Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews

journal homepage: www.elsevier.com/locate/rser

Smart grid and energy storage: Policy recommendations

⁎
Kenneth K. Zame , Christopher A. Brehm, Alex T. Nitica, Christopher L. Richard, Gordon
D. Schweitzer III
Center for Energy and Environmental Policy, University of Delaware, Newark, DE, USA

A R T I C L E I N F O A BS T RAC T

Keywords: Traditional energy grid designs marginalize the value of information and energy storage, but a truly dynamic
Energy policy power grid requires both. The authors support defining energy storage as a distinct asset class within the electric
Energy storage grid system, supported with effective regulatory and financial policies for development and deployment within a
Smart grid storage-based smart grid system in which storage is placed in a central role. This would enhance load and
Distributed generation
market operations through realization of the full range and value of services from storage technologies. Energy
Power distribution
storage technologies provide significant opportunities to further enhance the efficiency and operation of the
Power generation
grid. Its ability to provide application-specific energy services across different components of the grid make it
uniquely suited to respond quickly and effectively to signals throughout the smart grid. Therefore, energy
storage as a distinct asset class will increase the value of storage investments while enhancing the operation of
the smart grid. To further this goal, storage requires policy support. Research, development and demonstration
(RD & D) policies will increase operational experience and reduce costs; investment tax credits will accelerate
investment in storage projects; and continued market deregulation will augment revenue streams, enhance
competition, and provide more accurate prices for storage services.

1. Introduction unique opportunities to challenge the conventional model. To have a

fully dynamic and optimized grid, while transitioning to a low carbon
A shift to sustainable low carbon economy will require innovation energy system, planners need to embrace more fully the benefits of
and deployment of a range of low carbon technologies for providing renewable electricity generation and post-generation1 energy storage.
energy and other services. Rapid developing literature on renewable Realizing the full benefit of storage and smart grid technologies
energy and energy storage suggest electric power storage will facilitate requires establishing energy storage as a new asset class with a relevant
the deployment of renewable energy and also facilitate the develop- set of regulatory and financial policies to support its development.
ment of smart grids [1,2]. However, current conceptualization of the Although, this concept has been mentioned in a number of publications
role of electrical energy storage in smart grids does not well position [5–7], the specific role of electrical storage systems as a unique assert
storage to realize its full potential in facilitating a smart grid holisti- class in the structure of future electric gird value chain has not been
cally. sufficiently specified. This paper re-conceptualizes ESS as an asset class
Advancing smart grid technology and design requires that energy in a central role in the future structure of power markets and the grid;
system planning breaks from the business as usual understanding of thereby enabling the realization of the full potential of ESS in
energy storage to embrace a more efficient and sustainable framework. facilitating a smart grid. In the sections that follow, we review the
Energy storage systems (ESS) have been considered within the design literature on electric grid system; review integration of renewables into
of the energy grid but has had limited relevance because the high cost the grid, differentiate between conventional and smart grid models,
and inefficiency of many storage systems have undermined their value highlight the contributions or values of renewable energy to the electric
and functionality [3,4]. Additionally, existing perceptions of storage as grid. An overview of smart grid policy efforts is also provided. Next, we
little more than a bank to assist generators balance loads over time fails identify the limits to energy storage systems as a poorly defined asset
to capture the true potential of energy storage technologies [4]. class within the electric grid value chain, and demonstrate how creating
Improvements in smart grid designs and storage technologies present a new asset class for storage will both enhance the value of storage and

⁎
Corresponding author.
E-mail address: [email protected] (K.K. Zame).
1
Deal et al. defines energy storage as any technology, which is capable of storing energy after it has first been converted to electricity. For our purposes here, “post-generation” is used
to make a distinction of the placement of energy storage in the energy value chain in contrast to “pre-generation” fuel storage [72].

http://dx.doi.org/10.1016/j.rser.2017.07.011
Received 29 July 2016; Received in revised form 25 May 2017; Accepted 4 July 2017
1364-0321/ © 2017 Elsevier Ltd. All rights reserved.

Please cite this article as: Zame, K.K., Renewable and Sustainable Energy Reviews (2017), http://dx.doi.org/10.1016/j.rser.2017.07.011
K.K. Zame et al. Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

also provide significant benefits to the operation of the smart grid. and nuclear) continues towards more robust, renewable, and efficient
Finally, policy mechanisms in support of the rapid deployment of processes of connecting end-users with energy services [14]. A
energy storage systems are identified and discussed. disaggregated value chain [6] has created limits in the systematic
efficiency of the traditional grid system. In contrast, a smart grid design
2. Related literature allows for greater efficiencies by providing greater control of supply and
immediate usage feedback, which limits waste.
2.1. The electric grid system Looking to Makansi (2007), the traditional grid is easily defined
through five (5) asset classes: Source, Generation, Transmission,
The first electric system in history was Thomas Edison's Manhattan Distribution, and Delivery [6]. This classification of asset classes
Pearl Street Station in New York, which began operation in 1882. compartmentalizes the value chain and supports disaggregation as
Edison's electrical system was essentially a microgrid [8] consisting of a mentioned above. A depiction of Makansi's electricity value chain
100 V coal-fed generator supplying electricity to power a few hundred maintains the core components: Energy Source, Generation,
lamps. Thus at its beginning, the electric grid was small and highly Transmission, Distribution, and End User (Delivery), but to this the
localized [9]. In the early years of electric systems, power plants were modified model (as shown in Fig. 1) adds various forms of energy
usually situated both close to their fuel sources and to the end-users of storage and surplus (in the form of dump-loads). Notice the central
the power generated. However, with time, as more people began to feature of this grid structure: the mono-directional flow of energy and
demand for electricity smaller grids expanded into more complex information.
systems, and the electric grid evolved into large interconnected net- There are no feedback loops – just supply. It is only by occasionally
works as we have them now, connecting power generation stations, dumping excess loads that the energy is diverted from the end user.
transmission lines, distribution networks and load centers (end users). Since dump loads amplify revenue losses – in the sense of all costs and
The increasing complexity of power gird, coupled with growing electric no revenue – there are strong incentives for grid operators to minimize
demand, and the need for greater grid reliability and efficiency as well these occurrences while simultaneously maintaining delivery expecta-
as environmental and energy sustainability concerns continue to high- tions and the quality of energy services.
light the need for a considerable leap in making the grid “smarter.” The In contrast, a smart grid is often defined by the use of technology to
leap towards such a “smarter” grid is broadly referred to as “smart grid” relay time-sensitive energy data from end users back to the grid
[10,11]. Tuballa et al. (2016) noted that a smart grid does not have a operator in an effort to optimize the grid's efficiency while also
one universal accepted definition, however, simply put, the smart grid providing end users with real-time pricing from upstream generators.
is an intelligent grid. Whiles the traditional grid functions primarily by The same series of asset classes are considered, but now the flow of
transmitting and distributing electricity, the smart grid on the other information is bi-directional, as depicted in Fig. 2. This representation
hand is able to store, communicate and make decisions, thereby of the smart grid design replaces dump loads with a more direct flow of
functioning more cooperatively, responsibly and organically [11]. information back to generation. The result is that the energy generator
Differences between the traditional grid and the smart is well presented is provided with real-time demand, and supply can be matched more
by a number of publications [11–13]. In Table 1, we present a directly with consumer needs than by estimating the expected demand
compilation by Tuballa et al. (2016), of the general characteristics of at any given time. The drawback associated with this design is that the
the two grids. feedback loop applies only with the generation of energy, creating
inefficiencies in supply as generators ramp up and down in response to
2.2. Electric grid value chain: conventional vs. smart grid structures rapidly changing demand. Distributed generation technologies mitigate
this problem to a degree, but the scale of capacity is comparatively
Modern electric grid structures are often discussed through de- small (i.e. residential solar use).
scriptive terminologies, such as “traditional” or “smart.” Smart grid Differentiating the traditional energy grid from a smart grid design
technologies and real-time access to user information improve the focuses on greater efficiency by increasing knowledge. Better informa-
efficiency with which the production, transmission, and use of elec- tion leads to more efficient operation, while more stable and responsive
tricity occurs [7]. Through this improvement in efficiency, the roles and supply reduces consumer costs [15]. Note that energy storage enters
relationships of energy asset classes have begun to change. The this value chain in a supplementary rather than central role (see Fig. 2),
inadequacy of the traditional energy grid becomes increasingly appar- limiting the value of energy storage services and complicating regula-
ent as the transition away from stockpiled fuels (coal, natural gas, oil, tory and operational considerations for storage ownership.

