Electrical energy storage systems: A comparative life cycle cost analysis

Electrical energy storage systems: A comparative life cycle cost analysis

Renewable and Sustainable Energy Reviews 42 (2015) 569–596 Contents lists available at ScienceDirect Renewable and Sustainable Energy Reviews journa...

4MB Sizes 3 Downloads 46 Views

Renewable and Sustainable Energy Reviews 42 (2015) 569–596

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews journal homepage: www.elsevier.com/locate/rser

Electrical energy storage systems: A comparative life cycle cost analysis Behnam Zakeri n, Sanna Syri Department of Energy Technology, Aalto University, PL 14100, FIN-00076 Aalto, Finland

art ic l e i nf o

a b s t r a c t

Article history: Received 8 May 2014 Received in revised form 1 August 2014 Accepted 5 October 2014

Large-scale deployment of intermittent renewable energy (namely wind energy and solar PV) may entail new challenges in power systems and more volatility in power prices in liberalized electricity markets. Energy storage can diminish this imbalance, relieving the grid congestion, and promoting distributed generation. The economic implications of grid-scale electrical energy storage technologies are however obscure for the experts, power grid operators, regulators, and power producers. A meticulous technoeconomic or cost-benefit analysis of electricity storage systems requires consistent, updated cost data and a holistic cost analysis framework. To this end, this study critically examines the existing literature in the analysis of life cycle costs of utility-scale electricity storage systems, providing an updated database for the cost elements (capital costs, operational and maintenance costs, and replacement costs). Moreover, life cycle costs and levelized cost of electricity delivered by electrical energy storage is analyzed, employing Monte Carlo method to consider uncertainties. The examined energy storage technologies include pumped hydropower storage, compressed air energy storage (CAES), flywheel, electrochemical batteries (e.g. lead–acid, NaS, Li-ion, and Ni–Cd), flow batteries (e.g. vanadium-redox), superconducting magnetic energy storage, supercapacitors, and hydrogen energy storage (power to gas technologies). The results illustrate the economy of different storage systems for three main applications: bulk energy storage, T&D support services, and frequency regulation. & 2014 Elsevier Ltd. All rights reserved.

Keywords: Battery energy storage Cost of energy storage Electricity market Electricity storage Renewable energy integration Smart grid Techno-economic analysis

Contents 1. 2.

3.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Electricity storage for a flexible power system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Imperatives of electricity storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.1. Meeting demand and reliability in grid's peak hours . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2. Liberalized electricity markets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.3. Intermittent renewable energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.4. Distributed generation and smart grid initiatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Alternative solutions for increasing the flexibility of the power system. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . EES technologies: characteristics and costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. General considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Methodology in cost analysis of EES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.1. Total capital cost (TCC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.2. Life cycle costs (LCC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

570 571 571 571 571 571 571 572 572 572 572 572 573

Abbreviations: AA-CAES, advanced adiabatic compressed air energy storage; ALCC, annualized life cycle costs; BES, battery energy storage; BOP, balance of plant; CAES, compressed air energy storage; CRF, capital recovery factor; D-CAES, diabatic compressed air energy storage; DG, distributed generation; DOE, The US Department of Energy; DoD, depth of discharge; EES, electrical energy storage; FC, fuel cell; GT, gas turbine; IQR, interquartile range; LCC, life cycle costs; LCOE, levelized cost of electricity; LCOS, levelized cost of storage; NaS, sodium–sulfur (battery); Ni–Cd, nickel–cadmium (battery); O&M, operation and maintenance; PCS, power conversion system; PEM, polymer electrolyte membrane; PHS, pumped hydroelectricity storage; PSB, polysulfide–bromide (battery); RES, renewable energy source; RES-E, electricity from renewable energy source; SCES, supercapacitor energy storage; SMES, superconducting magnetic energy storage; T&D, transmission and distribution; TCC, total capital costs; TSO, transmission system operator; UPS, uninterruptible power supply; VRFB, vanadium-redox flow battery; VRLA, valve-regulated lead–acid (battery); ZEBRA, zero emission battery (NaNiCl2 battery) n Corresponding author. Tel.: þ 358 405007085. E-mail addresses: [email protected]fi (B. Zakeri), [email protected]fi (S. Syri). http://dx.doi.org/10.1016/j.rser.2014.10.011 1364-0321/& 2014 Elsevier Ltd. All rights reserved.

