Available via license: CC BY 4.0
Content may be subject to copyright.
Journal Pre-proof
Centralized vs. distributed energy storage systems: The case of residential solar PV-
battery
Behnam Zakeri, Giorgio Castagneto Gissey, Paul E. Dodds, Dina Subkhankulova
PII: S0360-5442(21)01691-1
DOI: https://doi.org/10.1016/j.energy.2021.121443
Reference: EGY 121443
To appear in: Energy
Received Date: 31 December 2020
Revised Date: 4 July 2021
Accepted Date: 6 July 2021
Please cite this article as: Zakeri B, Gissey GC, Dodds PE, Subkhankulova D, Centralized vs. distributed
energy storage systems: The case of residential solar PV-battery, Energy, https://doi.org/10.1016/
j.energy.2021.121443.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2021 The Author(s). Published by Elsevier Ltd.
Conceptualization: B.Z., G.C.G., P.E.D., D.S.
Methodology, software, data curation, and formal analysis: G.C.G., B.Z.
Writing - Original Draft: B.Z., G.C.G.
Writing - Review & Editing: B.Z.
Visualization: B.Z., G.C.G.
Supervision: P.E.D.
Funding acquisition: G.C.G., P.E.D.
Journal Pre-proof
Centralized vs. distributed energy storage systems: The
case of residential solar PV-battery
Behnam Zakeri a,b,c,d,*,¥, Giorgio Castagneto Gissey b,¥, Paul E. Dodds b, Dina Subkhankulova b
Distributed energy storage is a solution for balancing variable renewable energy such as solar
photovoltaic (PV). Small-scale energy storage systems can be centrally coordinated to offer different
services to the grid, such as balancing and peak shaving. This paper shows how centralized and
distributed coordination of residential electricity storage could affect the savings of owners of
battery energy storage and solar PV. A hybrid method is applied to model the operation of solar PV-
storage for a typical UK householder, linked with a whole-system power system model to account
for long-term energy transitions. Based on results, electricity consumers can cut electricity bills by
28-44% using storage alone, 45-56% with stand-alone solar PV, while 82-88% with PV-battery
combined. Centralized coordination of home batteries offers 10% higher benefits compared to
distributed operation. Under centralized coordination, consumers without onsite energy
technologies benefit almost double compared to PV-battery owners, because peak electricity prices
decline in the system for all consumers. Therefore, the economic benefits of aggregation may be
redistributed to incentivize prosumers with PV-battery to join such schemes, who can balance their
electricity demand even without coordination. The private value of distributed energy storage
declines as more storage owners join the coordination scheme.
Keywords: electrical energy storage; energy policy; energy system model; decentralized energy;
value of energy storage; smart energy systems
a Energy, Climate, and Environment Program, International Institute for Applied Systems Analysis (IIASA),
Austria
b UCL Energy Institute, University College London, UK
c Energy Efficiency and Systems, Aalto University, Espoo, Finland
d Sustainable Energy Planning, Aalborg University, Copenhagen, Denmark
* Corresponding Author: Behnam Zakeri; +43(0)6767073682; zakeri@iiasa.ac.at. Address: IIASA, Schlossplatz 1,
Laxenburg, Austria
¥ Authors with equal contribution.
Journal Pre-proof
1
1 Introduction
1.1 Distributed solar PV and energy storage
Many governments worldwide plan to increase the share of renewable energy for
environmental, economic, and energy security reasons. For achieving renewable energy targets,
different incentives and support schemes have been put in place to promote the deployment of
renewable energy through decentralized and distributed generation, e.g., through solar photovoltaic
(PV) at consumer sites.
Electricity generation from solar PV is not always correlated with electricity demand. For
example, in cold climate countries electricity demand peaks typically happen in the evenings when
there is no solar energy [1]. There are different solutions for increasing the consumption of solar PV
onsite, or so called “self-consumption”, which can maximize the benefits of distributed energy
generation and minimize the electricity bills of the PV owner [2]. One of the common solutions is to
export extra electricity from solar PV to the grid. However, in large-scale penetration of distributed
solar PV, the export of electricity from many buildings to the distribution grid at peak generation
times will cause contingencies and grid imbalances [3], resulting in additional costs for the system
[4]. Moreover, the value of self-consumption of solar electricity for the private owner is typically
much higher compared to the gains from exporting electricity to the grid, as export tariffs are
typically lower than purchasing electricity prices [5]. Therefore, the private owner of solar PV prefers
to find different ways to increase their self-consumption, e.g., by storing electricity via electrical
energy storage
1
(EES) systems such as batteries [6].
EES can balance the mismatch between onsite solar PV generation and electricity demand by
storing electric energy at hours of low demand in daytime and discharging that to meet evening
peaks. Different studies have shown that pairing solar PV with batteries (PV-EES) increases self-
consumption of solar energy onsite [7] and can offer significant cost savings to the private owner.
For example, Zhang et al. [8] shows that paring solar PV with a home battery in California and
Hawaii is a feasible investment with a payback period of less than 10 years for different building
types, while others demonstrate possible cost savings for PV-battery owners in high latitude
countries in Europe under different energy storage policies [9]. Also, from the system operator’s
perspective, distributed EES devices can contribute toward balancing the (distribution) grid by
reducing peak contingencies [10] and grid management costs [11]. This can offer the Transmission
and Distribution (T&D) grid operator significant cost savings for postponing T&D investments and
grid fortification measures at the low-voltage level [12,13].
1
The terms EES, “electricity storage”, “energy storage”, and “storage” are interchangeably used in this paper for
referring to technologies that can store electricity and discharge it back at a reasonable response time. Examples of such
technologies include secondary electro-chemical batteries, flow batteries, pumped hydropower storage (PHS), etc.
Journal Pre-proof
2
However, the cost of batteries are still at the start of their learning curves [14], which diminishes
the financial viability of investment in such technologies, from a private owner’s perspective [2].
Different studies show that a PV-EES system is not economically viable under current market
conditions in different countries without additional financial supports [15]or policy incentives
[16,17]. These policies are, for example, capital subsidies [8], enhanced time-of-use tariffs [18,19],
peer to peer trading [20], or provision of revenue stacking
2
[21]. Revenue stacking is considered as one
of the most effective support mechanisms for enhancing economic profitability of EES systems [22],
which can be possible by combining the onsite use of EES with offering grid services, such as
balancing the load and/or ancillary services as shown in [23,24].
1.2 Coordination of distributed solar PV-storage systems
Providing grid services in many power systems is regulated by the System Operator with some
technical requirements for candidate technologies. These requirements are commonly specified as
response time, availability, reliability, minimum capacity rating, etc. For example, the requirement
for an energy technology for providing balancing services in Finland is a minimum power output
of 5 MW [25]. These requirements leave many distributed technologies such as PV-EES systems with
a typical size of a few Kilowatts unqualified for entering such marketplaces. To overcome such
barriers of entry, the available capacity of many small-scale distributed technologies can be
aggregated and coordinated by aggregators, which are typically third-party companies benefiting
from control and transaction fees. Therefore, the owner of a PV-EES system can operate their asset
either independently mainly for managing their own generation and demand or, alternatively, they
could offer their available storage capacity to be coordinated with other small-scale EES units to
participate in wholesale electricity markets through aggregators.
