around 10% or less.
The losses for the non-bulk schemes are significant with one scheme at 43%. The
data collected for this report cannot accurately attribute these high losses, but it is
believed they occur due to thermal losses in internal pipework within buildings, which
is not always insulated to the same standards as the buried heat network pipes. The
kWh / m metric is based on the length of the buried network, and shows losses of
orders of magnitude higher for the non-bulk schemes. It is unlikely that the buried
network is an order of magnitude higher in terms of thermal loss, which indicates the
internal pipework could be the cause. It may be possible to quantify these losses
through examination of the heat output profiles and this could be conducted through
further analysis of each scheme. Internal losses from communal heating systems in
apartment blocks will occur irrespective of whether the building is connected to a
heat network, and therefore whilst these losses need considering for heat network
schemes, they should be considered a building services design and maintenance
issue. Some of the heat loss from internal pipework will provide useful heat gains to
buildings during the winter period but the heat loss may increase the risk of
overheating in buildings during the summer months.
Description of data captured
26
Operating temperatures are generally higher in the bulk schemes than the non-bulk
schemes. The cause of this cannot be directly determined, but it may be due to the
age of the networks, or due to the presence of intermediate heat sub stations (where
the heat is sold in bulk) which incur temperature drops, necessitating higher
temperatures on the primary network.
Parasitic electricity demand for pumping and operation of controls is generally low
ranging between 1% and 4% of the annual heat demand. However the higher costs
of electricity mean that this can have a disproportionate impact and the design of
networks will need to carefully optimise the pumping requirements versus pipe
sizing.
Table 6: Typical benchmarks describing the performance of heat networks.
Thermal storage
The values in Table 6 include thermal store capacity expressed as m
3
storage per MWh annual
heat demand, and thermal stores were present on two of the schemes. Both of these schemes
were consistent with around 0.015 m
3
of storage provided per MWh annual heat demand. Both
of these stores were approximately 120m
3
, and thus from similar size schemes.
Thermal storage is not a central component of heat networks, but can contribute to improving
network performance and heat supply performance in some circumstances.
OVERALL
PERFORMANCE
BENCHMARKS
AVERAGE AVERAGE MAX MIN
Number
AVERAGE MAX MIN
Number
% seasonal heat load
factor of network
% na 23% 28% 20% 4 17% 24% 13% 3
Distribution losses
1
% heat
generated
na 6% 11% 3% 4 28% 43% 12% 3
Distribution losses
1
per m
distribution network
kWh / m na 544 850 231 4 767 5942 416 3
Linear heat density MWh
demand per m network
2
MWh / m 7.6 9.7 18.3 5.9 4 4.7 7.8 3.1 3
Thermal store capacity
per MWh heat demand
3
m
3
/MWh
0.015 0.016 0.016 0.016 1 0.014 0.014 1
Length of internal
pipework per dwelling /
flat
4
m 13.3 0 13.3 20.8 9.2 3
Operating temperature -
flow
Degrees C 88 92 95 82 4 84 85 81 3
Operating temperature -
return
Degrees C 62 68 75 55 4 54 60 46 3
Network parasitic
electricity consumption
% heat
demand
na 2.9% 4.0% 1.0% 4 1.9% 2.0% 1.7% 2
1
Difference between heat generation and heat demand served divided by the heat generation
2
Average figure for heat demand served per m of buried pipe in the network
3
Average figures for schemes with thermal storage is provided.
4
Only applicable to non bulk schemes which serve individual dwellings/flats/units
BULK SCHEMES
27
Thermal storage helps to smooth the load profile, by providing a buffer between the heat supply
plant and the building loads. This is particularly true of smaller schemes with a limited customer
base, early phases of larger schemes where the heat plant may be initially oversized, and
schemes where there is limited diversity. By charging and discharging the thermal store
periodically, heat demand peaks can be reduced allowing the heat supply plant to be operated
more consistently with less peak load plant (typically gas boilers), and at times when operation
may be more economic. Thermal stores typically consist of large insulated water tanks, but the
heat network itself could also be used through variable temperature regimes, or the thermal
capacity and control systems within buildings could be exploited.
The need for, and volume of, thermal storage depends on both the thermal loads and the heat
source. Where the thermal capacity may be low cost or have no constraints on capacity
compared with load (for example, extraction of heat from a large power station), then thermal
storage may not be necessary. However when the thermal capacity may be expensive, and/or
benefit from continuous operation (for example, gas CHP engines), or where the capacity may
be constrained (such as a smaller waste heat source), then thermal storage can help improve
the load profile and make best use of the heat source.
The optimisation of the thermal store capacity will depend on a number of parameters which are
scheme specific, and suitable data has not been collected as part of this study to allow a
detailed analysis.
When stakeholders with schemes with no thermal storage were questioned about the absence
of stores, the response was often that the inclusion of stores would be preferable, but that
spatial or planning limitations prevented their installation. The height of thermal stores means
that they are better suited to external locations, and their appearance and height can be seen
as unacceptable in some areas.
The optimisation of thermal store capacity for one of the schemes with a store was discussed
with the scheme owner. Far from being a technically refined process, the capacity was
determined by (a) the height which was acceptable to the planners, and (b) the maximum
diameter which could be easily delivered by road for a pre-constructed tank.
The consensus based on the consultation and data provided therefore appears to be:
The use of thermal storage is generally of benefit to schemes where it complements
the heat source.