570

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

3.3. 3.4.

Methodology in review and collection of cost data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 573 EES technologies and related costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 577 3.4.1. Mechanical energy storage systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 577 3.4.2. Electrochemical battery energy storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 579 3.4.3. Electric and magnetic energy storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 580 3.4.4. Power to gas energy storage technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 581 3.4.5. Other electricity storage technologies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 582 4. Results and discussion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 582 4.1. Results of the review for individual cost items . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 582 4.2. Total capital cost (TCC) of EES systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 584 4.3. Life cycle costs (LCC) of EES systems and uncertainty analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 585 4.3.1. Sensitivity analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 588 4.4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 589 5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 589 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 590 Appendix A. Cost elements of different EES systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 590 Appendix B. Summary of technical characteristics of EES systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 592 Appendix C. Total capital cost of different EES systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 592 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 593

1. Introduction Power systems are on the threshold of a new transformation by the confluence of deploying variable renewable energy sources (RES) and free electricity markets. High share of variable RES intensifies the variability and intermittency of the power supply, disrupting the optimal operation of conventional power systems and grid reliability. Deregulated electricity markets introduce a competitive environment for power producers resulting in high capital cost requirement for meeting peak demands and volatile electricity prices. This new setting has imposed technical, economic, and environmental challenges for secure supply of electricity. Energy storage is deemed as one of the solutions for stabilizing the supply of electricity to avert uneconomical power production and high prices in peak times. The recent World Energy Outlook (2013) published by International Energy Agency (IEA) predicts a significant growth in the share of variable RES in total electricity generation, from 6.9% in 2011 to 23.1% by 2035 within the EU [1]. Accordingly, the European Commission has recognized electricity storage1 as one of the strategic energy technologies in SET-Plan in achieving the EU's energy targets by 2020 and 2050 [2]. The US Department of Energy (DOE) has also identified energy storage as a solution for grid stability, through the Energy Storage Systems Program (DOE OE/ESSP) for developing the energy storage technologies and systems [3]. A wide spectrum of studies address the technical features of electrical energy storage (EES) technologies. For instance, technical characteristics of different EES systems have been subject to study and review in a number of contributions [4–12]. There are other studies that have more thoroughly investigated operational features of certain EES technologies, including pumped hydroelectricity storage (PHS) [13], compressed air energy storage (CAES) [14], different types of batteries [15–17], flywheel energy storage [18,19], superconducting magnetic energy storage (SMES) [20], and supercapacitor energy storage (SCES) [21]. There is also a broad range of researches in modeling and optimization of EES in exemplary or real power systems [22–30]. The aforementioned and similar efforts have contributed to the better understanding of 1 The terms “electricity storage” and “electrical energy storage” are used interchangeably in the literature and are equal in this study, representing all the technologies that can store and then discharge back the electricity, with a reasonable response time.