Aggregators can offer the combined capacity of EES technologies in wholesale electricity
markets, to meet the needs of the System Operator for load management and ancillary services, e.g.,
for Fast Frequency Response (FFR) [26]. Different studies have shown that the aggregation of small-
scale EES systems could reduce the risk of higher electricity prices at peak times [27], improve social
welfare [28], and increase the integration of renewable energy in the grid [29], compared to
uncoordinated, independent management of such assests by their owners. As consumers are
unlikely to be able to provide such services and exploit arbitrage benefits simultaneously, they may
operate their resources in a way that minimizes their own electricity bills, irrespective of the
potential system-level benefits they could offer through aggregation [27]. Figure 1 illustrates the
main features of these two schemes for the operation of distributed energy storage, i.e., the
uncoordinated operation of EES by multiple owners for their private benefits (a), versus a centrally
coordinated operation of small EES systems through an aggregator.
2
Revenue stacking or aggregation of benefits means using an EES device for offering multiple services, such as
energy arbitrage, balancing services, and T&D support; and receiving revenues for each service.
Journal Pre-proof
3
Figure 1 Schematic representation of uncoordinated (a) and centrally coordinated (b) operation of distributed
electricity storage devices. The main characteristics of each mode of operation, including benefits for the system and the
private owner is depicted under each scheme.
1.3 Private and system-level value of solar PV and energy storage
The private value of solar PV and EES to consumers is the financial gain that a consumer can
obtain by reducing its electricity bills [30]. Wholesale electricity prices vary widely on an hourly or
half-hourly basis and are typically the largest component of electricity costs of consumers,
comprising nearly 40-60% of their electricity bills in Europe [20]. Most prosumers
3
have been
early adopters, environmental enthusiasts, looking for energy security by being independent from
the grid, and/or motivated by social and peer effects; not necessarily motivated purely by cost-
benefit analysis [31,32]. Yet the savings that prosumers with EES could achieve is a key indicator to
show if more widespread adoption of such distributed energy technologies is likely to occur in the
future or not.
Numerous studies have investigated the profitability of consumer investments in solar PV and
EES. Many studies have derived the cost of electricity and assessed the profitability of investments
by considering metrics such as the Net Present Value (NPV), Internal Rate of Return (IRR), or the
Return on Investment (ROI) of the investment. Other work adopts the “grid parity” concept to
evaluate the profitability of storage by considering the levelized cost of electricity [33]. These studies,
3
Prosumers are defined as consumers with the ability to produce electricity from solar PV.
Journal Pre-proof
4
however, do not take a whole electricity system approach for modelling the future electricity prices,
on which the economic profitability of PV-storage systems depends. A recent study considers the
impacts of a changing electricity system on the consumer savings, but does not account for potential
impacts of the development of demand-side technologies on the system [2]. This paper extends the
previous work by accounting for the impact of the EES on the system, which, if neglected, may
overestimate the potential benefits of the EES for the owner. Because the larger the capacity of EES
in the system offered by many private owners, the lower the value of arbitrage for each EES owner
as the price gap between peak and off-peak will diminish.
The value of solar PV-EES to consumers is different from the value they may offer to the wider
electricity system. Solar PV-EES and other distributed energy technologies could provide the
electricity system with different services, while offering energy security and cost savings to the
owner. However, maximizing the private value of distributed technology may not simultaneously
offer the highest system-wide value. Energy security has a private value to the consumer, whereas
the flexibility it offers to the system has a social value. The social (system) value of these resources
will depend on whether these resources are being operated to reduce electricity system costs, a
benefit for all consumers, or to minimize private electricity costs. Solar PV may reduce electricity
demand if it is subject to individual coordination by cost minimizing consumers, which would
reduce prices for all consumers in the system [27]. Privately coordinated EES could increase
electricity prices as there is potential that most of EES owners charge simultaneously at low price
hours resulting in significant increase of electricity demand and prices in those hours, affecting all
electricity consumers. But private EES devices could also reduce peak demand, hence prices, if they
were optimally operated in coordination, lowering electricity prices for all consumers [34].
Several studies focusing on EES in different countries have concluded that centralized
coordination of distributed energy resources could offer numerous system-level advantages. For
example, central coordination of EES can offer required flexibility in matching load and supply,
reducing the cost of procuring flexible capacity for the system [35,36]. The value of aggregation to
an electricity system has been shown to increase as more consumers are aggregated [37], with small
contributions by each customer leading to large reductions in electricity costs for all consumers [38].
It is also argued that distributed energy devices could improve social welfare under efficient
aggregation and coordinated operation of technologies [28]. Castagneto Gissey et al. [27]
investigated the impact of centralized and distributed scheduling of EES on electricity prices,
highlighting that a centralized coordination offers 7% lower mean electricity price and 60% lower
price volatility in the system. Sousa et al. [39] compares a peer-to-peer (P2P) versus a community
market for energy trade, concluding that P2P trade offers the highest social welfare. It is further
shown that the aggregator can control the capacity of distributed EES to manage the frequency
deviations in the grid in a more effective way [40]; another system-level benefit for all consumers.
In a recent study [41], a whole-system comparison of centralized versus decentralized electricity
planning is carried out, showing that coordinated planning can save between 7% and 37% of the
total system costs. Last but not the least, Ahmadi et al. [42] applies a two-stage optimal coordination
Journal Pre-proof
5
of central and local EES for showing the impact on system cost reduction and voltage profile
enhancement.
However, none of the reviewed studies investigate the impact of the aggregation of distributed
energy technologies (here PV-EES) on the private value of such technologies, i.e., the additional cost
or benefit that the owner bears for letting the aggregator coordinate their PV-EES. This is an
important question as the deployment of EES by consumers might be affected by the way the
technology is operated throughout the system. Answering this question could reduce the
uncertainty consumers face when investing in battery storage, thereby facilitating further
deployment of storage resources when needed. This would help the electricity system to reduce
costs and improve security of supply by making such resources available to provide multiple other
system services. In this respect, it is crucial to understand how the deployment of EES resources by
consumers could be affected as more EES is aggregated throughout the electricity system. Our study
investigates this too.
1.4 Objectives of this study
As mentioned earlier, pairing solar PV with EES can maximize the self-consumption of PV
electricity for consumers who adopt the technology and minimize their electricity costs. Yet it
remains unclear how the savings that these consumers can expect from their storage device might
be affected by the way of coordination of EES in the electricity system. This paper investigates how
aggregator-led and consumer-led operation of EES capacity might affect the private economic value
of solar PV and EES for a UK electricity consumer with typical domestic electricity consumption.
Different future developments of the energy system are explored to analyse the economic savings a
consumer can achieve from investing in PV and batteries. Finally, it is shown that how these savings
will be affected when more EES capacity is integrated into the electricity system through
aggregation. By identifying these three gaps in the literature, this paper aims to answer the following
research questions:
1. How would aggregator-led and consumer-led operations of EES in the electricity system
affect savings to a typical consumer who pairs solar PV with storage?
2. Which system evolutions or energy pathways are likely to explain the process by which EES
aggregation could affect savings to a consumer pairing solar PV with storage?
3. What is the relationship between savings from pairing solar PV with storage to a private
electricity consumer and the level of electricity system-wide storage aggregation? In other
words, how would additional aggregation of EES affect the savings to a typical consumer
pairing solar PV with storage?
The remainder of this paper is structured as follows. Section 2 provides the methodology and
describes the data used in this study. Section 3 reports our main results, which are discussed in
Section 4. Conclusions are drawn in Section 5.
Journal Pre-proof
6
2 Methods
Onsite, small-scale batteries and electric vehicle-to-grid storage are some examples of
distributed EES technologies for private consumers. The ever-growing electrification of transport,
heating and other sectors are expected to change the pattern and magnitude of electricity demand
over the coming decades [43]. Accurate modelling of electricity demand over such extended periods,
i.e., 20-30 years, is crucial to understand how consumer electricity prices will vary in the future and
how investment in distributed technologies will return economically. Also, transitions in the
electricity supply side will affect wholesale electricity prices. Higher shares of wind and nuclear
capacity in the power system will offer different electricity prices and price volatility compared to a
thermal power system relying on coal and gas. Hence, assessing the financial feasibility of
investment in distributed energy technologies with 20-30 years of lifetime needs to be informed by
a quantitative model of the overarching energy system for representing the increase in the use of
non-conventional energy resources and possible transitions in the energy system.