Table 1 2.3. Smart grid policy overview

Traditional electric grid versus the smart grid [11].
A number of existing published work have examined the social,
Traditional Grid Smart Grid
regulatory, and institutional issues relating to sustainable energy
Mechanization Digitization development in general, and smart grid development in particular
One-way communication Two-way real-time towards advancing the technological evolution of smart grids [1,16–
communication 18]. The introduction of various smart grid policies in many countries
Centralized power generation Distributed power generation
has stem from the need to formulate innovative policies to support
Radial network Dispersed network
Less data involved Large volumes of data involved competitive offerings and end-user pricing systems which in turn will
Small number of sensors Many sensors and monitors contribute to improving allocative and productive efficiency within the
Less or no automatic monitoring Great automatic monitors operation of the grid while minimizing risk in the power market
Manual control and recovery Automatic control and recovery thereby incentivizing investors [19,20].
Less security and privacy concerns Prone to security and privacy
issues
Various approaches to smart grid policy adoption in different
Human attention to system disruptions Adaptive protection countries and jurisdictions globally are linked to the state of the power
Simultaneous production and consumption Use of storage systems industry and market in these countries/jurisdictions [21]. Lin et al.
of energy/electricity (2013) explains that the U.S.A, for instance, prefers an “environmental-
Limited control Extensive control system
side policy” approach which focuses on “scientific and technical
Slow response to emergencies Fast response to emergencies
Fewer user choices Vast user choices development, financial, policy and public enterprise.” While China,
on the other hand, is focused on “supply-side policy” approach with

2
K.K. Zame et al. Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

Fig. 1. Traditional Electricity Grid Value Chain.

emphasis on “public enterprise, scientific and technical development deployment of smart grid technologies and energy storage to improve
and legal regulations.” To illustrate some of the smart-grid policies grid operation and management, and facilitation of the integration of
used in different places, the rest of this section presents a quick renewables is one of the country's piece of legislation that indicates
overview of smart-grid policies across a number of countries/jurisdic- China's commitment to smart grid development [23,28]. China's
tions. “Special Planning of 12th Five-Year Plan on Smart Grid Science and
With a call by the European Union (EU) for the installation of smart Technology Industrialization Projects” is one other policy that shows
meters in 80% of EU households by 2020, majority of member states of the country's commitment to smart grid development. This plan
the EU have instituted electricity smart metering [22,23]. Though identifies the following as the key drivers for China's smart grid
smart meter programs in many European countries are a core part of deployment: large scale grid-connected intermittent renewable energy
their energy policies, these policies are not exclusively designed to only technology; large scale energy storage systems; intelligent distribution
facilitate balancing of supply and demand within the electric grid. technology; intelligent transmission technology and equipment; grid
These policies serve to support other sustainable energy or climate information and communication technologies; flexible power transmis-
change policies [19,23,24]. In the United States for instance, the sion technology and equipment; and comprehensive smart grid inte-
country's Energy Policy Act of 2005 promotes the development of gration demonstrations [23,29].
smart meters and for that matter smart grid, as the Act directs all While a number of countries recognize the need for policy reform
utilities in the country to consider time-based rate schedule and time- towards the development of smart grid, the implementation of such
based metering upon the request of customers [25]. Also in the U.S, the policies usually face challenges or difficulties such as institutional
American Recovery Act of 2009 stipulates for the development of smart rigidity and inertia. Such institutional barriers are sometimes as a
grid technologies and provides smart grid investment grants towards result of financial constraints and/or security and privacy risks con-
this purpose [26]. cerns regarding smart grids. However, a country like Denmark has
Japan has a goal of reducing its emissions by 30% by 2030. To demonstrated that effective implementation of changes in complex
achieve this goal, the country plans to transform its energy system by country-specific social, economic, and political factors that impede the
building “the world's most advanced next generation interactive grid implementation of smart grid policies are relevant towards transform-
network.” Japan's objective is towards the realization of smart grids ing a traditional grid into a smart one.
and smart communities by the year 2020, and this plan is documented
in the country's Strategic Energy Plan of 2014 [27]. China's amended
Renewable Energy Law of 2009, which specifies the development and

Fig. 2. Smart Grid Value Chain.