technical characteristics, functional limitations, and possible operational strategies of EES systems. Yet, further research is required to address the barriers in large-scale deployment of EES systems in existing energy systems. In the absence of commercial, grid-scale adoption of the majority of EES technologies, their economic characteristics have remained obscure for energy system analyzers, power suppliers, grid operators, and policy makers. Moreover, cost analysis of the mature or commercial storage technologies, namely PHS and CAES, cannot be easily generalized as they are site-specific technologies [9,13,31]. According to different studies [27,32–34], this lack of adequate information regarding the economy of utility-scale EES systems is one of the major obstacles in the establishment of feasible business models, ownership structures, and required regulation strategies. In 2013, DOE announced four challenges in the widespread use of EES, of which cost-competiveness is to be addressed with focus on the life cycle costs (LCC) of EES systems [35]. To contribute in this regard, but not limited to that, this study provides an up-to-date, comprehensive, and comparative review of the available literature on cost analyses, capital cost data, and life cycle costs of different EES technologies. The focus is dedicated to recent publications considering their methodology, applied tools, and possible limitations. The LCC of EES systems is directly associated with the use case and its techno-economic specifications, e.g. charge/discharge cycles per day. Hence, the LCC is illustratively analyzed for three well-known applications; including bulk energy storage, transmission and distribution (T&D) support services, and frequency regulation. Since the cost data of EES systems are rather dispersed and varying in the literature, this study applies a robust uncertainty analysis in the determination of LCC of EES systems. This study is structured as follows. The main imperatives for the adoption of EES systems are briefly studied in Section 2. The cost analysis framework is established in Section 3, with describing the methodology for the representation of cost data. The cost elements of different EES technologies are discussed with respect to the recent publications in this field. Section 4 presents and discusses the results in three main parts: cost elements, total capital costs, and the LCC of EES systems. Conclusions are drawn in Section 5 supported with recommendations for the future work. This study focuses on stationary, utility-scale EES systems that are capable for supporting the grid at a reasonable response time. Indirect energy storage processes, smart electric vehicles, thermal

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

Nomenclature C BOP C cap C cap;a C DR C DR;a C FOM;a C LCC;a C O&M;a C PCS CR

cost of balance of plant (€/kW) total capital costs per unit of power rating (€/kW) annualized value of total capital costs (€/kW-yr) disposal and recycling costs (€/kW) annualized disposal and recycling costs (€/kW‐yr) fixed operational and maintenance costs (€/kW‐yr) annualized life cycle costs (€/kW‐yr) annualized operational and maintenance costs (€/kW‐ yr) cost of power conversion system (€/kW) replacement costs (€/kWh)

energy storage, and demand side management are excluded from this study. The EES technologies that are covered in this study include mechanical energy storage systems (PHS, CAES, and flywheel); secondary electrochemical batteries (lead–acid, sodium–sulfur (NaS), sodium–nickel chloride (ZEBRA), nickel–cadmium (Ni–Cd), and Li-ion); flow batteries (vanadium-redox flow battery (VRFB), zinc–bromine (Zn–Br), iron–chromium (Fe–Cr), polysulfide bromide battery (PSB), and metal–air batteries); electro-magnetic energy storage systems (SMES and SCES); and hydrogen-based energy storage systems. 2. Electricity storage for a flexible power system This section reviews the main imperatives for the adoption of EES technologies. Other alternatives for EES that can contribute to the grid stability and flexibility are also listed in Section 2.2. Comparing advantages and limitations of these alternatives is however beyond the scope of this study. 2.1. Imperatives of electricity storage 2.1.1. Meeting demand and reliability in grid's peak hours Electricity demand is inherently variable in momentarily, hourly, weekly, and seasonal time lags. It has been a tradition in power systems to maintain the production capacity large enough to meet the peak demands that occur just a few hours per year. This may result in oversized, inefficient, non-environmental, and uneconomical power systems. EES is an alternative to store the power in low demand time to be used later in the peak hours, diminishing the construction of extra power capacity. In some cases, for instance with high share of nuclear power, EES has been used to firm the production capacity to avoid part load operation or undesirable shutdowns, offering more economical baseload production [36–38]. Not only the generation capacity, transmission and distribution (T&D) systems are also constrained in peak hours. Since T&D networks are traditionally designed for one-way operation, they must be oversized to address the occasional peak hours. EES can relieve network contingency and reduce the risk of consequences of overloaded T&D network [39–41]. This can reduce large costs of grid management and reliability services. 2.1.2. Liberalized electricity markets The deregulation of electricity markets is another potential use case for EES systems by benefiting from price arbitrage, shifting electricity from low-demand periods to the peaks [32,40,42]. The profitability of EES in price arbitrage depends on the level of fluctuations in spot prices [43,44]. The use of EES in balancing markets and other deregulated ancillary services may stack the