The ability to install a thermal store can be severely limited by space requirements
and external considerations such as planning acceptability.
Where thermal stores are installed, the capacity may be determined by the desire to
have as large a volume as possible subject to space requirements and external
considerations, and not detailed technical optimisation.
It is suggested that DECC examine the use of thermal storage on heat network schemes in
more detail to assess the importance it has on improving efficiency and financial performance.
An outcome of this work could be to develop simple criteria which can be used to encourage
that appropriate levels of storage are included in schemes, and support heat network
developers in demonstrating the importance of thermal storage to other stakeholders such as
local authority planning departments.
Description of data captured
28
Characterisation of thermal storage requires load profile modelling to be conducted including
scheduling of heat supply plant
6
. It is often found that where storage is appropriate, the
inclusion of any thermal storage, and increases in capacity up to an optimal size, results in large
operational benefits which can outweigh the capital expenditure. Once the optimal size has
been achieved, further increases in capacity have diminishing returns and become uneconomic.
The aim of further work by DECC could be to examine the relation between the optimal
capacity, and different load types, including sensitivity testing against variations in load profile,
and heat supply characteristics.
Capital cost typical benchmarks
The capital costs for heat networks are shown in Table 7 opposite (normalised to MWh annual
heat demand,) and Table 8 on page 30 (normalised to other metrics). All costs are indexed to
2013/14 prices as described in section 2.
The costs of heat network connections and the heat network in Table 7 are shown at both an
aggregate (bold) and disaggregated (italics) level. It should be noted that the disaggregated
costs may not add up to the aggregated cost, due to the inclusion of different schemes for each
metric. For example, some schemes may have only provided information at an aggregate level,
whilst others may only have provided some items at a disaggregated level. To allow
comparison, the sum of disaggregated costs is also provided.
6
This modelling would typically be conducted on an hourly or half-hourly basis over a sample year/s.
29
Table 7: Capital cost typical benchmarks normalised to annual heat demand. (D – Domestic, ND – Non
Domestic, HIU – Hydraulic Interface Unit). The data presented in BOLD is provided as an aggregate where
available by the heat network scheme through the data collection process. All values presented in italics
is provided by heat network schemes at the disaggregated level where available. Therefore the aggregate,
and sum of disaggregated values are not always equivalent.
OVERALL
CAPITAL COSTS
AVERAGE AVERAGE MAX MIN
Number
AVERAGE MAX MIN
Number
Heat Connections £ / MWh na £25 £36 £8 4 £624 £1,179 £129 3
Cost substations ND and D
blocks
£ / MWh £16 £12 £16 £8 2 £19 £39 £5 3
Cost heat meters ND and D
blocks
£ / MWh £4 £3.7 £3.7 £3.7 1 £4.4 £8.0 £0.8 2
Cost HIUs Dwellings
1
£ / MWh £253 0 £253 £461 £46 3
Cost heat meters
Dwellings
1
£ / MWh £170 0 £170 £232 £107 2
Cost internal pipework
connection to HIUs
1
£ / MWh £492 0 £492 £782 £203 2
Cost for connections
prelims
2
£ / MWh
Overhead and profit
connections
2
£ / MWh
SUM OF DISAGGREGATED
CONNECTION COSTS
£ / MWh na £16 £20 £12 0 £939 £1,523 £362 0
Heat Network £ / MWh £150 £150 £239 £80 4 £150 £168 £132 2
Mechanical capital cost –
total
£ / MWh £70 £62 £102 £42 4 £80 £92 £72 3
Civil capital cost – total £ / MWh £74 £79 £136 £25 4 £64 £76 £52 2
Preliminary capital cost –
total
3
£ / MWh £8 £10 £13 £7 2 £3 £3 £3 1
Overhead and profit –
total
£ / MWh £10 £10 £13 £7 2 0
SUM OF DISAGGREGATED
HEAT NETWORK COSTS
£ / MWh £161 £160 £265 £80 £148 £171 £128
Ancillary plant associated
with network (eg
pumping, treatment, etc)
£ / MWh na £68 £68 £68 1 £137 £137 £137 1
Thermal store £ / MWh £14 £17 £17 £17 1 £12 £12 £12 1
1
Only applicable to non bulk schemes which serve individual dwellings/flats/units
2
None reported - 2 schemes indicated these costs are included but can't be disaggregated
3
Prelim costs are believed to be mainly included within mechanical, civil and/or total costs
BULK SCHEME
no data
no data
Description of data captured
30
Table 8: Capital cost typical benchmarks normalised to non MWh metrics.
Particular points of interest are:
Connection costs comprising internal pipework and substations are more than an order of magnitude
higher for the non-bulk schemes. This is due to the presence of components which do not feature in the
bulk schemes such as internal pipework, and individual hydraulic interface units (HIU) for dwellings. HIUs
are relatively expensive when normalised to MWh partially due to their large capacity to provide domestic
hot water, combined with relatively low annual heat demands for individual dwellings. The relatively high
cost of HIUs can be seen in
Table 8 with costs of around £1000 per dwelling. The costs of sub stations to blocks
of dwellings or larger commercial and public sector buildings are significantly lower
per MWh and kW.
Internal pipework costs in blocks of flats are relatively high. The cost will depend on
the quality of the pipework, installation requirements, and internal layouts of
buildings. However the data collected in this study shows that internal pipework
lengths can be significant at around 10 times greater than the heat network length.