3
K.K. Zame et al. Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

Big Island [42]. The ability of geothermal power to serve as a form of

baseload power and also to perform such ancillary services in the grid
make it suitable to support the increased grid integration of other
forms of renewable electricity like wind and solar PV.
Literature on renewable power generation indicates that the
integration of renewable electricity into the electric grid is not only
already happening now, but renewable power's share in the grid is sure
to continue to increase [33,43–45]. International Energy Agency's
(IEA's) “Sustainable Energy Future” scenario projects 57% of the
world's electricity to be generated from renewables by 2050 [43]. In
developing a low carbon scenario for Europe, the European Climate
Foundation (ECF) found that a 50% renewable electricity by 2030 is
achievable at a cost comparable to that of a business-as-usual scenario
[46]. Denmark, for instance, plans to increase its wind energy
penetration to 50% by 2025. A study in 2010 by the Danish Energy
Association and Energinet.dk (Denmark's largest utility) compared the
cost of using smart grid technologies to the cost of traditional grid
upgrades to accommodate this goal. Results from this study showed
that the net cost of the necessary smart grid upgrades would be DKK
1.6 billion, versus DKK 7.7 billion for traditional grid upgrades [44].
The study's assertion is that an intelligent power system – smart grid –
Fig. 3. Energy Flow in a Storage-based Grid.
is the most effective strategy for developing the power system [44]. In
2015, electricity generated by Denmark from wind turbines corre-
2.4. Integrating renewables into the grid
sponded to about 42% of the country's electricity consumption.
Denmark is able to sustain increasing levels in wind power deployment
Integration of renewable sources of electricity into the grid enables
over the years by adopting several smart grid technologies to effectively
the efficient use of renewable energies which is a key challenge for now.
integrate wind energy forecasting into its grid operations, and to
Due to the availability and environmentally friendly nature of renew-
efficiently supply electricity to users in Denmark and surrounding
ables, as well as the application of smart grid in renewable energy,
countries [45]. In a Dutch smart grid pilot project called
integration of renewables is viewed as valuable [30,31]. However, the
“PowerMatching City,” a domestic virtual power plant was used to
fluctuating nature of generation from renewable energy sources like
demonstrate that the flexibility in virtual power plants can be used to
solar and wind into distribution grids complicates the balancing of
dampen the intermittent behavior of wind and solar energy [37].
demand and supply, which in turn gives rise to the risk of grid
The significance of renewable sources and other forms of decen-
instabilities [32].
tralized generation in providing additional grid support services is
Many recent studies have examined high penetrations of renewable
enhanced via electrical storage systems (ESS). Centralized power plants
generation; especially from wind, and investigated approaches to
(originally to the traditional power grid) supply large quantities of
dealing with the increase in net-load variability and uncertainty that
electricity for transmission and distribution (as depicted by the
this brings into reliability management of the electric grid. Increasing
“conventional grid energy flow” arrow in Fig. 3); however, this model
demand elasticity or storing surplus power are considered general
of operation of the power grid does not adequately meet the needs of
approaches to addressing the challenge of fluctuating feed-in of
the electricity industry. Economically, the cost of transmission and
renewable energy resources into the distribution grid [33–37].
distribution facilities associated with centralized generation no longer
Matching fluctuating renewable electricity generation with electricity
offer efficiency gains through economies of scale associated with
demand in an economically efficient manner that is sustainable with
increased generation capacities. Decentralized generation creates gains
low losses and high levels of quality is enabled by smart grid
accruing from the aggregation of energy from many sources as a result
technologies [32,38].
of economies of scale [47]. For instance, appropriately sized and
A number of studies confirm that hydropower resources contribute
customer-targeted distributed generation assets such as solar photo-
significantly to the operation of the grid. Most hydropower facilities
voltaic systems are proving to be more commercially viable.
have the capability to cycle to help manage net load variability and
Additionally, distributed generation is able to offer end users (who
uncertainty which in turn enables high wind and solar power grid
could simultaneously be distributed generators) and utilities peak
integration [39,40]. With increased development and expansion in
management, emergency power and other capabilities via energy
integration of renewables into the grid, interest in coupling and
storage [48,49].
coordinating renewables with pumped storage hydropower is growing;
As illustrated in Fig. 3; an ESS-based grid system facilitates energy
enhancing the integration of renewables; particularly wind and solar
flow interactions among distributed generation, end users and other
[34–36,41]. In this way, pumped storage hydropower projects are used
grid elements via storage. The integration of distributed energy
by utilities to flexibly pick up or shed load very rapidly to ensure the
generation assets and storage is expected to become increasingly
stability and power quality of the electric grid.
important in the future for implementing the smart grid concept and
Although geothermal electricity generation is usually operated as a
balancing the growing production and grid deployment of decentra-
source of baseload power, the advancement of power plant and control
lized renewable energy sources. Energy storage at the core of the power
technology allow geothermal power plants to work in several variable
grid does not mean a single large facility, as this can jeopardize grid
modes [42]. Such variable modes or ancillary services include grid
reliability. The idea of “the energy bank”2 that combines various
support, regulation, load following, spinning reserve, non-spinning
storage systems to an “energy storage cloud” would ensure efficient
reserve, and replacement or supplemental reserve. Puna Geothermal
power distribution with a high share of distributed solar PV and wind
Venture plant in Hawaii which generates 38 MW, and contracted
16 MW of flexible capacity is one such geothermal power project that
provides ancillary services for grid support, and these services are 2
The “energy bank” concept of creating an “energy storage cloud” is inspired by the
identical to those of existing oil-fired peak generating resources on the
concept of cloud computing and financial banking [50].