C R;a C stor C V OM Ein Eout h i n r t T

ηsys

571

annualized replacement costs (€/kW‐yr) cost of storage section (€/kWh) variable operational and maintenance costs (€/kWh) input energy in one cycle (kWh) output energy in one cycle (kWh) discharge time (h) interest rate (  ) number of discharge cycles per year number of replacements replacement period (yr) lifetime (yr) overall efficiency of storage system (%)

benefits, resulting in more economic attractiveness [45,46]. Adopting an optimal strategy in charge/discharge scheduling and more improvements in price forecasting are two important parameters in increasing the revenues from EES in price arbitrage [47,48]. 2.1.3. Intermittent renewable energy National and regional energy policies endeavor to promote the use of renewable-based electricity (RES-E) to reduce carbon emissions and secure local power supply [1,49]. Inherent intermittency of variable RES, namely wind and solar PV, introduces new challenges in optimal operation of power systems, including frequency fluctuations, voltage flicker, and the cyclic operation of thermal power plants that are networked with high-level wind generation [50–52]. In the energy systems employing variable RES, other conventional generation plants are usually planned large enough to handle the maximum load without reliance on intermittent RES [53]. It is shown that EES is needed in relatively high shares of variable RES, even in the presence of an ideal, widely dispatched transmission system [54,55]. EES can be employed in different ways to enhance the use of RES-E. For instance, it can store extra, uncontrollable RES-E to be used at desirable time, eliminating power curtailment and oversized construction of power capacity [56]. With respect to wind power generation, EES can contribute in relieving the fluctuation suppression, low voltage ride through, and voltage control support, resulting in smooth power output [57–59]. The recent improvements in modeling and analysis of wind-storage systems have contributed in better understanding of the role of storage and its integration into the hybrid power systems [60–64]. It is further shown that EES favors other storage technologies, namely heat storage and gas storage, for large-scale wind integration [65]. EES can facilitate the use of RES for secondary applications, e.g. water desalination [66]. 2.1.4. Distributed generation and smart grid initiatives Distributed generation (DG) is one of the economic, reliable, and efficient ways for power supply in small scales, kW to a few MW. Not only DG dispatches the issue of power generation and transmission leading to further resource flexibility, it is known as a secure path for increasing the share of local RES-E [67,68]. By providing flexible power and reliability services, EES contributes in uninterruptible power supply (UPS) and overcoming voltage drops in these decentralized and inflexible power systems [69–72]. EES also facilitates the remote islands and microgrids with more RES integration, resulting in higher energy security and lower emissions [73]. In many smart grid schemes, which are seen to be a major step in achieving sustainable energy systems, EES is considered as an inherent solution [74–77].

572

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

2.2. Alternative solutions for increasing the flexibility of the power system While technical solutions are developing for power smoothing of variable RES at equipment level, e.g. controlling the kinetic energy of inertia, pitch angel, and DC link voltage in wind production [57], comprehensive solutions should deal with the problem from “system design” viewpoint. Demand side management or flexible power demand [78–80] and electric vehicles with smart charging [81–86] are two alternatives for shifting the peaks. Power system control, grid expansion, and more advanced methods in network management are other measures ensuring optimized power flow through the grid [87–91]. To provide the actors of power systems with proactive decisions in optimal planning and balancing of their energy systems, improved forecasting methods for production and consumption are crucial [92–96]. The optimal planning and control of power generation fleet and the adoption of combined heat and power (CHP) plants are other measures to increase the flexibility of the power system [97,98]. Electric-tothermal energy systems are other mechanism to capture the excess power to be utilized in the district heating networks [99]. Interconnected heat and power networks are one of the promising solutions for the integration of RES-E by applying energy storage as an inter-linkage [100–102].

3.1. General considerations In general, EES technologies include two main sections: power conversion system (PCS) and energy storage section. PCS is used to adjust the voltage, current, and other power characteristics of the storage based on the load requirements. PCS may consist of two separated units for charging and discharging with different characteristics. Energy storage section is the other part of EES that is designated to contain the storage medium, e.g. water reservoirs in PHS. Since PCS and energy storage units have inherent inefficiencies and losses, overall efficiency (AC-to-AC) of EES technologies is defined by Eq. (1), in which Eout and Ein are output and input electric energy, respectively. Eout ðkWh=kWhÞ Ein

To provide a uniform framework for cost comparison of different EES technologies, first, the scope of the cost analysis should be agreed. There are two main approaches in the literature in studying the cost of EES technologies: total capital cost (TCC) and LCC. In this study, both cost analyses are explained with respect to EES systems.