This results in high cost but also large potential losses as discussed earlier.
Heat meters are also a relatively expensive component for individual dwellings with
costs per MWh of up to 100 times the cost of meters in bulk heat supply substations,
due to the need for many smaller meters and meter reading systems. It should be
noted that the data collected in this study relates to non pre- payment meters, and
pre-payment meters can be more expensive. It should also be noted that the
operation costs of heat meters need to be taken into account when assessing their
viability, and a higher up front capital cost with improved data capture and reading
systems may provide lower operational costs.
Heat network costs are typically around £150 per MWh for both types of scheme,
split roughly equally into civil installation costs, and mechanical pipework costs. If a
OVERALL
CAPITAL COSTS
AVERAGE AVERAGE MAX MIN
Number
AVERAGE MAX MIN
Number
Network capital costs per
length (main network-
buried)
£/m £984 £1,242 £1,472 £874 4 £468 £514 £422 2
Network capital costs per
length (internal pipe)
£/m £169 0 £169 £244 £94 2
Substations cost per kW
capacity
£/kW £32 £28 £40 £15 2 £35 £53 £16 3
Domestic HIUs cost per
dwelling
£/dwelling £1,075 0 £1,075 £1,326 £738 3
Heat meter cost per
building (ND and bulk)
£/building £2,878 £1,949 £1,949 £1,949 1 £3,343 £6,136 £551 2
Heat meter cost per
dwelling
£/dwelling £579 0 £579 £668 £491 2
Thermal store cost per m3 £/m3 £962 £1,080 £1,080 £1,080 1 £843 £843 £843 1
BULK SCHEMES
31
50 year lifetime is assumed, this means the capital cost of the network in relation to
supplied heat is approximately £3 per MWh. The entire cost of the heat network
scheme including all the connections, and operation is of course much higher.
Heat network costs are approximately £1000 per m of main network length on
average across both types of scheme, although there is a wide range from £422 / m
to £1,472 / m. This range of costs will reflect the different scales of network (and
thus pipe sizes), and the type of installation (for example a hard dig situation in
existing utility-congested roads, or a soft dig on a lightly congested area). In
general, the bulk scheme costs are higher but this could be due to a number of
factors including scale and heat density of scheme and whether they are in newly
developed areas or not. Additional information has been received on both pipe sizes
and installation requirements for some schemes and has been provided to DECC for
further analysis.
Figure 2 provides a high level breakdown of the capital costs for each type of scheme. It
demonstrates that the costs for all components are broadly similar apart from the heat
connections costs.
Figure 2: Breakdown of capital costs for bulk and non-bulk schemes.
Description of data captured
32
Operational cost benchmarks
Information has also been obtained on operation costs as shown in Table 9 on the following
page.
In general, the operation cost associated with the heat network itself, rather than the associated
heat supply plant and energy centre buildings, is low. Operation cost data directly associated
with the heat network was hard to obtain from operators and in many cases no information was
available. The following points were noted from the data capture process:
Where information was available, it was often a simple estimate or nominal amount.
The heat network direct operation costs are often not itemised in annual budgets or
forecasts, suggesting they are low and not of importance.
A common response was that these costs are not itemised, and feature as part of
wider overheads. For example, a maintenance engineer may not allocate time to a
particular network or component, and is funded more centrally.
From the data obtained, maintenance costs for the heat network appear to be negligible and
unlikely to impact on the economic performance. It should be noted that where data was
available, it generally related to routine maintenance, and it is possible that networks are
operated on a responsive maintenance regime.
The highest maintenance costs appear to be related to domestic HIU units and heat meters,
although the range for both was large, and the sample small.
Metering and billing costs also had a large range and these are likely to be heavily influenced by
the organisational structure for conducting these services. For example they may be conducted
in house along with other billing services, or they may be externally contracted. The costs also
will depend on the type of meter reading system available.
Business rates were also identified as an operation cost. It is not clear whether these relate to
the heat network itself, or associated energy centre buildings or operation buildings. The
eligibility of heat network schemes for business rates partially depends on the nature of the
customers and location of the energy centre building and therefore there may be variation
across schemes
7
.
7
Valuation Office Agency Rating Manual Section 340: District Heating Undertakings.
33
Table 9: Operation cost typical benchmarks normalised to annual heat demand and capacity.
Heat sales
Heat prices for heat network schemes are unregulated and calculated using a number of
different methods. The price of heat is often determined by the heat source, which can result in
a wide range of costs and structures.
Information on heat sales prices was collected as part of the data capture exercise. The
following points were noted:
Scheme drivers. Some of the heat network schemes are not-for-profit and have
been developed with the prime driver of reducing fuel poverty and lowering heating
costs. The heat prices may therefore not represent a competitive price against the
alternative counterfactual prices (gas boiler heating for example), but may be lower
such that it is the minimum for the scheme to remain viable. It may therefore be
possible for some of these schemes to charge more for heat than the prices given
whilst still providing lower cost heating than alternatives.
Capital cost model. Many heat network schemes have adopted existing assets or
made use of significant capital grants, reducing initial investment. This means that
whilst the heat prices used can allow these existing schemes to operate
economically, they may be too low to be economically viable for schemes where all
capital investment is incurred.
To allow heat from a heat network scheme to be competitively priced for customers, the
counterfactual cost of heat needs to be understood. Whilst many customers may only consider
the fuel or energy cost (predominantly gas or electricity), the full counterfactual cost should
include the purchase, replacement, and costs of operation of the heat source.