2.1 A multi-level modelling framework
The modelling approach is based on soft-linking a national-level, electricity system
management model (ESMA) to a consumer cost optimization model. The input data of ESMA, i.e.,
electricity demand, power capacity mix, and fuel prices are based on the UK “future energy
scenarios” developed by the national energy regulatory, National Grid [44]. The electricity system
model ESMA is designed for evaluating the operation and dispatch of a given power system mix for
a time-period of one year (8760 h). It is ideally suited to generate wholesale prices under different
scenarios for EES and the rest of the system. Wholesale electricity prices are then converted into
retail electricity tariffs based on different tariff designs, i.e., time of use (ToU), static, and dynamic
tariffs. These tariffs are fed into an electricity private cost minimization model that optimizes the use
of solar PV and EES for a consumer with a typical electricity consumption profile. This framework
accounts for possible future evolutions of the energy system considering how EES deployments are
likely to affect savings of consumers. The electricity generation costs, e.g., future capital cost of
different power plants, are based on the output of the UKTM energy system model [45].
The modelling framework including the linkage between different models and modules to
derive consumer savings is illustrated in Figure 2. This framework has been previously applied to
calculate solar PV-battery consumer investments [2] and value of storage aggregation to the system
and electricity prices [27]. This is extended in this study by iterating electricity demand of
prosumers, which itself is based on the optimal scheduling of PV-EES according to retail prices, back
to the electricity dispatch model. With the updated electricity demand, the electricity dispatch model
generates a new set of hourly electricity prices, which will affect the retail price for all consumers,
both with and without onsite energy technologies. This process, highlighted in red in Figure 2,
continues until electricity prices converge in two consecutive iterations.
Journal Pre-proof
7
The ESMA model has been validated on an hourly basis against both the historical data and
future energy scenario developed by National Grid. The results of validation suggest that the hourly
demand curve modelled by ESMA stays within an acceptable level of agreement with historical data,
e.g., with an average correlations coefficient of 0.92 for 8760 hourly demand data points for the
reference year 2015. Similarly, the analysis of hourly electricity prices simulated by ESMA in
different season shows a high degree of agreement with historical spot prices, with an average
correlation of 0.83 in winter, while 0.91-0.93 correlation on other seasons. The comparison of the
ESMA’s future scenarios with those modelled by National Grid shows a very high degree of
agreement, yet some slight differences exist due to different modelling assumptions and limitations
of ESMA. A detailed analysis on validation of the model is represented in Chapter 5 in [46].
The applied modelling work has some limitations and shortcomings. Assuming fixed, average
fuel prices throughout each year, i.e., fixed gas or biomass prices, may not conform with reality
where fuel prices change by season. ESMA does not include electricity consumers under the
Economy 7 tariff who benefit from a lower night tariff, which may result in a slight demand and
price difference in winter. ESMA represents each technology as a large power plant which is
different from the strategy that each single power plant may adopt.
Figure 2. Relationship between different models used in this analysis
The model is run over a 26-year period, 2015–2040, initially with the objective to optimize the
consumer’s utility based on the lifetimes of distributed PV-EES systems. ESMA minimizes electricity
costs and calculates wholesale electricity prices under the assumption of centralized and distributed
coordination of demand-side EES technologies. Additional information on the modelling
Journal Pre-proof
8
framework and formulation is provided in the Supporting Information (Appendix A-C). Appendix
H summarizes main data sources and assumptions of the model.
Retail electricity prices are calculated by adding a time-dependent mark-up over the wholesale
prices, which is assumed to account for the electricity network management and distribution fees
[47] (see Appendix F for calculation of prices). Static and dynamic ToU electricity tariffs are
calculated based on retail prices, calibrated to historical tariff data (assuming same ratios between
tariffs and retail prices as today for future years).
2.2 Future energy scenarios
The evolution of the energy system over time will impact wholesale electricity prices, and hence,
consumer retail prices. A whole systems approach is adopted to account for these future transitions
systematically and consistent with the National Grid scenarios, which are based on a broad
stakeholder engagement and modelling. Four possible evolutions of the energy system are
considered according to National Grid’s Future Energy Scenarios [44]. These scenarios are chosen
as the basis of our analysis as they cover a wide range of future energy pathways represented across
two axes for green ambition and prosperity. The GB Office of Gas and Electricity Markets (Ofgem),
the National Regulatory Authority, has reviewed these scenarios, which gives them more merit for
our analysis.
These four energy transition pathways include: (i) Gone Green, which is the most ambitious
renewable expansion scenario, where the UK meets its renewable targets; (ii) Consumer Power, a
consumer-centred scenario with energy security and costs as main drivers of decisions; (iii) Slow
Progression, a scenario with low ambitions for decarbonization; and (iv) No Progression, where the
status quo persists and there is a negligible deployment of renewables and EES. Gone Green has the
highest ambition on renewables and storage capacity, while No Progression is similar to the present-
day energy system and has the lowest capacity of renewables from all four scenarios. Table A3 in
Supplementary Material shows the key developments of the power sector in 2030 under these future
scenarios. Figure 3 portrays the installed power capacities for each of the future energy scenarios.
More details of the share of each generation mode are provided in Appendix H, Table A4.
Journal Pre-proof
9
Figure 3. Electricity generation mix in each future energy scenario [44].
2.3 Consumer electricity cost optimization
Two cases of EES scheduling are examined, in which consumers respond to either distributed
or centralized coordination. Under the former, demand-side storage resources are autonomously
optimized by consumers. In a centralized scheduling system, an aggregator coordinates electricity
dispatch from EES by iterative negotiation with consumers, whose resources it does not know,
enabling them to participate in the wholesale market. Centralized coordination mimics the current
arrangements for large-scale EES technologies in the UK and major worldwide liberalized markets,
such as for PHS. Transaction costs relating to aggregation are neglected for simplicity. Distributed
coordination reflects the behaviour of consumers who individually schedule their flexible resources
to smoothen their own demand profiles and minimize their own electricity bills. More information
on our coordination algorithms is provided in Appendix E.
The financial viability of different combinations of investments in solar PV and EES for a typical
UK domestic electricity user
4
are examined under different energy scenarios. The household’s
electricity bill is dependent on the consumer’s load profile, and on the electricity generated from
solar PV, which exhibit intra-day, monthly and seasonal variations.
End users with onsite generation from PV are entitled for feed-in tariffs (FiTs) of £0.049 kWh-1
for electricity generation [47] and an export-to-grid tariff of £0.043 kWh-1. FiT payments are assumed
to cease after 20 years and to increase with the retail price index (RPI) of 3.4% p.a. [48]. An average
retail electricity tariff is considered based on UK National Statistics: a static tariff of £0.15 kWh-1 and
4
This user is represented by a three-bedroom dwelling with a load profile displaying mean percentage night
consumption of 30% and 55% under static and Economy7 ToU tariffs, respectively [47].
Journal Pre-proof
10
dynamic ToU tariff including on-peak £0.16 kWh-1 during the day (7:00-23:59) and off-peak tariff of
£0.07 kWh-1 at nights (0:00-6:59) [48]. Future developments of static tariffs are estimated based on
the average of wholesale electricity price in each season. We use the static tariff as the basis to derive
future values for day and night ToU tariffs (see Appendix H for more details).