4
K.K. Zame et al. Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

energy [50]. A well-defined energy storage asset class at the core of the an excellent starting point by proposing the following definition for an
power grid would best facilitate this. This not only promotes the smart energy storage asset class [56]:
grid but also advances a shift away from conventional sources of energy
such as coal and nuclear energy. 1. Has the ability to store (receive and supply back) a definable amount
of energy (joules or gigajoules) to an electrical network or electrical
grid
3. Energy storage: an asset class within the electric value 2. Has a definable rate of both storing and providing the stored energy
chain (watts and watt-hours respectively)
3. Has a definable calendar life (years) under specified conditions
3.1. Energy storage: the missing link in electric value chain 4. Has a definable cycle life (total kWh transferred) under specified
conditions
Energy storage is pinpointed as a key technological component that 5. Has definable maintenance criteria and schedule
can transform the current structure and operation of the power grid 6. Has a definable round trip efficiency (including parasitic losses) to
[51–53]. However, regulatory and economic barriers limit its feasibility be used for economic analysis
and deployment on the grid. This, in turn, limits the applications and 7. Can be designed for use in one or more specific applications to
sources of revenue of energy storage systems. The source of these limits optimize grid operation and energy economics.
originates in the poorly defined status of energy storage. Creating a new
asset class for energy storage systems will open storage technologies to Of particular importance is the seventh and final characteristic of
provide multiple services and generate greater value. Currently, energy the energy storage asset class. If energy storage is to be viable it is vital
storage operators are forced to “choose between two dichotomous that we consider its viability in relation to its intended use. Howard &
regulatory structures, both of which may result in inefficient storage Kamath (2007) argue in support of this point by saying: “The real
use and investment” [54, p. 8]. Existing regulatory structures lead to challenge for energy storage is not whether it is possible, but how it
energy storage being undervalued in both traditional, vertically inte- will be used. There is no question that storage represents an
grated electricity models and restructured markets. A rate based opportunity; it will take strategy and understanding of this opportu-
system inhibits the deployment of energy storage toward its full value nity to make sure it is exploited to its full potential” [57]. If that
because it limits the revenue from multiple market services such as potential is to be directed towards each of the various asset classes and
those for ancillary services. A purely market-based system does not stakeholders, the framework for using energy storage needs to change.
provide price signals for the full range of energy storage services such Because energy storage systems store electricity in a form and for a
as transmission and distribution relief, and it currently is incapable of duration determined by the technology itself, they can be designed for
pricing all energy services [54]. multiple applications as needed by the grid. This is the source of its
The problem with narrowly defining the role or form of energy value, and defining storage as a new asset class would allow owners and
storage is that it limits the full value of storage services. Energy operators to provide the highest-valued services across components of
management, bridging power, and power stability [15] are key market the grid.
roles, which are at present provided less effectively by other asset The benefits of energy storage depend on the flexibility in applica-
classes. Energy storage is capable of providing a “shock absorber” for tion inherent in system design and operation. A growing body of
the overall grid system [15]. The traditional consideration of the role of literature details the numerous benefits of storage with new technol-
energy storage as supplementary largely creates regulatory uncertain- ogies expanding the range of potential applications [4,58–62]. Storage
ties. Regulators are uncertain how energy storage costs and benefits services are generally classified based on applications for grid support
should be allocated among the main components of the grid, as energy (market), generation, transmission/distribution, and customer/end-
storage can provide multiple services for generation, transmission, and user applications [58,60]. Apart from the aforementioned benefits of
distribution [55]. storage through peak-shaving capabilities, which may be considered
With grid assets classified as generation, transmission and dis- either a generation or transmission benefit, storage is also effective for
tribution, services offered by storage facilities only fall under one of power quality and regulation, integration of distributed and intermit-
these three asset classes. Narrowly classifying energy storage as tent power resources, reserve capacity supply, local, regional and grid-
supplementary to existing asset class of the grid stymies its pivotal level power back-up, and demand or supply management, amongst
role in facilitating a smart grid and makes it difficult for energy storage others. The following table (Table 2) summarizes a number of the
to find a suitable place in regulation as currently there is no defined services provided at each point of the traditionally defined electricity
energy storage-pricing policy and no specific tariff-regulation for grid value chain [58,60].
storage. As an inexplicitly defined or even non-asset class in the electric Each of the above services represents potential market opportu-
grid system, energy storage cannot derive full valuation in the market- nities for generators, utilities, end-users, and other interests.
place as markets are not structured and regulated to recognize services Traditionally, many of these services have been provided through more
that do not fit in the current value-chain. The economic argument conventional technologies, principally through generation-based tech-
against energy storage of not being able to fully recover cost stems from nology [63]. Recent approaches include demand-side management and
the fact that energy storage facilities are not offered distinct asset status energy efficiency strategies, but these approaches, too, have limitations
and accorded the appropriate regulations to harness multiple revenue in terms of matching technology or strategy to service desired. For
streams. example, demand-side management strategies limit options to load
It is quite obvious that, in order to extract the full value of energy leveling or peak shaving (by contracting consumers who agree to limit
storage applications, energy storage must be defined as a distinct asset energy consumption during specific periods), but this option cannot
class from those that currently exist in the grid value chain. The also provide additional grid support services such as back-up power,
problem at this point is building a definition that sufficiently distin- transmission congestion relief, and system upgrade deferrals. Energy
guishes energy storage from other asset classes, especially generation. storage options provide applications and services that match technol-
DNV KEMA,3 an energy and environmental consulting firm, provides ogies to needs. Already, several reports indicate the technical and
economic benefits that storage has over conventional technologies,
3
DNV KEMA Energy & Sustainability is a global, leading authority in business and
particularly in ancillary service markets [61,63].
technical consultancy, testing, inspections & certification, risk management, and Another key benefit of the new storage asset class is that more
verification, along the energy value-chain (〈http://www.dnvkema.com/about/〉). revenue leads to more investment. Because energy storage is no longer