The average heat tariffs for a range of dwellings are shown in Table 10 (page 35) at 2014
prices. The dwellings and demands are indicative, but are used to demonstrate how the
effective heat tariff is dependent on dwelling annual heat demand. The heat costs are
OVERALL
OPERATION COSTS
AVERAGE AVERAGE MAX MIN
Number
AVERAGE MAX MIN
Number
Heat network
maintenance cost
£ / MWh £0.6 £0.4 £0.4 £0.4 1 £0.6 £0.9 £0.3 3
Heat network management
cost
£ / MWh
Substation maintenance
cost
£ / MWh
HIUs maintenance cost £ / MWh £9 0 £9 £16 £2 3
HIUs maintenance cost £/MW £820 0 £820 £1,039 £658 3
Heat meter maintenance
cost
£ / MWh £3.4 0 £3.4 £9.0 £0.1 3
Avg annual staff cost for
metering, billing and
revenue collection
£ / MWh £11.1 £2.5 £4.6 £0.4 2 £16.9 £34.8 £0.1 3
Annual business rates £ / MWh £6 £7 £8 £5 2 £6 £8 £2 3
BULK SCHEMES
no data
no data
Description of data captured
34
calculated assuming a 30 year lifecycle, with a 15-year boiler replacement, and costs typical of
current capital costs and annual service costs
8
. The fuel costs are averages taken from the five
large UK energy suppliers offering gas-only tariffs assuming standard variable rates and
monthly direct debit billing. The resultant prices are an average standing charge of 20.21 p /
day, and variable charge of 4.20 p / kWh. No energy price inflation is included, so that the
counterfactual heat price can be directly compared with the present-day heat prices from heat
network schemes obtained through this work. This analysis therefore does not attempt to
compare payback, but simply current cost comparisons.
The results show that for a larger dwelling, the boiler costs add approximately 1.1p / kWh to the
fuel cost, but for small efficient dwellings, this increases to 4.6 p / kWh giving a total heat cost of
just over 10 p / kWh. Heat network schemes are generally deployed in areas of high density
consisting typically of smaller high density houses or more usually flats. Therefore the
counterfactual heat cost appropriate for heat network schemes is likely to be at the higher end
of the range, typically above 7.2 p / kWh.
8
A capital cost of £2500 is assumed in the Green Deal Final Stage Impact Assessment. Reference figures for
operation costs vary, but are typically £150 - £200 per year, including servicing and repairs.
35
Dwelling size (examples)
Small and
efficient (eg
small modern
flat)
Older flat or
modern small
terraced
house
Older
terraced
house or
larger
modern
house
Older and / or
larger house
Heat demand (kWh)
5,000
10,000
15,000
20,000
Boiler ownership costs
Capex (£)
£2,500
£2,500
£2,500
£2,500
Replacement capex (£ at 15 years)
£2,500
£2,500
£2,500
£2,500
Annual opex (£ / yr)
£150
£150
£150
£150
Annualised capex (£ / yr) (3.5% discount rate)
£133
£133
£133
£133
Annualised opex (£ / yr) (3.5% discount rate)
£95
£95
£95
£95
Boiler costs per kWh heat demand (p / kWh)
4.57
2.28
1.52
1.14
Fuel costs
Fuel demand (at 85% efficiency)
5,882
11,765
17,647
23,529
Standing charge (£ / yr)
£74
£74
£74
£74
Annual variable (£ / yr)
£210
£420
£630
£840
Fuel costs per kWh demand (p / kWh)
5.68
4.94
4.69
4.57
Total costs
Total annual cost
£512
£722
£932
£1,142
Average heat cost (p / kWh)
10.24
7.22
6.21
5.71
Table 10: Comparison of effective heat tariffs for different size dwellings (p / kWh costs for boiler related
expenditure, fuel expenditure, and total heat tariff are shown in italics for clarity)
The heat prices obtained from the 7 heat network schemes in this study are provided in Table
11 (see overleaf). The prices from bulk scheme are lower than non-bulk scheme as to be
expected. The lowest price across both types is approximately 5p / kWh, but this increases to a
10p / kWh maximum for non bulk schemes, i.e. for domestic customers.
Description of data captured
36
Heat prices from schemes
All schemes
Bulk schemes
Non-bulk
schemes
Mean average (p / kWh)
6.43
5.77
7.52
Minimum (p / kWh)
4.64
4.94
4.64
Maximum (p / kWh)
9.88
6.89
9.88
Table 11: Summary of heat prices from the heat network schemes.
Through comparison of the current heat prices obtained and the calculated counterfactual
prices, it can be seen that the current prices are generally low compared with the counterfactual
for smaller domestic properties. In particular, the lowest price of around 5p / kWh may be
providing up to 50% discount on heating bills to customers compared with individual gas boilers.
The issues here are complex and the counterfactual is not a simple all in cost. For example,
whilst the cost is presented including boiler repairs and maintenance, in a social dwelling, the
tenant may only pay the gas costs and not boiler repair and maintenance, and so a lower heat
cost is more appropriate. However this means that there are offset costs elsewhere which
should be taken into account when assessing the viability of heat networks, such as the social
housing provider’s annual maintenance bills.
These results demonstrate that in certain circumstances, heat network schemes could obtain
higher heat revenues to help fund investment and operation of the network than is currently
practised, whilst still offering a cost reduction to customers based on the counterfactual of
individual gas boilers. Variable heat tariffs for different sizes of customers could allow heating
cost reductions across different dwelling sizes whilst providing an adequate income.