The objective of a residential PV, EES, or PV-EES owner is to minimize the private costs of
electricity bills. Under ToU tariffs, the lower rate during the off-peak period is suitable for charging
the storage system. When the consumer operates PV, a 4-kW PV system is considered; and for EES,
a 6.4 kWh–3.3 kW battery, with a lifetime of 13 years or 5,000 cycles (Li-ion batteries) [49]. The battery
capacity degradation and efficiency losses are taken into account as described in Appendix B. A
discount rate of 5% p.a. is assumed, based on the recommendations of the UK Committee on Climate
Change. Appendix G reports the details on the consumer PV-EES optimization model and the data
used for modelling PV-EES technologies.
The electricity costs are calculated for four consumer technology combinations: (i) no
technology; (ii) an EES system (EES-only); (iii) a solar PV system (PV-only); and (iv) both a solar PV
and an EES system (PV-EES). We show the value of EES, which is derived by comparing annual
electricity costs in the PV-EES scenario relative to the PV-only scenario. The base case scenario for
deriving the relative savings of other scenarios is the no-technology case with static electricity tariffs.
3 Results
Two types of energy storage coordination, i.e., coordinated and distributed, are considered for
calculations. The results are based on the data of annual electricity costs and savings, averaged over
the modelling period of 2015–2040. The results are reported relative to a base case scenario, i.e., the
No Progression scenario under static tariff and with no onsite energy technology investments.
The results show that the evolution of the energy system and the scheduling coordination
regime have meaningful impacts on annual savings by the consumer. Distributed coordination
generally induces 4–11% lower savings than centralized coordination, whereas the system’s
evolution accounts for changes in savings by 1–27%. The largest savings occur in scenarios with high
storage and renewable capacity. The impact of additional storage capacity in the electricity system
on the savings to the consumer when aggregated to participate in the wholesale market is explored
too.
3.1 Private savings under centralized and distributed coordination
The results of the centralized coordination is presented in Table 1. The annual electricity bills
and potential savings in the electricity bill are compared for consumers whose EES capacity in the
electricity system is coordinated by an aggregator and scheduled centrally. The results are illustrated
for four different technology options under static and ToU tariffs and for each future energy
scenario.
Table 1. Annual electricity bills and possible savings (£ p.a.) for a typical consumer under centralized coordination.
Journal Pre-proof
11
Centralized
coordination
No Progression
Slow Progression
Consumer Power
Gone Green
Tariff
Technology
Bill
(£ p.a.)
Savings1
(£ p.a.)
Bill
(£ p.a.)
Savings
(£ p.a.)
Bill
(£ p.a.)
Savings
(£ p.a.)
Bill
(£ p.a.)
Savings
(£ p.a.)
Static
No
technology
574
-
541
33
449
125
470
104
EES
574
0
541
33
449
125
470
104
PV
363
211
342
232
284
290
297
277
PV-EES
107
467
98
476
78
496
82
492
ToU
No
technology
540
34
515
59
420
154
449
125
EES
405
169
389
185
321
253
339
235
PV
307
267
298
276
244
330
260
314
PV-EES
92
482
87
487
68
506
73
501
1 The savings are shown as difference relative to the base scenario, i.e., consumers having “No technology” onsite, static tariffs, and
under the business-as-usual scenario (No Progression).
The results show that the consumer savings is dependent on the future energy scenarios for the
entire energy system. Consumer Power scenario, in which future policies are consumer-centred and
promoting distributed generation offers the highest savings for all technology combinations. Gone
Green and Consumer Power scenarios offer 18% and 22% annual savings, respectively, even in the
case when the consumer has no investment in distributed technologies, i.e., “No technology”. This
is due to higher renewable energy in these scenarios, larger share of electricity storage, and lower
electricity prices compared to No Progression.
Figure 4 compares the average annual savings in the electricity bill in the centralized
coordination for two different types of tariffs. The results show that PV-battery offers the highest
savings for consumers ranging between 81-86% depending on the future scenario.
Journal Pre-proof
12
Figure 4. Annual electricity bill savings for a typical consumer with different distributed energy technology options
in centralized coordination, under (A) a static and (B) Time-of-use (ToU) tariff, and for different future energy scenarios.
The values are the average of 2016-2040 and show % change in savings relative to the base case, which is “No-
technology” under a static tariff, and No Progression Scenario with the annual cost of 574£.
However, battery alone offers no higher benefits compared to the no-technology case, as under
the static tariff there will be no potential for price arbitrage by EES, as electricity prices are constant
for the consumer. The annual savings of the consumer from investing in solar PV alone (without
EES) varies between 37% and 51% of the base case costs, with the lower range for No Progression
scenario and the highest savings for Consumer Power.
The results for the battery-alone case show significant higher savings under a ToU tariff. When
the consumer electricity prices differ between off-peak and peak hours, battery can offer electricity
cost savings between 29-41%, depending on the future energy scenario. Moreover, investing on solar
PV under the ToU tariff improves the annual cost savings by 6-10%-point compared to the static
tariff (~ 56£ per year). A PV-battery system offers the highest savings under ToU as well, with a
slight improvement compared to the static tariff (i.e., 1-3%-point). Also, the results show that the
benefits of the PV-battery options are the least sensitive technology investment to future energy
scenarios, offering savings ranging between 84% and 88% for the four energy scenarios. Table 2
summarizes the results of centralized coordination for different tariffs, technology choices, and the
future scenarios.
Under centralized scheduling of the consumer’s energy technologies in the electricity system,
the typical electricity consumer gains substantially larger annual savings compared with the
decentralized scheduling. This is valid for all combinations of technologies, tariffs and future energy
scenarios. The consumer is able to accumulate greater savings in the centralized case by between 4–
8% when operating no technology, by 3–11% with EES alone, by 2–5% with PV alone, and by 0–2%
with both PV and EES. More notably, the higher savings in the centralized coordination compared
to the distributed scheme decline as the consumer operates more onsite technologies. Operating
more technologies implies greater electricity self-sufficiency, hence, a lower exposure to the risk of
Journal Pre-proof
13
changes in retail electricity prices, which itself is affected by the type of scheduling coordination of
EES by other consumers in the system (see Figure 5).
Figure 5. Centralized coordination versus distributed scheduling of consumers’ energy technologies under time-of-
use the (ToU) electricity tariff. The values show the % savings of centralized coordination minus that of distributed
scheduling relative to the base case (hence, positive values show that centralized coordination offers greater savings).
Consumers with “No technology” make higher electricity bill savings in the centralized
coordination scheme due to the system operator being able to improve the balancing of load and
flexibility resources, which results in lower peak electricity prices in the system. The lower wholesale
electricity prices benefit all consumers, including those without investment in any distributed
technology. Distributed storage scheduling results in substantially lower integration of EES capacity
in the electricity supply. Through arbitrage, storage minimizes the differential between on- and off-
peak prices, thereby reducing electricity system costs. Less aggregated storage capacity implies a
lower ability for the system operator to reduce electricity prices. Hence, in all scenarios, greater
private electricity costs and lower private savings are observed relative to centralized scheduling.
Table 2 summarizes the findings for the distributed scheduling.
3.1.1 Consumer’s choice of technology and electricity tariffs
The lowest electricity cost in the no-technology case occurs under centralized coordination,
Consumer Power and ToU tariffs (£420 p.a.), while the highest costs occur under distributed
scheduling, Slow Progression, and static tariffs (£569 p.a.). With ToU tariffs, the EES system can
provide 2–3% greater savings relative to static tariffs under distributed coordination compared with
centralized coordination. Under ToU, the savings in the EES-only case are £99–126 under centralized
coordination versus £101–140 under distributed coordination compared to “No technology” in the
respective future scenario. This shows approximately 7% larger savings in distributed scheduling.
