Greenpeace 2030 Public Sept2015 Lowres

2030 Energy Scenarios
Dr. Daniel Quiggin

Max Wakefield
c 2015 Dr Daniel Quiggin

Copyright
WWW. DEMANDENERGYEQUALITY. ORG /GP REPORT
Licensed under the Creative Commons Attribution-NonCommercial 3.0 Unported License (the License). You may not use this file except in compliance with the License.
You may obtain a copy of the License at http://creativecommons.org/licenses/
by-nc/3.0. Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an AS IS BASIS , WITHOUT WARRANTIES OR
CONDITIONS OF ANY KIND , either express or implied. See the License for the specific
language governing permissions and limitations under the License.
First printing, September 19, 2015
Contents
Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.1
Introduction
1.1.1
1.1.2
The challenge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
The model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.2
Greenpeace:2030
1.2.1
1.2.2
Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.3
Conclusion
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.1
Time for change
15
2.2
Framing the UKs challenge
16
2.2.1
2.2.2
2.2.3
Non-renewable alternatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Opportunities: The falling cost of renewables . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Making it work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.3
The question & the test
2.3.1
2.3.2
2.3.3
The question . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
The test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
The SHED model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
The Greenpeace 2030 Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.1
Aims & Objective of the Greenpeace 2030 scenario
25
3.2
The Dispatch Protocol
26
3.2.1
3.2.2
3.2.3
Pragmatic & Climate driven sub-scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

CHP and base-load nuclear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Surpluses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
12
14
22
3.3
Inputs
27
3.3.1
3.3.2
3.3.3
Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Supply & balancing mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Emissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.4
Outputs & Results
3.4.1
3.4.2
3.4.3
3.4.4
3.4.5
3.4.6
3.4.7
Emissions . . . . . . . . . . . . . .
Demand . . . . . . . . . . . . . . .
Supply . . . . . . . . . . . . . . . . .
Load Factors . . . . . . . . . . . .
Balancing . . . . . . . . . . . . . .
Surplus power . . . . . . . . . . .
Importance of heating targets
Economics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.1
Costs of similar scenarios
59
4.2
Scenario divergences
59
Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.1
Electrification of heating
5.1.1
5.1.2
5.1.3
The role of traditional electricity demand reduction . . . . . . . . . . . . . . . . . . . . . . . 63

Post 2030 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Importance of meeting space heating targets . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
5.2
Domestic DSM
64
5.3
Decentralisation
65
5.3.1
Summer CHP & emergency CHP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
5.4
Batteries
65
5.5
Interconnectors
65
5.5.1
5.5.2
SHED shortcoming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Exports & surpluses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
5.6
Pragmatic & climate driven sub-scenarios
66
5.7
Dispatch Protocol
66
5.8
Limitations of hourly time-steps
67
5.9
A final note
67
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
6.1
Decarbonisation
69
6.2
CCS, nuclear and technical feasibility
70
6.3
Electrification
70
6.4
Balancing supply & demand
70
6.4.1
The importance of heating demand reduction targets . . . . . . . . . . . . . . . . . . . . . 71
43
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
43
44
45
49
50
56
57
63
5
6.5
Domestic demand side management (DSM)
71
6.6
Economic feasibility
72
6.6.1
6.6.2
Cost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Dispatchable generator load factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
6.7
The possibility & the imperative
The SHED model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
A.1
Overview
77
A.2
Renewables modelling
78
A.3
Dispatch protocol
78
A.4
Demand Modelling
78
A.4.1
Heating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
A.5
Electric Vehicles
82
A.6
Line Losses
82
A.7
Non-domestic Loads
82
Weather stations & renewable modelling . . . . . . . . . . . . . . . . . . . . . 83
73
Introduction
The challenge
The model
Greenpeace:2030
Design
Findings
Conclusion
1. Executive Summary
1.1
Introduction
"The UKs consumption [of energy] cannot continue to rise indefinitely.... if it
is to make an effective contribution to a global reduction in greenhouse gas
emission"
(Energy and Climate Change Committee 2012)
This report is a direct response to the increasingly urgent need for industrialised
countries to rapidly decrease their carbon emissions to prevent a worsening climate
crisis.
It positions itself within the context of the UKs policy and legislation: we treat the
Climate Change Act 2008 requirement for an 80% cut in emissions on 1990 levels
by 2050 as a starting point for a more ambitious carbon reduction plan. This reflects
the latest climate science published by the UNFCCC. The cumulative effect of carbon
emissions guarantee that action taken sooner will have a far greater chance of avoiding
the worst outcomes of a climate crisis; this report therefore concerns itself with 2030
targets.
Our report aims to make a significant contribution to enabling the Committee on
Climate Changes (CCC) preference for a transition to a low carbon economy, spearheaded by electrification of demand traditionally delivered through other means; and
decarbonisation of the power sector that provides electrical supply to meet such demand.
Only through the former can the environmental benefits of the latter be extended to
demand types currently delivered through the direct combustion of fossil fuels: whether
the burning of gas in a domestic boiler, or of diesel in a family car. And only through
that extension can a comprehensive decarbonisation of the overall UK economy be
achieved.
This contribution takes the form of employing an advanced modelling process to
design, test and iterate a 2030 energy scenario - in collaboration with Greenpeace UK that can demonstrably overcome the specific technical, infrastructural and engineering
problems associated with migrating to a radically decarbonised power sector. Indeed, a
power sector that we know will have significant extra demands made on it in the future
as a consequence of electrification and population growth.
Chapter 1. Executive Summary
Figure 1.1: Power supplied by gas and electricity grid over one week in January. Source
: Department of Energy and Climate Change (2012b).
1.1.1
The challenge
Both electrification and decarbonisation, as policy priorities, presents their own challenges. By unpacking each in turn we can see how, when combined, these challenges
magnify one-another.
Electrification
Electrification refers to the intention to move substantial proportions of energy demand,

primarily space heating and transport requirements, onto the electricity network by 2030
- with more to follow before 2050. The extent to which this is achieved has very significant
knock-on effects for the size and nature of electrical demands that the National Grid will
be expected to service.
Figure 1.1 shows the difference between the overall total UK electricity demand in
comparison to the demand for gas, which is predominately used for space heating via
gas boilers.
The size of the gas demand, in comparison to the total electricity demand, should
make clear why electrification provides a problem. Even taking a proportion of this
demand (this report will suppose 25 %) and transferring it onto the electricity network is
a new and very sizeable challenge.
Furthermore, demand is not flat (i.e. is not equally spread across each hour of a
24-hour cycle). This is particularly true in the case of domestic space heating which
varies according to time of day, producing large peaks in demand in the mornings and
evenings of, predominantly, winter days. Thus, by transferring a proportion of heating
demand onto the electricity network, we also transfer a proportion of the associated
peaks in demand onto the electricity network. These in turn combine with existing peaks
driven by traditional electricity demand1 (also generally in mornings and evenings) to
1
Traditional electricity demand is the electrical demand from both domestic and non-domestic electricity
consumers, with no contribution from electric vehicles, heat pumps or economy seven. It is the future
electrical demand that is similar to current electrical demands, but with economy seven or resistive heating
removed.
1.1 Introduction
produce larger overall peak demands on the electricity network.
Figure 1.2: Transfer of heat and hot water demand from gas to electrical network.
Source : Wilson et al. (2013).
Larger peaks in demand require larger reserves of generation capacity in order

to meet that demand - failure to do so results in blackouts or brownouts. The main
consequence of electrification is the need to deal with increased demand peaks.
Figure 1.2 shows the changes to electrical demand with 30% of non-daily metered
(NDM) heating transferred to either resistive heating technologies (COP1) or heat pumps
(COP3), where the electrical demand is scaled by the coefficient of performance (COP)
of those heat pumps. Wilson et al. (2013) find that if electrification of 30% of NDM
heating demands were provided solely provided via heat pumps daily demand would
increase by around 25%.
Key Information 1.1 Throughout this report a 2050 energy scenario developed by
the DECC which is described by its strap line as; "Higher renewables, more energy
efficiency", is utilised to compare the inputs and outputs of GP:2030. This DECC
scenario is referred to as DECC:2050:Renew and it should be noted that it is a
2050 rather than 2030 scenario. DECC:2050:Renew was utilised as a comparator or
benchmark as;
1. SHED has previously been used to model DECC:2050:Renew, hence the
inputs and outputs are available to the authors.
2. DECC:2050:Renew is a scenario that is characterised by high renewable
integration and comparable demand reduction.
Decarbonisation
Decarbonisation refers to the reduction to as close to zero as possible of the carbon

intensity of energy delivery. The carbon intensity of certain energy uses is fixed by
the form of delivery: the fuel burned in a cars internal combustion engine emits a
certain amount of carbon when burnt. To decarbonise car travel, therefore, the form
by which energy is delivered, fuel, must be substituted for electricity, which is able
to be decarbonised. We are therefore concerned with the overall carbon intensity of
10
electrical power delivered. The carbon intensity of the UKs power sector in 2015 is
500geq CO2 /k W h - the CCC has identified a target of 50-100geq CO2 /k W h by 2030.
This report aims to reach as close to 50geq CO2 /k W h by 2030, in light of worsening
climate science and the clear benefits of mounting earlier action.
Decarbonisation, by a factor of over
10, is no small task. Broadly, it involves
the substitution of centralised, dirty, fossil
fuel generators by cleaner, decentralised
renewable generators, such as wind and
solar. Depending on your method of analysis, renewables are either considered
zero-carbon (if life-cycle considerations
are not made), or low carbon (if they are).
Figure 1.3 shows the relative life-cycle
analysis (LCA) carbon densities of different electricity generation technologies.
Unless reliant on unproven carbon
capture and storage technology (CCS)2 ,
or expensive and controversial nuclear
power stations (which this report chooses
not to consider for reasons elaborated on
in Section 2.2.1), a future decarbonised
electricity system will necessarily involve
a high degree of renewables deployment.
This, in turn, brings its own peculiar challenges - challenges which a "traditional"
electricity system is protected from.
With a high level of renewables operational within a system the nature of
supply - and control over, and expectation
of, supply - changes. Under a traditional
system, operators rely on the knowledge Figure 1.3: Emission factors per technolthat, more or less, the full extent of the ogy. Source : Intergovernmental Panel on
electricity networks generation capacity Climate Change (2011)
can be called on as and when needed to
match demand. Under a renewables-dominated scenario this is no longer the case:
supply varies depending on the weather (sunny, cloudy; windy, still) and time of day supply is therefore "variable" or "intermittent", as opposed to "dispatachable"3 There is
no guarantee that peak supply will be simultaneous with peak demand; herein lies the
problem.
Combining the two
Consider a future in which we continue to generate our electrical power in much the
same way it is today but choose to electrify significant portions of heating and transport
2
Carbon Capture and Storage (CCS); a technology currently being developed designed to be fitted to
fossil fuel generators that captures CO2 emissions from those generators in order to reduce the emission
intensity factor of those generators. The captured carbon is then transported and stored to prevent
emissions entering into the atmosphere.
3
Dispatchable generator; an electrical generator capable of varying its output in accordance to the
demand for electrical power, these are generally large scale centralised power stations.
1.1 Introduction
11
demand. Such a choice would require the building of additional traditional, centralised
and dispatachable generation capacity, ready to meet the increased demand peaks
resulting from electrification.
Now consider a future in which we continue to demand electrical power in much
the same way as we do today, i.e. do not electrify significant portions of heating
and transport demand, but do decide to move to a predominantly renewable power
generation system. In this context the difficulties associated with balancing supply and
demand using variable generation (renewables) would be mitigated partially by peak
demands remaining at historical levels.
Both scenarios attempted alone would contain their own difficulties - but when
combined the challenges inherent in each multiply those of the other: electrification
increases the size of demand peaks on the electricity network; while decarbonisation
(via renewables) in turn decreases the predictability of supply intended to meet those
(now increased) peaks.
This is the dilemma facing the transformation of the UKs power infrastructure. It
must be resolved in order for the UK to make good its obligation to rapidly decarbonise
across the board. It is to this specific problem that this report turns its attention.
1.1.2 The model
The model used in this report, SHED (Smart Household Energy Demand), was built
by Dr. Daniel Quiggin during his PhD in the Centre for Doctoral Research in Energy
Demand, a joint collaboration between the Energy Institute at UCL and Loughborough
University. SHED was built on previous research by Dr John Barton who created the
Feasibility of Energy System Assessment Tool (FESA), utilised in the modelling of the
Transition Pathways to a Low Carbon Economy project energy scenarios (Barton et al.,
2013) and utilised in widely cited academic modelling of energy scenarios. SHED was
developed initially to analyse three 2050 energy scenarios produced by the Department
for Energy and Climate Change (DECC), and subsequently to meet gaps in the toolkit
available to academics, energy planners and policy makers in designing future energy
systems:
First, it implements new methods to accurately model heating demand, enabling the implications of heat electrification to be far more accurately appreciated.
Second, it draws on hourly data for all demand inputs (including e.g traditional electricity demand), and on hourly weather data. This enables it to
match modelled hourly demand with renewable supply, which is determined using
weather inputs.
Third, it incorporates demand and supply data from a period of 11 years,
enabling rigorous testing of a given energy scenarios resilience.
Fourth, it enables the requirements of Demand Side Management (DSM) to
be modelled at a national level, before being disaggregated to (and quantified at) the household level. This enables the impact of flexibility in the demandside of future energy scenarios, assumed in other modelling exercises, to be
accurately assessed for its technical utility and likely social acceptability
By introducing each of these features into the modelling environment, SHED permits
the following:
1. an accurate assessment of the ability of a modelled electricity system to enable
supply-demand balancing over a period of over a decade
2. a measure of such a systems carbon intensity
3. the requirements that would be placed upon households to shift or reduce
12
their demand during given periods, depending on the priority given to DSM in an
energy scenario
In doing so it can help to answer the question of what an electrified and decarbonised
energy system will need to look like, and what our priorities should be in building it.
1.2
1.2.1
Greenpeace:2030
Design
GP:2030 was designed by Greenpeace UK and Demand Energy Equality with the
following objectives:
1. decarbonisation within the CCC targeted 50 - 100geq CO2 /k W h emission intensity factor, whilst;
2. achieving this figure in the absence of carbon capture and storage technology
deployment, or the building of new nuclear power stations;
3. being technically feasible;
4. electrifying a substantial proportion of transport and heating to deliver emissions
reductions in those sectors;
5. balancing supply and demand - ensuring the same, or an improved, guarantee
of security of supply as is currently enjoyed in the UK;
6. ensuring the worst-case scenario impact of demand-side management on household consumption of energy is, nevertheless, likely technically and socially plausible, and;
7. being economically feasible
In creating GP:2030 decisions were made regarding both the expected (and targeted)
nature of demand in the scenario, and the make-up (and sequencing) of generation and
balancing technologies - including DSM.
Figure 1.4(a) shows the annual demand for GP:2030, the Department for Energy
& Climate Change (DECC) 2050 Renewables scenario (DECC:2050:Renew), and
historic demand over 11 years between 2001 - 2011. It demonstrates the major demand
reductions assumed across both scenarios, particularly in domestic space heating.
Grappling with the challenge of significant demand reduction in this area will hold great
sway over the success or failure of a strategy of decarbonisation via electrification.
Figure 1.4(b) shows the generation capacities of GP:2030 and DECC:2050:Renew,
with the notable absence of new nuclear (some existing nuclear capacity is expected
still to be operational in 2030) and CCS from the GP:2030 scenario. GP:2030s total
generation capacity is roughly double that of current levels.
1.2.2
Findings
Using SHED to model GP:2030, we conclude that:
Radical decarbonisation of the power sector in the UK is possible. GP:2030
achieves either 77.9 geq CO2 /k W h (Pragmatic) or 51.2 geq CO2 /k W h (Climate
driven), depending on the extent to which gas-fired power stations are called on to
supply net demand after renewable supply and balancing mechanisms.
Achieving this by 2030 is technically feasible - no significant assumptions are
made about the innovation of new technologies in GP:2030; instead informed
assumptions were made regarding the application of predominantly developed
technologies.
1.2 Greenpeace:2030
13
(a) GP:2030 generation capacity, compared to (b) Demand parameters from GP:2030 and
DECC:2050:Renew.
DECC:2050:Renew, compared to historic demands. Annual demand in T W h/yr .
Figure 1.4: Overview of supply & demand
Electrification of substantial proportions of transport and heating is possible, while maintaining constant supply-demand balancing. GP:2030 experiences
no periods of deficits (blackouts of brownouts).
Load factors on combined-cycle gas turbine (CCGT) generators are maintained at an economically viable level of 23.6 % within the pragmatic GP:2030
sub-scenario. In 2013 load factors for CCGTs fell to a record low of 28%.
In achieving such outcomes:
Ambitious domestic heating demand targets must be met. Without a -57.2 %
change in annual demand by 2030, supply-demand balancing will not be feasible
and blackouts and brownouts will occur.
Domestic DSM plays a modest role in mitigating periods of deficit . Fewer
than 7 periods in which total demand shifting exceeds 3GW occur on average
each year - however questions remain about the feasibility and social acceptability
of DSM roll-out.
Figure 1.5(a) shows the increase in demand within GP:2030 if heating targets are
not met for January and February weekdays - in this case we would expect 47 hours
of deficits over 11 years, totaling 141.9 GW h. Additionally, we would expect domestic
DSM requirements to become onerous on households (and thus unworkable). This
illustrates starkly the importance of meeting heating reduction targets.
Figure 1.5(b) shows total domestic contribution to balancing per month in GW h via
DSM. We find that, in a scenario with significant combined heat and power (CHP)4
supply (21.5GW ), the periods with the most significant balancing requirements are in
the summer months when CHP generation - following heat demand - is low.
Finally, by comparison with a fully costed high-renewables scenario developed by
Poyry (2011) we indicate that GP:2030 is economically feasible.
4
Combined heat and power (CHP); a type of generator that produces heat simultaneously to electrical
power, there are many forms of CHP units fuelled by different liquid or gas fuels.
14
(a) Increase in electricity demand due to a fail- (b) Domestic DSM contribution towards balancure to meet heating targets in GP:2030 for Jan- ing per month in GWh.
uary/February weekdays through 2030, based
on 2001:2011 data.
Figure 1.5: Overview of importance of meeting heating targets and monthly role of
domestic DSM
1.3
Conclusion
We argue that building an ambitious power sector able to take up the gauntlet laid down
by the urgency of decarbonising the UKs economy is certainly achievable. Renewable
technologies, falling in price, are proven and deployable. Demand reductions, though
testing of our often profligate use of energy, are realistic with sufficient political will and
appropriate policy-making. Increased information flows throughout the grid can enable
more rapid responses to fluctuations in demand and supply.
However, results indicate that scenarios (such as GP:2030) with a high degree of
heat electrification, combined with intermittent renewable generation, are susceptible to
additional supply-demand balancing requirements. In consequence, meeting heating
demand targets is a prerequisite for achieving electrification targets: even modest levels
of heat electrification result in large increases in peak electrical demand. The average
electrical peak demand, across published energy scenarios treated in Quiggin (2014),
rises by 15.3% when heating demand remains at historic levels, compared to when
targets are met. Electrification and decarbonisation are not simple substitutions for
existing energy forms. If electricity is the medium by which a reliable and clean energy
future is to be delivered, then heating demand reduction must be achieved alongside
heating electrification.
This report wishes to emphasise above all else, then, that our relationship with
energy must change: at a national, organisational, household and personal level. In
particular we must use less energy to heat our homes sufficiently for safety and comfort.
Failure to achieve this will deny us all the possibility of building a progressive and
ambitious energy infrastructure we can rely on into the coming century. In this report we
do not treat that as an option.
Time for change

Framing the UKs challenge
Non-renewable alternatives
Opportunities: The falling cost of renewables
Making it work
The question
The test
The SHED model
2. Introduction
2.1
Time for change

The UKs energy system is on the cusp of, arguably, its first real revolution since the
inception of what was to become the National Grid in 1926. Its infrastructure, control,
operation, governance, ownership, technologies and outcomes must - and will - change
rapidly until, and beyond, 2030 if an energy system fit for the future is to be realised.
The immense complexity of securing this - and the overlapping technical, political,
social and economic issues (and interests) such a task incorporates - ensure that
no single pathway to an energy future is yet the subject of either political or public
consensus. Perhaps this is unlikely to ever be the case. Yet, this is a crucial moment:
the Paris UNFCCC COP21 climate talks will challenge world governments to commit
themselves to the era-defining task of preventing the worst excesses of a climate crisis;
while the UK is legally bound to meet its own 80% CO2 reductions on 1990 levels by
2050 - a target already outdated by emergent climate science. Furthermore, public
trust in the major UK traditional utilities - the "Big 6" - is at an all-time low, as consumer
energy prices have continued to grow well in advance of inflation.
Across the industrialised world signs of a seismic shift in the way energy can be
delivered is underway: the price of renewables, and solar in particular, continues to
plummet; radically changing the traditional economics of energy generation.
"The reason solar-power generation will increasingly dominate: its a technology, not a fuel. As such, efficiency increases and prices fall as time goes
on. The price of Earths limited fossil fuels tends to go the other direction."
(Bloomberg, 2014)
The implications are startling, and traditional centralised mega-utilities are unable to
participate too keenly in the coming renewables surge without undermining their existing
business models.
However - there is no silver switch. The technical implications alone of creating
energy systems able to meet legally binding CO2 reductions targets (and, indeed targets
that necessarily go further) are significant, and will stretch our engineering and scientific
expertise. Officially, the need to decarbonise the UKs power sector (leading the rest of
the economy), is accepted across the political spectrum. Yet proposals for achieving this
16
Chapter 2. Introduction
are often dependent on imperfect modelling exercises, unable to demonstrate strategies

competencies in delivering the reliability of supply the UK population expects, and that
its economy requires. This is unsurprising: the size and quality of data sets, and model
refinement, required is substantial, and evolving over time.
Without clear and rigorous assessments of the opportunities and tasks ahead neither
the public, nor local campaigners, NGOs or Government will be able to campaign, plan
or act for the changes we urgently need.
2.2 Framing the UKs challenge

Successful decarbonisation of the UKs economy, in-line with or in advance of stated
targets, depends on the decarbonisation of the UK power sector - energy that is delivered
to consumers through the national electricity grid. The majority of what scientists and
society classify as "renewables" produce energy in the form of electricity, so the primary
focus on the electricity system for the UKs decarbonisation is both logical and intuitive.
However, its role in decarbonisation becomes even more vital - and difficult - when
the implications of decarbonising other elements of the UKs energy use, notably the
heating of domestic and industrial properties and motorised transport are considered.
Currently, 83% of UK homes are heated by gas boilers (Baker, n.d.), supplied
through a national gas grid - and 60% of the average UK household annual energy bill
comprises paying for them. This is despite the fact that boilers are predominantly used
to heat homes during the winter months. Burning gas in domestic boilers cannot be
decarbonised - but homes must be heated. So a substitution for an alternative energy
form, electricity, must eventually be achieved.
The largest delivered1 domestic heating peaks reach 160GW in the mornings and
evenings of the coldest winter days (Quiggin, 2014). The UKs total electricity generation
capacity in January 2015 stood at 85GW . Even accounting for the performance gains
enabled by heat-pump technologies, to substitute electricity for gas as the principle
energy form for domestic heating delivery requires over a 60% increase in our entire
electricity generation capacity. Given the majority of that capacity is fossil fueled and
will need to be taken off-line to meet carbon reduction targets, this task is, in fact, more
formidable than it first appears.
Key Information 2.1 The terms "dispatchable" and "non-dispatchable" generators are
used throughout this report. So too are the terms "rampable" and "intermittent". A
dispatchable generator is an electrical generator capable of varying or "ramping"
its output in accordance to the demand for electrical power, these are generally
large scale centralised power stations. A non-dispatchable generator is an electrical
generator whose output is dependant on weather and climatic conditions, its output
is referred to as "intermittent" or variable as these generators do not respond dynamically to electrical demand and their output varies in time with changes in the natural
resource (e.g. wind speed or solar radiation intensity).