The analysis also demonstrates the need for transparency when setting heat tariffs and the
need to educate customers about the true cost of heat, taking into account both fuel and
equipment costs.
Carbon dioxide emissions
Calculation of CO
2
emissions
The CO
2
savings provided by a heat network scheme can be calculated by comparing the
scheme with a counterfactual heating source. All energy supplies to the heat network scheme
should be included such as the prime heat source fuel consumption, gas consumption in back-
up boilers, and electricity consumption for pumping energy and ancillary uses.
For schemes with CHP, the CO
2
intensity of heat is determined by calculating the CO
2
emissions arising from fuel and electricity consumption, and then subtracting the displaced
electricity emissions from CHP electricity generation. Electricity generated by CHP can be
valued the same as the grid average CO
2
intensity, or at a different value, such as the long run
marginal factor (which is lower than the supplied average factor).
The main counterfactual technology appropriate to heat network schemes is the condensing
gas boiler which is predominant in urban areas. The efficiency of domestic gas boilers is
typically recorded as above 90% based on manufacturer’s testing (SEDBUK ratings). When
calculating the actual carbon and bill savings a householder could expect to see it is appropriate
to look at the in-situ performance of gas boilers as this is the level of performance actually
37
experienced by householders. The key finding of EST’s 2009 field trial of condensing gas
boilers was that the efficiency of condensing boilers at home was typically 5 percent less than
published SEDBUK ratings
9
. An efficiency of 85% is therefore assumed in the economic model
for calculation of CO
2
emissions.
CO
2
emissions from heat sources
The CO
2
intensity of heat from a range of heat networks and counterfactual technologies can be
compared to provide an indication of the relative performance of each technology. Figure 3
overleaf, shows the CO
2
intensity of heat supplied from different technologies for a range of
different electricity grid CO
2
intensities. For technologies which either consume electricity or
generate electricity, the CO
2
intensity of heat varies with the electricity emission factor. The
current average electricity grid CO
2
emissions factor is circa 0.54 kg CO
2
/ kWh
10
. Figure 3
demonstrates that if electricity from gas fired CHP is assumed to displace grid electricity at this
intensity it can deliver heat at a negative CO
2
intensity, and is the lowest carbon form of heat
generation of all technologies shown. This assumes that the electricity generated from the CHP
unit is valued at the same CO
2
intensity as electricity supplied to other technologies. In heat
network schemes, the CHP plant will only be providing a proportion of the heat, with the
remainder from gas boilers. This, in combination with distribution losses, means that a negative
CO
2
emissions factor for heat is unlikely to be achieved in practice. The chart assumes a gas
boiler efficiency of 85%, and CHP efficiencies of 37% electrical and 40% thermal (as indicated
in the key).
As the electricity CO
2
emissions factor reduces, the heat CO
2
emissions factor from gas fired
CHP increases due to the reduced benefit of displacing grid electricity. Below a grid emissions
factor of 0.5 kg CO
2
/ kWh, heat from biomass boilers becomes the lowest carbon source, but
CHP remains lower carbon than heat pumps until the grid emissions factor reduces to around
0.35 kg CO
2
/ kWh. For gas boilers or direct electric heating to become lower carbon than gas
fired CHP, the grid emissions factor needs to reduce to circa 0.2 kg CO
2
/ kWh.
However, comparisons between different heat sources are complicated by the fact that grid
emissions factors for electricity displaced by CHP will not be the same as grid emissions factors
for electricity consumed in electrode boilers and heat pumps. The carbon intensity of electricity
displaced by CHP will depend upon its operating costs in comparison to the cost of power from
other generation types, the proportion of the CHP’s output which is sold to the wholesale market
versus that used locally displacing retail price electricity and the extent to which the CHP is
embedded within the distribution network. LCP have recently modelled
11
competition of gas
CHP with other generating capacity for DECC including modelling the effect of these issues.
This work concluded that, on average, under DECC’s central grid decarbonisation trajectory
operation of gas CHP will continue to deliver annual carbon savings until 2032.
9
http://www.energysavingtrust.org.uk/northernireland/Organisations/Technology/Field-trials-and-monitoring/Field-
trial-reports/Condensing-boilers-and-advanced-room-thermostats-field-trials
10
DEFRA UK Government Conversion factors for company reporting. Generation average of 0.494 kg/kWh, and
transmission and distribution losses of 0.0432 kg/kWh, providing a total supplied factor of 0.537 kg/kWh.
11
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/389070/LCP_Modelling.pdf
Description of data captured
38
Figure 3: Comparison of heat CO
2
intensity from a range of heat sources. Heat pumps are shown for both
a COP of 2.5 and 3.
CO
2
emissions from heat network schemes
The CO
2
emissions associated with a heat network scheme are more complex than the
emissions factor of heat from a single technology. In general there is a mix of heat sources
(typically baseload CHP plant combined with peak load gas boilers) which need to be
accounted for, efficiency losses, and other energy supplies such as parasitic electricity
demands.
The results in Table 12 (see next page) show the CO
2
emissions from a heat network scheme
constructed from benchmarking the data collected in this study (this includes operational data
for heat supply plant not presented in this report). These are presented for a range of electricity
emissions scenarios:
2014 values. The represents a 2014 single year calculation based on current grid
emissions. These are the savings which would be achieved by the benchmark
scheme operating at the current time.