As the distributed coordination scenario implies a less smoothened system demand, this leaves a
greater ability for the consumer to take advantage between peak and off-peak price differentials.
Table 2. Annual electricity bills and possible savings (£ p.a.) for a typical consumer under distributed scheduling.
The savings are relative to the base case: No technology, static tariff, and No Progression scenario.
Journal Pre-proof
14
Distributed scheduling
No
Progression
Slow
Progression
Consumer
Power
Gone Green
Tariff
Technology
Bill
(£
p.a.)
Savings
(£ p.a.)
Bill
(£ p.a.)
Savings
(£ p.a.)
Bill
(£
p.a.)
Savings
(£ p.a.)
Bill
(£ p.a.)
Savings
(£ p.a.)
A.
Static
No technology
588
0
569
19
476
112
516
72
EES
588
0
569
19
476
112
516
72
PV
378
210
359
229
301
287
327
261
PV-EES
116
472
103
485
83
505
91
497
B.
ToU
No technology
559
29
541
47
442
146
491
97
EES
419
169
406
182
341
247
370
218
PV
321
267
309
279
256
332
281
307
PV-EES
101
487
91
497
71
517
80
508
The largest savings recorded in the EES-only case occurs under centralized coordination,
Consumer Power, and ToU tariffs (£321 p.a.). Conversely, the lowest savings arise under distributed
coordination, Slow Progression, and static tariffs (£569 p.a.).
If the consumer operates solar PV without EES (PV-only), the electricity bill will decline by 37-
57% compared to the no-technology case, and by between 13-37% relative to EES-only. The lowest
electricity costs for PV-only relate to centralized scheduling, Consumer Power and ToU tariffs (£244
p.a.), whereas the largest costs arise under distributed scheduling, Slow Progression, and static
tariffs (£359 p.a.).
The combination of solar PV with EES implies a reduction in annual electricity costs by 81-88%,
or by £476-506 annually. Therefore, the consumer reduces electricity costs by at a substantial rate of
60% compared to the PV-only case (£176-256 further savings annually). On average across the future
energy system scenarios, ToU tariffs imply 12% larger savings relative to static tariffs for the
consumer. In this case, annual electricity costs are between £68-73 p.a. and £71-80 p.a. in the
centralized and distributed cases, respectively. When operating a PV-EES system, the consumer
achieves maximum savings under centralized coordination, Consumer Power and ToU tariffs (£68
p.a.), whereas the lowest savings occur when scheduling occurs on a distributed basis, under Slow
Progression and static tariffs (£103 p.a.).
Overall, for different technology mixes, a distributed coordination of energy storage in the
electricity system, as well as Slow Progression, and static tariffs tend to minimize annual savings by
the consumer. Conversely, central energy storage coordination, Consumer Power and ToU tariffs
maximize savings.
3.2 Future energy scenarios
The results suggest that the centralized coordination of EES resources in the electricity system
is always lead to greater savings (up to 11%) for a typical consumer, irrespective of the future
evolution of the energy system. Yet the order of magnitude by which savings under centralized
coordination are larger depends on the relationship between variable renewable energy capacity –
mostly includes wind and PV generation – and flexible supply capacity, such as gas plants. If
Journal Pre-proof
15
resources are mostly centrally coordinated, consumers can reduce annual electricity costs by 8–11%
in Gone Green, by 4–5% in Slow Progression, and by 4–6% in Consumer Power, relative to
distributed coordination.
The impact of centralized coordination of storage resources on the consumer’s annual electricity
costs generally increases with the level of variable renewable generation capacity in the electricity
system while inversely related to level of flexible supply capacity. Savings to the consumer under
centralized coordination are double in Gone Green relative to Slow Progression due to the higher
variable renewable generation in the former case, which requires an aggregated storage for
balancing variations.
Table 3 reports the ratio of variable renewable capacity to each unit of flexible generation
capacity, as well as the change in the consumer’s annual electricity costs (%) resulting from storage
aggregation in the electricity system. There is a positive relationship between the share of variable
renewables in the system, and the change in electricity prices due to centralized coordination. By
dividing the latter by the former, a relatively constant relationship is observed, between 3–4%.
Demand-side flexibility will be most valuable when supply is inflexible, leading to greater savings
in the consumer’s annual electricity cost under a more system-efficient coordination of storage
resources. Yet the change in the electricity cost from coordination is small relative to the ratio
between renewables and flexible supply.
Table 3. Ratio of variable renewable to flexible supply capacity (excluding storage), and relationship with savings
from demand coordination.
Future energy
scenario
Ratio of renewable energy capacity to
flexible supply capacity
Change in annual electricity costs
under central coordination (% p.a.)
Gone Green
2.62
-8.8%
Consumer Power
1.97
-5.3%
Slow Progression
1.81
-4.5%
3.3 Impact of additional storage deployments on private savings
Figure 6 shows how additional electricity storage capacity is likely to affect savings from storage
to a consumer with EES. In this specific analysis, we consider ToU tariffs only as they are shown to
maximize the savings that storage can provide to consumers with solar PV. Additional (aggregated)
storage capacity operating in the electricity system can decrease the differential between on- and off-
peak electricity wholesale prices, which could in turn reduce the retail tariff on- and off-peak
differential.
Journal Pre-proof
16
Figure 6. Savings to the typical consumer due to their electricity storage relative to the installed electricity storage
capacity in the electricity system. This analysis considers the centralized case with ToU tariffs.
There is a quasi-exponential fall in the private savings as more electricity storage is installed
and aggregated in the wider electricity system. An increase in aggregated storage capacity from 3
GW to 17 GW implies a 20% reduction in the private annual cost savings from storage to the
consumer. These results do not hold if considering distributed coordination, as non-aggregated
storage capacity has no effect on the marginal savings from private storage capacity.
4 Discussion
4.1 Private savings from storage and control scheme
This paper shows that the savings that a typical UK electricity consumer can achieve from their
EES device could increase if most consumers in the electricity system allow an aggregator to
coordinate their storage resources. When consumers’ storage capacity is operated to minimize the
private costs of these consumers, herding behaviour occurs, leading to charging the consumer EES
devices at the same time of the day leading to higher electricity prices relative to centralized
coordination. These results are shown to hold true for different types of technologies and evolutions
of the energy system.
Our findings confirm those of [50], [28], and [35] who reported that social welfare increases if
storage resources are centrally scheduled. Similarly, Castagneto Gissey et al. [27] compared
centralized and distributed coordination and suggested that consumers could be nudged into giving
away control of their storage devices to provide system benefits. They found that aggregation of EES
has a lower electricity system cost compared to private operation by consumers. Our study enhances
this work by considering how the private savings that consumers can expect from investing in
storage could be affected by the way other consumers operate their storage devices.
Journal Pre-proof
17
The results also show the distributional effect of the centralized coordination on consumers.
Those consumers owning flexible technologies such as EES and providing the aggregator with the
capacity of their device for load balancing, make relatively lower bill savings compared to those
consumers with “No technology”. For example, PV-EES owners make 0-2% additional savings in
the centralized scheme while consumers with no technology 2-10%. This is mainly due to the lower
electricity prices for all consumers in the centralized coordination compared to a distributed
scheduling, which benefits the most consumers under static tariffs with no technology. Therefore,
the regulator should put a policy in place for redistributing some of the system-level benefits back
to the EES providers in the centralized coordination. In other words, the positive externality of
aggregating distributed EES can be calculated, including lower electricity prices at peak times and
lower grid congestion management fees, and a part of that can be used to incentivize EES owners
participating in the aggregation scheme. The lack of such incentives can deteriorate the economic
attractiveness of centralized coordination schemes for consumers [51–53].