5
K.K. Zame et al. Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

Table 2 3.3. Benefits of energy storage in the smart grid

Storage Services by Application.
Just as smart grids are essential in increasing the implementation
Generation Transmission & End-User Grid Support
Generation of ESS, so is ESS complementary in meeting the goals of an efficient
smart grid. The proposed benefits of smart grids to utility companies
Peak-shaving Peak Capacity Back-up Power Voltage and the electricity system as a whole include improved reliability with
Support Support
less costly interruptions, deferred capital spending on costly transmis-
Capacity Deferral Peak-shaving Demand Frequency
Management Regulation sion and generation assets, and increased efficiency of power delivery
Reserve Capacity Congestion Relief Price Arbitrage Energy due to lower distribution losses [59]. Energy storage meets these
Balancing benefits by negating the need for extra peaking generation through load
Distributed & Upgrade Deferral Resource leveling, deferring the transmission and distribution upgrades needed
Intermittent Adequacy
to meet load growth with a smaller sized energy storage investment,
Source
Integration reducing transmission congestion fees in deregulated markets by
adding energy storage to distribution substations, and providing the
load following capabilities that will help improve the intermittency of
restricted to supplementing other asset classes, it can derive revenue renewable energy sources [66]. Electric energy storage as a key enabler
from the services it provides to each. Furthermore, storage operators and enhancer of dispatchability of renewables; provides options to
are free to dispatch their systems according to the highest-valued use, offset the mismatch between demand and supply and to operate the
thus generating a higher return on their systems than they would as a distribution system in a more efficient, economic, and environmentally
supplementary asset. Storage derives its greatest returns through its sound manner [11,32,38,52,67]. Lastly, with smart grid technologies
ability to provide multiple services for all currently existing asset and energy storage in place, benefits to residential consumers will
classifications [56]. This creates a natural (rather than subsidy-based) include cost savings from peak load management, energy efficiency,
incentive for additional investment in storage. Of course, such invest- and increasingly affordable distributed renewable energy systems [59].
ments require appropriate market and regulatory guidelines to be in On the other hand, smart grid in turn also provides opportunities for
place, and operators would need access to accurate, real-time price and load control and dispatch of storage units making renewable sources
demand information to dispatch their systems in the most efficient such as solar PV and wind more valuable to the grid.
manner.
Regulatory, economic and other challenges that inhibit further 3.4. Benefits of the smart grid to energy storage
development and deployment of energy storage in the power grid can
best be surmounted through the classification of storage as a distinct The availability of utility information through the smart grid
asset. The marketplace would be sufficiently receptive and responsive facilitates achieving the maximum benefit in the operation of an ESS
for storage to realize its most efficient value. The support for appro- [64]. Heightened visibility enhances price forecasting techniques,
priate regulations and incentives for storage would also attract invest- ultimately optimizing gains in short term energy markets through
ments in energy storage systems. arbitrage [67]. Currently, grid operators would use strategies, such as
back-casting (using historical data to predict economically desirable
deployment schedules) to apply energy storage. This strategy does not
completely capture arbitrage value due to near time weather and usage
3.2. Asset class position and role of energy storage within the smart variations (only 85%) [67]. Real time information exchanges allows for
grid a more responsive grid, achieving near perfect forecasting. Maximizing
these gains increases both return on investment for ESS and competi-
As utility networks are transformed into smart grids, interest in tiveness with other energy systems. [67]
energy storage systems is increasing within the context of aging One of the advantages of the smart grid is that it allows for a wider
generation assets, heightening renewable energy penetration, and more array of technologies. Fast response storage technology, such as
distributed sources of generation [64]. Along with this growing flywheels, can respond to intermittency issues, improving the overall
interest, smart grids are developing to effectively and intelligently reliability and power quality of the grid. Additional information can
integrate the functions of various asset classes to more efficiently also lead to increasing storage deployment, ultimately reducing the
deliver sustainable, secure, and economic electricity supplies [65]. standby efficiency losses.
These systems become mutually beneficial. For example, storage Likewise, as the proportion of low voltage, feed-in renewable energy
systems require the information advantages of the smart grid to realize generators grow, grid complications in supply and demand will
their full potential. Smart meters and real-time access to demand, increase, potentially damaging electronic equipment and causing
supply, power quality, and price levels provide the signals that power outages [32]. Smart grid coupled with energy storage systems
incentivize the use of storage across various points of the power supply increases demand elasticity while also disconnecting the simultaneity
chain. of production and consumption. Together, these services balance
Therefore, placing the energy storage asset class at the nexus of the supply and demand while allowing a continual increase of renewables
electricity value chain would emphasize the role that energy storage on the grid. [32]
technologies are able to play in the implementation of smart grid Although ESS can provide these various grid services along with
systems. The conceptualization of placing an explicitly defined energy others noted in Table 2, it is unrealistic that ESS can deliver every
storage as a distinct asset class in a central role to generation, potential benefit at all times. The smart grid, however, can facilitate the
transmission, distribution and end-user segments of the electric grid delivery of the most needed technical benefit to the grid, co-optimizing
system is illustrated in Fig. 4. The bi-directional arrows indicate a bi- between revenue streams such as arbitrage and ancillary markets [67].
directional flow of information. With energy and information flow Ultimately, the integration of ESS into a smart grid expands the
directly to the energy storage asset class, storage can respond more exchange of both information and services between asset classes,
effectively to the needs of the grid. ultimately optimizing usage and value of ESS.
The following sections provide examples of how the classification of Placing the energy storage asset class at the nexus of the value chain
energy storage as a separate asset class within the smart grid improves emphasizes the role that energy storage technologies are able to play in
the performance of both. the implementation of smart grid systems and vice versa. However, the

6
K.K. Zame et al. Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

Fig. 4. Information Flow in an Energy Storage-based Smart Grid.

current capacity of energy storage on the grid is wholly inadequate. In beginning with government-backed demonstration projects followed by
order for it to reach sufficient capacity to support smart grid operation, utility and independent investment in larger installations should the
energy storage systems require policies that will enhance their deploy- pilot prove successful [54]. Although many emerging energy storage
ment in the near term. We therefore explore and recommend policies technologies exhibit this progressively scaled schedule, there are
with the most potential at facilitating the transition to a storage-based current existing limits to their physical storage capacity that prevent
smart grid. their application for larger scale energy management operations that
are typically reserved for technologies such as pumped hydro storage
and single-cycle natural gas-fired power plants.
4. Policy recommendations
On the technical side, the most immediate requirements are
financing research to improve performance of the most market-ready
4.1. Research, development and demonstration (RD & D)
or highest-valued storage systems. According to a United States
Department of Energy (DOE) report that conducted an electricity
Continued research and development of new energy storage
market analysis for emerging energy storage applications such as
technologies, as well as larger scale applications of existing energy
flywheels and NaS batteries, current RD & D efforts for energy storage
storage technologies, is crucial for promoting the increased develop-
should focus on improving round-trip efficiency and reducing capital
ment of energy storage within a smart grid framework. Many advanced
costs [62]. For instance, the current energy-limited nature of flywheels,
energy storage technologies are in a pre-commercialized stage, and
or short dispatch time, limits the amount of time they can provide
thus the costs of energy storage devices are expected to decrease in the
ancillary services such as regulation services and receive regulation
future [68]. RD & D reduces the relatively high capital costs of energy
revenues [62]. Increased research and development into improving the
storage technologies, enhances the possible revenue streams of energy
technical characteristics of existing and emerging technologies will
storage technologies through increased applications in ancillary ser-
allow them to play a greater role when improved ancillary service
vices markets, and helps to show the viability of advanced and next-
markets are created.
generation energy storage technologies. Research and development
funding for energy storage includes such diverse areas as the pilot
projects for emerging energy storage technologies, for the continued 4.2. Investment tax credit
scale up in physical storage capacity of existing energy storage
technologies, and improving technical capabilities of existing technol- Implementing short-term policies addresses the high cost of capital
ogies with new designs and materials. associated with many energy storage options. Investment tax credits
Technology risks are a critical barrier to the deployment of energy (ITCs) are an extremely effective method of reducing capital costs and
storage technologies, and numerous technically feasible energy storage limiting exposure to technological and capital risk. ITCs promote a
technologies have seen delayed deployment because developers are more rapid increase in storage capacity for services such as frequency
reluctant to be the first to undertake projects with new systems [54]. regulation. ITCs also enhance wide-spread deployment of technology
Pilot and demonstration projects are especially critical in order to show by bolstering market demand and promote the cost-competitiveness of
the viability of newer technologies, and successful demonstrations new technologies against conventional or established ones. Research
reduce the risk of investing in these technologies and help in securing has shown that with a possible 20% federal ITC for storage in the
private investor funding for large-scale energy storage systems. United States over a ten-year period, total capacity could triple
Examples of technologies that have shown to be technologically feasible compared to a scenario without the ITC [69]. In the same analysis, it
through small-scale demonstration projects include compressed air was shown that within seven years nearly 2500 MW of new compressed
energy storage (CAES), flywheels, and battery technologies such as air energy storage capacity is possible with 20% ITC as compared with
sodium-sulfur (NaS) batteries, lithium ion batteries, and liquid elec- only 700 MW without any ITC [69]. Furthermore, renewable energy
trolyte low batteries such as vanadium redox batteries [54]. Typically, markets have demonstrated the effectiveness of tax break incentives,
energy storage technologies tend to follow this pattern of deployment including ITC regimes over the past years [70].