And this is true for more reasons than one. With the necessity to bring more and
more renewables online comes the challenge of intermittency. While much of the UKs
existing generation capacity lies in "disptachable"2 or "rampable" centralised power
1
"Delivered" energy; this is the energy demand at the point of use, rather than the energy in the fuel
consumed.
2
Dispatchable generator; an electrical generator capable of varying its output in accordance to the
17
stations, a future generation portfolio dominated by renewables will be subject to the

whims of Britains famously changeable weather. This provides the possibility for both
under and over-supply of power into the grid depending on simultaneous demand. While
over-supply is easier to deal with (notwithstanding economic consequences that will be
considered later in this report), under-supply can lead to brownouts and blackouts3 - a
political non-starter and a serious economic disrupter.
2.2.1
Non-renewable alternatives
Several published energy scenarios (Electricity Networks Strategy Group, 2010; Department for Energy and Climate Change, 2009; Department of Energy and Climate
Change, 2010; Ofgem, 2010a; Ault et al., 2008; Foxon, 2013; Department of Energy
and Climate Change, 2010; Barnacle et al., 2013; Department of Energy and Climate
Change, 2010) reserve large portions of electricity generation capacity for technologies
that can supply power in a more traditional, predictable and dispatchable manner. In a
low-carbon scenario the permissible technologies are nuclear or CCS fitted fossil fuel
power stations. However, this report chooses to disregard both in designing future-proof
ed energy supply systems due in part to Greenpeace policy (Greenpeace, 2008), but
also because of substantial delivery risks that apply to both technologies. It should be
noted that the UK currently intends to build at least one new nuclear power station at
Hinkley Point in Somerset (see below), with the political consensus generally agreed
that further nuclear capacity will need to be built also. In this respect this report assumes
a major divergence from current political trajectories.
Carbon capture and storage
CCS is not, despite its regular inclusion in major party policy portfolios and industry
narratives, a mature technology - and is not in commercial operation, at scale, anywhere worldwide to the authors knowledge. Nor does this possibility appear to be
something we can expect in the near future. Even if it were, putting its potential cost to
one side, the carbon intensity reductions available through use of the technology are
unimpressive. IPCC rate CCS gas at 245gCO2 eq/k W h, compared to its unmitigated
cousins rating of 469gCO2 eq/kWh - under a 50% reduction (for coal, unsurprisingly,
the figures worsen: for CCS fitted we find 396 CO2 eq/kWh, as compared to unfitted at
1001CO2 eq/kWh) (IPCC, 2012). Inclusion of an untested, uncosted and even theoretically under-performing technology in energy scenarios can only be justified in the
absence of recourse to workable alternatives. Our findings suggest strongly this is not
the case.
Nuclear
Nuclear, unlike CCS, is a mature technology: the first large scale power station was built
in the UK in 1956. Analysed purely in terms of its carbon intensity, nuclear performs well
at 16gCO2 eq/kWh (compared to solars life-cycle carbon intensity of 45gCO2 eq/kWh,
and winds 12gCO2 eq/kWh). However, it remains an extremely contentious form of
electricity generation, for several reasons.
First, it produces highly toxic, radioactive waste that presents a serious risk to human
populations if mismanaged. Storage of this waste, while feasible within the lifetimes of
those who benefit from the power that produced it, is nonetheless both expensive, and
demand for electrical power, these are generally large scale centralised power stations.
3
Blackouts and brownouts; are regional or local losses of power due to demand exceeding supply
resulting in automatic relays cutting off consumers
18
not morally insignificant when the burdens - financial and otherwise - placed on future
generations (and far beyond) are considered. In 2012 the yearly spend on nuclear
waste containment comprised 42% the DECC budget (The Guardian, 2012). While,
with notable exceptions, the UK nuclear industry has demonstrated an ability to operate
nuclear power generation without major disasters occurring - many believe the risk
remains.
Second, nuclear is an extremely expensive technology to implement, with very long
lead-in times. The most recent announcement, in October 2013, on new nuclear build in
the UK - 3.2GW at Hinkley Point C, Somerset - revealed the governments commitment
(on behalf of the taxpayer) to guarantee a "strike price" (minimum price) of 92.50/MW h
(UK wholesale electricity prices at the time of announcement were 48/ MW h) to EDF,
the French company proposing to build and operate the plant. The plant is not planned to
open until 2023, and will cost an estimated 25bn to build (at time of writing). This does
not include costs associated with decommissioning and ongoing waste management.
Third, while nuclear can provide a predictable base-load supply of electricity into the
grid, it is not a technology designed to be ramped - and the cost-effectiveness of nuclear
declines rapidly if it is regularly required to reduce or increase its production levels. In
a scenario with a significant renewables contribution (even with the presence of CCS
or nuclear), and increased peak magnitudes due to electrification of space heating,
non-rampable centralised power technologies are less attractive.
Finally, the arguable relationship between the civil nuclear industry and the military
nuclear industry provides another point of moral objection for many.
For these reasons this report similarly chooses to demonstrate that less problematic,
risky and expensive alternatives to centralised nuclear power exist and are capable of
meeting the UKs energy needs unaided by new nuclear capacity.
2.2.2
Opportunities: The falling cost of renewables

CCS and new nuclear have been excluded from this reports energy scenario on
the basis of better - and feasible - alternatives. This is not an assessment that has
been made solely on the basis of carbon intensity values, but also the economic and
infrastructural implications of committing to given technologies in real economic, social
and economic contexts.
There are clear trends in the the cost of renewables which are shifting the economics
of future energy provision in favour of investment in solar, wind and other technologies
previously thought to be too expensive and intermittent.
From the late 1970s, when solar first emerged as an applicable power source, its
cost - according to Bloomberg - has fallen by 99%, from just under 50/W to just under
0.50/W (authors currency conversion). In 2014 the cost of new wind power in the
US had fallen to 2.5 cents/kWh - the cheapest of any new energy source in America
(Shrubsole and Cameron, 2014).
As a case in point, the fall in the cost of wholesale electricity prices in Germany
is causing serious problems for major utilities firms, which hold enormous fossil fuel
assets. They are unable to embrace new technologies, for various reasons, as easily
and with such overwhelming market dominance as they have been accustomed to (not
least because doing so would undermine their existing, fossil fuel dependent, assets).
The rapidly worsening economics of traditional electricity supply in Germany caused
E.On to announce it would be splitting into two companies - one owning its old, fossil
assets - the other focusing on power distribution and investments in renewables. This is
likely a sign of things to come across the developed world - and a clear signal of the
19
disruptive power of renewables when the appropriate policies are in place.

However, it should be noted that the cost of renewables technologies, and rapid
decreases in them (most notably solar over the last 5 years) is not the complete picture.
As renewables claim a greater share of overall generation capacity from traditional
dispatchable generators the associated grid costs, required to integrate variable power
production, also increase. As soon as renewables move beyond providing marginal
power supply into the grid their variability begins to have a serious impact on supplydemand balancing. This requires investment in further technologies such as storage
and smart grid ugrades. The costs of storage, though falling, are not falling as fast as
the costs of, say, solar - and thus will have a negative impact on solars total cost as
long as it continues to be integrated into national grid infrastructures. The quest for low
cost, high density storage technologies will continue as long as renewable shares of
national generation portfolios are intended to increase.
Despite the economics of energy being vulnerable to the volatile global pricing of
oil - which can spell death for both extraction projects on the carbon frontier and green
investments alike - there are clear, and accelerating, trends towards the continued
reduction in costs for widespread renewable deployment to the detriment of traditional
forms of power generation. This report explicitly builds on the opportunities provided by
the changing economics of power to build a scenario technically workable, economically
defensible and socially desirable.
2.2.3
Making it work
In a scenario with significantly increased demand on the electricity network, due to heat
and transport electrification, and an electricity supply system characterised by greater
intermittency, the task of balancing supply and demand on a second-by-second basis
year-round within politically, socially and economically acceptable boundaries takes on
new meaning.
That we currently have access to such a secure supply of nationally delivered energy,
notwithstanding inequalities in access to that system, is a staggering engineering
achievement - but it is one aided by the availability of traditional, dispatchable, rampable
power generation.
Making renewables-based systems work will require large reductions in energy
demand for space heating to enable the transfer of a portion of space heating from
gas to electrical heat pumps. The average figure for this required reduction across six
prominent published energy scenarios is 47.5% within the domestic sector (Quiggin,
2014) - and even they are unable to demonstrate balanced supply and demand. Huge
investment in energy saving programmes (such as adequate insulation in both new
and old builds), improvement in technologies operating efficiencies and - crucially behaviour change will all be required to make any low carbon power system function
acceptably.
According to The Economist;
"Improvements in energy efficiency since the 1970s in 11 IEA member
countries that keep the right kind of statistics (America, Australia, Britain,
Denmark, Finland, France, Germany, Italy, Japan, the Netherlands and
Sweden) saved the equivalent of 1.4 billion tonnes of oil in 2011, worth $743
billion. This saving amounted to more than their total final consumption in
that year from gas, coal or any other single fuel. And lots of money is being
invested in doing even better: an estimated $310-360 billion was put into
energy efficiency measures worldwide in 2012, more than the supply-side
20
investment in renewables or in generation from fossil fuels."

(The Economist, 2015)
Demand Side Management
Demand side management (DSM, sometimes referred to as "demand side response",

or DSR) has been popular within policy and academic circles for some time as a
technological innovation able to mitigate some of the supply-demand balancing issues
presented by highly electrified, highly renewable energy scenarios. Traditionally supplydemand balancing would have been achieved through simply managing supply up or
down dependent on the nature of demand. Within a dispatchable scenario this would
be (assuming sufficient overall generation capacity) predominantly enabled through
the ability to ramp generators when needed and, at times of particular stress, invoking
pumped storage held in reservoirs. Within a renewables-dominated scenario, the ability
to manage supply upwards to meet peaks in demand is diminished significantly in an
effort to reduce reliance on fossil fuelled power. Nor is the level of supply, and therefore
extent of a possible deficit, able to be predicted as reliably given the intermittency of
large portions of the generation mix. There are several possible responses to supplydemand imbalances within a renewables scenario (which will be addressed in turn in
the report), but DSM is unique in that (instead of sourcing additional supply) it seeks
to incentivise (or in extreme cases, oblige), reductions in demand at peak times from
power consumers, whether domestic or industrial.
Industrial application of DSM is already in operation. Flexitricity is the first and
largest UK provider of national supply-demand balancing services enabling access
to the energy market for commercial clients and small generators. This is achieved
via aggregation and smart two-way communication (Flexitricity website, 2011) such
that "relatively small units down to individual pumps or compressors can participate".
National scale balancing services interact via the National Grid Short Term Operating
Reserve (STOR). Signals are passed to the Flexitricity Edinburgh based control room
which communicates with generators and loads around the country (Flexitricity website,
2011). Fast load shedding is performed on a minutely basis utilising large commercial
consumers who can temporarily shut down loads, such as air conditioning units and
large freezers. In return Flexitricity clients earn money for reducing demand at particular
times.
The technical, administrative and logistical feasibility of interacting with corporate
and large scale power users in this way has not been matched, thus far, in a domestic
setting. It should be clear why - the National Grid cannot negotiate contracts to provide
demand reduction at peak times with every UK household, let alone communicate
directly with each when needed. A process by which an entity aggregates the demands
of many households and acts as a node for the National Grid (or another entity) to
communicate with is needed instead.
This is enabled through new domestic DSM technologies: principally smart meters.
A basic smart meter acts as a two-way communication device between the grid and
a household unit: the grid is able to know the demand from a household, while the
household is able to receive rapid (15 minute) price signals for electricity, reflecting
the supply-demand balance on the grid in, almost, real-time. The assumption is that
householders will move demand forward or backward in time in order to access a
lower unit price. If enough households, aggregated, react in concert to price signals
peak demands can be "smoothed", in turn reducing the amount of spare capacity, both
renewable and non-renewable, needed to prevent shortages. Smart meter roll-out is
already government policy by 2020. Future smart meters will be able to automatically
21
manage demand from larger household appliances (e.g. fridge-freezers) in response to

price signals.
Figure 2.1: Illustration of national DSM facilitating reduced peak demand by shifting
demand to off peak periods, achieved using variable tariff structures.
Figure 2.1 illustrates the principle of DSM where peak demand is shifted in time,
forward or backwards, as consumers respond to a reduced electricity unit price either
side of the peak period, which is relatively more expensive. These variable pricing tariffs
are generally referred to as Time Of Use (TOU) tariffs. The illustration in Figure 2.1 is of
the aggregate response from many households. Individual consumers will, of course,
each respond differently in order to provide the aggregate response.
The Department of Energy and Climate Change (DECC) said in 2010:
"The amount of flexible demand assumed in different 2050 analyses varies
but the assumption used is generally between 20% and 30%."
(Department of Energy and Climate Change, 2010)
This indicates they expect 20 - 30% of demand will be available for DSM. A number
of questions arise in order to assess the likelihood households will respond positively
to this new role, how to prepare households, and indeed if such expectations are even
technically feasible:
How regularly will DSM be required?
Will particular times of the year be more onerous on households?
Is this "20 - 30%" of demand in one particular hour or over several hours?
What will DSM mean for individual households, of different socio-economic status?
In the existing published literature of energy scenarios, the above questions are not
quantified and answered. SHED, and GP:2030, attempt to answer them. In doing so a
more robust representation of the future electricity system is offered.
This is imperative, as it is unclear from current research how households are going
to respond to the need to their change demand patterns, via DSM. Certain studies have
shown that consumers who are made aware of consumption during peak times will
shift demand to less expensive periods (Lindley, 2010). TOU tariffs via smart meters,
in the Italian ENEL Telegestore project, have been shown to reduce consumption by
22
5-10% (Lindley, 2010). The Energy Demand Research Project (EDRP) trials of smart
meters in the UK has shown up to 4% energy savings, with loads shifted in time via
TOU tariffs by up to 10% (Ofgem, 2010b). Other studies that have investigated the
effectiveness of feedback on demand have shown 5-15% reductions (Darby, 2006).
Recent research also offers contrary evidence that there is a strong resistance by early
adopter households to change behaviours based on information from smart meters
(Hargreaves et al., 2010). Research by Hargreaves et al. (2010) on the EDRP, indicates
that certain households found smart meters to create a sense of "fatalism, despondency,
anxiety and even guilt". Untested at scale, DSM is the largest unknown of this energy
scenario; despite being modelled beyond what is available in other published scenarios.
Our approach to managing the risk of modelling such an unknown, to produce high-utility
results, is treated in Section 3.3.2.
2.3
2.3.1

The question
The reality of constructing a radically decarbonised energy system in the UK, beginning
with the power sector, thus presents many challenges. First, though, the task of
designing that system must be addressed, with urgency.
While the complexity of a feasible energy system cannot be downplayed, we can
ask three simple questions of any energy scenario purporting to hold the key to future
energy demands:
1. "Given the necessity to electrify forms of energy delivery currently
supplied otherwise, principally space heating, can the scenario be
reasonably thought to balance supply and demand at the tail end of a
January or February anti-cyclone?"
2. "Further to this; if combined heat and power (CHP) units are a major component of electrical supply, will summer months (when it is
uneconomic to run CHP units), result in a scenario supply - demand
imbalance?"
3. "Finally, is the scenario within politically, socially and economically
acceptable boundaries?"
Nevertheless, while the line of questioning can be simplified - beginning to respond
it is a formidable exercise.
2.3.2
The test
Consider why: a model fit for the purpose must be able to:
Incorporate a sufficient number of the many inputs to both national supply and
demand - and do so with a sufficient degree of accuracy; to, in turn
Assess the supply-demand balancing outcomes of a theoretical future scenario with sufficient "temporal resolution" (using data points close enough
together in time) to accurately characterise the demand peaks and supply variations that form the crux of problem; and
It must use a sufficiently vast data-set, representing several years of historical
weather and demand values, to be a useful predictive tool for the purposes of
designing such vital infrastructure.
2.3 The question & the test
23
Finally, a good model will be able to comment on the reality that workable scenarios
entail for the people energy systems affect most - citizens - particularly in light of the
inevitable introduction of domestic DSM.
Historically, energy models have struggled to meet the complexity of the task, while
data sets of appropriate quality, relevance and size have often been unavailable. What
makes this report significant is its presentation of results produced by a model that
meets each of the criteria listed above.
For example, before privatisation in 1990, the Central Electricity Generating Board
(CEGB) assumed that 85% of national generating capacity would be available during
winter months. The CEGB required a 24% capacity margin4 (calculated using equation
2.1), facilitating a Loss of Load Probability (LOLP) of 9%.
capacity margin (%) =
total available capacity peak demand

100
peak demand
(2.1)
Strbac et al. (2007) showed that to maintain reliability of supply at historic levels,
with increased non-dispatchable generation, significant reserve generating capacity of
dispatchable generators will be required. If the installed capacity of wind was 25GW ,
a minimum additional dispatchable reserve of 4.6GW would be required. This type
of methodology underpins the way in which many energy scenarios calculate the
generation capacity required to maintain reliability of supply.
Yet only recently has the electrification of heating been considered as having significant impacts on maintaining security of electrical supply (Wilson et al., 2013); consequently, it has not been considered simultaneously with an increase in non-dispatchable
generation. Historic capacity margins used within much of the literature to determine
reliability of supply are thus outdated: historically heating has not been electrified to the
extent required to decarbonise the overall UK energy system. This report meets this
problem head-on.
2.3.3
The SHED model

In this report we explore the outcomes of 2030 energy scenarios, designed by Greenpeace UK in a process facilitated by Demand Energy Equality, when subjected to
rigorous testing using the Smart Household Energy Demand (SHED) model.
SHED was created by Dr Daniel Quiggin during his PhD at Loughborough University
in the Centre for Doctoral Research in Energy Demand, based in the department of
Civil and Building Engineering. The centre is a joint collaboration between the Energy
Institute at UCL and Loughborough University, whose directors are Prof Kevin Lomas
and Prof Bob Lowe.
Dr Quiggin was supervised by Dr Richard Buswell, with oversight from Prof Kevin
Lomas. Dr Quiggin built on previous research by Dr John Barton who created the
Feasibility of Energy System Assessment Tool (FESA). FESA was utilised in the modelling of the Transition Pathways to a Low Carbon Economy project energy scenarios
(Barton et al., 2013), whose distinguished academics include; Prof Geoffrey Hammond,
Prof Peter Pearson, Prof Goran Strac, Prof Simon Watson, Prof Graham Ault and Prof
David Infield. In addition to the Transition Pathways scenarios, FESA was used in the
modelling of the Centre for Alternative Technologys Zero Carbon Britain report (Centre
4
The capacity margin of the electricity system is the percentage by which national available electricity
generation capacity exceeds the maximum expected level of demand (peak demand)
24
For Alternative Technology, 2014) - indeed the Zero Carbon Britain team consulted Dr
Quiggin on elements of their modelling process.
In creating SHED Dr Quiggin took FESA and built on the work by Dr Barton, expanding and improving on a model already held in high regard and utilised in widely
cited academic modelling of energy scenarios. SHED was constructed to analyse, in
particular, three 2050 energy scenarios produced by DECC. Dr Quiggin therefore has
a detailed understanding of third party energy scenarios and the implications of using
highly specialised modelling techniques to construct them.
Dr Quiggin received his PhD in December 2014 having passed his viva, which
was examined by two highly distinguished academics; externally, Prof David Infield
(University of Strathclyde) and, internally, Prof John Wright (Loughborough university).
Dr Quiggin completed a Masters in Physics at the University of Birmingham, and
worked on the ATLAS detector at the particle physics laboratory at CERN before moving
into the field of climate science, completing a second masters at the University of Bristol
in Earth System Science.
SHED improves significantly on other energy scenario modelling tools by:
First, implementing new methods to accurately model heating demand, enabling the implications of heat electrification to be far more accurately appreciated.
Second, drawing on hourly data for all demand inputs (including e.g traditional electricity demand), and on hourly weather data. This enables it to
match modelled hourly demand with renewable supply, which is determined using
weather inputs.
Third, incorporating demand and supply data from a period of 11 years,
enabling rigorous testing of a given energy scenarios resilience.
Fourth, enabling the requirements of Demand Side Management (DSM) to
be modelled at a national level, before being disaggregated to (and quantified at) the household level. This enables the impact of flexibility in the demandside of future energy scenarios, assumed in other modelling exercises, to be
accurately assessed for its technical utility and likely social acceptability.
The report will demonstrate, using SHED, that a radically decarbonised energy
system is possible by 2030 - not just possible to plan, but possible - with the right
policies and political leadership - to fund and build; confident it will meet the strenuous
demands placed on it by future populations.
This document and the research underpinning it has been reviewed by Prof David
Infield (University of Strathclyde);
"This is a useful report dealing with the complex issue of absorbing high
penetrations of renewable power generation in line with achieving challenging
reductions in carbon emissions."
(Prof David infield)
Key Information 2.2 It should be noted that although an hourly time step model such
as SHED is an improvement on the majority of energy scenario modelling, an hourly

time step fails to capture variations in supply and demand at the sub hourly level.
Covering variations within the hour will require some additional reserve margin and
this will have some added economic and carbon costs unless met by additional
storage or demand side management. SHED and modelling undertaken within
GP:2030, is however, a significant improvement when compared to other published
energy scenarios.
Aims & Objective of the Greenpeace 2030

scenario
Pragmatic & Climate driven sub-scenarios
CHP and base-load nuclear
Surpluses
Inputs
Demand
Supply & balancing mechanisms
Emissions
Outputs & Results
Emissions
Demand
Supply
Load Factors
Balancing
Surplus power
3. The Greenpeace 2030 Scenario
3.1
Aims & Objective of the Greenpeace 2030 scenario