Mid projection. This uses a 15-year projection of grid emissions developed by the
Building Research Establishment (BRE) for SAP 2012 long term modelling
12
.
12
Emission factors and primary energy factors (15-year projection 2013-2027).
http://www.bre.co.uk/filelibrary/SAP/2012/Emission-and-primary-factors-2013-2027.pdf
39
IAG 15 year average. This takes a 15 year projection of grid emissions based on
the Interdepartmental Analysts’ Group (IAG) guidance on greenhouse gas
valuation.
13
For each scenario, the emissions are modelled with the displaced emissions factor equal to the
supplied emissions factor, and with a lower marginal factor.
Scenario
2014 values
Mid projection 15 year
average
IAG 15 year average
Grid factors
used
Supplied =
displaced
Marginal
displaced
Supplied =
displaced
Marginal
displaced
Supplied =
displaced
Marginal
displaced
Grid supplied
factor (kg /
kWh)
0.537
0.537
0.381
0.381
0.259
0.259
Grid
displaced
factor (kg /
kWh)
0.537
0.318
0.381
0.274
0.259
0.240
% CO
2
savings on
heat supply
14
75%
-1%
23%
-12%
-18%
-24%
Heat
emissions
factor (kg /
kWh)
0.05
0.22
0.17
0.24
0.25
0.27
Table 12: Comparison of CO
2
savings from the benchmark scheme under a range of emissions factor
scenarios.
The results demonstrate that under current grid emissions, up to 75% CO
2
reduction in heat can
be achieved if the electricity generated by CHP is valued at the current grid average. For CHP
operating at peak periods, the displaced electricity factor may be even higher if older less
efficient fossil electricity generation is offset, and so the CO
2
savings could be higher. The CO
2
intensity of heat is 0.05 kg / kWh, lower than any alternative forms of heat generation identified
in Figure 3 apart from biomass boilers.
If the lower marginal factor is used for the displaced electricity, the CO
2
savings are removed
and the heat has approximately the same CO
2
intensity as heat from the counterfactual gas
boiler. In this scenario, the CHP would need to be displacing electricity from renewable or low
carbon forms of generation including renewable sources, nuclear power, or fossil generation
with carbon capture. With grid decarbonisation, it is expected that a proportion of electricity
13
It is worth noting that DECC is revising the way its treat CHP displaced emissions factors in-line with its research
in https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/389070/LCP_Modelling.pdf
14
The CO
2
calculation takes the CO
2
associated with fuel burnt in the CHP engine, and subtracts displaced
emissions for the counterfactual heat source and displaced electricity. The savings are expressed as a % of the
counterfactual emissions.
Description of data captured
40
generation will remain as unabated gas generation, which is most likely to be displaced by the
CHP output, helping to justify the use of a higher marginal factor.
In the mid projection, and IAG projection scenarios, the CO
2
savings are reduced further. Under
the mid conditions, a 23% CO
2
saving is achieved when the displaced grid factor is equal to the
supplied factor. Under all other scenarios, the CO
2
emissions are predicted to increase due to
the low displaced factors.
The analysis demonstrates that a more suitable framework is required for assessing CO
2
emissions and savings from heat network schemes. By taking simple grid averages, the
potential benefits of different generation technologies operating at different times are not
realised. Conversely, during off-peak periods when excess low carbon generation is available,
the use of heat pumps on heat network could provide a suitable load and source of low carbon
heat. Figure 3 (page 38) indicates that heat pumps with a COP of around 3 could deliver lower
carbon heat than gas fired CHP with a grid emissions factor of around 0.35 or less. The
combination of heat pumps and gas fired CHP on the same heat network scheme could be
optimised to operate at appropriate peak and off-peak periods to maximise the CO
2
reduction
potential, helping to balance electricity loads on the national grid.
To enable the benefits of different heat supply technologies to be assessed and allow a robust
long-term assessment of heat network schemes, it is recommended that a framework is
developed for CO
2
calculations. This should take into account the future electricity grid
emissions projections, the power generation technologies which will be displaced under
different load conditions by decentralised generation, and the operation regime for decentralised
technologies. Without this, the lifecycle CO
2
emissions of heat network schemes cannot be
suitably assessed.
41
4. Conclusions
Data collection and availability
This study has identified a number of existing heat network schemes from which data has been
collected from a representative sample of 7 schemes and used to assess the costs and
performance of heat networks. Additional data from feasibility studies and an online survey has
helped validate the data from existing schemes.
The data collection exercise has highlighted the difficulty in obtaining accurate and consistent
data across existing heat network schemes in the UK, and the need to develop a framework in
which the schemes can be better monitored and understood. The fragmented and nascent
nature of the heat network industry in the UK means that data is either not collated, or not easily
accessible, but could be used if available to help understand and support further heat network
development.
This study has also identified the complexity of heat network schemes, with many technical and
financial approaches. Understanding the boundary of the heat network schemes is important -
whilst some heat network schemes are relatively simple with all heat supply, distribution and
customer connections financed and operated by the same organisation, the majority are more
complex with different organisations responsible for different elements of the scheme.