4.2 Potential impact of system variables on the consumer savings
EES could provide numerous services to the electricity system [54,55], and the possibility for
storage capacity to be aggregated can reduce the cost of electricity systems by decreasing peak
demand and the need for expensive peaking plants. A few studies have shown the value of storage
in high-renewable, inflexible power systems [12,34,56]. Studies considering the role of storage in the
electricity system generally do not make a distinction between private and system benefits from EES,
which we instead consider by considering the impacts of distributed and centralized coordination.
Our work suggests that storage will be more valuable to energy storers if variable renewable
capacity is on average larger than the capacity of flexible supply resources such as gas power plants
in the power system. When variable renewable capacity is large relative to flexible supply capacity,
there is a shortage of flexibility on the supply-side, meaning that a system able to centrally
coordinate more demand-side storage resources will be more valuable, and would produce more
savings to consumers from their storage technology. Yet these insights must be checked against the
possibility of distributed energy storage coordination to account for the likely scenario in which
storage resources belonging to consumers are operated in a way that does not necessarily benefit the
system, so long as it benefits the cost-minimizing consumer.
Many consumers would prefer to dispatch their storage resources to reduce their own electricity
bills rather than to reduce costs to the wider system. Hence, previous studies may have tended to
overestimate the utility of storage in reducing electricity prices by assuming large amounts of
demand-side energy storage aggregation. As additional storage capacity is deployed, the lower gap
in peak and off-peak electricity prices diminishes the potential benefits, sending a discouraging
signal to the market for new investments. Hence, policymakers should closely monitor the flexibility
requirements of the system and the willingness of consumers to provide flexibility services to the
system. This can be done by internalizing the system-level benefits of EES, through introducing
incentives for investment in EES. From modelling method perspective, this implies that models of
Journal Pre-proof
18
the electricity system should account for the trade-offs between private and system benefits of
energy storage aggregation.
Yet it is unlikely that consumers will allow an aggregator to control their resources at all unless
they are paid a financial incentive to do so [57]. The decision by consumers to forego control of their
storage resources could meaningfully reduce electricity wholesale prices [27]. This also entails the
installation of smart meters and the access to the energy consumption data of private consumers,
which they might be unwilling to share. The ability of aggregators and the System Operator to nudge
consumers into providing such information could be key to the successful operation of aggregators.
The private savings that consumers can gain from their storage device will depend on the
evolution of the electricity and energy systems. Consumers contemplating to invest in EES should
not only be aware of the quantity of storage capacity deployed in the electricity system but should
also monitor the level of renewables that this aggregated storage capacity is likely to meet. This
information is important because it affects the operational savings from storage by consumers, hence
the probability of them investing in the technology. This could also be a reason for the complexity
of cost-benefit calculations by consumers and hence the current lack of EES deployments by
domestic users [17,58].
Providing consumers with an understanding of how savings from their storage devices could
be affected by numerous energy system conditions could improve consumer confidence in the
technology and might facilitate deployments. It is more likely for such information to be useful if
provided in the form of a software integrated into an easily accessible website that calculates savings
from storage based on high temporal and spatial resolution models of the electricity system. Such a
model would consider where on the system the consumer is based, as well as the consumer’s
electricity consumption patterns, among other factors. This would help inform the consumer’s
decision as to whether a financial case to invest in storage exists in their specific case, and to
understand the relationship between their investment on distributed technologies and their overall
support for any future energy pathway.
4.3 Additional storage in the electricity system and consumer savings
We demonstrate that a consumer could expect lower savings from their storage technology if a
large amount of storage installed throughout the electricity system. Yet this only occurs if this
capacity is subject to aggregation. Annual electricity cost savings from storage to a typical UK
consumer could fall by more than 20% if EES capacity were to increase from 3 GW to 17 GW in the
system.
The policy implication here is that the system operator should provide the data of the existing
capacity of storage in the system, planned new storage installations, and the level of aggregation of
these assets. This information should ideally be made public together with statistics about the
fraction of these resources that are centrally coordinated as this is likely to impact the savings of
consumers, lowering that compared to the case no storage deployments or aggregation occurred.
Journal Pre-proof
19
4.4 Drawbacks and future work
This paper focused on arbitrage using EES, and the value of storage to consumers in providing
non-economic benefits such as energy security has not been considered. Similarly, the value that
consumers could extract from their storage device by providing balancing or ancillary services to
the grid have also been neglected. As electricity systems evolve, it will become increasingly
important to assess the value of security and the potential provision of grid services through
aggregation, as these are effectively substitutes to one another, while having synergies with energy
arbitrage [23]. We simplified the representation of domestic consumers by considering a typical
domestic electricity consumer with a representative solar PV production and electricity
consumption pattern. Yet these factors may largely vary across consumers and geographical areas.
Furthermore, we focused on the role of coordination in the determination of wholesale electricity
prices. Yet to uncover the changes in retail tariffs, our modelling work would benefit from an
analysis where prices are made depending on capital, fuel, and networks costs in relation to each
consumer in the electricity system.
5 Conclusions
This study investigates the potential economic savings to a UK electricity consumer as a
function of energy storage coordination scheme, i.e., central vs. distributed, as well as the system-
wide impact of deployment of such storage devices. As more consumers, and the wider electricity
system, adopt electricity storage technologies, herding behaviour could occur: many cost-
minimizing consumers with an incentive to shift electricity demand to the same periods of low
electricity prices, which will ultimately lead to an increased electricity demand and price peaks.
Storage technologies already face multiple market barriers today. Hence, it is crucial to understand
the impact of electricity market design on potential financial benefits of a storage owner (storer).
This paper examines the possible economic impact of owning a demand-side energy storage on
the savings to a typical domestic consumer equipped with a solar PV microgeneration system. We
conclude that pairing solar PV with storage could reduce electricity bills for a typical UK consumer
by 80-88%. Yet the value of storage device is likely to increase if most electricity consumers allow an
aggregator to coordinate their storage resources, thereby, reducing peak electricity demand
resulting in more affordable electricity for all consumers. Our study shows that the benefits of
consumers investing in energy storage is partly dependent on the ratio of variable renewable energy
capacity to flexible supply capacity in the system. This ratio tends to improve savings from storage
when the need for flexibility grows in the system.
This paper further investigates the relationship between savings to a typical UK electricity
consumer using energy storage only for arbitrage versus the amount of aggregated storage capacity
deployed by the electricity System Operator. A five-fold increase in the level of aggregated storage
capacity can potentially lead to 20% lower savings to the consumer from their energy storage device.
We show that consumers should expect diminishing marginal savings to the private utility of their
Journal Pre-proof
20
storage device because of additional aggregated storage capacity if they pay time-dependent
electricity tariffs, such as dynamic ToU tariffs. To maximize the value of the storage resources, the
system operator should reduce the uncertainty in investing in storage by providing the consumers
with the information about amount of deployed storage resources in the system, either centrally or
individually coordinated. The scale of reduction in electricity bills of consumers depends on future
electricity system evolutions too.
6 Acknowledgments
This research was funded by the UK Engineering and Physical Research Council (EPSRC)
through the Realising Energy Storage Technologies in Low-carbon Energy Systems (RESTLESS)
project (EP/N001893/1), for which the authors are very grateful. The contribution by BZ have been
partly supported by International Institute for Applied Systems Analysis (IIASA), the RE-INVEST
project “Renewable Energy Investment Strategies – A two-dimensional interconnectivity approach”
funded by Innovation Fund, Denmark, and the STEEM Project, Aalto University, Finland. The
authors would also like to thank Professor Richard Green (Imperial College London) for useful
suggestions.