The Greenpeace 2030 scenario ( GP:2030) was designed with the set of precise aims
of;
1. Decarbonisation within the CCC targeted 50 - 100geq CO2 /k W h emission intensity factor, whilst;
2. Achieving this figure in the absence of carbon capture and storage technology
3. Being technically feasible;
4. Electrifying a substantial proportion of transport and heating;
5. Balancing supply and demand - ensuring the proposal offered the same, or an
improved, guarantee of security of supply as is currently enjoyed in the UK;
7. Being economically feasible
In this section, we explore the decisions that were made in creating GP:2030. These
decisions apply not only to choices of energy generation technologies (such as offshore
wind) - and the extent of their deployment - but also to the deployment of "balancing"
technologies (such as battery storage) and to the ambition of demand reduction targets
in areas such as traditional electricity1 consumption and domestic space heating.
Of course, each decision in any area of a system has knock on effects for other
elements - the process of refining these choices (or "model inputs") was one of iteration:
making reasonable assessments and estimates, running the model, reassessing, and
so forth.
The final inputs represent the result of a number of compromises - between ambition
and cost; and between optimism and pragmatism.
1
Traditional electricity demand is the electrical demand from both domestic and non-domestic electricity
consumers, with no contribution from electric vehicles, heat pumps or economy seven. It is the future
electrical demand that is similar to current electrical demands, but with economy seven or resistive heating
removed.
Chapter 3. The Greenpeace 2030 Scenario
26
3.2

In considering the utility of GP:2030 it is important to understand how the UKs electricity
system would operate if the model were to be faithfully replicated in real life. This
requires us to look at the dispatch protocol which defines the order technologies are
deployed in, or processes initiated, as the system attempts to respond to a given
demand level at a given time.
The dispatch protocol, in this scenario, is broken down into five distinct phases:
Phase 1. Non-dispatchables: Uncurtailed renewable generation, base-load
nuclear (90% of capacity) and heat-led CHP
Phase 2. Non-fossil fuel dispatchables: Geothermal and hydro
Phase 3. Balancing mechanisms (1): domestic DSM, pumped storage, emergency CHP (EmCHP), centralised battery storage and interconnectors (import)
Phase 4. Fossil & nuclear dispatchables: combined-cycle gas turbines (CCGTs)
and rampable nuclear2 (10% capacity)
Phase 5. Balancing mechanisms (2): commercial DSM, decentralised battery
storage (DBS), further domestic DSM and interconnectors (export)
The logic for this dispatch protocol should be, in the main, self-explanatory once the
aims of GP:2030 are revisited: by invoking all non-fossil fuel supply options, and certain
balancing mechanisms, before calling on CCGTs to deliver extra power, the overall
carbon intensity of the system is lowered. However, in order to maintain economically
favourable load factors3 for CCGTs (essentially the extent to which the turbines are
actually used, and therefore the extent to which building them is attractive), Phase 3
only seeks to reduce net demand (after Phases 1 and 2) to as close to the generation
capacity of the installed CCGTs as possible - 19.5 GW. In instances where this is not
possible, Phase 5 is initiated to deal with residual demand (or, in the case of export
interconnectors, residual surplus). For further discussion of load factor implications,
please see 5.6.
3.2.1
Pragmatic & Climate driven sub-scenarios

This report will refer to two sub-scenarios within GP:2030: "Pragmatic"
(GP:2030:Pragmatic) and Climate driven (GP:2030:ClimateDriven). In the former the
dispatch protocol will follow the logic outlined above; in the latter Phase 3 will seek to
reduce demand (after Phases 1 and 2) to the lowest value possible (i.e. potentially
under the total generation capacity of installed CCGTs). This, in turn, reduces the loads
placed upon CCGTs which has two effects:
1. A reduction in the overall carbon intensity of the power sector, but simultaneously;
2. A reduction in the average load factors for CCGTs.
The reduced load factors result in unfavourable economics for the operation of the
CCGTs needed, reducing the chance of them being built (at least under traditional
build arrangements). For this reason the sub-scenario is labeled "Climate driven".
The comparison between the resultant load factors in each sub-scenario is covered in
Section 3.4.
2
see section 3.3.2 for discussion on the viability of ramping nuclear

Load factor; is the average load placed on a dispatchable generator divided by the generation capacity
of that generator
3
3.3 Inputs
3.2.2
27
CHP and base-load nuclear

The position of CHP and base-load nuclear generation in the dispatch protocol should be
noted. It may seem counter-intuitive to place such technologies alongside renewables
as "non-dispatchables": however, this decision is determined by the economics of
their operation. For CHP plants, it is uneconomical to run them following demand for
electricity - instead CHP units follow heat demand (which is supplied directly through
local heat networks). The electricity produced (effectively, following this logic, as a
by-product) and fed into the grid then affects the size of the surplus or net demand
of electrical power available at any given time, but cannot be called on to increase or
decrease its production (except in "emergencies", see Section 3.3.2). CHP plants thus
help to balance an electricity system servicing electrified heat demand twice over: first,
by reducing the overall heat demand on the electricity system via direct heat network
supply and, second, by contributing electrical power to meet the subsequent demand
experienced by the electricity system.
Nuclear power stations do have the ability to ramp their production levels up and
down, but doing so is expensive. For this reason, nuclear base-load runs at 90% of the
installed capacity. The final 10% can be called on in a dispatchable manner, in Phase 4.
For more information on the rampable nature of nuclear please see Section 3.3.2.
3.2.3
Surpluses
With a large capacity of renewable generation comes, as with deficit periods, increased
periods of surplus power. As storage of electricity is challenging at scale, both technically
and economically, careful thought within GP:2030 has been given to how surplus periods
are dealt with, as modelled under SHED, in a general sense. Precise details are given
in the following sections, under the relevant technologies utilising surpluses, such as
batteries. Surplus power can be stored in batteries (both centralised and de-centralised),
used to supply demands that are shifted in time due to DSM, used to pump water uphill
in pumped storage facilities to be used at a later time, or exported and sold to other
countries via interconnectors. In GP2030 exports only occur once any surpluses have
been utilised to the greatest possible extent domestically.
3.3
3.3.1
Inputs
Demand
The inputs to the model, as already mentioned, do not solely pertain to supply and
balancing elements. The first task of designing, and modelling, a future energy scenario
is to model the demand the system will be expected to supply.
This depends on initial, predictive, estimations of the expected:
Prevalence of and usage of electric vehicles, heat pumps and solar thermal
technologies (due to solar thermals impact on reducing heating demand);
Annual space heating demand (or, rephrased, the expected or desired level of
space heating demand reduction by 2030);
Annual water heating demand (or, rephrased, the expected or desired level of
water heating demand reduction by 2030);
Increase or decrease in traditional annual electricity demand, and;
Reisdual economy seven electricity demands.
SHED combines these input decisions with eleven years of hourly historical data.
At a high level (and from a simplified perspective), these annualised demand inputs
28
scale (up or down) the historic data such that eleven years of hourly future demands are
modelled, representing eleven consecutive years of 2030. SHED also calculates the
impact that, for example, solar thermal installations will have on the demand for heating.
The result is the predicted national hourly demands for a period of 11 years, against
which the scenarios generation capacity and balancing mechanisms can be tested for
their ability to deliver sufficient and reliable supply. For technical detail on the operation
of the model, see Appendix A and Quiggin (2014).
Table 3.1 shows the values assumed in GP:2030 for various demand inputs, and
factors that impact on demand, alongside the values assumed in the DECC 2050
renewables (DECC:2050:Renew) energy scenario (which does not, tested with SHED,
balance sufficiently):
Key Information 3.1 Throughout this report a 2050 energy scenario developed by
the DECC which is described by its strap-line as; "Higher renewables, more energy
efficiency", is utilised to compare the inputs and outputs of GP:2030. This DECC
scenario is referred to as DECC:2050:Renew and it should be noted that it is a
2050 rather than 2030 scenario. When comparing demand parameters ti should also
be noted that DECC:2050:Renew assumes 40 million households to GP:2030s 30
million, meaning that energy use per household is significantly higher in GP:2030.
DECC:2050:Renew was utilised as a comparator or benchmark because;
1. SHEDa , hence the inputs and outputs are available to the authors.
2. DECC:2050:Renew is a scenario that is characterised by high renewable
integration and comparable demand reduction.
Smart Household Energy Demand (SHED); the model used in this report to construct and test the
Greenpeace 2030 energy scenarios. For detail please see 1.1.2
Demand Parameter
Traditional Electricity (TWh/yr)
% households on economy 7 tariffs
Dom space heating (TWh/yr)
Non-Dom space heating (TWh/yr)
Dom water heating (TWh/yr)
Non-Dom water heating (TWh/yr)
Electric Vehicles (TWh/yr)
% Heat supplied by Heat Pumps
Number Households (millions)
Historic
329.4
5%
257
93.9
67
11.2
26
DECC:2050:Renew
304.5
10%
101.8
57.8
65.3
15.3
45
90
40
GP:2030 (%change on historic)

260 (-21.1 )
0 (-)
110 (-57.2 )
60 (-36.1 )
65 (-3 )
15 (34.2 )
32 ) (-)
25 (-)
30 (15.4 )
Table 3.1: Demand input parameters for GP:2030 and DECC:2050:Renew and the
change in demands relative to historic values. Note that heating values are delivered
energy rather than the embodied energy within the fuel supplying that energy.
Key Information 3.2 It should be noted that throughout this report, including within
Table 3.1 and Figure 3.1, space and water heating demand within the domestic and
non-domestic sectors is defined in terms of "delivered" energy demands. This is
the energy demand at the point of use, rather than the energy in the fuel consumed.
Quantification of space and water heating in this sense is integral to calculating
electrified heating demands as technologies, such as heat pumps, electrify the
3.3 Inputs
29
energy demand not the demand for fuel (such as gas). Hence the historic values
quoted in Table 3.1 and Figure 3.1 may be lower than the reader would anticipate if
the metric was to consider the fuel consumed.

Figure 3.1 focuses on annual demand values (over the 11 modelled years) divided
between different demand types, and compares them to the averaged historical data
(without any demand reduction assumptions) and the values chosen for the DECC 2050
Renewables scenario.
Figure 3.1: Demand parameters from each energy scenario. Annual demand in
T W h/yr .
In both the GP:2030 and DECC:2050:Renew scenarios significant average annual
demand reductions are assumed in all categories, except non-domestic and domestic
water heating. Each of the main demand categories are now explored in turn.
Space heating
Most notably, very significant reductions are assumed in domestic space heating where
GP:2030 envisages a -57.2 % change. This is in line with the average annual domestic
space heating reduction figure for DECC:2050:Renew for 2050. Being a 2030 scenario,
the GP:2030 space heating targets are significantly more ambitious: but this report will
repeatedly emphasise the enormous importance of reducing space heating demand
- particularly in the domestic sector - as a pre-requisite for a workable decarbonised
power sector when heat electrification is attempted in parallel.
The focus on non-domestic space heating demand is required due to the difficulty
of shifting the time of such demand, and thus reducing peaks, using DSM: ultimately
spaces need to be heated at certain times. If that fact cannot be substantially altered,
then the demand occurring at those times must be reduced more so than when dealing
with more mobile demand types. This explains why the space heating demand reduction
assumed for the non-domestic sector is not as radical: this demand already occurs
outside of the morning and evening peak periods.
Overall we argue that these national demand reduction assumptions must become
priority national targets in the near future.
30
Water heating
GP:2030 assumes a very modest change of -3 % in domestic water heating demands,

in line with DECC:2050:Renew. Domestic water heating demand is determined primarily
through behavioural choices, such as the temperature of basin water and whether baths
or showers are used for washing. Compounded by the fact that building characteristics
do not significantly impact water heating demands, it is therefore unlikely that large
reductions in domestic water heating demand will be made.
Heat pumps
The GP:2030 scenario assumes, in contrast to DECC:2050:Renew, that the prevalence

of heat pumps will be significantly lower: with 25 % of heat delivered via heat pumps
compared to 90%. Even if achieving 90% of heat delivery from heat pumps by 2050 is
feasible, 25% by 2030 is ambitious, given the relative infancy of the UKs heat pump
industry. Skills and understanding in the supply chain, combined with improved public
communication and project execution, will be key to overcoming consumer resistance.
Thus our assessment is based on current (and fairly low) rates of deployment and the
recognition of the cost and disruption currently associated with installation, as well as
the challenges highlighted by the Energy Saving Trusts 2010 field study (EST, 2010).
It is notable that the Renewable Heat Incentive (RHI) is currently unfunded beyond
2016; this will need to be addressed to ensure supply chain investor confidence. 24%
of total 2030 heat demand in the medium abatement scenario of the CCCs Fourth
Carbon Budget is expected to be delivered via heat pumps (Committe on Climate
Change, 2008), a target considered substantial but achievable with appropriate policy
action by WWF in 2014 (WWF, 2014). In both GP:2030 and DECC:2050:Renew it is
assumed that ground or water, as opposed to air, sourced heat pumps are employed
(to guarantee performance during winter) and that they provide 3 times as much heat
as they consume electrical energy, in line with the literature. This factor of 3 is termed
the coefficient of performance and is modelled as stable over all seasons, for more
information on the modelling of heat pumps see Appendix A and Quiggin (2014). No
resistive heating is included in GP:2030, to which air-sourced heat pumps revert when
they cannot function normally.
Traditional electricity demand
Traditional electricity demand is the electrical demand from both domestic and nondomestic electricity consumers, with no contribution from electric vehicles, heat pumps or
economy seven4 . For SHEDs predictive purposes, future traditional electricity demand
is similar to current electrical demands, but with economy seven and resistive heating5
removed. Traditional electricity demand is the single largest energy demand category,
within GP:2030, at 260 TWh/yr. Indeed GP:2030 assumes a more optimistic outcome
for demand reduction than DECC:2050:Renew based on increasing energy efficiency
of appliances and increased awareness of consumption.
Modelled in SHED the higher DECC:2050:Renew traditional electricity demands
contribute to deficits that cannot be closed, leading to supply - demand deficits. Hence
it is integral to maintaining supply - demand balance that traditional electricity demand
is reduced from current levels.
4
Economy seven; is a differential tariff which is cheaper at night time - encouraging consumers to move
demand to periods where demand is low. This type of tariff has been in operation since the 1980s in the
UK
5
Resistive heating; refers to heating provided by electricity, many current households on economy
seven tariffs have storage resistive heaters
3.3 Inputs
31
Electric Vehicles
Electric vehicles (EVs) will be a part of the future given the clear desire of the British
public to have access to private transport. The precise extent of their roll-out will depend
on improvements in technology, investment in national infrastructure (such as charging
points) and price competitiveness vs. internal combustion engine vehicles (which in turn
depends on battery development and the volatile price of oil). GP:2030 EV demand
is set at 32 TWh/yr, in line with DECC:2050:Renew which projects 45TWh/yr of EV
demand by 2050. 32 TWh/yr equates, (based on CCC vehicle number projections
(Committe, 2014); Government average mileage data (Department for Transport, 2014);
and estimates of power usage from Tesla and the International Energy Agency (IEA,
2013)) to 12.6 million electric vehicles travelling 12,640k m a year, using 0.2k W h/k m. It
has been assumed that EVs are not capable of providing power back to the grid, hence
EVs are unable to provide balancing services, which many studies have suggested
would be possible, but expensive (Kempton and Tomic, 2005). The driving and charging
profile used to model hourly EV demands is a modified DSM profile, under which DSM
has been accounted for (Acha et al., 2011). Hence the EV demand profile assumes EVs
are charged overnight and during the day between morning and evening peak demand
periods. EVs thus help to reduce or eliminate surpluses in the system more often than
they contribute to net demands. If however a non-DSM charging profile is assumed
within the modeling then the charging times would be more likely to coincide with times
of supply-demand stresses.
Solar thermal
Although solar thermal units do not consume or produce electricity it is important to

consider their impact on heating demands given these heating demands are later
electrified via heat pumps. GP:2030 is more optimistic about the potential for growth
than DECC:2050:Renew. While the latter assumes that only 1 in 40 properties will have
solar thermal installed by 2050, with an average installed power rating of 2kW, GP:2030
assumes that 1 in 10 will have the technology installed. Under the RHI currently 4,000
properties have solar thermal installed, representing 1 in 6,500 households. It must be
noted that physical limitations on the roll-out of solar thermal exist given the growth of
solar PV: roof space already occupied by solar PV panels prevents installation of solar
thermal in the future. In order for the growth of solar thermal to continue the RHI (or
equivalent policy) must be continued beyond the end of 2016, when it is currently set to
expire.
Number of households, build rates & energy efficiency
The UK is currently building between 125,000 - 130,000 new homes per year. These new
homes are required under building regulations to have a much greater energy efficiency
than the current building stock. Indeed, building regulations are under review and in
2016 it is likely that developers of new builds will be required to further improve heating
efficiency standards. There has been discussion of incorporation of level 6 of the "The
Code for Sustainable Homes" (CSH) within building regulations, which would mean new
builds would be required to reduce CO2 emissions to net zero. Hence space heating
demands would necessarily be reduced substantially in all new builds. It is unlikely that
by 2016 level 6 of the CSH will be implemented within the building regulations but a
significant move in that direction is anticipated. The likely implementation of building
regulations is that Fabric Energy Efficiency Standards (FEES) will be substantially
increased.
With energy efficiency standards increasing for new build properties, a substantial
32
component of the domestic space heating reduction targets of GP:2030 could be

achieved. Indeed GP:2030 anticipates the number of households within the UK will
rise from its current number of 26 million to 30 million, an increase of 15.4 %. This
increase would require 267000 households to be built per year, assuming a constant
yearly build rate. This build rate should be considered within the context of the current
build rate (above), but also the UK Governments aspiration for 300,000 per year, as
announced by Business Secretary Vince Cable in May 2014. Hence the build rate
required within GP:2030 is within Government policy aspirations and indeed roughly
double the current build rate. It should also be noted that within DECC:2050:Renew 40
million households are anticipated by 2050.
3.3.2
Supply & balancing mechanisms

With demand inputs specified, hourly demands are quantified for the eleven years of
data within SHED, against which supply must be matched - enabling testing of the
scenarios overall ability to balance supply and demand.
The supply profile is generated in SHED by combining hourly historical weather data
for the period 2001 - 2011 (for more information on weather data and the modelling of
renewables see Appendix A and B, as well as Quiggin (2014)) with decisions on:
Generator capacities for all generation technologies;
Pumped storage capacity;
Other storage capacity (e.g. centralised or decentralised batteries);
The dispatch protocol sequence, and;
Demand side management
The hourly historic weather data enables us to predict the expected supply of
uncurtailed, non-dispatchable, variable renewable power. Hourly capacity factors are
generated from the historic weather data, regional resource constraints (such as sea
bed depth for offshore wind) and technology technical characterisations (such as power
curves for wind turbines, which define the wind turbine output at various wind speeds).
Key Information 3.3 The capacity factor for a given renewable generator is the ratio
between its actual output over a given period of time, and its potential output if it were
able to operate at rated nameplate generation capacity.
Detailed modelling descriptions of hourly capacity factors are not given within the
report, for this information please see Quiggin (2014). However the mean capacity
factor for each renewable generator, over the 11 years of historic data, is given below.
Depending on modelled demand, a net demand or surplus will be defined for any
given hourly period. In the case of a net demand, the dispatch protocol is initiated, as
detailed at the beginning of this section, until the net demand is reduced to zero. In the
case of surplus, excess power is redirected to DSM, various forms of storage, exported
and finally turned into hydrogen for hydrogen vehicles - according to the prioritisation of
importance for surplus absorbers described above. This, amongst other things, defines
the "state of charge" of storage (either pumped or battery), which in turn defines the
ability of storage to reduce future net demands. This is why winter anti-cyclones provide
such a challenge to renewable scenarios: if the wind doesnt blow for 5 days the ability
of storage to reduce net demands driven by peaks in heating demand at the end of that
period is severely diminished, in the absence of opportunities to re-charge.
Table 3.2 shows the input values for different categories of supply and balanc-
3.3 Inputs
33
ing technologies for both GP:2030:Pragmatic and GP:2030:ClimateDriven, alongside

DECC:2050:Renew. It should be noted that both the GP:2030 sub-scenarios project the
same installed capacities of generators and balancing mechanisms as well as identical
demand targets. The difference between the two sub-scenarios arises from balancing
mechanism alterations, as described in Section 3.2.1.
Key Information 3.4 It should be noted that all generators, be they renewable or
CCGTs and nuclear power stations, are assigned an "availability factor". The availability factor is the amount of time that a generator is able to produce electricity over
a certain period, divided by the amount of the time in the period. All renewables
are assigned an availability factor of 85%, meaning that at any moment in time
only 85% of the national capacity is available for generation, due to maintenance
and equipment failure. The availability factors for each technology can be found in
Quiggin (2014). In certain cases the availability factor is stated if the technology has
a particularly low or high availability factor. It should be noted that the values in Table
3.2 and Figure 3.2 do not include availability factors.