3.2.1. Total capital cost (TCC) TCC evaluates all costs that should be covered for the purchase, installation, and delivery of an EES unit, including costs of PCS, energy storage related costs, and balance of power (BOP) costs [104]. PCS costs of the EES system are typically explained per unit of power capacity (€/kW). Energy related costs include all the costs undertaken to build energy storage banks or reservoirs, expressed per unit of stored or delivered energy (€/kWh). In this manner, cost of PCS and storage device are decoupled to estimate the contribution of each part more explicitly in TCC calculations. For instance, turbo-machinery related costs of a PHS with a certain power capacity can be addressed without considering the construction cost of reservoirs, which itself is a complex function of volume and geological characteristics of the site. BOP costs can be expressed per unit of power (€/kW) or energy (€/kWh), or a certain fixed amount depending on the technology and application [104]. BOP includes costs for project engineering, grid connection interface and integration facilities (e.g. transformers), Table 1 Main items in initial capital cost analysis of EES [39,105].

3. EES technologies: characteristics and costs

ηsys ¼

3.2. Methodology in cost analysis of EES

TCC elements

Cost element

Example/Notes

Power conversion system (PCS)

Power interconnections

Converter, rectifier, turbine/pump (PHS)

Storage section

Containment vessels

Cabling and piping

Construction and excavation Balance of plant (BOP)

ð1Þ

Kaldellis et al. [103] proposes separated efficiency calculations for power output/input and energy output/input. In that case, overall efficiency is the product of the two items. Fig. 1 illustrates the main sections of a typical EES system and the associated losses.

a

Battery banks, air tanks (CAES) Cavern, reservoir

Project engineering Grid connection and system integration ESS isolation and protective Switches, DC brakes, and devices fuses Construction managementa Land and access Buildings and foundationa HVAC system Air-conditioning and vacuum pumps Monitoring and control Voltage and frequency systems control Shipment and installation Applicable to other costs sections

Those not included directly in PCS and storage related costs.

Fig. 1. Main sections of EES systems and energy losses.

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

construction management including cost of land and accessibility, in addition to other services and assets required that are not included in the scope of PCS and storage related costs [39]. A summary of general cost elements of TCC analysis is provided in Table 1. More technologyspecific cost components are addressed in the corresponding section of each EES technology. TCC can be calculated per unit of output power rating, presented as (C cap ) in Eq. (2). While C PCS , C BOP , and C stor represent unitary costs of PCS, BOP, and storage compartment (€/kWh), respectively, h is the charging/discharging time. C cap ¼ C PCS þ C BOP þ C stor nh ð€=kWÞ

ð2Þ

Considering the discharge time of the asset, C cap can be interchangeably presented per unit of power rating or storage capacity (€/kWh). According to [9], cost per kWh per cycle offers a better indicator for the cost evaluation of EES systems, as it also accounts for the life cycle numbers of EES. 3.2.2. Life cycle costs (LCC) From ownership perspective, LCC is more important indicator to evaluate and compare different EES systems. LCC accommodates all the expenses related to fixed operation and maintenance (O&M), variable O&M, replacement, disposal and recycling, in addition to TCC. LCC can be presented in levelized annual costs (€/kW yr), which is the yearly payment that the operator should maintain for all services of EES, including repayment of the loan and upfront of the capital costs. Schoenung and Hassenzahl [106] propose the term revenue requirement in (¢/kWh), which can be used by an energy supplier to calculate all the operating and ownership costs required for discharging each unit of stored energy in kWh. In some other studies, the price of electricity (and natural gas for CAES) is excluded from cost analysis of EES. For instance, Poonpun and Jewell [107] suggest a methodology to calculate the added cost by storing electricity. LCC calculations can be performed, first, by annualizing TCC ðC cap Þ, presented by (C cap;a ) in Eq. (3). Based on the present value of money the capital recovery factor (CRF) is calculated by applying Eq. (4), subject to the interest rate (i) during the lifetime (T) [108]. C cap;a ¼ TCC  CRF ð€=kW  yrÞ CRF ¼