7 References
[1] Pfenninger S, Staffell I. Long-term patterns of European PV output using 30 years of validated
hourly reanalysis and satellite data. Energy 2016;114:1251–65.
https://doi.org/10.1016/j.energy.2016.08.060.
[2] Castagneto Gissey G, Zakeri B, Dodds PE, Subkhankulova D. Evaluating consumer
investments in distributed energy technologies. Energy Policy 2021;149:112008.
https://doi.org/10.1016/j.enpol.2020.112008.
[3] Shrivastava A, Saini DK, Pandit M. Distribution grid parameter variation due to solar PV
power integration. Int J Renew Energy Res 2020;10:1125–32.
[4] Fernández G, Galan N, Marquina D, Martínez D, Sanchez A, López P, et al. Photovoltaic
generation impact analysis in low voltage distribution grids. Energies 2020;13.
https://doi.org/10.3390/en13174347.
[5] Sweco. Distributed electricity production and self-consumption in the Nordics. 2019.
[6] Bagalini V, Zhao BY, Wang RZ, Desideri U. Solar PV-battery-electric grid-based energy
system for residential applications: System configuration and viability. Research 2019;2019.
https://doi.org/10.34133/2019/3838603.
[7] Klingler AL. Self-consumption with PV + Battery systems: A market diffusion model
considering individual consumer behaviour and preferences. Appl Energy 2017;205:1560–70.
https://doi.org/10.1016/j.apenergy.2017.08.159.
[8] Zhang J, Cho H, Luck R, Mago PJ. Integrated photovoltaic and battery energy storage (PV-
BES) systems: An analysis of existing financial incentive policies in the US. Appl Energy
2018;212:895–908. https://doi.org/10.1016/j.apenergy.2017.12.091.
Journal Pre-proof
21
[9] Zakeri B, Cross S, Dodds PE, Gissey GC. Policy options for enhancing economic profitability
of residential solar photovoltaic with battery energy storage. Appl Energy 2021;290:116697.
https://doi.org/10.1016/j.apenergy.2021.116697.
[10] Mulenga E, Bollen MHJ, Etherden N. A review of hosting capacity quantification methods for
photovoltaics in low-voltage distribution grids. Int J Electr Power Energy Syst 2020;115.
https://doi.org/10.1016/j.ijepes.2019.105445.
[11] do Nascimento ÁDJ, Rüther R. Evaluating distributed photovoltaic (PV) generation to foster
the adoption of energy storage systems (ESS) in time-of-use frameworks. Sol Energy
2020;208:917–29. https://doi.org/10.1016/j.solener.2020.08.045.
[12] Denholm P, Jorgenson J, Jenkin T, Palchak D, Kirby B, Malley MO, et al. The Value of Energy
Storage for Grid Applications. Natl Renew Energy Lab 2013:37.
[13] Akhil AA, Huff G, Currier AB, Kaun BC, Rastler DM, Chen SB, et al. DOE/EPRI 2015
electricity storage handbook in collaboration with NRECA. Sandia Natl Lab 2015.
[14] Schmidt O, Hawkes A, Gambhir A, Staffell I. The future cost of electrical energy storage based
on experience rates. Nat Energy 2017;2:1–8. https://doi.org/10.1038/nenergy.2017.110.
[15] Uddin K, Gough R, Radcliffe J, Marco J, Jennings P. Techno-economic analysis of the viability
of residential photovoltaic systems using lithium-ion batteries for energy storage in the
United Kingdom. Appl Energy 2017;206:12–21.
https://doi.org/10.1016/j.apenergy.2017.08.170.
[16] Diezmartínez C V. Clean energy transition in Mexico: Policy recommendations for the
deployment of energy storage technologies. Renew Sustain Energy Rev 2021;135:110407.
https://doi.org/10.1016/j.rser.2020.110407.
[17] Hossain E, Faruque HMR, Sunny MSH, Mohammad N, Nawar N. A Comprehensive Review
on Energy Storage Systems: Types, Comparison, Current Scenario, Applications, Barriers, and
Potential Solutions, Policies, and Future Prospects. Energies 2020;13:3651.
https://doi.org/10.3390/en13143651.
[18] Linssen J, Stenzel P, Fleer J. Techno-economic analysis of photovoltaic battery systems and
the influence of different consumer load profiles. Appl Energy 2017;185:2019–25.
https://doi.org/10.1016/j.apenergy.2015.11.088.
[19] Parra D, Patel MK. Effect of tariffs on the performance and economic benefits of PV-coupled
battery systems. Appl Energy 2016;164:175–87.
https://doi.org/10.1016/j.apenergy.2015.11.037.
[20] Rodrigues DL, Ye X, Xia X, Zhu B. Battery energy storage sizing optimisation for different
ownership structures in a peer-to-peer energy sharing community. Appl Energy
2020;262:114498. https://doi.org/10.1016/j.apenergy.2020.114498.
[21] Stephan A, Battke B, Beuse MD, Clausdeinken JH, Schmidt TS. Limiting the public cost of
stationary battery deployment by combining applications. Nat Energy 2016;1.
https://doi.org/10.1038/nenergy.2016.79.
[22] Gardiner D, Schmidt O, Heptonstall P, Gross R, Staffell I. Quantifying the impact of policy on
the investment case for residential electricity storage in the UK. J Energy Storage
2020;27:101140. https://doi.org/10.1016/j.est.2019.101140.
Journal Pre-proof
22
[23] Pusceddu E, Zakeri B, Castagneto Gissey G. Synergies between energy arbitrage and fast
frequency response for battery energy storage systems. Appl Energy 2021;283:116274.
https://doi.org/10.1016/j.apenergy.2020.116274.
[24] Zakeri B, Syri S. Value of energy storage in the Nordic Power market - Benefits from price
arbitrage and ancillary services. Int. Conf. Eur. Energy Mark. EEM, vol. 2016- July, 2016.
https://doi.org/10.1109/EEM.2016.7521275.
[25] Fingrid. Reserves and balancing power - Fingrid 2020. https://www.fingrid.fi/en/electricity-
market/reserves_and_balancing/ (accessed December 12, 2020).
[26] Merten M, Rücker F, Schoeneberger I, Sauer DU. Automatic frequency restoration reserve
market prediction: Methodology and comparison of various approaches. Appl Energy
2020;268:114978. https://doi.org/https://doi.org/10.1016/j.apenergy.2020.114978.
[27] Castagneto Gissey G, Subkhankulova D, Dodds PE, Barrett M. Value of energy storage
aggregation to the electricity system. Energy Policy 2019;128:685–96.
https://doi.org/10.1016/j.enpol.2019.01.037.
[28] Jia L, Tong L. Renewables and Storage in Distribution Systems: Centralized vs. Decentralized
Integration. IEEE J Sel Areas Commun 2016;34:665–74.
https://doi.org/10.1109/JSAC.2016.2525638.
[29] Chaminda Bandara WG, Godaliyadda GMRI, Ekanayake MPB, Ekanayake JB. Coordinated
photovoltaic re-phasing: A novel method to maximize renewable energy integration in low
voltage networks by mitigating network unbalances. Appl Energy 2020;280.
https://doi.org/10.1016/j.apenergy.2020.116022.
[30] Borenstein S. Private net benefits of residential solar pv: The role of electricity tariffs, tax
incentives, and rebates. J Assoc Environ Resour Econ 2017;4:S85–122.
https://doi.org/10.1086/691978.