Supply parameter
Non-dispatchables (Phase 1)
Nuclear (at 90% capacity)
CHP Biomass
CHP Gas
Wind Offshore
Wind Onshore
Solar PV
Tidal
Wave
Non-fossil dispatchables (Phase 2)
Geothermal
Hydro
Balancing Mechanisms (1) (Phase 3)
Domestic DSM
Pumped Storage
CHP emergency
Centralised Battery Storage (GWh)
Interconnectors (imports)
Dispatchables (Phase 4)
CCGT no CCS
CCGT with CCS
Coal no CCS
Coal with CCS
Biomass
Nuclear (rampable 10%)
Balancing mechanisms (2) (Phase 5)
Commercial DSM
Decentralised battery storage
Interconnectors (exports)
GP:2030
DECC:2050:Renew
1.2
4
17.5
55
22
28
8
0.5
15.7
0
0
54
28.4
14.1
3.6
5.8
2
2
0
2.1
4 (27.8 GWh)
3
10
17.2 (400 GWh)

0
30
19.5
0
0
0
0
1.2
0
8.1
0
5.1
0.6
15.7
1 in 10 households equivalent
10
30
Table 3.2: Supply inputs, all in GW unless stated.

Figure 3.2 shows the total generation capacities of GP:2030 and DECC:2050:Renew.
GP:2030 has a larger overall generation capacity than DECC:2050:Renew and a higher
34
Figure 3.2: GP:2030 generation capacity, compared to DECC:2050:Renew.
overall total of renewable generation capacity. It contains 1/3r d more fossil fuel
generation capacity (as distinguished from use), in the form of non-CCS CCGTs. In
2013, the total UK generation capacity was 85GW, down 4.9% from the 89.5GW at
year-end 2012. Hence the total installed generation capacity in GP:2030 of 159.7 GW
is roughly double that of 2013. For comparison DECC:2050:Renew has a total installed
capacity of 137.5 GW.
Demand Side Management (Domestic and Non-Domestic)
The potential for DSM has already been discussed at some length in Chapter 2, and is
expected to be enabled by the installation in households of smart meters and, in some
cases, intelligent appliances that can turn on or off in response to price signals from the
grid. Here we discuss both domestic and non-domestic DSM specifically in terms of the
modelling and input decisions to SHED.
While most "balancing mechanisms" - mechanisms for reducing the net demand that
are not generation technologies (such as battery storage) - work by providing increased
supply into the grid, DSM (either domestic or industrial), does so by shifting demand
away from peaks. The decision where to place domestic and non-domestic DSM in the
dispatch protocol defines to what extent technologies succeeding them are called on to
provide extra supply, or reduce or shift demand.
There are two principle dimensions of the DSM algorithm that are defined by the
scenario designers:
the size of the "DSM window" - the time period either side of demand peaks
into which it is deemed reasonable to expect consumers to shift their demand
backwards or forwards in time;
a lower limit, if implemented, below which DSM will not reduce net demand
before calling on successive dispatchable generators, or additional balancing
mechanisms.
Figure 3.3 is a diagrammatic representation of the national DSM algorithm used for
3.3 Inputs
35
both domestic and non-domestic DSM.
Figure 3.3: Representation of the generalised DSM algorithm for domestic and nondomestic consumers.
The domestic DSM algorithm allows participation during any hour of the day, with no
maximum constraint. An Ofgem (2012) report has been used to define non-domestic
DSM dynamics beyond the generalised algorithm in Figure 3.3. The extreme Ofgem
non-domestic DSM scenario was followed in SHED, defining the winter week day load
flexibility during peak demand as 4.4GW . It should be noted however that non-domestic
DSM is one of the last components of the dispatch protocol and hence is called upon
far less than domestic DSM.
In GP:2030 the window within which demand must be moved in time to is kept
to 3 hours pre and post the deficit periods for both domestic and non-domestic DSM.
Obviously electrified heating demand can only be shifted forward in time with the aid
of storage or highly heat efficient homes, whereas traditional electricity can be shifted
forwards or backwards in time. Renewable surpluses or spare generating capacity
within that window must exist to supply the demand. Demand is only shifted within the
3 hour window if there is sufficient supply either side of the the peak to absorb it. Thus
the size of surpluses either side of peaks disciplines the degree to which demand can
be shifted.
Within the non-domestic DSM algorithm, demand is removed in equal proportions
within the hours of the deficit period. This contrasts with the domestic DSM algorithm
where the largest instance of deficit is given priority. Given the 4.4GW maximum
demand constraint does not apply to domestic DSM, the only constraining factor to
36
unlimited demand shifting is the spare CCGT capacity and surplus renewable supply,
within the 3 hour window, either side of the deficit period.
Once defined, the algorithm runs as part of the supply-demand balance modelling.
Results are obtained that show the frequency and size of DSM requirements within
the scenario, at a national level, in order to maintain system balancing over the 11
modelled years. Later in this report we present the results of the disaggregated DSM
requirements, enabling us to picture what extent and frequency of demand shifting
would be asked of individual households of different socio-economic types. For detailed
information on this disaggregation process see Quiggin (2014), as a description of this
methodology is beyond the scope of this report.
As outlined above, decisions affecting the type and extent of domestic DSM expectations, and its position in the dispatch protocol, affect the overall supply-demand
balancing of the system.
Key Information 3.5 DSM placement in dispatch protocol. The placing of domestic
DSM higher up the dispatch protocol may appear to place an unnecessary onerous
burden on households, requiring them to be the first line of defence in balancing
the national grid if there is still net demand after non-dispatchable and non-fossil
dispatchable generation. However, this was decided by Greenpeace to enable:
1. The worst case scenario of domestic DSM to be quantified; modelling the
most demanding scenario means that the most that might ever be reasonably
expected of households is clearly defined. This enables GP:2030 to place
boundaries on what domestic DSM can be considered capable of
2. Reduce reliance on more expensive storage technologies, such as batteries
and compressed air: if domestic DSM can be kept within reasonable boundaries despite being primary in the dispatch protocols balancing mechanisms,
GP:2030 is overall less reliant on projected price movements of storage technologies explored below.
Hence domestic DSM occurs straight after renewable supply to reduce net demand (Phase 3), whilst non-domestic DSM occurs as the final balancing mechanism
after CCGTs to soak up any residual net demand (Phase 5). During the iterative
process of modelling GP:2030 a second phase of domestic DSM was left within
Phase 5 of the dispatch protocol - as the very last balancing mechanism, but as will
be shown later is not called upon.

Wind
Wind power, both offshore and onshore, will provide the largest contribution to renewable
supply into the grid. This expectation is reflected across all major energy scenarios in
the literature: this is simply where the greatest opportunity lies to harvest energy from
our surroundings in the UK.
Key Information 3.6 Across all eleven years of data, the mean hourly capacity factors
modelled within SHED for onshore and offshore wind generation are 30.1% and
44.3% respectively.

In 2013 the installed capacity across the UK of onshore wind was 7, 994MW
across 4,804 turbines ( 1.7MW average capacity), the offshore installed capacity
was 4, 049MW across 1,184 turbines (Renewable UK, 2013). Wind generation is expanding rapidly in the UK, currently there is a total of 5, 651MW and 9, 795MW of
onshore and offshore wind projects with planning consent, respectively. There are
3.3 Inputs
37
1, 322MW of onshore wind projects under construction and 1, 005MW of offshore under
construction (Renewable UK, 2013).
The total for offshore and onshore wind capacity in GP:2030 is 77 GW; compared to
the current total built, under construction and consented (offshore and onshore) at the
time of writing of 30GW - a 157% increase required by 2030.
Total onshore wind built, under construction or consented is 15GW, requiring a
47% increase (assuming all consented projects are built). This is a highly achievable
target over the next 15 years. However, offshore capacity must be increased by around
270% by 2030 (assuming all consented projects are built). This is then an ambitious,
but necessary, target.
In calculating the total possible available resource for offshore wind, the constraints
differ to onshore. There are no National Parks and cities to consider: but there are sites
of special scientific interest, shipping lanes and protected areas. The main constraint,
however, is the depth of sea bed (bathymetry) where the depth is required to be less
than 40m, out to 30km. A detailed description of assumptions made in calculating
various regional resource constraints can be found in the DTI (1998) report. Please
see the maps of onshore and offshore wind weather stations used to determine wind
capacity factors within Appendix B, further detailed descriptions of wind modelling can
be found in Quiggin (2014).
The UK has the largest wind resource in Europe and urgent focus needs to be
put on enabling its integration as the largest renewable contributor into a future power
supply system; in particular onto the very rapid scaling up of offshore wind capacity.
This is within our reach. Without it meeting carbon targets within reasonable timescales
becomes very likely impossible.
Solar PV
There has been major growth in Solar PV installations since the introduction of the feedin tariff in 2010, rising from 80MW installed capacity at year end 2010 to 5GW year
end 2014 (Department of Energy and Climate Change, 2015b). These numbers reflect
feed-in tariff subsidised installations as well as those installed under the Renewables
Obligation (RO).
Key Information 3.7 Across all eleven years of data, the mean hourly capacity factor
modelled within SHED for solar PV is 11.4%.
Though it cannot be assumed that growth achieved under a favourable subsidy

regime will be matched year on year for the next 15, the UK Government predicted in
2012 that solar installation capacity could reach 22GW in 2020 located on 4 million
households, on the basis of falling costs of fabrication and installation. The prospect
of reaching 28 GW of installed capacity by 2030, as required by GP:2030, is therefore
possible and achievable. Especially considering the projected installed capacity in
GP:2030 would require 1.5 GW per year for the next 15 years, and 1.2GW was installed
in 2014 alone (Department of Energy and Climate Change, 2015b).
Combined heat & power
There are many forms of CHP units fueled by different liquid or gas fuels. In 2012 the
technical potential of CHP was calculated by DECC to be 29.4GW , rising to 33.8GW
by 2030 (Ricardo-AEA, 2013). In 2011 the installed capacity of "Good Quality CHP"
was 6.1GW (Ricardo-AEA, 2013). The GP:2030 target of 21.5 GW is therefore 64% of
DECCs technical potential estimate for 2030, and a 350% increase on 2011 installed
38
"Good Quality" capacity. It appears therefore well in line with official forecasts and
expectation.
The approach taken here is that CHP units are treated as community scale heating
systems, where power is produced simultaneously with heat (rather than many dispersed
micro CHP units, due to the high capital costs (Lund et al., 2010)). There are several
forms of district heating, from gas, biomass, biogas and geothermal plants - yet only
5% of installed CHP units in 2011 were fired by renewable fuel sources (Ricardo-AEA,
2013). Under GP:2030 just under 23% of installed CHP would be renewably fueled, the
remainder burning natural (fossil fuel) gas. While it would be theoretically possible to
increase this proportion beyond 23% (to lower CHPs overall carbon intensity), doing
so would require either turning over an unrealistic amount of UK agricultural land to
biomass production; or importing vast quantities from abroad, which carries its own
risks and emissions drawbacks.
In Phase 3, emergency CHP (EmCHP) is invoked after pumped storage. This simply
enables the system to call on the full generation capacity (21.5GW ) of installed CHP
regardless of heat demand at that time. This doesnt however account for the availability
factor for CHP, defined at 90%. Before Phase 3, heat demand drives the electrical output
of CHP in GP:2030, which may not require it to operate at full capacity. However given
the impact of space heating requirements on the electricity supply through deployment
of heat pumps, it will often be, in fact, heat demand driving peaks - in which case Phase
1 CHP will be operating near or at capacity, and the ability of EmCHP to reduce those
peaks will accordingly be impaired.
Tidal
Tidal patterns correlate to the motion of the moon, not to weather patterns, apart from
during extreme storm surges. Tidal generation is calculated in a similar method to
Mackay (2009), where output is proportional to the tidal stream velocity cubed, on a
lunar cycle of 29.5 days, and generating power on both the ebb and flow tides. Hence
the tidal stream energy, rather than lagoon or barrage types of tidal power is calculated.
modelled within SHED for tidal generation is 24.4%.
There is no aggregation over multiple independent sites to form a national capacity

factor. Rather it is assumed there is one main scheme. This therefore results in four
distinct peaks in generation per 25 hour period. The ratio between spring and neap tide
capacity factors is just greater than 4. Within GP:2030 8 GW of tidal power is projected
to be installed by 2030.
It should be noted that the current momentum behind tidal is focused on tidal lagoon
facilities rather than tidal stream or barrage. There are currently no operating tidal
lagoons known to the authors, and the supply profile of tidal lagoons will be different to
tidal stream.
Hydro
In 2007 hydroelectric power stations in the UK generated 4TWh/yr, 1.4% of total

electricity demand from 1.3GW of installed capacity. Within GP:2030 a total generating
capacity of 2 GW is projected to be installed, achieved though refurbishment of existing
schemes and micro-hydro sites. Most renewable generators are given a 90% availability
factor within SHED, hydro is given a lower 60% value due to the 2 GW of capacity
being comprised of many micro-hydro sites. Hydro is useful as it can load follow and is
3.3 Inputs
39
classified as a dispatchable generator, Phase 2 in the dispatch protocol.

Wave
Being an immature technology, there is currently no commercial wave power in operation

in the UK, hence the minor contribution the technology is assumed to make to the general
energy mix, even by 2030.
modelled within SHED for wave generation is 21.1%.
While GP:2030 only expects 0.5 GW of wave generation capacity to come online by
2030 this still represents a challenge, given the infancy of the technology at applied scale.
As wave power generation systems are still under development in the UK, resulting in
no clear technology leader, two systems were identified to produce power ratings and
capacity factors in SHED; the Pelamis system (Henderson, 2006) and a generic system.
The Pelamis system, was until recently, the closest to achieving commercial viability and
its power curve can be found in Quiggin (2014). The generic device has an efficiency
of 47% for all sea states, this modelling derives from the methodology developed in
FESA (Barton et al., 2013). It should be noted that Pelamis went into administration in
December 2014 (BBC, 2014a), during the modelling process for GP:2030.
Nuclear
The total nuclear installed capacity in GP:2030 is simply the amount of current installed
capacity due to be online in 2030, set at 1.2GW (i.e. Sizewell B). In GP:2030 nuclear is
assumed to run at 90% capacity, enabling only 10% of total capacity to be called on in
peak periods. Being more ambitious with respect to the rampable portion of installed
nuclear generation capacity relies on assumptions about what would be commercially
acceptable (with or without Government subsidy), and involves additional engineering
strain on critical components. The higher the reserved rampable portion of capacity, the
less economical the station would be, meaning Government would likely need to further
subsidise nuclear power operators.
Currently nuclear power stations provide base-load power to the electricity system
with no dispatchable element, as ramping the generators puts large stresses on the
reactor core causing expensive maintenance costs. Under current European Utilities
Requirements nuclear power stations must be able to cycle daily between 50% and
100% of their rated power, with ramp rates of between 3-5%. As the OECD Nuclear
Energy Agency points out (Nuclear Energy Agency OECD, 2011) the current lack of
nuclear plant demand following is an economic rather than technical barrier.
GP:2030 does not include any new nuclear build (see Chapter 2 for more detail).
Geothermal power
The DECC has assessed the upper limit technical potential for geothermal within the
UK to be 4GW (Department of Energy and Climate Change, 2011) by 2030. In a
2011 report for WWF the projected installed capacity of geothermal was 5GW (Garrad
Hassan, 2011). Geothermal power has high utility as a renewable source of energy as
it is dispatchable, hence it can follow demand, and it can act as base load in a similar
way to nuclear power stations. Within GP:2030 the siting of geothermal installations, be
them in the UK or in Iceland with interconnection, is not defined.
40
Pumped storage
Pumped storage facilities pump water up a hill to a reservoir, increasing its gravitational
potential energy, and releases the stored water, when required, through a turbine,
generating electricity. Water is pumped up hill when electrical demand is low and
released when required. Currently the UK has a total pumped storage capacity of
2.78GW with a storage capacity of 24.9GWh. The DECC:2050:Renew scenario
forecasts significant increases in pumped storage generation capacity from 2.78GW to
17.2GW and storage capacity increases to 400GWh.
Proposals to build new pumped storage do currently exist. Scottish and Southern
Energy are in negotiation over the Coire Glas 0.6GW facility (Scottish Canals, 2013)
with further installations being considered at Sloy (0.06GW ) and Bamacaan (up to
0.6GW ) (Lannen, 2012), representing a significant increase in capacity. Within GP:2030
the capacity of pumped storage is increased slightly to 4 GW and storage capacity to
27.8 GWh, in line with those sites that have already be identified.
Centralised battery storage
Electricity storage is currently expensive and technically challenging at scale. It is

also very limited in terms of UK installed capacity, the majority of which is pumped
storage. Nevertheless grid-level battery storage is assumed to be a useful part of
a 2030 energy scenario, despite its omission from many scenarios in the literature.
The largest battery in Europe (6MW /10MW h) was switched on at Leighton Buzzard,
Bedfordshire in advance of a 2 year trial in December 2014 (BBC, 2014b). A major
advantage of battery storage is its ability to provide power back into the grid almost
instantaneously in order to reduce net demand. Pumped storage is limited in this respect
by the the generation capacity of the hydro plants tasked with converting pumped waters
gravitational potential energy back into electrical energy. For example, in GP:2030 there
is 27.8 GW h of pumped storage capacity proposed but the total associated disptach
power is only 4 GW , meaning that it would take a minimum of just under 7 hours to
deliver the full stored energy capacity if called upon.
The levelised generation costs of power from CCGTs in 2012 was 80/MW h and
134/MW h for offshore wind DECC, 2012b. The cost of battery production has fallen
and is predicted by many analysts to fall further in the near future. Within the next 7-8
years Citigroup estimates battery storage to fall to 149/MWh, based on technology
cost reductions due to increased numbers of electric vehicles and policy and legislative
changes within the EU (Pitt et al., 2015). GP:2030 projects 3 GWh of grid-connected
battery storage, less than the 5GWh that utility company Oncor is currently seeking
investment for in Texas, USA.
It should be noted that although GP:2030 projects 3 GWh of centralised battery
storage, which sounds significant, net demand post renewables could often exceed 3
GW. In this case the store of power in the batteries would last only 60 minutes. Given
that when net demands exists the duration of net demand often extends over many
hours, this battery storage is limited in its capability of providing a high utility balancing
function.
Interconnectors
There are currently three high voltage direct current (HVDC) links from the National
Grid to France, Holland and Ireland. As an example, the UK-France interconnector is
70km in length with 45k m of under-sea cable. The combined capacity of the current
three interconnectors is 3.5GW and the National Grid is currently working on further
interconnector projects with Belgium, Norway, Denmark and a second interconnector
3.3 Inputs
41
with France (National Grid Company, 2013). Of these projects the UK-Belgium interconnector is the most developed, with completion anticipated in 2018, adding 1GW of
capacity (National Grid Company, 2013).
GP:2030 envisages just over a doubling in interconnector capacity between 2018
and 2030; not unreasonable taking into consideration other planned projects listed
above. With the European Commission increasingly setting targets and legislation that
encourages or mandates (CEC, 2008a,b,c,d) European countries to move towards
renewable-heavy power supplies (therefore introducing intermittency issues) there is
greater impetus for countries to be able to export surpluses and reduce net demands
with imports. As will become clear the UK will, at times, produce power surpluses due
to uncurtailed renewable supply (even after battery, pumped storage and EV charging) in these instances interconnectors enabling power exports have the potential to provide
economic benefits.
Supply-demand dynamics at the non-UK end of the interconnectors are not considered, thus no consideration is given to the power available to import or export via
interconnectors at any given time; instead it is assumed that the maximum carrying
capacity of the interconnectors can be called at any time. This is a limitation of SHED,
but one that is reflected throughout the literature. Recent DECC and Ofgem analysis
of reliable supply from interconnection has concluded that availability of supply from
continental Europe is relatively high, although the specifics will depend on conditions in
other countries, which gives confidence that firm supply of 10GW is not unreasonable
(Department of Energy and Climate Change, 2015a).
Closed cycled gas turbines
In order to maintain an acceptably reliable power supply, in line with current official
lower limits of loss of load probability (LOLP, see Chapter 2) some rampable CCGTs
are included in GP:2030. They are not fitted with CCS technology, unlike the majority of
fossil generation capacity in DECC:2050:Renew.
Key Information 3.10 In line with Greenpeace policy (Greenpeace, 2008), the CCGT
generators projected within GP:2030 are not fitted with CCS. This consequently
results in higher emissions, but means the scenario is not reliant on unproven
technology with potentially harmful effects. The emission intensity factor used
for CCGT power stations originates from the IPCC, which estimates the life-cycle
emissions from CCGTs to be 469gCO2 eq/k W h, rather than 245gCO2 eq/k W h for
CCS fitted CCGTs (IPCC, 2012).