ð3Þ

ið 1 þ iÞ T

ð4Þ

ð 1 þ iÞ T  1

Total annual O&M costs (C O&M;a ) can be expressed by adding annualized costs of fixed O&M (C FOM;a ), and variable O&M (C VOM ) multiplied by yearly operating hours, as presented in Eq. (5). C O&M;a ¼ C FOM;a þ C VOM  n  h

ð€=kW  yrÞ

ð5Þ

The price of electricity, as well as fuel costs (for CAES system) can be included in variable O&M costs or separately addressed. Number of discharge cycles per year (n) is one of the applicationbased parameters in cost calculations. To accommodate the replacement costs for replaceable EES systems, e.g. batteries, the future cost of replacement (C R ) in €/kWh and replacement period (t) in years should be known. Annualized replacement costs (C R;a ) can be calculated by using Eq. (6), given the number of replacements (r) during the application lifetime [106]. ! r X CR  h ð1 þ iÞ  kt  C R;a ¼ CRF  ð€=kW  yrÞ ð6Þ Zsys k¼1 Discharge time ðhÞ and overall efficiency (ηsys ) are given for one full cycle at the rated depth of discharge (DoD) of the batteries. All the losses during the charge, discharge, and other losses in storage part due to self-discharge and DoD should be reflected in the overall efficiency. Disposal and recycling costs ( C DR ) are other cost items that are usually neglected in the LCC analysis of EES in the

573

literature. Annualized disposal and recycling costs ( C DR;a ) can be calculated by applying interest rate (i) and lifetime of the plant (T), as explained in Eq. (7) i ð€=kW  yrÞ ð7Þ ð1 þiÞ T  1 The annualized LCC costs (ALCC) of EES systems, presented by C LCC;a in Eq. (8), is determined by stacking the previously discussed cost items. C DR;a ¼ C DR 

C LCC;a ¼ C cap;a þ C O&M;a þ C R;a þ C DR;a

ð€=kW  yrÞ

ð8Þ

The levelized cost of electricity (LCOE) delivered by EES systems can be then calculated by applying Eq. (9), knowing the annual operating hours of the system in question. LCOE ¼

ALCC C LCC;a ¼ ð€=kWhÞ yearly operating hours n  h

ð9Þ

If the cost of charging electricity would be deducted from the LCOE delivered by EES, the net levelized cost of storage (LCOS) itself can be realized (Eq. (10)). This way, the cost of employing EES can be calculated despite the price of electricity, which is inherently market-specific. LCOS ¼ LCOE 

price of charging power ð€=kWhÞ overall efficiency

ð10Þ

The majority of the references in the literature have reported the costs of EES based on TCC, for instance [8,9,109]. This may be based on the notion that LCC analysis cannot be adequately established in the absence of long-term utilization and field experiences for the majority of EES technologies. For instance, useful lifetime and replacement costs of emerging battery technologies in large-scale applications are unclear and different from different suppliers. Moreover, O&M costs are heavily depended on the operational regime of the EES system, e.g. charge/discharge cycles per day and DoD. 3.3. Methodology in review and collection of cost data Estimating the cost of EES systems includes levels of uncertainty and complexity. Except of some mature technologies, the use of large-scale EES systems is scarce and the economic performance of the existing sites is not widely reported in the literature. The cost data are scattered, from different times and power markets, and calculated/estimated based on different methods. Since most of the EES technologies are in the early stages of development and demonstration, their cost data cannot be conveniently scaled for the larger or smaller sizes. For those cost data that are merely reported based on the power rating of EES, the comparison and generalization may entail errors, as the storage size can be different for the same power rating. In this study, the authors have collected the cost data that are accompanied with required technical data, e.g. storage size, efficiency, and lifetime. Those reports not applicable to grid-scale services, e.g. small-scale batteries, are excluded from cost analysis. The cost figures are grouped with respect to the corresponding technical configuration of each EES system. For instance, the cost of underground and aboveground CAES is reported separately. Publications that provide cost estimations of EES systems comprise a wide range of different approaches and tools. While some publications [110,111] provide the cost data based on the inputs from vendors and manufacturers, others [12,112] present the cost data based on the review and update of different sources. While TCC are reported frequently, the LCC of EES systems is studied in a more limited number of publications, for example [14,105,107,111,113], resulting in fewer samples for the calculation of O&M and replacement costs. The data for the capital cost are combined from both reference groups resulting in more samples. The authors have endeavored to evaluate the source of the cost