7
K.K. Zame et al. Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

An effective ITC program promotes energy storage expansion in the Storing electricity is expensive, especially when compared against
short run while accelerating long-run capital cost reductions. A the cost of generating power with coal, gas, or other conventional
dynamic energy storage market characterized by stable supply and resources. Deploying storage one-dimensionally – for peak-shaving
low costs is essential to reach the full potential of an effective smart and load balancing – limits its value and viability as a technology.
grid as envisioned in this paper. Several aspects are important in Expanding markets for energy services that match energy storage
designing an ITC program [70,71]. Among the most important design applications is an appropriate mechanism to realize storage technolo-
aspects are that ITCs have a long-term program horizon to prevent gies’ full value. The deregulation of utilities and energy markets and
cyclical market volatility, are simple and clear for market investors, and support for greater regulatory and market structure will enhance
have a predictable, gradual phase-out schedule. In addition, ensuring participation by storage owners and allow greater third-party and
that high-quality and reliable equipment is deployed, ITC-eligibility independent access to energy markets. Energy markets, too, will
should be predicated on a set of performance standards. An effective become more diversified in terms of the services available for exchange
set of standards will reinforce technical reliability, bolster consumer and eligible participants. Allowing storage to capture its full value in
confidence and smooth out supply volatility by ensuring manufacturing competitive markets may be among the most effective tools in
quality. promoting the development and deployment of storage technologies.
Production incentives are preferred from a policy standpoint
because there are fewer opportunities for misuse of the credits, but, 5. Conclusion
because energy storage does not produce power but rather energy
services, such a policy tool is impractical. A different policy approach is In conclusion, we reaffirm the proposition for the re-conceptualiza-
therefore required to ensure energy storage has adequate influence in tion of the traditional power grid model. The conventional model relies
power markets. on upstream generators to forecast demand and operates based on
these forecasts. This results in inefficient operation of the grid, with
4.3. Market formation and support for energy storage surplus loads being dumped while insufficient energy scenarios result
in expensive courses of action to prevent more serious grid problems.
As was previously discussed, a key barrier to widespread deploy- This model is one of unidirectional energy and informational flows,
ment of energy storage results from its traditional role in the conven- with limited feedback systems to enhance operation. The emergence of
tional grid system. When deployed as a one-dimensional means of load smart grid technologies provides ample opportunity to improve grid
balancing and peak shaving, the value of grid-level and distributed operation by increasing the bi-directional flow of information between
storage systems is limited to energy arbitrage: transferring energy upstream and downstream parties. However, there are still limitations
produced during low-price, off-peak hours to hours of high demand to this model, as all energy services rely primarily upon generation
and prices. This severely circumscribes the full value of the services assets.
that different types of energy storage technologies provide. With the Energy storage technologies provide significant opportunities to
rise of deregulated markets, significant opportunities exist to expand further enhance the efficiency and operation of the grid. Its ability to
markets to more fully and efficiently exploit energy storage services on provide application-specific energy services across different compo-
the grid. This should further incentivize national regulatory entities or nents of the grid make it uniquely suited to respond quickly and
agencies such as the Federal Regulatory Commission (FERC) in the effectively to signals throughout the smart grid. Therefore, energy
U.S. that regulates the transmission in the nation's interconnection storage as a distinct asset class in a central role will increase the value
regions to work towards developing and supporting new and existing of storage investments while enhancing the operation of the smart grid.
markets that promote energy storage. To further this goal, storage requires policy support. RD & D policies
Deregulated electricity markets offer the best potential for devel- would increase operational experience and reduce costs; investment
oping storage services. For instance, in the U.S., a number of inter- tax credits will accelerate investment in storage projects; and continued
connection regions and independent service operators such as PJM (a market deregulation will augment revenue streams, enhances competi-
regional transmission organization (RTO) that coordinates the move- tion, and more accurately price storage services.
ment of wholesale electricity in 13 States and the District of Columbia),
and the New York Independent System Operator (NYISO) have already Acknowledgment
established vibrant ancillary services markets for services such as
frequency regulation and voltage support [61,62]. For example, The authors would like to thank the anonymous reviewers for their
markets for frequency regulation in NYISO generated nearly $100 invaluable comments. We are grateful to Dr. John Byrne, Dr. Lado
million in service payments, and as the price for regulation services has Kurdgelashvili, Mr. Yeng-Chieh Tsai of the Center for Energy and
steadily risen so have incentives for growth in investments and Environmental Policy as well as Dr. William Latham of the University
increased deployment of storage technologies. Enhancing the value- of Delaware and Mr. J. Mack Wathen of Pepco Holdings, Inc. for their
generating potential of energy storage through expanding access to contributions to this paper. This research did not receive any specific
ancillary and other markets is a powerful mechanism for building a grant from funding agencies in the public, commercial, or not-for-profit
storage-based smart grid network. sectors.
In order to further this effort, agencies such the FERC should
continue supporting the process of deregulation in energy and power References
markets. Furthermore, regulatory agencies, interconnection adminis-
trators, and utilities should take the necessary steps to ensure stable [1] Wolsink M. The research agenda on social of distributed generation in smarts grids:
market development. Regulatory agencies can develop rules and renewable as common pool resources. Renewable Sustain Energy Rev
2012;16:822–35.
protocol to monitor storage operations and evaluate cost-of-service [2] Hledik R. How green is the smart gird. Electr J 2009;22(3):29–41.
rates that will enable owners of storage assets to secure revenue [3] Rahbar K, Moghadam MRV, Reindl T, Panda SK. Shared energy storage manage-
streams and access to capital for provision of ancillary storage services ment for renewable energy integration in smart grid, in: Innovative Smart Grid
Technologies Conference (ISGT), 2016 IEEE Power & Energy Society; 6–9
[63]. RTOs and utilities that participate in wholesale markets can September 2016.
enhance market participation and access to independent interests such [4] Walter CJ. High-value energy storage for the grid: a multi-dimentional look. Electr
as storage owners by creating interconnection standards and compen- J 2010;23(10):59–71.
[5] Makansi J, Energy Storage: The Sixth Dimension of the Electricity Production and
sation schemes for distributed or small-scale storage units. This, in Delivery Value Chain; 18 Dec 2002. [Online]. Available: 〈http://www.
turn, furthers market competition.