CCGTs are a mature technology, delivering a significant amount of the UKs current
power supply. During non-ramping operation CCGTs are more efficient than open cycle
gas turbines (OCGTs), but are less suitable for ramping rapidly during times of stress
on supply than OCGTs. However, CCGTs are now the standard form of gas turbine
due to significant investment in them over previous decades. Consequently, within
SHED emission ramping penalties are applied: as generation from CCGTs varies more,
emissions accounted for rise also. For more information on these ramping penalties see
Quiggin (2014).
GP:2030 envisages a reduction in the current capacity of CCGTs to near the
minimum level required (19.5 GW ), in conjunction with non-fossil generation capacity
and existing nuclear capacity, to ensure a reliable power supply.
42
Decentralised battery storage
Alongside an increase in the prevalence of battery storage at grid level, GP:2030

envisages the proliferation of decentralised battery storage within household and commercial or public units. This follows the logic employed in assessing the potential for
grid level storage: with battery costs declining and decentralised renewables on the
rise it will likely become increasingly common to include storage alongside installations. Barclays estimates that battery costs have declined from $17,000 in 2009 to
$3,700 in 2014 for a self-sufficient household system (Rob, 2014). Users will benefit from being able to reserve surplus power for use during peak periods, effectively
being able to swap surplus prices for peak prices on certain occasions. The logic in
favour of such investments increases under future energy conditions where electrical
demand peaks are larger, supply more intermittent and batteries cheaper to install.
GP:2030 assumes that the equivalent of
1 in 10 households will have a single battery, of average 90Ah capacity, installed
by 2030. Given the government expects 4
million households to host solar PV installations by 2020 (see Section 3.3.2), and
GP:2030 assumes there will be 30 million households in total by 2030; it would
take only three quarters of those 4 million households to purchase batteries by
2030 (assuming no more solar PV was
installed) in order for 1 in 10 to become a
reality. While ostensibly ambitious, given
the combination of factors encouraging
household economic decision-making in
this direction we consider it feasible. Furthermore, this assessment does not include battery storage investment by commercial or public sector organisations in
combination with solar arrays, likely well
before 2030 given similar economic calculations apply.
3.3.3
Emissions
"The UKs consumption [of
energy] cannot continue to
rise indefinitely.... if it is to
make an effective contribution
to a global reduction in greenhouse gas emission"
(Energy and
Climate Change Committee
2012)
Figure 3.4: Emission factors per technology. Source : Intergovernmental Panel on

Climate Change (2011)
The majority of emission intensity factors for generators are taken from the Intergovernmental Panel on Climate Change 2012 report on "Renewable Energy Sources
and Climate Change Mitigation" (Intergovernmental Panel on Climate Change, 2011).
3.4 Outputs & Results
43
Where a better or more UK relevant source could be obtained, it has been. The emission
intensity values taken as inputs to SHED are given in Figure 3.4(a). In relation to fossil
fuel generators, the numbers quoted in Figure 3.4(a) do not include start up, part loaded,
or ramping penalties which all increase emissions (Lew et al., 2012).
This is particularly true in the presence of high penetrations of renewables, resulting
in a high frequency of load variation on those flexible fossil fuel generators. For CCGTs
and coal power stations these penalties are a 15.6% and 5.1% increase respectively
(Lew et al., 2012) (Note GP:2030 does not project any coal capacity in 2030). These
penalties are applied in SHED if the national generation from these generator types fall
below 50% installed capacity.
Emission intensity calculation and Life-Cycle Analysis
The CCCs 50 - 100geq CO2 /k W h target by 2030 is based on non Life-Cycle Analysis
(LCA) numbers for all generators except for those that burn fuel, including biomass. This
means offshore and onshore wind, solar PV, wave, tidal, geothermal etc are assigned a
zero value for emissions associated to those technologies. Although this methodology
by the CCC is adequate it does mean that, as the electricity system moves from a fuel
dominated emission paradigm to an infrastructure dominated future, LCA numbers will
become increasingly important in assessing the UK economys overall carbon intensity.
The authors wish to state that CCC methodology therefore makes comparisons between
present and future emission intensity factors challenging.
For completeness, emission intensity factors presented in Section 3.4 have been
calculated using the full LCA numbers, shown in Figure 3.4, and using the CCC
methodology with all but fuel burning generators emissions intensity factors being set
to zero.
3.4
Outputs & Results

The aim of this modelling process was to explore the possibility of reaching a 2030
electricity system which;
1. Decarbonised within the CCC targeted 50 - 100geq CO2 /k W h emission intensity
factor, whilst;
3.4.1 Emissions
Of the listed objectives above, the objective of meeting the CCCs CO2 targets has
been achieved, reaching an emission intensity factor of 77.9 geq CO2 /k W h. We also
considered the impact of balancing mechanisms working to further reduce emissions
44
irrespective of the impact of balancing dispatch on CCGT load factors. This further
reduced the emission intensity factor to 51.2 geq CO2 /k W h
3.4.2
Demand
Figure 3.5(a) illustrates the hourly average national daily demand profile during January/February weekdays of; traditional electricity, heat pump demand, electric vehicle
demand and resistive heating. As is the case with current national electrical demand,
traditional electricity demand peaks in the early evening, the profile shape resembling
current demand profiles. The national demand profile changes with the higher morning
electrified heating demand, resulting in national demand exhibiting two distinct peaks.
The overriding change to total scenario demand, relative to historic demand, is the
inclusion of the morning peak demand, almost equivalent in magnitude to the evening
peak demand. This results in increased variability of demand, the morning peak period
is 8 9am and the evening peak period 7 8pm.
(a) Average electricity demand from traditional (b) Increase in electrical demand due to heat
electricity, heat pumps and electric vehicles pumps during a typical January day.
for January/February weekdays through 2030
based on 2001:2011 data.
Figure 3.5: GP:2030 evolution of electrical demand due to heat and transport electrification.
Figure 3.5(b) shows the increase in electrical demand during the 17th January based
on 2001 data. Whereas Figure 3.5(a) shows the average national electrical demands
for January and February weekdays, Figure 3.5(b) illustrates the real time series results.
The morning total demand can be seen to peak at 44GW and the evening peak at
45GW , demonstrating the increased importance of the morning period as electrical
demands increase due to heat electrification.
Figure 3.6(a) shows mean annual demands across the eleven years of SHED.
Line losses amount to 21.5T W h/y r , significantly less than the 31.3T W h/y r within
DECC:2050:Renew. This is due to the lower level of heat electrification and traditional
electricity demand within GP:2030.
GP:2030 projects 25% of heat to be delivered by heat pumps, whilst
DECC:2050:Renew projects 90% (by 2050 rather than 2030), resulting in 15.1T W h/y r
of heat pump electrical demand in GP:2030 compared to 73.1T W h/y r within
DECC:2050:Renew. It should be noted that DECC:2050:Renew forecasts 10% heating
to be delivered by resistive heating, whereas GP:2030 forecasts none. These factors
45
combine to result in a GP:2030 total annual demand of 328.6T W h/y r compared to

477.8T W h/y r within DECC:2050:Renew. Heat pump electrical demands therefore
represent 4.9% of total annual demand within GP:2030, compared to 16.4% within
DECC:2050:Renew. This, as will be discussed later, results in significantly more
manageable supply - demand balancing as peak demands are not driven to the extreme
levels of DECC:2050:Renew, by heat pumps.
(a) Mean GP:2030 demands using forecast (b) Cyclical nature of traditional electricity deheating demands across the eleven years of mand within GP:2030, based on 2001:2011
data (2001:2011).
data.
Figure 3.6: GP:2030 evolution of electrical demand, annual and hourly breakdowns.
Figure 3.6(b) illustrates the modelled hourly national total electrical demand for
all eleven years of data. As is expected, the cyclical nature of demand is apparent,
with winter periods exhibiting the greatest demand values. Demand can be seen to
peak at 62.3GW , which compares to a historic National Grid peak of 59.6GW over
the same period from which the underlying data is drawn. GP:2030 peak demands
are comparable to historic levels, meaning that the reduction in traditional electricity
demand has offset increased demand due to electrification of transport and heating though only in the context of achieved heating reduction targets.
3.4.3
Supply
Figure 3.7 illustrates the average monthly output from all non-dispatchable renewable
generators. As hydro and geothermal are dispatchable they are not shown here. The
residual inflexible (90%) nuclear generation of Sizewell B, anticipated to still be open in
2030, can be seen to generate a small amount of base-load power, at the very bottom
of the graph. The greatest seasonal variation in generation originates from CHP, dipping
to a summer low when following heating demands.
46
Figure 3.7: Non-dispatchable monthly supply through 2030 (mean across all years of
data)
Seasonal variations
Figure 3.8(a & b) illustrate hourly generation for non-dispatchable generators for January
(based on 2001) (Figure 3.8(a)), and a May - June period (Figure 3.8(b)). The summer
period has been selected on the basis that the nighttime of May 26th , into the early
hours of May 27th (2005 data), represents the minimum renewable generation output
of all the 11 years modelled. This period is the most challenging for supply - demand
balancing and therefore the most onerous domestic DSM is experienced during this
period (explored further in Section 3.4.5). CHP has a biomass and gas fired combined
capacity of 19.4GW , which takes into account their availability, set at 90%. Within the
winter period (Figure 3.8(a)) CHP generation regularly reaches it maximum output as
heat demands are high, however in the summer this is not the case and inter-day CHP
generation varies much more. In the late night period of May 26th the combined output
of all renewables drops to 5.5GW , compared to a 13.2GW minimum in the January
winter period. The other major contributor to this fall in renewable output is offshore
wind, whose output during this summer low drops to 3.5GW . The mean and variation
in output in the summer and winter are comparable, but the summer minimum forces
the total renewable generation to problematic lows.
47
(a) January (2001 data).
(b) May - June (2005 data).
Figure 3.8: Non-dispatchable GP:2030 supply during particular periods of the year.
48
Balancing mechanisms
Figure 3.9 illustrates how dispatchable CCGT power stations and the balancing mechanisms of domestic DSM, pumped storage, batteries and interconnectors (IC) and so
on respond to net demand post renewable supply. The legend of Figure 3.9 follows
the order of dispatch within SHED. As can be seen interconnectors and batteries (both
centralised grid connected and decentralised low-voltage network connected), do not
provide power back to the network to aide balancing in this particular instance (Figure
3.9 late May (2005 data)). This is due to dispatch order in which these balancing
mechanisms have been modelled.
After renewable supply from both intermittent non-dispatchable and dispatchable
renewable generators (hydro and geothermal) the first balancing mechanism to be initiated is domestic DSM, which in the first instance shifts demand to periods where excess
renewable supply would otherwise become surplus supply. In GP:2030:Pragmatic this is
only initiated if the CCGT generators would otherwise not be able to meet net demand,
preventing uneconomical low load factors on CCGTs. Pumped storage, then centralised
batteries and then interconnectors follow the first phase of domestic DSM, which are
also only initiated if CCGTs would otherwise fail to meet net demand. Hence by the
point that batteries would be called upon (in this instance - Figure 3.9 late May (2005
data)) the net demand has been reduced to a level that can be supplied by the CCGTs
and as such batteries and interconnectors do not play a role.
This methodology, sequentially speaking (see Section 3.2), prioritises domestic
DSM over other storage and balancing mechanisms (though which is actually relied
on most over an annual period is not necessarily determined by this hierarchy). This
approach was taken as large scale electricity storage, either in the form of batteries,
compressed air or hydrogen-electrolysis processes, are not currently cost effective
and is yet to be proven economically at scale (despite the expectations indicated in
Section 3.3.2). Indeed the extra pumped storage of2.9GW h with a capacity of 1.2GW
is at the limit of current appropriate sites for pumped storage. By ensuring that the
level of domestic DSM is manageable, even with households playing the greatest role
they would ever have to play, it is possible to treat this GP:2030 scenario as practically
achievable without having to rely heavily on unproven large scale electricity storage, or
interconnectors.
From Figure 3.9 it is possible to observe that the challenging period of late evening
of May 26th into the early morning of May 27th both pumped storage and domestic DSM
play a critical role in maintaining supply - demand balance. The contribution of domestic
DSM to balancing peaks at 4.8GW , and pumped storage 4GW during this period.
Further to the net demand periods, Figure 3.9 also shows periods of excess or surplus
supply when renewable supply exceeds demand. This excess power is utilised within
the first instance of domestic DSM as power to supply the demand that is shifted away
from net demand periods, and secondly to charge the store of energy within storage
facilities such as pumped storage and batteries. In Section 3.4.6 further attention will be
given to the surplus supply.
As can be seen, when a substantial amount of CHP capacity is envisaged - in order
to help reduce heating demand on the electricity supply - the most challenging periods
for supply-demand balancing actually occurs in the summer, when CHP generation is
low, following seasonally reduced heating demands. Critics of renewable based systems
often focus on the dangers of the lights going out because of winter anticyclones. In
practice, as we have shown it is summer periods that provide the biggest difficulties,
in GP:2030. Winter anti-cyclones are not unproblematic, nonetheless. System man-
49
Figure 3.9: Non-dispatchable GP:2030 net demand post renewables and the balancing
mechanisms and CCGTs that fill that net demand during the summer minimum in
renewable supply.
agement during those periods relies, within GP:2030, on the effective combination of
all balancing mechanisms; DSM, storage and fossil fuel generators working in concert. As such, a solution in one area of the scenario creates problems, though not
insurmountable, elsewhere.
3.4.4
Load Factors
In 2013, with power generation from CCGTs at its lowest level since 1996, load factors
for CCGTs fell to a record low of 28%. In 2008 CCGT load factors were at an eight
year high of of 71.0% per cent. The fall was due to an increase in power generation
from coal fired power stations, whose load factors reached 58% in 2013. In GP:2030
CCGTs without CCS are the only dispatchable centralised fossil fuel generators, with
an average load factor of 23.6 %.
Climate driven sub-scenario
Turning to GP:2030:ClimateDriven, where the balancing mechanisms work to minimise

deficits and CO2 emissions, without consideration of maintaining reasonable load factors
on the CCGTs. The load factor is found to fall by an order of magnitude to 2.1 %.
50
3.4.5
Balancing
Figure 3.10 shows the annual average contribution from all balancing mechanisms.
The order of the balancing mechanisms from left to right in Figure 3.10 follows the
order of the dispatch protocol within SHED. For comparison, offshore wind supplies
on average 181.3T W h/y r and CCGTs 36.3T W h/y r . Pumped storage is the
largest contributor to balancing at 0.7T W h/y r with emergency CHP the second most
vital balancing mechanism at 0.6T W h/y r . Imports from interconnectors are also
critical to balancing, supplying 0.1T W h/y r . The contribution from domestic DSM
and grid connected batteries is significantly lower at 0.1T W h/y r and 0.01T W h/y r
respectively. Household batteries supply even less at 0.4GW h/y r , thats giga-watts
rather than the tera-watt units of the previous balancing mechanisms. Non-domestic
DSM supplies a similar amount of balancing to household batteries at 0.4GW h/y r .
So although the first phase of domestic DSM sequentially occurs prior to pumped
storage and interconnector imports; the requirement for surplus renewable power within
a 3 hour window of the net demand period prevents domestic DSM playing a greater role.
It should be noted that although domestic DSM has been included in a second phase,
occurring after CCGT dispatch, this second round of domestic DSM is not required
within GP:2030. During periods when this second phase of domestic DSM would have
been required CCGTs were running at maximum capacity, i.e. there was no spare
renewable capacity either side of the peaks in question for net demand to be shifted to.
In eventuality, all domestic DSM in GP:2030 occurs as a result of shifting demand away
from periods where CCGTs are unable to supply all of net demand to periods where
there is excess renewable supply.
Figure 3.10: Annual average energy supply from balancing mechanisms
In order to assess the viability of these balancing mechanisms contributing to the

annual contributions shown in Figure 3.10, historic data was obtained for pumped
storage, interconnectors and non-domestic DSM. In 2014 pumped storage contributed
3T W h/y r back into the National Grid (Department of Energy and Climate Change,
2014), significantly more than the 0.7T W h/y r utilised within GP:2030. In the real
51
world balancing mechanisms are used not only during periods when fossil fuel generators are unable to supply power due to capacity constraints, but also when it is
economically viable and profitable, thus the large difference. As pumped storage is able
to react efficiently and quickly to surges in demand, the facilities tend to supply peak
power even if fossil fuel generators could, as it is cheaper. Also in 2014, interconnectors
to the continent supplied 16.6T W h/y r into the UK, 10.3T W h from France and
6.3T W h from the Netherlands. This is again a much greater supply of energy than
the 0.1T W h/y r found by SHED under GP:2030, for the same reasons as were
highlighted for pumped storage.
Finally it is worth considering the current amount of energy shifted in time due to
DSM within the non-domestic sector, where demand-side mechanisms do currently
operate. As the development of non-domestic DSM services is a recent addition to
balancing mechanisms the National Grid can call upon, data is difficult to obtain. The
National Grid classifies different contracts with balancing services, depending on the
response time from the service provider, into "Frequency Response", "Fast Reserve"
and "Short Term Operating Reserve (STOR)". The National Grid states;
STOR in particular has seen strong recent volume growth on the demand
side. However significant volumes of those STOR Services are in fact
delivered by back-up generators depressing demand, rather than "true" load
reduction.
Focusing on STOR, where the strongest demand-side balancing growth has been
witnessed, the total STOR capacity utilised over 2013 was 167.2GW h (The National
Grid Company, 2014). The authors have been unable to obtain a breakdown of the
fraction of this which is non-domestic DSM - however 4% of the instantaneous capacity
is non-domestic DSM. This is 4% of the capacity of STOR that is available at any
given time, akin to a power rating. Making the assumption that this 4% translates from
instantaneous capacity into annual supply, we can assume 6.7GW h was supplied in
2013, again significantly greater than the 0.4GW h/yr within GP:2030.
Domestic DSM
Although domestic DSM plays a significantly smaller role in balancing than pumped
storage, emergency CHP or interconnectors the complexities of householders attitudes
and responses to DSM means careful attention should be paid to its expected role.
Whereas the other balancing mechanisms face technical and economic barriers to their
implementation, domestic DSM faces social and psychological barriers. It has been
shown by numerous studies (Darby, 2010; Lindley, 2010; Ofgem, 2010b; Hargreaves
et al., 2010; The Commission for Energy Regulation, 2011) that households are likely
to respond negatively to frequent requests for DSM above 10% of demand. Other
technical and feasibility studies into the integration of high penetration of renewables,
such as Poyry (2011), have assessed the technical and economic feasibility of balancing
mechanisms. A unique aspect of SHED is its ability to assess domestic DSM down to
the household level. For a detailed description of how SHED disaggregates national
domestic DSM dynamics and requirements to the household level see Quiggin (2014),
as the methodology is extensive and beyond the limitations of this report.
Figure 3.11 illustrates the average occurrence of domestic DSM at a national level
in one year, across different levels of demand shifting. As can be seen the majority of
these occurrences are below 0.2GW , occurring on average 77 times per year, across
the eleven years of data that SHED runs over. Figure 3.11 also demonstrates that the
52
requirement for balancing from domestic DSM declines exponentially as the requirement
increases. Hence the number of occurrences where the requirement is 3GW is on
average only 6.8 instances per year.
Figure 3.11: Number of required domestic DSM occurrences in one year across different
levels of DSM
Figure 3.12 shows that the requirement of domestic DSM falls predominantly within
summer months. As was discussed in Section 3.4.3, this is mainly as a result of lowered
CHP power output, following heating demand, during summer months. Consequently,
the following analysis will focus on summer months in interpreting the impact DSM will
have on individual households.
Impacts on individual households
As previously indicated, the methodology used to disaggregate the national DSM