Historic investment and payback on infrastructure also need to be considered when assessing
heat networks. A large number of existing schemes have developed over a number of years,
and some operators have been able to adopt existing heat supply and distribution infrastructure
for little or no charge, effectively a grant to the scheme. Some other more recent schemes have
received large capital grants for infrastructure investment. This means that not only is historic
cost information difficult to obtain, more importantly it does not feature in the operator’s
economic model. Whilst this may enable existing schemes to operate in a financially viable
manner, it is not a replicable model for new schemes where all investment costs will need to be
included.
Data benchmarking
The data collected in this study has been used to produce a set of benchmarks or typical
values. Key conclusions are as follows:
Performance:
Distribution losses range widely with averages of 6% and 28% for each type of
network. The losses for the bulk schemes are in line with commonly used
assumptions with losses of around 10% or less although are representative of the
more urban schemes examined in this report, and could be higher in less dense
schemes. The losses for the non-bulk schemes are significant with one scheme at
43%. Evidence suggests that high losses can be experienced through internal
Conclusions
42
distribution pipework within buildings, and it will be important that the specification
and design of internal pipework and operating parameters are controlled to reduce
losses on schemes.
Capital costs:
Heat network buried pipe costs are typically around £150 / MWh annual, whilst the
connection costs range significantly from £25 / MWh for bulk schemes to £624 /
MWh for non-bulk schemes. The key area of sensitivity is therefore around the heat
connections and configuration. In particular, individual dwelling connections in the
form of HIUs and heat meters appear to dominate the capital cost of domestic
schemes with individual connections.
Buried heat network pipework costs average around £1000 / m over a heat network,
although this ranges from £422 / m to £1472 / m depending on the size and nature
of the schemes.
Operation costs:
Operation costs associated with the heat network appear to be low, and not
considered significant by organisations. However there will be higher operation
costs associated with other elements of the heat network scheme, such as heat
sources and energy centre buildings.
The highest operation costs appear to be associated with HIU and heat meter
maintenance, and meter reading / billing activities.
Heat sales
This report has assessed the heat sales prices used by the 7 schemes, and compared these
with counterfactual prices for the domestic sector. The analysis suggests that in low heat
demand dwellings, the counterfactual prices can be relatively high once boiler maintenance and
replacement is included, and current heat sales prices from heat networks may be under-
valuing the heat.
This undervaluation may allow economic operation of some current heat networks which have
benefited from historic investment or grant funding, but higher prices may be required for future
networks which incur all capital expenditure. The analysis suggests that higher prices may be
obtained, whilst still offering consumers a discount over counterfactual heating costs.
CO
2
emissions
The CO
2
emissions from heat distributed by heat networks depend on the type of heat source
and the fuel used. Larger heat networks predominantly source heat from gas fired CHP at
present although other sources such as energy from waste are also used.
The analysis demonstrates the sensitivity of the CO
2
calculations to assumptions for grid
electricity emissions, and the need for improved advice and guidance for these calculations. In
particular, the valuation of grid CO
2
at an average or marginal generation factor may not be
suitable. It is recommended that a framework is developed for CO
2
calculations. This should
take into account the future electricity grid emissions projections, the power generation
technologies which will be displaced under different load conditions by decentralised
43
generation, and the operation regime for decentralised technologies. Without this, the lifecycle
CO
2
emissions of heat network schemes cannot be suitably and consistently assessed, making
investment in heat networks and heat supply technologies higher risk. DECC is addressing this
by investigating CHP operation in further detail.
15
15
https://www.gov.uk/government/publications/bespoke-natural-gas-chp-analysis
44
Appendix: Description of metrics
Appendix A provides a description of the metrics presented as typical performance and cost
benchmarks in this report. An explanation is only provided where deemed necessary.
PERFORMANCE
BENCHMARKS
(Tables 1 and 6)
DESCRIPTION
% seasonal heat load
factor of network
%
The fraction of the annual heat demand to the heat generated by the network if it
was run continuously at peak capacity, calculated by:
total MWh heat demand/ (peak load X 8760)
Distribution losses
% heat
generated
These are the difference between heat generated from the energy centre and the
building annual heat demand served, as a percentage of total heat generated. For
bulk schemes, the losses relate to the main distribution network, but do not
include internal distribution losses for pipework not included in the analysis. In
non-bulk scheme, the losses are included from internal pipework up to the final
customer connection points.
Distribution losses
per m
distribution network
kWh / m
These are the distribution losses (as defined above) divided by the length of the
network over which they are incurred.
Linear heat density
MWh demand per m
network
2
MWh / m
Annual heat demand served per metre length of buried pipework.
Thermal store capacity
per MWh heat demand
3
m
3
/MWh
Size of thermal store per annual heat demand.
Length of internal
pipework per dwelling /
flat
4
m
Only relevant in non-bulk schemes. This is the total length of internal pipework
on the scheme divided by the total number of dwellings served by the internal
pipework.
Operating temperature -
flow
Degrees C
Supply temperature of the water on leaving the energy centre. This can vary
throughout the year and the value provided is based on design winter operation.
Operating temperature -
return
Degrees C
Temperature of the water on return to the energy centre after heat take-off by
customers. This may vary throughout the year.
Network parasitic
electricity consumption
% heat
demand
Electricity consumption for network operation including pumping requirements.
45
CAPITAL COSTS
(Tables 2 and 7)
DESCRIPTION
Heat Connections
£ / MWh
Aggregate cost as supplied by scheme.
Cost substations ND and
D blocks
(Note:
D – Domestic,
ND – Non Domestic).