574

Table 2 Different publications examined in this study for analyzing the capital cost and LCCs (life cycle costs) of electricity storage systemsa. Affiliation/ country

Reference ID

EES technologiesb

Abrams et al. (2013)

California Energy Commission/ USA

[114]

Lead–acid, Li-ion, flywheel

Akhil et al. (2013)

DOE-EPRI/ USA

[111]

[115]

Auer and Keil (2012)

DB Research/ Germany

Future cost estimation

Applicationc

Sensitivity analysis

Probabilistic model

Method for cost estimation

Notes

– Total capital cost (per unit of power and energy) – O&M costs

No

T&D support and investment deferral, RES integration

No

No

Data from manufacturer and review, analyzed by their model

Develops an analytical cost-benefit framework and tool for evaluating both cost and benefits of EES

PHS, CAES, flywheel, lead–acid, NaS, NaNiCl2, Li-ion, Zn–Br, VRFB

– Total capital cost (per unit of power and energy) – Cost of PSC and storage – Fixed and variable O&M

No

Bulk energy storage

No

No

Historical data from plant operators and developed modeling framework

Including cost analysis framework and modeling datasheets

PHS, CAES, hydrogen (methane)

– Total capital cost – Storage (operating) cost

No

Bulk energy storage

No

No

Model by DB Research and information from system operators

Analysis of power to gas routes. No details about modeling approach and cost elements

Cost items

Battke et al. (2013)

Swiss Federal [113] Institute of Technology (ETH)/ Switzerland

Lead–acid, NaS, Li-ion, VRFB

– Costs of PCS and storage – BOP cost – O&M costs

No

Energy time shift, T&D support, frequency regulation, userlevel storage

Yes

Yes

Data from review of other studies and manufacturers. LCC modeling based on uncertainty analysis (Monte Carlo method)

Taking into account the uncertainty in input data. Assuming fixed values for cost of PCS and O&M costs

Chen et al. (2009)

University of Leeds/UK

[9]

– Total capital cost (per unit of power, energy, and cycle)

No

Energy storage

No

No

Review of extensive references

No O&M costs

Connolly (2010)

University of Limerick/ Ireland

[116]

PHS, CAES, flywheel, lead–acid, NaS, NaNiCl2, Ni–Cd, Li-ion, Zn–Br, VRFB, PSB, SMES, SCES, hydrogen PHS, CAES, flywheel, lead–acid, NaS, Ni–Cd, Zn–Br, VRFB, PSB, SMES, SCES, hydrogen

– Cost of PCS and BOP No – Cost of storage part – Fixed and variable O&M

Bulk energy storage, T&D support, RES integration

No

No

Review of other references

A descriptive review of the costs

Danish Energy Agency (2012)

Energi styrelse/ Denmark

[117]

PHS, CAES, NaS, VRFB, hydrogen

– Capital cost of pump part – Total capital cost – Fixed O&M costs

Energy time shift ancillary services

No

No

Based on different sources and references

Grid related costs are not included

Díaz-González et al. (2012)

Energy Research Inst./Spain

[12]

PHS, CAES, flywheel, lead–acid, NaS, Ni–Cd, Liion, Zn–Br, VRFB, PSB, SMES, SCES, hydrogen PHS, CAES, flywheel, lead–acid, NaS, Li-ion, VRFB, SMES

No – Total capital cost (per unit of energy)

Energy storage related to wind integration

No

No

Review of other references

No O&M costs

– Total capital costs

No

Bulk energy storage

No

No

Review of other sources

The market potential of EES in the future

PHS, CAES, flywheel, lead–acid, NaS, Ni–Cd, Liion

– Capital cost of storage unit

No

Energy storage

No

No

From plant operators

Cost of PCS is not included in TCC. No O&M costs

– Total capital cost (per unit of power

No

No

No

From vendors and operators

Include project contingency, substation and interconnection costs

Energy Research Partnership ERP/UK (ERP) (2011)