dynamics to the household level can be found within Quiggin (2014). Broadly the
methodology:
Takes real hourly household heating and electricity consumption, and;
Forms a pool of 1000 households from which an aggregate demand profile can
be built, before;
Passing a proportion of the national DSM requirement to this aggregate pool of
households based on its hourly consumption, before;
Assuming each individual household participates on an equal basis in shifting its
demand to fulfill the aggregate requirement.
Households are protected from reducing demand below 100 watts and are not
required to reduced their demand by more than 80%. In essence this means those
households that are consuming more power within a given period will be required to
shift more of their electrical demand.
When considering the impacts of the required DSM, the differential response across
household socio-economic groups is important, particularly as studies have shown
households respond in different ways to smart meters (Hargreaves et al., 2010). Identifying households within the aggregate pool that are typical of particular socio-economic
53
Figure 3.12: Domestic DSM contribution towards balancing per month in GWh
groups aids in defining how different households will be required to participate in DSM.
Work by Druckman and Jackson (2008) has been used to define the characteristics of
these "typical" households. Druckman and Jackson (2008) identified "typical" households by utilising UK National Output Area Classifications (OAC) to segment households
into seven "supergroups"6 which exhibit different socio-demographic characteristics
(Office for National Statistics, 2005; Vickers and Rees, 2007). Within SHED the selected
supergroups are; Constrained by Circumstance, Typical Traits and Prospering Suburbs.
Further information on the specifics of each supergroup can be found in Vickers et al.
(2005).
The Prospering Suburb household is a detached mortgaged property consuming on
average 18, 500k W h/y r of gas and 5, 000k W h/y r of electricity, both of which are above
the national average. The property is 30-75 years old with two adults aged between
46-55 with no children and one adult at home during the day. The lead participant is
employed earning more than 75, 000 per year.
The Constrained by Circumstance household is a local authority rented flat consuming on average 7, 100k W h/y r of gas and 1, 200k W h/y r electricity, both of which are
below the national average. The house is 10-30 years old with two bedrooms. The lone
occupant is retired within an income of less than 15, 000 and is at home during the
day.
Finally the Typical Traits household is a terrace privately rented property consuming
on average 15, 400k W h/y r of gas, and 4, 900k W h/y r of electricity, both of which are
roughly equivalent to the national average. The house is 5-10 years old with three
bedrooms. There are two adults and one child living at the property, none of whom
are at home during the day. The lead participant is 36-45 of age, employed earning
30 50, 000 per year.
Figure 3.13 illustrates the average DSM percentage demand reduction, each hour
of the day, during July/August weekdays for traditional electricity and heat pump DSM.
6
The Supergroups were designed for the Office for National Statistics (ONS)
54
(a) Traditional Electricity.
(b) Heat Pump.
Figure 3.13: Typical households average percentage reduction in demand due to DSM
during Jul/Aug weekdays
% DSM
% DSM10
% DSM>10
ProsSub
22
17.2
4.7
Trad
TypTrait
23
16.4
6.6
ConCirm
16.2
10.9
5.3
ProsSub
3.9
3.1
0.8
HP
TypTrait
4.1
3.3
0.8
ConCirm
0
0
0
Table 3.3: Probability of DSM participation requirement during the evening peak (9pm 11pm) demand periods, and of those DSM periods which are greater or less than 10%
of demand for each typical household, in Jul/Aug weekdays.
Table 3.3 shows that the majority of DSM reduction periods for traditional electricity
demand are below 10% of demand for all three typical households. Figures 3.13 (a &
b) shows the mean reduction is between 0.9 1.3% for the three typical households.
To put this in context; the Prospering Suburbs and Typical Traits households consume
on average 600 watts of power during these DSM periods. A 10% DSM demand
reduction would therefore equate to 60 watts of power, this is equivalent to a higher
power consumption laptop. Given that consecutive periods of DSM 10% of demand
can last up to 1.4 hours in GP:2030 scenario this would mean unplugging a laptop up
to that maximum time period, which households may be unwilling to do. Alternatively
a household could turn off a typical A + + rated fridge freezer for 30 minutes with no
impact on the items inside the fridge or freezer EA Technology (2011) and save 30
watts of the total 60 watts required. This can be achieved with simple smart appliance
control, with no engagement from the household required.
It should be noted that within Table 3.3 the Constrained by Circumstance household
is not required to participate in heating DSM as its heating demand is very low during
these periods, and is hence protected from participation due to the rules of the DSM
algorithm.
The maximum period of time a household would be required to participate in DSM
where that DSM reduction is 10% of demand is shown in Figure 3.14, where the DSM
reduction peaks at 35.7% for the Typical Traits household and lasts for 5 hours.
55
Figure 3.14: Most Challenging DSM period
It should be noted that the period of DSM represented in Figure 3.14 occurs only
once in the 11 years that SHED runs over. Indeed out of the 96408 hours of the 11
years of SHED, there are only 132 instances of traditional electricity DSM demand
reduction exceeding 10% for the Prospering Suburbs household, 152 for the Typical
Traits household and 108 for the Constrained by Circumstance household. As was found
at the national level, as the DSM demand reduction requirement increases the number
of occurrences falls off exponentially. As such the number of instances exceeding 30%
(as is the case in Figure 3.14) is only 12, 17 and 6 for the Prospering Suburbs, Typical
Traits and Constrained by Circumstance households, respectively. Of the 17 instances
of 30% DSM demand reduction for the Typical Traits household, 2 of those hours are
shown in Figure 3.14 and the rest occur where the duration of the DSM requirement
is either a singular hour or two hours, but not the consecutive 5 hours duration as in
Figure 3.14. It is this duration and magnitude in combination that makes this late May
period the most challenging.
It should also be noted that although Figure 3.14 peaks at 35.7%, this could be
achieved by simply delaying, for example, the washing or tumble dryer cycle. This
can be achieved automatically, using smart control interfaces between the washing
machine and the households smart meter. This would not require the household to be
consciously aware of any interruption of their daily patterns of behaviour.
However, on careful inspection of the energy being consumed by the two households
during this period it was found that the mean energy demands were 450W h and
290W h for the Prospering Suburbs and Typical Traits households, respectively. This
level of demand is suggestive of minimal activity within the household at this time, with
perhaps some lighting on and devices charging. Consequently, turning off a washing
machine or tumble dryer would not be an option: in this extreme case the household
would be required to turn off all non essential appliances.
Although this sounds extremely onerous, requiring households to turn off lighting and
non-essential loads, this occurrence is, as previously mentioned, very rare occurring
only once in 11 years. Further to this it should be noted that the dispatch protocol
56
within GP:2030 has been designed with the intention of exploring the most onerous
DSM requirement on households. By placing domestic DSM at the top of the dispatch
protocol households are therefore required to participate much more than if they were
placed at the bottom. These results can be thought of therefore as the worst case
scenario for domestic DSM. If other balancing mechanisms within Figure 3.10 were to
increase their role in balancing then the role households play would be lessened.
Key Information 3.11 As was highlighted in Section 3.4.5 the balancing mechanisms
of pumped storage, non-domestic DSM and interconnectors supply within GP:2030,

on an annual basis, far less than currently. For instance in 2014 interconnectors to
the continent supplied 16.6T W h/y r into the UK, 10.3T W h from France and
6.3T W h from the Netherlands. Much greater than the 0.1T W h/y r found by
SHED under GP:2030. This is due to the following reasons;
1. Domestic DSM is placed higher up the dispatch protocol in order to asses
the worse case scenario that households will be confronted with, in doing so
domestic DSM takes a greater burden of balancing responsibility.
2. Currently all balancing mechanisms respond to market price signals, under
SHED it is the net demand each mechanism responds to.
3. The dispatch protocol was designed to minimise CO2 emissions.
The authors have assessed that domestic DSM quantified within GP:2030 are
within achievable levels, however these other balancing mechanisms could remove
some of the balancing responsibility from domestic DSM.

3.4.6
Surplus power
Figure 3.15(a) illustrates the average energy exported via the interconnectors over the
eleven years of SHED. The monthly exported power drops to a summer low in August,
again due to the reduction in CHP output following lower heating demands in the
summer. Once power has been exported up to the capacity limit of the interconnectors
the remaining power is surplus to requirements within the electricity system. Figure
3.15(b) illustrates this surplus energy per month. Within GP:2030 this surplus energy is
assumed to be available for hydrogen production for transport. There are other uses for
this surplus power, other than hydrogen production, such as the production of methane.
It is beyond the scope of SHED and this report to detail the conversion type and use
thereafter of this surplus power. However due to the greater conversion efficiencies, both
straight to the gas type and back to electricity, it is assumed hydrogen production would
be preferred to methane. Following through with the assumption that electrical surpluses
are used for hydrogen production and that hydrogen is used for hydrogen vehicles, a
current typical car-driver uses 40k W h of energy per day (Mackay, 2009). If we now
assume pessimistic conversion efficiencies in order to give conservative estimates, the
wheel-to-wheel efficiency of hydrogen is set at 24% (Bossel, 2006), and the efficiency of
electricity to hydrogen at 60%. Then taking the average monthly production of surplus
energy from Figure 3.15 as 3.5T W h, this equates to supplying 438, 000 hydrogen
cars with a continuous daily supply. This would be expensive to achieve due to the high
costs of electrolysis, hence these numbers should be treated as possibilities rather than
specifications.
(a) Exported Power.
57
(b) Surplus electrical power for hydrogen production.
Figure 3.15: Monthly power exported and surplus power.
3.4.7

A failure to meet heating reduction targets will have significant impacts on electricity
demand. Figure 3.16(a & b) illustrates changes in the demand profile during January/February weekdays due to heating demands remaining at historic levels, for both
GP:2030 and DECC:2050:Renew. Heat pump electrical demand becomes more pronounced, with morning and evening peaks now more dominant. This is exemplified
within DECC:2050:Renew due to high heat pump electrification, ambitious reduction
targets and 10% of heat being delivered by resistive heating. A failure to meet heating
targets results in morning peak electrical demand dominated by heat pumps rather than
traditional electricity demand, which further emphasises the double diurnal demand
peaks.
(a) Greenpeace.
(b) DECC Renewable.
Figure 3.16: Increase in electricity demand due to a failure to meet heating targets for
January/February weekdays through 2030, based on 2001:2011 data. Note difference
in scale for Y axis between graphs.
58
Key Information 3.12 In GP:2030 there are no deficit periods, meaning that reliability
of supply is consistently maintained and blackouts and brown outs are totally avoided.
If heating targets are not met then over the course of eleven years there would be 47
hours of deficits totaling 141.9 GW h. Further to this domestic DSM would increase to
onerous levels for households, further illustrating the importance of meeting heating
reduction targets.

4. Economics
4.1

It is beyond the scope of this analysis and report to investigate detailed costing of the
GP:2030 scenario. Consideration has, however, been given to the cost implications
by drawing on published work by the CCC. In The Renewable Energy Review (CCC,
2011) the CCC draws on analysis by Poyry (2011) regarding the cost implications for an
electricity network with a high penetration of renewable generators and electrification of
heating and transport. A comparison of the GP:2030 scenario and that of Poyrys "very
high 2030" scenario is given in Table 4.1, in which comparable generation capacities of
renewable generators between GP:2030 and Poyry:2030 are evident.
Poyrys analysis shows that up to a 65% (2030) and 80% (2050) share of generation
from renewables results in additional intermittency costs to consumers of 1 pence per
kWh. The GP:2030 scenario projects 76% penetration of renewables, whilst the
Poyry:2030 scenario contains 68%. As the CCC acknowledges, there is considerable
uncertainty over the future costs and investment required to manifest any scenario,
mainly due to the uncertainty of future costs of any particular technology (Committee
on Climate Change, 2011). The CCC expect that with a 65% renewables share, the
average electricity generation cost in 2030 would be 8.2 13.8p/k W h, requiring
126 227bn of investment.
4.2
There are clearly differences between the Poyry:2030 "very high" renewable scenario
and GP:2030. Table 4.1 shows total electricity demand within the Poyry scenario is
409T W h/y r , compared to GP:2030 of 327T W h/y r . As was discussed in Chapter 3
electricity demand is necessarily low to enable robust balancing of supply - demand
when domestic heating is electrified, which drives peak electrical demand due to diurnal
heating peaks.
Levels of renewable generation are comparable between the two scenarios (Table
4.1); an additional 8GW of offshore wind is projected under GP:2030 whilst onshore
wind and solar PV remain at comparable levels. "Marine" generation is set at 8GW in
Poyry:2030 which is comparable to the 8.5GW of tidal and wave combined in GP:2030.
Chapter 4. Economics
60
The capacity of balancing mechanisms is very similar between the two scenarios
(pumped storage is precisely the same), whilst the Poyry scenario projects an additional
5GW of interconnectors and GP:2030 an additional 3GW of centralised battery storage.
There is a significant divergence in the two scenarios when it comes to nuclear, CHP,
coal and CCGTs. The Poyry scenario projects 35GW of CCGTs; GP:2030 only
19.5GW . The GP:2030 scenario also projects no new build of nuclear (remaining
at 1.2GW as Sizewell B remains open) whilst Poyry:2030 projects 11GW of nuclear
capacity. Further to this Poyry:2030 projects 4GW of unabated coal power stations
to be open in 2030, whilst GP:2030 predicts none. These three scenario divergences
result in Poyry:2030 projecting an additional 29.3GW of centralised generation in the
form of nuclear, unabated coal and CCGTs compared to GP:2030. This is however
roughly balanced against GP:2030 projecting an additional 21.5GW of CHP (gas and
biomass).
Total Electrical Demand (TWh/yr)

Non-dispatchables (Phase 1)
Nuclear
CHP Biomass
CHP Gas
Wind Offshore
Wind Onshore
Solar PV
Marine
Tidal
Wave
Non-fossil dispatchable (Phase 2)
Geothermal
Hydro
Balancing Mechanisms (1) (Phase 3)
General Extra Storage
Additional Pumped Storage
Centralised Battery Storage (GWh)
Interconnectors
Dispatchables (Phase 4)
CCGT no CCS
CCGT with CCS
Coal no CCS
Coal with CCS
GP:2030
327
Poyry:2030 (Very High)

409
1.2
4
17.5
55
22
28
8
0.5
11
Not stated
Not stated
47
21
25
8
-
2
2
Not stated
Not stated
1.2 (2.9 GWh)

3
10
1.2 (2.9 GWh)

1.2 (2.9 GWh)
Not stated
15
19.5
0
0
0
35
0
0
4
Table 4.1: Comparison of GP:2030 to Poyry:2030. Source : (Poyry, 2011)
In summary, while the costs of GP:2030 are not modelled within SHED, by comparison with a fully costed scenario of similar dimensions we are able to estimate the
order of magnitude of costs, and impacts on average electricity generation cost. Two
things should be borne in mind, nevertheless: first, accurately predicting future average
prices for any given scenario is fraught with difficulty. Second, revolutionising the UKs
energy infrastructure is, naturally, a highly expensive enterprise. Nevertheless, costs
must be borne, whether or not our existing infrastructure is replicated (if climate change
4.2 Scenario divergences
61
science were ignored) - the question is to what extent the right investments now will
avoid spiralling future costs as a result of ecological damage and resource scarcities.
Electrification of heating
The role of traditional electricity demand reduction
Post 2030
Importance of meeting space heating targets
Domestic DSM
Decentralisation
Summer CHP & emergency CHP
Batteries
Interconnectors
SHED shortcoming
Exports & surpluses
Pragmatic & climate driven sub-scenarios
Dispatch Protocol
A final note
5. Discussion
In any model of the real world, especially one such as SHED which looks into the future,
assumptions must be made. The implications of some of the assumptions made within
SHED and the decisions made within GP:2030 are discussed here, alongside some of
the relevant important outputs of GP:2030.
5.1 Electrification of heating

5.1.1 The role of traditional electricity demand reduction
Many published energy scenarios foresee a rise in traditional electricity consumption
in the coming decades. Within GP:2030 a change of -21.1 % is projected as the
electrification of 25 % of heating, as well as 32 TWh/yr from EVs, requires demand
reductions within traditional electricity consumption as well as reductions in heat demand.
In the absence of such reductions, supply-demand balancing would fail to achieve
reliability of supply.
5.1.2
Post 2030
If further decarbonisation of the UK energy system is to be achieved post 2030, which
is necessary from a climate change perspective, this is likely to take the form of increased electrification of heating. In order to maintain reliability of supply with increased
electrification there are a number of options:
1. Increase the capacity of fossil fuel generators, which will lead to a rise in emissions.
2. Increase the capacity of renewables in conjunction with increased storage and
DSM.
3. A greater emphasis on heating demand reductions, via increased heating efficiency and behaviour change. This too will be challenging, but ultimately confronts
the problem at its root cause: the demand for heating in our poorly insulated
housing stock.
The authors favour a further reduction in heating demand as the preferable solution
to the incorporation of more electrified heating into the electricity network, in order to
maintain reliability of supply.
64
5.1.3
Chapter 5. Discussion
Importance of meeting space heating targets

As was discussed in Section 3.4.7, if GP:2030 heating demand reduction targets are not
met then peak demands increase resulting in 47 hours of deficits across the 11 years
of the model, with domestic DSM becoming more onerous. This would likely result in a
negative feedback loop where households, too regularly required to participate in DSM
in order to prevent blackouts, begin to respond negatively to price signals (to the extent
that clusters of households stop participating), further compounding the 47 hours of
deficits. For these reasons it is imperative that any 2030 scenario, such as GP:2030,
ensures that heating demand reduction is prioritised in equal measure to renewables
integration.
5.2
Domestic DSM
Domestic DSM within GP:2030 was intentionally placed early on in the dispatch protocol.
This choice was made for a number of reasons, principally due to a need to investigate the most conceivably onerous requirement on households; and to manage that
accordingly. The outputs show that summer months (due to the reliance on CHP units
following heating loads) are the most challenging, but have been kept within manageable
levels. By this we mean that households are not expected to participate above the
10% of demand level too often. The level at which households are likely to respond
negatively (Darby, 2010; Lindley, 2010; Ofgem, 2010b; Hargreaves et al., 2010; The
Commission for Energy Regulation, 2011). It is the Prospering Suburbs household
which experiences the most onerous DSM, being required to participate at 10% on
4.7 % of July/August weekdays, with a further 17.2 % of those days requiring 10% of
demand to be shifted in time. Considering the studies cited in this regard this degree of
demand shifting is not considered unreasonable.
As a consequence of domestic DSM being initiated immediately after non-dispatchable
renewables within the dispatch protocol, the results represent the worst case scenario
for households. This requirement from households could be reduced with non-domestic
DSM, as well as storage, playing a larger role.
The most vulnerable household is the Constrained by Circumstance household,
where the lone occupants income is 15, 000, is at home during the day, retired, and
living in a a local authority rented property. The age of the occupier, and the occupation
of the property during the daytime, makes this household vulnerable to lulls in internal
temperature, and the income and tenure indicates this household is less likely to be
able to implement interventions to help automate or shift demand, in response to DSM
requirements. This household is required to participate less often in DSM and their
heating isnt impacted at all during the challenging summer months.
Looking further at the OAC supergroup classifications (Vickers et al., 2005) it is
possible to identify areas of the UK that these typical households represent. The
Prospering Suburbs household could be found across England with concentrations in
the North-West, the Midlands and around London. The Constrained by Circumstance
household is associated with city areas, and the Typical Traits household can be found
across the whole of the UK. For further details on the distributions of these household
classifications across the UK see Quiggin (2014).
Despite GP:2030s success in maintaining a reasonable level of quantified reliance
domestic DSM for system balancing, it should be remembered that the deployment of
this technology is not comparable to building wind turbines or installing solar panels
- it only works if deemed socially acceptable by households and citizens. Given this,
5.3 Decentralisation
65
questions of fairness as to whom is required to shift what and when are highly relevant.
It is beyond the scope of this report to comment more here, except to say that this can
be seen not only as a problem for technological implementation - but perhaps instead as
an opportunity for addressing distributive questions within the context of energy access.
5.3
Decentralisation
GP:2030 is a highly decentralised energy scenario; not only do renewables represent
76% of generation capacity but 4 GW of biomass and 17.5 GW of gas fired district
CHP units are projected to be installed by 2030. In 2011 the installed capacity of "Good
Quality CHP" was 6.1GW (Ricardo-AEA, 2013), however other scenarios also project
high levels of CHP, such as 52.5GW under Transition Pathways Thousand Flowers
(Barton et al., 2013).
Decentralisation in turn provides new questions, and opportunities, regarding the future operation and ownership of an electricity system, which will need to be incorporated
into any policies designed to realise it.
5.3.1
Summer CHP & emergency CHP

With high levels of installed CHP it is important to note that as SHED models CHP
following heating demands, this results in the summer net demand periods that domestic
DSM is then forced to close. As previously mentioned it is costly to run CHP units
without the heat they generate being utilised. Hence emergency CHP follows domestic
DSM and pumped storage in the dispatch protocol, see Section 3.2.
5.4 Batteries
Both centralised and decentralised battery storage have been modelled as playing a role
in balancing within GP:2030. Although the economics of large scale battery storage is
expected to improve in the short term, it remains a relatively expensive option; and has
been placed low down in the dispatch protocol. Hence although 3 GWh of centralised
battery storage is forecast by GP:2030, only 0.01 TWh/yr is supplied back into the
grid for balancing purposes. As the economics of batteries improve greater proportions
of the role domestic DSM currently plays in GP:2030 in balancing can be moved onto
battery capacity. While battery costs remain high it would be unwise to rely too heavily
on expensive batteries in modelling exercises.
5.5 Interconnectors
5.5.1 SHED shortcoming
It should be highlighted that SHED does not model the availability of power from the
non UK end of the 10 GW of interconnectors projected under GP:2030. This is a
modelling limitation, which is difficult to overcome as the whole European electricity
system would need to be considered and modelled to quantify the availability of power at
the non UK end of each interconnector. To the authors knowledge there is currently no
European wide electricity system model that achieves this. It should however be noted
that GP:2030 places interconnectors low down in the dispatch protocol and hence only
0.1 TWh/yr is imported through those interconnectors, compared to 0.7 TWh/yr from
pumped storage. In 2014, interconnectors to the continent supplied 16.6T W h/y r
66
Chapter 5. Discussion
into the UK, 10.3T W h from France and 6.3T W h from the Netherlands. This is
a much greater supply of energy than found by SHED under GP:2030, this is due to
interconnectors under SHED supplying periods of net demand post other balancing
mechanisms rather than those periods that are economically profitable to do so, as is
the case within the current electricity market.
5.5.2
Exports & surpluses

Whilst imported energy via interconnectors averages 0.1 TWh/yr, exported power
averages 12.2 TWh/yr. Hence the UK becomes a net, large, exporter of power. It
should be noted that this exported power is only exported once all UK demands are
met and storage facilities are at capacity. Even once the maximum amount of power
(determined by the interconnector capacity) has been exported the UK is able to utilise
the rest of the monthly surplus for hydrogen production for transport, totaling on average
42.8 TWh/yr. As was discussion in Section 3.4.6 this equates to supplying 438, 000
hydrogen cars with a continuous daily supply. Detailed considerations of surplus power
utilisation are beyond the scope of this report, hence this hydrogen production and use
should be treated as a guide to the possibilities, rather than a specification.
5.6 Pragmatic & climate driven sub-scenarios

Two sub-scenarios have been modelled, GP:2030:Pragmatic and GP:2030:ClimateDriven,
both of which have the same demand and supply inputs. Under GP:2030:Pragmatic
all balancing mechanisms work in collaboration with CCGTs to maintain an economically favourable average load factor on CCGT generators of 23.6 %. Under
GP:2030:ClimateDriven the balancing mechanisms work to reduce net demand as
much as possible prior to CCGTs to maximise CO2 reductions, hence the load factors
drop to 2.1 %. Although GP:2030:ClimateDriven reduces emissions further, the low
load factors would be unlikely (without additional subsidy) to result in investment in new
CCGTs. However, a substantial portion (PROBABLY NEED A NUMBER OFF DOUG
HERE) of the projected 19.5GW gas CCGT capacity in GP:2030 has already been
built. Additionally, the advent of the Governments Capacity Market, though less than
perfect, means that fixed costs associated with new CCGTs can be met by capacity
payments, as well as commercial returns via the energy market. Still, the authors
consider GP:2030:ClimateDriven predominantly as a useful guide until the scale of
reduction in average load factors has been taken into account within future plans for the
Capacity Market, or other subsidy and public finance mechanisms.
5.7
Dispatch Protocol
Currently generators bid into the energy market on various time scales based on supplydemand forecasts, the price of fuel and many other factors. The dispatch protocol
modelled here assumes the overriding objective governing generator dispatch is the
minimisation of CO2 emissions, particularly within GP:2030:ClimateDriven. In either
case, this would require the energy market to be reformed and preference given to
those generators with the lowest associated CO2 emissions (as is done with solar in
Germany), which are generally non-dispatchable. This therefore assumes that by 2030
the UK government will reform the energy market in order to achieve such an outcome,
which in turn produces the issues identified above regarding load factors on CCGT gas
turbines.
5.8 Limitations of hourly time-steps
5.8
67