8
K.K. Zame et al. Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

energystoragecouncil.org/1%20-%20Jason%20Makansi-ESC.pdf〉. [Accessed 15 wind power fluctuations with a domestic virtual power plant inEurope's first smart
October 2012]. grid. IEEE Trondheim Power 2011.
[6] Makansi J. Lights out: the electricity crisis, the global economy, and what it means [38] Dincer I. Smart energy solutions. Int J Energy Res 2016;40(13):1741–2.
to you. Hoboken, NJ: John Wiley & Sons, Inc; 2007. [39] Loose VW. Electric grid: role of hydropower in existing markets. Albuquerque, New
[7] Mohd A, Ortjohann E, Schmelter A, Hamsic N, Morton D. Challenges in integrating Mexico,U.S.A: Sandia National Laboratories; 2011.
distributed energy storage systems into future smart grid, in IEEE International [40] EPRI, Quantifying the value of hydropower in the electric grid: final report, Electric
Symposium on Industrial Electronics; 2008. Power Research Institute (EPRI), Palo Alto, CA; 2013.
[8] Asmus P. Microgrids, virtual power plants and our distributed energy future. Electr [41] Kapsali M, Kaldellis JK. Combining hydro and variable wind power generation by
J 2010;23(10):72–82. means of pumped storage under economically viable terms. Appl Energy
[9] Monaco A, Edison’s Pearl Street Station recognized with milestone: IEEE honors 2010;87(11):3475–85.
the world’s first central power station; 27 July 2011. [Online]. Available: 〈http:// [42] Matek B. Firm and flexible power services available from geothermal facilities.
theinstitute.ieee.org/tech-history/technology-history/edisons-pearl-street-station- Geothermal Energy Association; 2015.
recognized-with-milestone810〉. [Accessed 18 February 2017]. [43] IEA, Energy technology perspectives. International Energy Agency; November
[10] Moslehi K, Kumar R. Smart Grid – a reliability perspective, in: Innovative smart 2012.
grid technologies (ISGT),; 19–21 January 2010. [44] Energinet.dk, Smart grid in Denmark; 2010.
[11] Tuballa ML, Abundo ML. A review of the development of smart grid technologies. [45] Energinet.dk, New record-breaking year for Danish wind power; 15 January 2016.
Renew Sustain Energy Rev 2016:710–25. [Online]. Available: 〈http://energinet.dk/EN/El/Nyheder/Sider/Dansk-
[12] Cunjiang Y, Huaxun Z, Lei Z. Architecture design For smart grid. Energy Procedia vindstroem-slaar-igen-rekord-42-procent.aspx〉. [Accessed 24 February 2017].
2013;17:1524–8. [46] European Climate Foundation, Power Perspectives 2030: on the road to a
[13] Sendin A, Berganza I, Arzuaga A, Ororio X, Urrutia I, Angueira P. Enhanced decarbonised power sector, European Climate Foundation; 2011.
operation of electricity distribution grids through smart metering PLC network [47] Byrne J, Glover L, Takahashi K, Baker S, Kurdgelashvili L. Policy options to support
monitoring, analysis and grid conditioning. Energies 2013;6:539–56. distributed resource options. A report to conectiv power delivery. Newark, DE:
[14] Smalley RE. Future global energy prosperity: the terawatt challenge. MRS Bull Center for Energy and Environmental Policy; 2005.
2005. [48] Byrne J, Letendre S, Govindarajalu C, Wang Y. Evaluating the economics of
[15] Makansi J, Abboud J. Energy storage: the missing link in the electricity value chain. photovoltaics in a demand-side management role. Energy Policy
Energy Storage Counc 2002. 1996;24(2):177–85.
[16] Giordano V, Fulli G. A business case for smart grid technologies: a systemic [49] Byrne J, Nigro R, Wang Y-D. Photovoltaic technology as a dispatchable peak-
perspective. Energy Policy 2012;40:252–9. shaving option. Public Uti Fortn 1995, Sept. 1.
[17] Krishnamurti T, Schwartz D, Davis A, Fischhoff B, De Bruin WB, Lave L, Wang J. [50] F. Röhr, The energy bank: a smart grid business model inspired by cloud computing
Preparing smart grid technologies: a behavioral decision research approach to and financial banking; 2011.
understanding consumer expectations about smart meters. Energy Policy [51] Ozdemir E, Ozdemir S, Erhan K, Aktas A. Energy storage technologies opportu-
2012;41:790–7. nities and challenges in smart grid, in Smart Grid Workshop and Certificate
[18] Lund H, Anderson AN, Ostergaard PA, Mathiesen BV, Connolly D. From electricity Program (ISGWCP); 2016.
smart grids to smart energy systems: a market operation based approach and [52] Bocklisch T. Hybrid energy storage systems for renewable energy applications. In:
understanding. Energy 2012;42:96–102. Proceedings of the 9th International Renewable Energy Storage Conference,
[19] Clastres C. Smart grids: another step towards competition, energy security and Dusseldorf; March 2015.
climate change objectives. Energy Policy 2011;39(9):5399–408. [53] Lindley D. The energy storage problem. Nature 2010;463(7277):18–20.
[20] Katina PF, Keating CB, Zio E, Gheorghe AV. A criticality-based approach for the [54] Sioshansi R, Denholm P, Jenkin T. Market and policy barriers to deployment of
analysis of smart. Technol Econ Smart Grids Sustain Energy 2016:1–20. energy storage. Econ Energy Environ Policy 2012:47–63.
[21] Lin C-C, Yang C-H, Shyua JZ. A comparison of innovation policy in the smart grid [55] DOE Electricity Advisory Committee, Bottling electricity: storage as a stragegic tool