[31] Balta-Ozkan N, Yildirim J, Connor PM, Truckell I, Hart P. Energy transition at local level:
Analyzing the role of peer effects and socio-economic factors on UK solar photovoltaic
deployment. Energy Policy 2021;148:112004.
https://doi.org/https://doi.org/10.1016/j.enpol.2020.112004.
[32] Mundaca L, Samahita M. What drives home solar PV uptake? Subsidies, peer effects and
visibility in Sweden. Energy Res Soc Sci 2020;60:101319.
https://doi.org/https://doi.org/10.1016/j.erss.2019.101319.
[33] Tu Q, Mo J, Betz R, Cui L, Fan Y, Liu Y. Achieving grid parity of solar PV power in China-
The role of Tradable Green Certificate. Energy Policy 2020;144:111681.
https://doi.org/https://doi.org/10.1016/j.enpol.2020.111681.
[34] Carbon Trust. Energy storage report: can storage help reduce the cost of a future UK electricity
system? | Carbon Trust 2016. https://www.carbontrust.com/resources/energy-storage-report-
can-storage-help-reduce-the-cost-of-a-future-uk-electricity-system (accessed October 17,
2020).
[35] Varghese S, Sioshansi R. The price is right? How pricing and incentive mechanisms in
California incentivize building distributed hybrid solar and energy-storage systems. Energy
Policy 2020;138:111242. https://doi.org/10.1016/j.enpol.2020.111242.
[36] de Sisternes FJ, Jenkins JD, Botterud A. The value of energy storage in decarbonizing the
Journal Pre-proof
23
electricity sector. Appl Energy 2016;175:368–79.
https://doi.org/10.1016/j.apenergy.2016.05.014.
[37] Katz, ML; Shapiro C. Network externalities, competition, and compatibility. Am Econ Rev
1985;75:424–40.
[38] Burger S, Chaves-Ávila JP, Batlle C, Pérez-Arriaga IJ. A review of the value of aggregators in
electricity systems. Renew Sustain Energy Rev 2017;77:395–405.
https://doi.org/10.1016/j.rser.2017.04.014.
[39] Sousa T, Soares T, Pinson P, Moret F, Baroche T, Sorin E. Peer-to-peer and community-based
markets: A comprehensive review. Renew Sustain Energy Rev 2019;104:367–78.
https://doi.org/10.1016/j.rser.2019.01.036.
[40] Obaid ZA, Cipcigan LM, Muhssin MT, Sami SS. Control of a population of battery energy
storage systems for frequency response. Int J Electr Power Energy Syst 2020;115:105463.
https://doi.org/10.1016/j.ijepes.2019.105463.
[41] Carvallo JP, Zhang N, Murphy SP, Leibowicz BD, Larsen PH. The economic value of a
centralized approach to distributed resource investment and operation. Appl Energy
2020;269:115071. https://doi.org/10.1016/j.apenergy.2020.115071.
[42] Ahmadi M, Adewuyi OB, Danish MSS, Mandal P, Yona A, Senjyu T. Optimum coordination
of centralized and distributed renewable power generation incorporating battery storage
system into the electric distribution network. Int J Electr Power Energy Syst 2021;125:106458.
https://doi.org/10.1016/j.ijepes.2020.106458.
[43] Sanders D, Hart A, Ravishankar M, Strbac G, Aunedi M, Pudjianto D, et al. An analysis of
electricity system flexibility for Great Britain 2016.
[44] National Grid. Future Energy Scenarios | National Grid ESO 2017.
https://www.nationalgrideso.com/future-energy/future-energy-scenarios (accessed October
26, 2020).
[45] UCL. UK TIMES | UCL ENERGY INSTITUTE MODELS - UCL – University College London
2020. https://www.ucl.ac.uk/energy-models/models/uk-times (accessed December 13, 2020).
[46] Subkhankulova D. Exploring future opportunities and challenges of Demand Side
Management with Agent Based Modelling. University College London, 2018.
[47] Ofgem. Feed-In Tariff (FIT) rates | Ofgem 2020. https://www.ofgem.gov.uk/environmental-
programmes/fit/fit-tariff-rates (accessed October 26, 2020).
[48] National Statistics. Quarterly Energy Prices - GOV.UK 2020.
https://www.gov.uk/government/collections/quarterly-energy-prices (accessed October 26,
2020).
[49] Rampazzo M, Luvisotto M, Tomasone N, Fastelli I, Schiavetti M. Modelling and simulation
of a Li-ion energy storage system: Case study from the island of Ventotene in the Tyrrhenian
Sea. J Energy Storage 2018;15:57–68. https://doi.org/10.1016/j.est.2017.10.017.
[50] He X, Delarue E, D’haeseleer W, Glachant J-M. A novel business model for aggregating the
values of electricity storage. Energy Policy 2011;39:1575–85.
[51] Sorrell S. Reducing energy demand : A review of issues , challenges and approaches. Renew
Sustain Energy Rev 2015;47:74–82. https://doi.org/10.1016/j.rser.2015.03.002.
Journal Pre-proof
24
[52] Winfield M, Shokrzadeh S, Jones A. Energy policy regime change and advanced energy
storage: A comparative analysis. Energy Policy 2018;115:572–83.
https://doi.org/10.1016/j.enpol.2018.01.029.
[53] Child M, Haukkala T, Breyer C. The role of solar photovoltaics and energy storage solutions
in a 100% renewable energy system for Finland in 2050. Sustain 2017;9.
https://doi.org/10.3390/su9081358.
[54] Kalkbrenner BJ. Residential vs. community battery storage systems – Consumer preferences
in Germany. Energy Policy 2019;129:1355–63. https://doi.org/10.1016/j.enpol.2019.03.041.
[55] Mejia C, Kajikawa Y. Emerging topics in energy storage based on a large-scale analysis of
academic articles and patents. Appl Energy 2020;263:114625.
https://doi.org/10.1016/j.apenergy.2020.114625.
[56] Conejo AJ, Sioshansi R. Rethinking restructured electricity market design: Lessons learned
and future needs. Int J Electr Power Energy Syst 2018;98:520–30.
https://doi.org/10.1016/j.ijepes.2017.12.014.
[57] Obi M, Slay T, Bass R. Distributed energy resource aggregation using customer-owned
equipment: A review of literature and standards. Energy Reports 2020;6:2358–69.
https://doi.org/https://doi.org/10.1016/j.egyr.2020.08.035.
[58] Castagneto Gissey G, Dodds PE, Radcliffe J. Market and regulatory barriers to electrical
energy storage innovation. Renew Sustain Energy Rev 2018;82:781–90.
https://doi.org/https://doi.org/10.1016/j.rser.2017.09.079.
Journal Pre-proof
1
Highlights
Centralized coordination vs. distributed operation of residential solar PV-battery is
discussed.
Centralized coordination offers greater savings to prosumers, especially, under time of use
tariffs.
Value of home batteries is dependent on the need for flexibility in the energy system in long
term.
Consumers with no energy technology benefit more from coordination compared to battery
owners.
Benefits of storage aggregation drops by 20% if aggregated storage devices increase five-fold.
Journal Pre-proof
Article title: Centralized vs. distributed energy storage systems: The case of residential solar PV-battery
Reference: EGY 121443
Declaration of interests
☒ The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
☐The authors declare the following financial interests/personal relationships which may be considered
as potential competing interests:
Behnam Zakeri on behalf of the authors 08 July 2021
Dr. Behnam Zakeri
Research Scholar (Energy, Climate, and Environment Program)
International Institute for Applied Systems Analysis (IIASA)
Schlossplatz 1, A-2361 Laxenburg, Austria | www.iiasa.ac.at
Tel: +43 (0)2236807532; Publications: Google Scholar
Journal Pre-proof