Despite SHED utilising a large and relatively high-resolution data set in order to model
energy scenario outcomes it is, like any model, imperfect. One significant area for
improvement will be the reduction of hourly time-steps to smaller periods enabling
variations within the hour steps to be identified and modelled. Ensuring (currently
hidden) demand variations within the hour steps are met with sufficient demand will
require some additional reserve margin, unless met by additional storage or demand
side management, and this will have some added economic and carbon costs.
5.9
A final note
Due to electricity being a high grade form of energy enabling the use of new delivery
technologies, together with the principles and interpretation of mainstream economics,
policy makers and politicians have been guided towards utilising and planning for an
electric future. The interpretation of decoupling principles suggest that by substitution for
other energy forms, electricity could bring about both lowered emissions and increased
energy security. This substitution may not be straightforward, with challenges in supplydemand balancing likely to become an increasing problem in the short term. This is
important to remember when published energy scenarios visioning the energy systems
of the future forecast the aforementioned electrification and decarbonisation of the
energy system.
Decarbonisation
CCS, nuclear and technical feasibility
Electrification
The importance of heating demand reduction
targets
Cost
Dispatchable generator load factors
6. Conclusions
The stated aims of GP:2030 were:

1. Decarbonisation within the CCC targeted 50 - 100geq CO2 /k W h emission intensity factor, whilst;
Each of these aims has been achieved, as demonstrated in Section 3.4 and Chapter
4.
6.1
Decarbonisation
In GP:2030, we modelled two sub-scenarios. The first, GP:2030:Pragmatic, only
sought to reduce net demand after renewable supply, via balancing mechanisms, to
the level of the maximum capacity of fossil fuel dispatchable generators (see Section 3.2). These were then called on to meet as much of net demand as possible at this point. As explained, this sequence was designed to balance the economics of power station operation with the imperative to reduce the grids carbon
intensity. GP:2030:Pragmatic achieves a carbon intensity of 77.9 geq CO2 /k W h, falling
from the 1990 level of 770geq CO2 /k W h and from a contemporary level of around
500geq CO2 /k W h.
The second sub-scenario, named climate driven, differed only insofar as balancing
mechanisms were not prevented from reducing net demand after renewable supply
below the maximum capacity of fossil fuel dispatchable generators (i.e. they were
70
Chapter 6. Conclusions
allowed to reduce net demand at this point by as much as possible, see Section 3.2).
This arrangement prioritised maximum decarbonisation over economic considerations
of power station operation. GP:2030:ClimateDriven achieves a carbon intensity of 51.2
geq CO2 /kW h.
As stated in Section 3.3.3, non-Life Cycle Analysis values are used to calculate the
carbon intensity values stated above. It is worth noting for completeness that, using LCA
values, the carbon intensity of GP:2030:Pragmatic rises to 89.9 geq CO2 /k W h, while
that of GP:2030:ClimateDriven rises to 63.4 geq CO2 /kW h. Both still remain within the
target range of 50-100geq CO2 /kW h.
6.2 CCS, nuclear and technical feasibility

As is evident from Table 3.1, no new nuclear, or CCS fitted, generation is included
in GP:2030, for reasons relating to maturity, performance, risk and cost outlined in
more detail in Chapter 2. Otherwise, efforts have been made to restrain the technical
requirements of GP:2030 with a focus on mature technologies, or those operating in
some form already prior to being scaled: GP:2030 does not rely on the expectation of
major technological innovation, or invention, to be successful. It does however expect
the costs of certain existing technologies, such as batteries, to fall in order to help enable
the transformation it envisages - though such technologies have been de-prioritised
versus those where no such expectation is required, e.g. pumped storage. The success
of GP:2030 from a supply-demand balancing perspective helps vindicate the logic for
omitting costly or problematic technologies, such as CCS and nuclear, in the initial
stages.
6.3
Electrification
Ambitious but achievable decisions were made about the required electrification of heat
and transport - a vital component of the UKs broader decarbonisation.
As discussed in Section 3.3.1, 25 of space heating is expected to be electrified, via
heat pumps, by 2030. This then equates, in SHED, to moving 25 % of domestic space
heating demand onto the electricity system. Given the size of the UKs annual domestic
heating demand, this is a very substantial shift.
Electric vehicles are expected to create an additional annual demand of 32 T W h/y r .
Combined, these expectations mean GP:2030 achieves a significant level of electrification of demand currently supplied through direct combustion of fossil fuels. The
crucial role of the power sector in the UKs overall decarbonisation program depends on
being able to move demand onto the electricity network in this fashion.
6.4

Decarbonisation of the UKs economy via electrification of heating and transport demands is only available if the newly decarbonised power sector can absorb additional
demand while maintaining security of supply in a context of increased intermittency. In
( GP:2030) 76% of installed generation capacity is composed of non-dispatchable
renewable generators.
As is outlined in Chapter 2, it is to the question of balancing supply and demand in
this new world of energy supply and consumption that this report turned its full attention.
6.5 Domestic demand side management (DSM)
71
As stated in Section 3.4 GP:2030 experiences no black or brownouts for the 11

years of data, thus achieving a loss of load probability of 0. Given the imperfection of
modelling such complex future outcomes, the authors of this report do not wish to claim
that GP:2030 would be unable to experience brown or blackouts if built. For example,
we assume that the modelled domestic DSM requirements are in fact realised. However
it is clear that despite the electrification of heating and transport within the scenario,
and the hugely increased deployment of non-dispatchable renewable supply, GP:2030
achieves a supply that is at least as secure as is presently mandated. The Secretary of
State currently requires a LOLP of 9%, i.e. 9 instances of deficit over 100 years.
The majority of balancing is enabled by storage technologies, such as pumped
storage and batteries at both grid and household level. Storage of electricity is therefore
critical to maintaining a balanced system; however GP:2030 assumes storage levels
that are within reasonable expectations and does not rely on technologies not already
in operation.
6.4.1
The importance of heating demand reduction targets

It is one of the principle concerns of this report to reiterate the enormous importance
of meeting heating demand reduction targets in order to maintain supply-demand
balancing in a highly-renewable and electrified energy scenario. GP:2030 requires
a -57.2 % change in annual space heating demand by 2030. This is more ambitious
than the average of 47.5% reduction required across six prominent 2050 scenarios
in the literature (Quiggin, 2014). However, it should be noted that those scenarios do
not achieve acceptable levels of security of supply when tested in SHED, principally
because their heating demand reduction targets are insufficient and electrification levels
extremely high. Hence annual heating demands result in much greater peak electrical
demands driven by electrified heating. GP:2030 avoids these pitfalls by emphasising
heating reductions and containing heat electrification at lower, more manageable levels.
If heating targets are not met in GP:2030, and annual heating demands remain at
historic levels, then over the course of 11 years there would be 47 hours of deficits
totaling 141.9 GW h.
6.5

GP:2030 performs very well in balancing supply and demand, and is further strengthened by its demonstrable, quantified and reasonable reliance on domestic DSM to do so.
This is extremely important for two reasons. First; previous models have only been able
to assume a given flexibility in the demand-side of modelLed energy scenarios without,
for example, investigating whether there is sufficient supply either side of a peak period
to allow demand to be shifted. Second; domestic DSM, though technically feasible, is
as yet untested at scale and is therefore an unknown quantity given it relies on human
behaviour change and thus social conditions. A heavy, and un-modelLed, reliance on
domestic DSM within any energy scenario is consequently problematic because it both
expects much from an untested technology and fails to specify exactly what it actually
expects.
In GP:2030 domestic DSM is kept within reasonable limits, preventing onerous
responsibility on households to provide demand side balancing to the electricity system.
The majority of instances where households are required to shift their demand are below
0.2GW (at a national level), occurring on average 77/y r across the eleven years of data
that SHED runs over. There are only 6.8 instances per year where 3GW is required
72
at a national level from domestic DSM. The majority of balancing in GP:2030 is enabled
by storage mechanisms, as DSM requires conditions in which there is sufficient supply
within a 3 hour window on either side of a problematic demand peak - this is not always
the case.
6.6
6.6.1
Cost
In order to assess the economic feasibility of GP:2030 this report has chosen to compare
it to a full costed high-renewables scenario produced by Poyry in 2011 for the Committee
on Climate Change. It can be seen in Chapter 4 that the two scenarios are technically
similar enough to draw reasonable comparisons between, regarding potential costs.
Poyrys analysis shows that up to a 65% (2030) and 80% (2050) share of generation
from renewables results in additional intermittency costs to the consumers of 1 pence
per kWh. As stated above, GP:2030 assumes a 76% share by 2030. The CCC expect
that with 65% of renewables penetration, the average electricity generation cost in 2030
to be 8.2 13.8p/k W h, requiring 126 227bn of investment, roughly equivalent to
between one and two times the annual NHS budget, but spread over 15 years.
6.6.2
Dispatchable generator load factors

A key element of the operational economic feasibility of the power sector is the average
load factor for dispatchable fossil fuel generators called on to reduce net demand after
renewable supply and balancing mechanisms have been invoked. In order for the
building of power stations to be economically favourable, traditionally average load
factors must be maintained at levels that do not obscure the possibility of recouping
initial capital investment.
In 2013, with power generation from CCGTs at its lowest level since 1996, load
factors for CCGTs fell to a record low of 28%. In 2008 CCGT load factors were at an
eight year high of of 71.0% per cent. The intervening fall was due to an increase in
power generation from coal fired power stations, whose load factors reached 58% in
2013.
In GP:2030:Pragmatic mean load factors on CCGT generators are maintained at
an economically favourable level of 23.6 %. Within GP:2030:ClimateDriven this falls
substantially to 2.1 %. It should be clear from this that serious consideration was given
in the modelling process to this element - it is for consideration of the reader that each
sub-scenario with its attendant carbon intensities and load factors is presented.
It should be noted that dispatchable generation is a necessary component of an
electricity system that is required to balance each second of each day of each year. The
extent to which it is required is influenced by the overall demand peak values and how
far balancing mechanisms are enabled to reduce net demand after non-dispatchable
supply. If achieving the lowest technically feasible carbon intensity for the power sector
is the over-riding priority for the design of an energy scenario, consideration must be
given to how to fund, or incentivise the building of, economically unrewarding CCGT
power stations (or how to reduce demand in advance of GP:2030s projections). The
UKs newly inaugurated electricity capacity market is an initial, if uncertain step, in
addressing this issue and must be updated within the context of conclusions drawn from
modelling exercises such as this which project heavily reduced average load factors.
6.7 The possibility & the imperative
6.7
73

What GP:2030 shows is the real possibility of an ambitious carbon reduction plan for
the UKs power sector - and thus its economy more broadly - eschewing new nuclear
and CCS technology dependence in favour of non-dispatchable, highly renewable and
increasingly decentralised supply.
The huge scale of such a transformation ensures that its success will rely, amongst
other factors, on far-sighted political leadership. The replacement of the majority of the
UKs current electricity generation capacity, the roll-out of new, and relatively untested,
technologies into households, the rewriting of the electricity markets rulebook and shifts
in public attitudes and behaviour are but some of the prerequisites for a workable energy
future.
But what other choices are there? There will be choices between different national
energy plans, but the central themes of this report - the dilemma of increased electrical
demand peaks and increased supply intermittency - will affect any proposal that is
serious about taking responsibility for the UKs role in precipitating a climate crisis. We
either continue as normal, or we get serious about reducing space heating demand
rapidly; scaling up and innovating key technologies; and changing our relationships up
and down the scale with the production and consumption of energy.
Such systemic flux also presents exciting opportunities: far greater domestic energy
sufficiency, homes that are better insulated and cheaper to live in and community and
citizen ownership of national energy infrastructure - to name a few.
The authors wish to end this report by reiterating the urgency of the need to take
action on climate change. Even with a global temperature rise limited to 2C - the current
political consensus - the earths ecosystem will enter unknown territory, risking global
society in an unprecedented manner. As a nation it is the UKs responsibility to act
within the best of its knowledge and abilities, for the sake of every global citizen, to
prevent such unthinkable outcomes.
This requires enacting the most ambitious carbon reduction plans we think achievable. The Greenpeace:2030 report is dedicated to such an effort.
Definition of terms
Here terms that are used regularly throughout the document are defined;
Dispatchable generator; an electrical generator capable of varying its output in
accordance to the demand for electrical power, these are generally large scale
centralised power stations.
Non-dispatchable generator; an electrical generator whose output is dependant
on weather and climatic conditions, its output is referred to as intermittent or
variable as these generators do not respond dynamically to electrical demand
Capacity factor; the ratio between an electrical generators actual output over
a given period of time, to the potential output of that generator if it were able
to operate at its rated nameplate generation capacity. Capacity factors can be
calculated for both dispatchable and non-dispatchable generators.
Load factor; is the average load placed on a dispatchable generator divided by
the generation capacity of that generator
Availability factor; is the amount of time that a generator is able to produce
electricity over a certain period, divided by the amount of the time in the period.
Combined heat and power (CHP); a type of generator that produces heat simultaneously to electrical power, there are many forms of CHP units fuelled by
different liquid or gas fuels.
Traditional electricity; the electrical demand from both domestic and non-domestic
electricity consumers, with no contribution from either electric vehicles or heat
pumps
Carbon capture and storage (CCS); a technology currently being developed
designed to be fitted to fossil fuel generators that captures CO2 emissions from
those generators in order to reduce the emission intensity factor of those generators. The captured carbon is then transported and stored to prevent emissions
entering into the atmosphere.
Smart meter; is an energy (gas or electric) meter that records consumption of
energy in intervals of an hour or less communicating that information back to the
energy supplier for monitoring and billing purposes. Smart meters enable two-way
communication between the meter and the central system.
Time of Use tariffs (TOU); a type of metering and billing that employs smart
76
meters which are programmed to determine energy consumption at intervals
throughout the day, allowing energy suppliers to changes rates and charges based
supply and demand.
Overview
Dispatch protocol
Demand Modelling
Heating
Electric Vehicles
Line Losses
Non-domestic Loads
A. The SHED model
A.1
Overview
Here an overview of SHED is presented with further detailed descriptions given in
the following sections. This overview is designed to give a contextual understanding
of how the modelling components fit together. SHED is a hybrid top-down national
supply-demand model with a bottom-up household demand and DSM model, here an
overview of the top-down modelling methodology is presented, for a full description
please see Quiggin (2014). The top-down modelling is comprised of hourly historic
weather patterns, demand data and installed generator capacities. Supply-demand
modelling methodologies developed within the FESA model (Barton et al., 2013) have
been utilised and adapted.
Figure A.1: High level schematic of SHED modelling components. HP = heat pump, EV
= electric vehicle, DSM = demand side management
78
A.2
Chapter A. The SHED model
The renewables modelled include; wave, solar PV, tidal stream, onshore and offshore
wind. Hourly renewable supply is composed of historic weather data, the specific
technology characteristics and sub-division of the country into weighted regions, to form
hourly national capacity factors1 for each renewable technology. The scenario specific
national generator capacities are combined with these capacity factors to form scenario
specific hourly renewable generation. The weightings between sub regions are based
on the total possible available resource, which originate from UK government estimates
(DTI (1998)). Hourly weather data, to model renewable generation, was obtained from
the British Atmospheric Data Centre (BADC) (UK Meteorological Office, 2011) with the
exception of wave data, which the Met Office (UK Meteorological Office, 2013) supplied.
SHED utilises eleven years of data enabling a variety of weather patterns and variations
in demand to be represented, those years being 2001 2011.
A.3
Dispatch protocol
Currently electricity generators bid into the market on various time scales based on
demand forecasts, the price of fuel and many other factors. The simplified dispatch
protocol within SHED assumes the overriding objective governing generator dispatch is
the minimisation of CO2 emissions. This would require the energy market to be reformed
and preference given to those generators with the lowest associated CO2 emissions.
Under this methodology renewable generators are left to generate uncurtailed. So too
are CHP generators, which are modelled as following heat demands.
A.4
Demand Modelling
Figure A.2 illustrates the demand model components, these components are described
in further detail in the following sections. Historic national half hourly electricity demand
data from ELEXON (The NETA Web Site, 2013) forms the basis of deriving traditional
electricity demands. Traditional electricity demand is the electrical demand from both
domestic and non-domestic electricity consumers, with no contribution from electric
vehicles, heat pumps or economy seven. It is the future electrical demand that is
similar to current electrical demands, but as energy scenarios specify the level economy
seven or resistive heating this is removed and re-introduced later. The ELEXON data
encompasses all national domestic and non-domestic loads, as well as economy seven
water and space heating. This data is scaled by each scenarios annual demand
projections once economy seven water and space heating are removed. Historic
hourly temperature data from three Met Office Integrated Data Archive System (MIDAS)
weather stations form the basis of calculating the number of Heating Degree Hours
(HDH), which are scaled by the national annual delivered space heating demands2
(DSHD) data, to give hourly DSHD. Alongside hourly DSHD, hourly delivered water
heating demands are calculated enabling hourly heat pump electricity demands to be
found, along with solar thermal and CHP outputs. The approach taken within SHED
is to model solar thermal water heating demands first, such that heat pumps follow a
reduced net heating demand. Electric vehicle (EV) charging profiles (Acha et al., 2011)
1
Capacity factor; the ratio between an electrical generators actual output over a given period of time, to
the potential output of that generator if it were able to operate at its rated nameplate generation capacity.
2
Delivered space heating demand is the energy demand at the point of use, rather than the energy in
the fuel consumed
A.4 Demand Modelling
79
are utilised to derive the scenario specific EV hourly demands. Finally the resistive
heating demands are calculated utilising current economy seven water and space
heating profiles.
The final hourly electricity demand is, at a high level perspective, the sum of traditional electricity, electrical vehicle, heat pump, resistive heating and resistive heating
demands.
Figure A.2: Detailed representation of the components of the demand model elements
of SHED. Target output highlighted in yellow. HP = heat pump, EV = electric vehicle, E7
= economy seven
A.4.1
Heating
Only recently in publications by Sansom and Strbac (2012) and Wilson et al. (2013) has
the electrification of heating been considered to drive peak electrical demands to levels
80
where reliability of supply is potentially compromised. Wilson et al. (2013) quantified

daily demands, whilst Sansom and Strbac (2012) utilised half hourly gas demands to
derive heating demands.
To calculate hourly electrified heating demands, hourly delivered heating demands
are required and are comprised of both domestic and non-domestic water and space
heating. Daily demand profiles for unrestricted space and water heating originate from
heat flow measurements within a district heating scheme of a social housing complex
(Woods and Dickson, 2004). The normalised profiles are shown in Figure A.3. Hourly
water heating demands are derived from the 24 hour July unrestricted heating profile
and scaled by the annualised national delivered water heating demand. For space
heating (psp (t)) it is assumed there is no demand during high summer (July and August).
These profiles are normalised across the 24 hours of the day such that, for instance,
8.4% of the total daily space heating load occurs at 8am (see Figure A.3). Hourly
temperature data forms the basis of calculating the number of HDH which are scaled by
the annual delivered space heating demands and unrestricted profile of Figure A.3, to
derived hourly DSHD.
(a) Unrestricted heat demand profiles. Source :

Woods and Dickson (2004).
(b) Economy seven heat demand profiles.
Figure A.3: Heat demand profiles: percentage of demand occurring each hour of a
typical day.
A degree hour is the number of degrees Celsius by which the hourly average outside
temperature is below a no heating temperature. The no heating temperature, TNH , is
the temperature at which no heating is required to maintain sufficient inside temperature,
and is dependent on; building characteristics, heating equipment used, number of
occupants and their behaviour, TNH = 15.5o C. The smoothed temperature, Ts , given
by the moving average temperature of Equation A.1 is taken as the outside ambient
temperature from which the number of degree hours, DH , is found (Equation A.2). This is
the difference between the smoothed temperature, Ts , and the no heating temperature,
TNH , where DH is zero if Ts is greater than TNH .
A.4 Demand Modelling
81
Ts (t) =
T (t)
t23 24
(A.1)
where; T (t) is the hourly outside ambient temperature
DH (t) =
(
TNH Ts (t)
0
if TNH > Ts (t)

if TNH Ts (t)
(A.2)
The majority of UK housing, commerce, and industry is located in England. The

method of using an area enclosing the majority of the building stock is common within
representative temperature measurements (Parker et al., 1992). The daily Hadley
Centre Central England Temperature (HadCET) record is representative of a triangular
area enclosed by Lancashire, London and Bristol (Parker et al., 1992). As hourly, rather
than daily mean temperatures are required, HadCET is insufficient for the purposes of
SHED. FESA utilised the Rutherford Appleton Laboratory (RAL) dataset in Oxfordshire
to derive hourly values, and calibrated the data against HadCET. The HadCET daily
mean temperatures are derived from three equally weighted weather stations (Parker
and Horton, 2005); Rothamsted near London, Pershore on the boarder of south Wales
and Stonyhurst just north of Blackburn in Lancashire. The MIDAS dataset contains
hundreds of land based weather stations, so hourly temperatures were obtained from
weather stations as close to the HadCET stations as possible.
Hourly DSHD, S(t), is the product of the annual DSHD (S) and the HDH, normalised
by the total number of HDH in the year to give an effective degree hour capacity factor,
shown in Equation A.3. Equation A.3 gives the hourly flat DSHD with the unrestricted
profile (Figure A.3(a)) not accounted for.
S(t) = S.
DH (t)
n1 DH (t)
(A.3)
where; S(t) is the hourly flat DSHD;

S is the annual delivered space heating annual demand;
DH (t) is the number of degree hours; and,
n is the number of hours in the year.
To account for the daily unrestricted heating profile the total flat DSHD each day is
proportioned across the 24 hours by the unrestricted profile. The resulting unrestricted
demand profile is given by Equation A.4. The hourly flat (or restricted) DSHD is used to
determine non-domestic hourly heating demands.
82
24
S p (t) = S(t).
1
psp (t)
24
1 psp (t)
(A.4)
where; S p (t) is the hourly unrestricted DSHD;

psp is the space heating demand profile of Figure A.3; and,
S(t) is the hourly flat DSHD of Equation A.3.
Heat pumps
HP (t)) has been modelled as following hourly delivered
Heat pump electricity demand (Esc
space and water heating demands. The scenario defined fraction of heat supplied by
heat pumps (f HP ) and the coefficient of performance of the heat pumps (COP HP ), give
Equation A.5.