industry across the pacific: china and the USA. Energy Policy 2013;57:119–32. for managing variability and capacity concerns in the modern grid; 2008.
[22] Uribe-Pérez N, Hernández L, de la Vega D, Angulo I. State of the art and trends [56] KEMA D. KEMA Energy Storage Policy Briefing; 16 July. [Online]; 2009. Available:
review of smart metering in electricity grids. Appl Sci 2016;6(68). 〈http://www.kema.com/Images/KEMA%20Energy%20Storage%20Policy
[23] Brown MA, Zhou S. Smart-grid policies: an international overview. WIREs Energy %20Briefing.pdf〉.
Environ 2013;2(2):121–39. [57] Howard MW, Kamath H. Energy storage breakthroughs: an evolving technology for
[24] Woods E, Strother N. Smart meters in Europe. Navigant 2012. managing the grid; 2007 [Online] Available: 〈http://energycentral.fileburst.com/
[25] U.S GPO, Enegy policy act of 2005, United State Government Publishing Office, EnergyBizOnline/2007-4-jul-aug/Tech_Storage.pdf〉. [Accessed 25 June 2016].
Washington, DC; 2005. [58] Pauker S. State electricity storage policies. Clean Energy States Alliance 2012.
[26] U.S DOE, Recovery act smart grid document collection, SmartGrid.Gov; 2016. [59] EAC, Smart grid: enabler of the new economy report recommendations.
[27] Ministry of Economic Trade and Industry, Strategic Plan, Minstry of Economic Department of Energy, Washington, DC; 2011.
Trade and Industry; 2014. [60] Miller G, Corey G, The impacts of regulation, policy, advanced technologies, and
[28] Ministry of Commerce People’s Republic of China, Renewable energy law of the market dynamics on the deployment of energy storage processes; October 2011.
people’s republic of china; 20 December 2013. [Online]. Available: 〈http://english. [Online]. Available: 〈http://www.sandia.gov/eesat/2011/papers/ESSAT_Miller_
mofcom.gov.cn/article/policyrelease/Businessregulations/201312/ Monday_Paper_Oct_2011_Final.pdf〉. [Accessed October 2012].
20131200432160.shtml〉. [Accessed 6 March 2017]. [61] Allen D, Brown C, Hickey J, Le V, Safuto R. Energy storage in the New York
[29] Ministry of Science and Technology of the People’s Republic of China, Special electricity markets. NYISO; 2010.
planning of 12th five-year (2011–2015) on smart grid major Science and [62] Walawalkar R, Apt J. Market analysis of emerging electric energy storage services.
Industrialization Projects, in Smart-grid policies: an international overview, WIREs Washington, D.C.: Department of Energy; 2008.
Energy and Environment; 2013, p. 121–39. [63] Electricity Storage Association (ESA), Investment Tax Credit (ITC) for Energy
[30] Hossain MS, Madlool NA, Rahim NA, Selvaraj J, Pandey AK, Khan AF. Role of Storage, ESA; 2011.
smart grid in renewable energy: an overview. Renew Sustain Energy Rev [64] Wade N, Taylor P, Lang P, Jones P. Evaluating the benefits of an electrical energy
2016:1168–84. storage system in a future smart grid. Energy Policy 2010;38:7180–8.
[31] Köktürk G, Tokuc A. Vision for wind energy with a smart grid in Izmir. Renew [65] Group ENS. A smart gird vision. UK; November 2009.
Sustain Energy Rev 2016;73:332–45. [66] Geurin SO, Barnes AK, Carlos BJ, Smart Grid Applications of Selected Energy. In:
[32] Römer B, Reichhart P, Kranz J, Picot J. The role of smart metering and Innovative Smart Grid Technologies (ISGT), 2012 IEEE PES, Washington, D.C.;
decentralized electricity storage for smart grids: the importance of positive 2012.
externalities. Energy Policy 2012;50:486–95. [67] Roberts B, Sandberg C. The role of energy storage in development of smart grids.
[33] IRENA, Smart Grid and Renewables: a guide for effective deployment," Proc IEEE 2011;99(9).
International Renewable Energy Agency; November 2013. [68] Sioshansi R, Denholm P, Jenkin T, Weiss J. Estimating the value of electricity
[34] Haque A, Rahman MA. Study of solar PV-powered mini-grid pumped hydroelec- storage in PJM: arbitrage and some welfare effects. Energy Econ 2009;31:269–77.
tricity storage and its comparison with battery storage, in: Proceedings of the 7th [69] KEMA D. Market evaluation for energy storage in the United States. Fairfax:
International Conference on Electrical and Computer Engineering, Dhaka, KEMA, Inc.; 2012.
Bangladesh; December 2012. [70] Gyuk I. Electrical energy storage: U.S. policy issues. Paris: IEA; 2011.
[35] Wen S, Lan H, Fu Q, Yu DC, Zhang L. Economic allocation for energy storage [71] Sawin J, Flavin C. Policy lessons for the advancement & diffusion of renewable
system considering wind power distribution. IEEE Trans Power Syst 2015;30(2). energy technologies around the world. New Delhi: The Energy Research Institute;
[36] Jamii J, Saouidi Y, Fouizi M. Management of pumped storage hydro plant coupled 2004.
with wind farm, In: Proceedings of the 7th International Renewable Energy [72] Deal M, Churchill S, Chaset L, Villarreal C. Electric energy storage: an assessment
Congress (IREC); March 2016. of potential barriers and opportunities. Public Utilities Commission of California;
[37] Roossien B, van den Noort A, Kamphuis R, Bliek F, Eijgelaar M, de Wit J. Balancing 2010.