£ / MWh
Cost of heat exchangers and associated equipment for large customers and bulk
supply points per annual heat demand. A substation comprises all the equipment
to allow heat exchange from the primary to the secondary network, including heat
exchanger, and all associated pipework, valves, and controls. Heat meters are
assumed to be separate (see below).
Cost heat meters ND and
D blocks
£ / MWh
Cost of heat meters for large customers and bulk supply points per annual heat
demand.
Cost HIUs Dwellings
1
£ / MWh
Cost of individual Hydraulic Interface Units for dwellings (not including heat
meters) per annual heat demand.
Cost heat meters
Dwellings
1
£ / MWh
Cost of individual heat meters for individual dwelling connections per annual heat
demand.
Cost internal pipework
connection to HIUs
1
£ / MWh
Cost of internal pipework within domestic blocks from bulk supply substation (if
present) to individual dwelling connection points per annual heat demand.
Cost for connections
prelims
2
£ / MWh
Prelims associated with connection costs above per annual heat demand. Prelims
allow for construction overheads, management fees, and other related expenses
not directly attributable to the DH components installation.
Overhead and profit
connections
2
£ / MWh
Overhead and profit associated with connection costs above per annual heat
demand.
SUM OF
DISAGGREGATED
CONNECTION COSTS
£ / MWh
Summation of the disaggregated costs above per annual heat demand. NOTE –
THIS MAY NOT BE THE SAME AS FOR THE AGGREGATE FIGURE PROVIDED BY THE
SCHEME AND PROVIDED IN BOLD ABOVE.
Heat Network
£ / MWh
Aggregate cost for all components making up the heat distribution network per
annual heat demand.
Mechanical capital cost
– total
£ / MWh
Cost of pipes and mechanical installation per annual heat demand.
Civil capital cost – total
£ / MWh
Cost of civil installation works per annual heat demand,
Preliminary capital cost –
total
3
£ / MWh
Prelims associated with network installation costs above per annual heat demand
Prelims allow for construction overheads, management fees, and other related
expenses not directly attributable to the DH components installation.
Overhead and profit –
total
£ / MWh
Overhead and profit associated with network installation costs above per annual
heat demand
SUM OF
DISAGGREGATED HEAT
NETWORK COSTS
£ / MWh
Summation of the disaggregated costs above per annual heat demand. NOTE –
THIS MAY NOT BE THE SAME AS FOR THE AGGREGATE FIGURE PROVIDED BY THE
SCHEME AND PROVIDED IN BOLD ABOVE.
Ancillary plant
associated with network
(eg pumping, treatment,
etc)
£ / MWh
Ancillary plant cost per heat demand
Costs for ancillary plant associated with the network operation, but not
included as part of the distribution pipework. These may include equipment
such as pumping, water treatment, controls, etc. Where cost data has been
provided, little additional information has been provided by schemes. However
the data sheet requested these costs to be additional to the Heat Network costs
above and so additive.
Thermal store
£ / MWh
The cost of the thermal store which is usually a water tank per annual heat
demand served.
46
CAPITAL COSTS
Tables 3 and 8)
DESCRIPTION
Network capital costs
per length (main
network-buried)
£/m
Cost of main heat distribution network divided by length of network. Excludes
internal pipework.
Network capital costs
per length (internal
pipe)
£/m
Cost of internal distribution pipework divided by length of internal pipes (non-
bulk schemes).
Substations cost per kW
capacity
£/kW
Cost of heat exchangers and associated equipment for large customers and bulk
supply points Expressed as £ per kW capacity of substations on the network.
Domestic HIUs cost per
dwelling
£/dwelling
Cost of individual Hydraulic Interface Units for dwellings (not including heat
meters)
Heat meter cost per
building (ND and bulk)
£/building
Cost of individual heat meter per building for nondomestic building or for bulk
schemes.
Heat meter cost per
dwelling
£/dwelling
Cost of individual heat meter per dwelling.
Thermal store cost per
m3
£/m3
Thermal store cost per storage volume of thermal store.
OPERATION COSTS
DESCRIPTION
Heat network
maintenance cost
£ / MWh
Cost per annual heat demand.
Maintenance costs directly associated with heat network and ancillary
equipment. Excludes heat substations and heat supply plant.
Heat network
management cost
£ / MWh
Cost per annual heat demand.
Management costs directly associated with heat network and ancillary
equipment. Excludes heat substations and heat supply plant.
Substation maintenance
cost
£ / MWh
Cost per annual heat demand
Maintenance cost of heat exchangers and associated equipment for large
customers and bulk supply points.
HIUs maintenance cost
£ / MWh
Cost per annual heat demand.
Maintenance cost of individual Hydraulic Interface Units for dwellings (not
including heat meters)
HIUs maintenance cost
£/MW
Cost per annual heat demand Maintenance cost of individual Hydraulic Interface
Units for dwellings (not including heat meters) Expressed as £ per kW capacity of
substations on the network.
Heat meter maintenance
cost
£ / MWh
Cost per annual heat demand across the entire network. Information was not
provided on individual meter heat delivery.
Avg annual staff cost for
metering, billing and
revenue collection
£ / MWh
Cost per annual heat demand.
Annual business rates
£ / MWh
Cost per annual heat demand.
© Crown copyright 2015.
Department of Energy & Climate Change
3 Whitehall Place
London SW1A 2AW
www.gov.uk/decc
URN 15D/022 - Assessment of the Costs, Performance, and Characteristics of
Heat UK Networks