HP
Esc
(t) = Ssc (t) + Wsc (t) .f HP /COP HP
A.5
(A.5)
Electric Vehicles
It has been assumed that EVs are not capable of providing power back to the grid, hence
EVs are unable to provide a balancing services to the grid, which many studies have
suggested would be possible but expensive (Kempton and Tomic, 2005). The driving
and charging profile used to model hourly EV demands is a modified DSM profile (found
in Quiggin (2014)), under which DSM has been accounted for (Acha et al. (2011)).
A.6
Line Losses
The electricity system is considered as a single node, as such transmission and distribution losses are accounted for by a simple increase of 7%, regardless of local generation
levels but assuming a small increased efficiency of the electrical network by 2030.
These losses are applied to the total electrical demands. Typical historic transmission
losses as recorded by the National Grid Company are less than 2% (National Grid,
2009). In 2000/01 the average losses reported to Ofgem across all 14 Distribution
Network Operators was 7%, with a range between 5.4% to 9.1% (Ofgem, 2003).
A.7
Non-domestic Loads
The traditional electricity loads of both domestic and non-domestic are treated together
under the scaled Elexon data. Electrified space and water heating are treated in a similar
way to domestic demands, but a daily flat (or restricted) heating profile is applied rather
than the unrestricted profile. The non-domestic DSM dynamics are treated separately
to domestic DSM, for further information see Quiggin (2014)
B. Weather stations & renewable modelling
For a full description of renewable modelling methodologies applied within GP:2030

please see Quiggin (2014).
Figure B.1 shows the location of those weather bouys, three of which are located
off the south west and Wales coastline; Turbot Bank, Seven Stones and Aberporth and
two are situated off the west coast of Scotland and Ireland; K4 and K5. The country
was then subdivided into two regions; North and South West based on the resource
assumptions given in the DTI Marine Resource Atlas DTI, 2011 (Thorpe, 1999), such
that variation in generation across the country was accounted for. This resulted in a
North region weighting of 0.77 and consequently a South West weighting of 0.23.
Capacity factors for solar PV are calculated using hourly global irradiation data from
32 Met Office weather stations (UK Meteorological Office, 2011). Due to data quality
across the eleven years many potential weather stations had to be discarded. It is
assumed the majority of installations are roof mounted, which follows the methodology
of the DTI (1998) report. As such UK sub regions are weighted by building and,
more accurately, urban area distributions, for which statistics were obtained from the
Department for the Environment Food and Rural Affairs (2003) and The Office for the
Deputy Prime Minister (2004) (ODPM). As was found by Forrester (2005) 73% of the roof
mounted solar resource is situated in England, which is then divided into 9 sub-regions.
Scotland, Wales and Northern Ireland are classified as individual sub-regions resulting
in a total of 12 regions. These subregions can be seen along with corresponding
weather stations in Figure B.1. The weightings between regions, along with the raw
data, can be found in Quiggin (2014). In essence the weightings can be thought of as a
percentage of the total UK buildings that each region contains.
For onshore and offshore regional weightings, and capacity factors, to be calculated the available resource and associated constraints are required. Assumptions
and methodology from the DTI (1998) are again utilised. The breakdown of onshore
subregions can be seen in Figure B.2, with the total available resource and weightings
given in Quiggin (2014). England and Wales constitute one region as does Northern
Ireland, with fourteen subregions in Scotland. The breakdown of offshore subregions
can be seen in Figure B.3, with one square representing a 1GW wind farm. The total
available resource and weightings per region are given in Quiggin (2014). For onshore
84
Chapter B. Weather stations & renewable modelling
Figure B.1: Location of wave buoys and solar irradiation weather stations.
resource allocations, the variables within the DTI (1998) report include socio-economic
factors, along with size and spacing of the onshore sites. Also included are protected
areas such as National Parks and green-belt land as well as the location of urban areas,
agricultural land and lakes, and other geographic constraints. The DTI (1998) report
utilised wind atlases to determine those areas with annual mean wind speeds (AMWS)
of 7ms1 or more. This is the minimum mean wind speed at which wind generation
becomes economically viable with modern wind turbines. With the majority of the wind
resource contained within Scotland, this region was subdivided into further smaller
regions, with consequently more weather stations found for these areas.
In calculating the total possible available resource for offshore wind, the constraints
differ to onshore. There are no National Parks and cities to consider but there are sites
of special scientific interest, shipping lanes and protected areas. The main constraint,
however, is the depth of sea bed or bathymetry where the depth is required to be less
than 40m out to 30km. Again a detailed description of assumptions made in calculating
the regional resource can be found in the DTI (1998) report. The practical available
resource including all constraints was calculated to be 100T W h/y, whereas the total
possible available resource with no constraints other than sea bed depth, utilised
85
within SHED in calculating the capacity factors, is cited within the DTI (1998) report
as 2851T W h/y . This no constraint total resource was used rather than constrained
resource. As the wind speeds at hub heights of the 2MW and 5MW turbines meet the
required AMWS of 7ms1 in nearly all regions of the UK coastline, the main constraint is
simply the bathymetry. This then means the map shown in Figure B.3 broadly represents
the sea bed depth, rather than any constraints from shipping lanes etc.
Figure B.2: Location of wind weather stations.
86
Figure B.3: Location of offshore wind weather stations.
Bibliography
Acha, S., Green, T. C. and Shah, N. (2011), Optimal Charging Strategies of Electric
Vehicles in the UK Power Market, in Innovative Smart Grid Technologies (ISGT) IEEE
PES, pp. 18.
Ault, G., Frame, D., Hughes, N., Strachan, N., Watson, J. and Pollitt, M. (2008), Electricity
Network Scenarios for Great Britain in 2050 Final Report for Ofgems LENS Project,
Technical Report 157.
Baker, W. (n.d.), Off-gas consumers. Consumer Focus, Technical report.
Barnacle, M., Robertson, E., Galloway, S., Barton, J. and Ault, G. (2013), Modelling
generation and infrastructure requirements for transition pathways, Energy Policy
52, 6075.
URL: http://linkinghub.elsevier.com/retrieve/pii/S030142151200331X
Barton, J., Huang, S., Infield, D., Leach, M., Ogunkunle, D., Torriti, J. and Thomson, M.
(2013), The evolution of electricity demand and the role for demand side participation,
in buildings and transport, Energy Policy 52, 85102.
URL: http://linkinghub.elsevier.com/retrieve/pii/S0301421512007227
BBC (2014a), Jobs go after no buyer found for Pelamis wave business.
URL: http://www.bbc.co.uk/news/uk-scotland-scotland-business-30560980
BBC (2014b), Leighton Buzzard electricity facility tests largest battery.
URL: http://www.bbc.co.uk/news/uk-england-beds-bucks-herts-30476591
Bloomberg (2014), While You Were Getting Worked Up Over Oil Prices, This Just
Happened to Solar.
URL: http://www.bloomberg.com/news/articles/2014-10-29/while-you-were-gettingworked-up-over-oil-prices-this-just-happened-to-solar
Bossel, B. U. (2006), Does a Hydrogen Economy Make Sense?,in Proceedings of the
IEEE October 2006, Vol. 94.
88
CEC (2008a), Commission Staff Working Document: The support of electricity from
renewable energy sources, Technical report, European Commission, Brussels.
CEC (2008b), Impact Assessment: document accompanying the package of implementation measures for the EUs objectives on climate change and renewable energy for
2020, Technical report, European Commission, Brussels.
CEC (2008c), Proposal for a directive on the promotion of the use of energy from renewable sources. Technical report, Technical report, European Commission, Brussels.
CEC (2008d), The support of electricity from renewable energy sources, Technical
report, European Commission, Brussels.
Centre For Alternative Technology (2014), Zero Carbon Britain, Technical report.
Committe (2014), Meeting Carbon Budgets 2014 Progress Report to Parliament,
Technical Report July.
Committe on Climate Change (2008), The Forth Carbon Budget Reducing emissions
through the 2020s; Chapter 6: Power sector decarbonisation to 2030, Technical
report.
Committee on Climate Change (2011), The Renewable Energy Review, Technical
Report May.
Darby, S. (2006), The effectiveness of feedback on energy consumption. A Review for
DEFRA of the Literature on Metering, Billing and direct Displays, Technical Report
April, DEFRA.
Darby, S. (2010), Smart metering: what potential for householder engagement?,
Building Research & Information 38(5), 442457.
URL: http://www.tandfonline.com/doi/abs/10.1080/09613218.2010.492660
Department for Energy and Climate Change (2009), Smarter Grids: The Opportunity,
Technical Report December.
Department for the Environment Food and Rural Affairs (2003), Land by agricultural
and other uses.
URL: http://www.defra.gov.uk/environment/statistics/land/lduse.htm
Department for Transport (2014), National Travel Survey : England 2013, Technical
Report July.
Department of Energy and Climate Change (2010), 2050 Pathways Analysis, Technical
Report July.
Department of Energy and Climate Change (2011), Department of Energy and Climate
Change Review of the generation costs and deployment potential of renewable
electricity technologies in the UK Study Report, Technical Report 215030.
Department of Energy and Climate Change (2012a), Electricity Generation Costs,
Technical Report October.
Department of Energy and Climate Change (2012b), The Future of Heating : A strategic
framework for low carbon heat in the UK, Technical Report March.
89
Department of Energy and Climate Change (2014), Digest of UK Energy Statistics,
Technical report.
Department of Energy and Climate Change (2015a), Electricity Market Reform Announcement of de-rating methodology for interconnectors in the Capacity Market,
Technical Report February.
Department of Energy and Climate Change (2015b), Monthly deployment of all solar
photovoltaic capacity in the United Kingdom..
URL: https://www.gov.uk/government/statistics/solar-photovoltaics-deployment
Department of Trade and Industry (1998), New and Renewable Energy: Prospects in
the UK for the 21st Century: Supporting Analysis , ETSU R-122, Technical report.
URL: http://webarchive.nationalarchives.gov.uk/+/http://www.dti.gov.uk/renewables/publications/pdfs/supp
Department of Trade and Industry (2011), Marine Resource Atlas.
URL: http://www.dti.gov.uk/renewables/renew_atlaspages.htm
Druckman, a. and Jackson, T. (2008), Household energy consumption in the UK: A
highly geographically and socio-economically disaggregated model, Energy Policy
36(8), 31773192.
EA Technology (2011), Delivering the Benefits of Smart Appliances, Technical Report
September, London: DEFRA.
Electricity Networks Strategy Group (2010), A Smart Grid Routemap, Technical Report
February, UK Government.
EST (2010), Getting warmer: a field trial of heat pumps The Energy Saving Trust,
Technical report.
Flexitricity website (2011).
URL: http://www.flexitricity.com
Forrester, M. (2005), Design Methodology and Operational Working of a Spreadsheet
Tool for Aggregation of Renewable Resource Yields, Undergraduate research project,
Technical report, Centre for Renewable Energy Systems Technology, Loughborough
University, Loughborough.
Foxon, T. J. (2013), Transition pathways for a UK low carbon electricity future, Energy
Policy 52, 1024.
Garrad Hassan (2011), UK Generation and Demand Scenarios for 2030 (WWF), Technical Report 0.
Greenpeace (2008), Carbon Capture and StorageCarbon Capture and Storage.
http://www.greenpeace.org/international/en/campaigns/climateURL:
change/coal/carbon-capture-and-storage/
Hargreaves, T., Nye, M. and Burgess, J. (2010), Making energy visible: A qualitative
field study of how householders interact with feedback from smart energy monitors,
Energy Policy 38(10), 61116119.
URL:
http://linkinghub.elsevier.com/retrieve/pii/S030142151000460X
http://dx.doi.org/10.1016/j.enpol.2010.05.068
90
Henderson, R. (2006), Design, simulation, and testing of a novel hydraulic power takeoff system for the Pelamis wave energy converter, Renewable Energy 31(2), 271283.
IEA (2013), Vehicle energy efficiencies, Technical report.
Intergovernmental Panel on Climate Change (2011), Renewable Energy Sources and
Climate Change Mitigation, Cambridge University Press.
Kempton, W. and Tomic, J. (2005), Vehicle-to-grid power fundamentals: Calculating
capacity and net revenue, Journal of Power Sources 144(1), 268279.
Lannen, N. (2012), New Pumped Storage Proposals, Technical report.
URL:
http://www.theiet.org/communities/powergen/all-energy/pumpedstorage.cfm?type=pdf.
Lew, D., Brinkman, G., Kumar, N., Besuner, P., Agan, D. and Lefton, S. (2012), Impacts
of Wind and Solar on Fossil-Fueled Generators Preprint, Technical Report August.
Lindley, D. (2010), Smart grids: The energy storage problem., Nature 463(7277), 18
20.
URL: http://www.ncbi.nlm.nih.gov/pubmed/20054372
Lund, H., Mller, B., Mathiesen, B. and Dyrelund, a. (2010), The role of district heating
in future renewable energy systems, Energy 35(3), 13811390.
URL: http://linkinghub.elsevier.com/retrieve/pii/S036054420900512X
Mackay, D. (2009), Sustainable Energy without the hot air.
National Grid (2009), Investigation Into Transmission Losses on UK Electricity Transmission System, Technical Report April 2005, National Grid Company.
National Grid Company (2013), Interconnectors Interconnectors Potential future
interconnector opportunities, Technical Report February.
http://www.nationalgrid.com/NR/rdonlyres/44AEE96A-A93F-499A-9587URL:
E419A1ABE051/58992/Interconnectors130214_v12.pdf
Nuclear Energy Agency OECD (2011), Technical and economic aspects of load following
with nuclear power plants, Technical report.
URL: http://www.oecd-nea.org/ndd/reports/2011/load-following-npp.pdf
Office for National Statistics (2005), Area classification for Output Areas.
URL:
http://www.ons.gov.uk/ons/guide-method/geography/products/areaclassifications/ns-area-classifications/index/cluster-summaries/outputareas/index.html
Ofgem (2003), Electricity distribution losses A consultation document, Technical report,
Ofgem.
URL: http://www.ofgem.gov.uk/Networks/ElecDist/Policy/DistChrgs/Documents1/136203distlosses.pdf
Ofgem (2010a), Creating Britains Future Today, Technical report.
URL: www.ofgem.gov.uk/pages/moreinformation.aspx?docid=123&refer=media/factsheets
91
Ofgem (2010b), Energy Demand Research Project Fifth progress report, Technical
report.
Ofgem (2012), Demand side response in the non-domestic sector, Technical Report
May.
Parker, D. and Horton, B. (2005), Uncertainties in central England temperature 18782003 and some improvements to the maximum and minimum series, International
Journal of Climatology 25(9), 11731188.
URL: http://doi.wiley.com/10.1002/joc.1190
Parker, D., Legg, T. and Folland, C. (1992), A new daily Central England Temperature
Series, 1772-1991, Int. J. Clim 12, 317342.
Pitt, A., Buckland, R., Antonio, P. D., Lorenzen, H. and Edwards, R. (2015), Investment
Themes in 2015, Technical Report January.
Poyry (2011), Analysing Technical Constraints On Renewable Generation To 2050. A
report to the Committee on Climate Change, Technical Report March.
Quiggin, D. (2014), Modelling the expected participation of future smart households in
demand side management, within published energy scenarios, PhD thesis, Loughborough University.
URL: https://dspace.lboro.ac.uk/2134/16220
Renewable UK (2013), UK Wind Energy Database (UKWED).
http://www.renewableuk.com/en/renewable-energy/wind-energy/uk-windURL:
energy-database/index.cfm
Ricardo-AEA (2013), Projections of CHP capacity and use to 2030 : Report for DECC,
Technical Report 1.
Rob, W. (2014), Barclays Has The Best Explanation Yet Of How Solar Will Destroy
Americas Electric Utilities.
URL: Barclays Has The Best Explanation Yet Of How Solar Will Destroy Americas
Electric Utilities Read more: http://www.businessinsider.com/barclays-downgradesutilities-on-solar-threat-2014-5#ixzz3SxxTNCy0
Sansom, R. and Strbac, G. (2012), The impact of future heat demand pathways on the
economics of low carbon heating systems, in BIEE 9TH ACADEMIC CONFERENCE, Oxford.
Scottish Canals (2013), Coire Glas Application, Technical report.

URL: http://www.sepa.org.uk/water/water_regulation/advertised_applications/idoc.ashx?docid=8920d513
2e39-410b-b597-8558e1e33dc0&version=-1.
Shrubsole, G. and Cameron, A. (2014), The Big Six On The Run: Friends Of The Earth,
Technical report.
Strbac, G., Shakoor, a., Black, M., Pudjianto, D. and Bopp, T. (2007), Impact of wind
generation on the operation and development of the UK electricity systems, Electric
Power Systems Research 77(9), 12141227.
92
The Commission for Energy Regulation (2011), Electricity Smart Metering Customer
Behaviour Trials Findings Report - Part 1, Technical report.
URL:
http://www.cer.ie/en/information-centre-reports-andpublications.aspx?article=5dd4bce4-ebd8-475e-b78d-da24e4ff7339
The Economist (2015), Invisible fuel.
URL: http://www.economist.com/news/special-report/21639016-biggest-innovationenergy-go-without-invisible-fuel
The Guardian (2012), How much do we spend on nuclear waste?.
URL: http://www.theguardian.com/environment/2012/nov/16/nuclear-waste-sellafield
The National Grid Company (2014), STOR End of Year Report, Technical report.
The NETA Web Site (2013), ELEXON.
URL: www.bmreports.com
The Office for the Deputy Prime Minister (2004), Generalised Land Use Database
Statistics for England.
Thorpe, T. (1999), A Brief Review of Wave Energy, ETSU R-120, Technical report.
UK Meteorological Office (2011), MIDAS Land Surface Stations data (1853-current):
Available from http://badc.nerc.ac.uk/view/badc.nerc.ac.uk__ATOM__dataent_ukmomidas, NCAS British Atmospheric Data Centre .
URL: www.badc.rl.ac.uk/
UK Meteorological Office (2013), Met Office Wave Data, Technical report, UK Meteorological Office.
Vickers, D. and Rees, P. (2007), Creating the UK National Statistics 2001 output area
classification, Journal of the Royal Statistical Society: Series A (Statistics in Society)
170(2), 379403.
URL: http://doi.wiley.com/10.1111/j.1467-985X.2007.00466.x
Vickers, D., Rees, P. and Birkin, M. (2005), Creating the national classification of census
output areas : data, methods and results, Technical Report July.
Wilson, I. G., Rennie, A. J., Ding, Y., Eames, P. C., Hall, P. J. and Kelly, N. J. (2013),
Historical daily gas and electrical energy flows through Great Britains transmission
networks and the decarbonisation of domestic heat, Energy Policy .
Woods, P. and Dickson, J. (2004), The contribution CHP can make to energy efficiency
in buildings, in IMechE Seminar, Energy Efficiency in Buildings.
WWF (2014), Warm homes , not warm words : How the UK can move to a low-carbon,
Technical report.

Greenpeace 2030 Public Sept2015 Lowres

Uploaded by

Copyright:

Available Formats

Greenpeace 2030 Public Sept2015 Lowres

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Greenpeace 2030 Public Sept2015 Lowres

Uploaded by

Copyright:

Available Formats

2030 Energy Scenarios

Dr. Daniel Quiggin

c 2015 Dr Daniel Quiggin

Time for change

Framing the UKs challenge

The question & the test

The Greenpeace 2030 Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Aims & Objective of the Greenpeace 2030 scenario

The Dispatch Protocol

Pragmatic & Climate driven sub-scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

Outputs & Results

Costs of similar scenarios

The role of traditional electricity demand reduction . . . . . . . . . . . . . . . . . . . . . . . 63

Summer CHP & emergency CHP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

Pragmatic & climate driven sub-scenarios

Limitations of hourly time-steps

CCS, nuclear and technical feasibility

Balancing supply & demand

The importance of heating demand reduction targets . . . . . . . . . . . . . . . . . . . . . 71

Domestic demand side management (DSM)

The possibility & the imperative

The SHED model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

Weather stations & renewable modelling . . . . . . . . . . . . . . . . . . . . . 83

Chapter 1. Executive Summary

Electrification refers to the intention to move substantial proportions of energy demand,

produce larger overall peak demands on the electricity network.

Larger peaks in demand require larger reserves of generation capacity in order

Decarbonisation refers to the reduction to as close to zero as possible of the carbon

Chapter 1. Executive Summary

Chapter 1. Executive Summary

Figure 1.4: Overview of supply & demand

Chapter 1. Executive Summary

Time for change

Time for change

are often dependent on imperfect modelling exercises, unable to demonstrate strategies

2.2 Framing the UKs challenge

2.2 Framing the UKs challenge

stations, a future generation portfolio dominated by renewables will be subject to the

Opportunities: The falling cost of renewables

2.2 Framing the UKs challenge

disruptive power of renewables when the appropriate policies are in place.

investment in renewables or in generation from fossil fuels."

Demand side management (DSM, sometimes referred to as "demand side response",

2.2 Framing the UKs challenge

manage demand from larger household appliances (e.g. fridge-freezers) in response to

The question & the test

2.3 The question & the test

capacity margin (%) =

total available capacity peak demand

The SHED model

as SHED is an improvement on the majority of energy scenario modelling, an hourly

Aims & Objective of the Greenpeace 2030

3. The Greenpeace 2030 Scenario

Aims & Objective of the Greenpeace 2030 scenario

Chapter 3. The Greenpeace 2030 Scenario

The Dispatch Protocol

Pragmatic & Climate driven sub-scenarios

see section 3.3.2 for discussion on the viability of ramping nuclear

CHP and base-load nuclear