[53]

Electricity Storage Association (2013)

ESA/USA

[5]

Electric Power Research Institute (EPRI) (2010)

EPRI/USA

[110]

Yes up to 2050

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

Author (publisher), date

Renewable integration/energy time shift

[118]

PHS, CAES, battery, flow battery

Yes – Total capital cost (per unit of power and energy) – Fixed O&M costs – Variable O&M costs

Transmissionconnected bulk energy storage

Yes

No

Using EPRI Energy Storage Valuation Tool (ESVT)

Inputs to the tool are provided by California Public Utility Commission (CPUC) with support from energy storage and utility stakeholders

European Association for EASE-EERA/ Storage of Energy (EASE) EU and the European Energy Research Alliance (EERA) (2013)

[119]

PHS, CAES, flywheel, lead–acid, NaS, Ni–Cd, Liion, VRFB, SCES, hydrogen

Yes up to – Total capital cost (per unit of energy) 2030 and other investments needed for pilot plants

Long- and shortterm energy storage

No

No

A study related to the future market, technical, economic, and societal aspects of EES

Prospects of EES in 2020 and 2030. Costs are not provided for all the technologies

Evans et al. (2012)

Macquarie University/ Australia

[4]

PHS, CAES, flywheel, lead–acid, NaS, Ni–Cd, ZEBRA, Li-ion, VRFB, Zn–Br, Fe–Cr, SMES, SCES

– Total capital cost (per unit of power and energy)

No

Energy storage

No

No

Review and summary of other references

No O&M costs

Hittinger et al. (2012)

Carnegie Mellon University/ USA

[120]

flywheel, NaS, Li-ion, SCES

– Cost of PCS and storage – Total capital cost – Fixed O&M costs

No

Frequency regulation, peak shaving, wind integration

Yes

No

Review of other sources

Prioritizes the effective parameters in LCC for each technology

Inage (2009)

International [121] Energy Agency (IEA)/ OECD Institute for [122] Energy and Transport/JRC European Commission National [14] Technical University of Athens/ Greece

PHS, CAES, flywheel, lead–acid, NaS, Ni–Cd, Liion, SMES, SCES

– Total capital cost (per unit of power)

No

Power quality, bulk No energy storage

No

Review of other sources and manufacturer reports

Develops an analysis for simulating storage in power systems

PHS, CAES, flywheel, lead–acid, NaS, Ni–Cd, ZEBRA, Li-ion, VRFB, Zn–Br, SMES, SCES, hydrogen CAES, hydrogen-fuel cell

– Total capital cost (per unit of power and energy)

No

Energy storage

No

No

Review of other references

– Total capital costs of No electrolysis, storage, and fuel cells (PEM) – O&M costs – Replacement costs

Energy storage in island areas

No

No

From technology developer and manufacturers

Kintner-Meyer et al. (2010)

Pacific [112] Northwest National Lab./ USA

PHS, NaS, Li-ion

Yes – Total capital cost (per unit of power) – Fixed O&M costs – Variable O&M costs

Energy arbitrage, balancing services

Yes

No

Review of other sources Including the combination of PHS and cost analysis with other technologies when framework developed by needed for balancing services authors

Kintner-Meyer et al. (2011)

Pacific [123] Northwest National Lab./ USA

PHS, NaS, Li-ion

– Cost of PCS and BOP No – Total capital costs – Fixed and variable O&M costs

Power balancing

No

No

Review of other sources

Lund and Salgi (2009)

Aalborg University/ Denmark

CAES (underground)

No – Total capital cost – Fixed O&M costs – Variable O&M costs

No

No

Typical costs for compressor and turbine, in combination with [125,126]

Costs are for one plant with different capacities for compressor and turbine

Poonpun and Jewell (2008)

Wichita State [107] University/ USA

PHS, flywheel, lead acid (VRLA), NaS, Zn–Br, VRFB

– Costs of PCS and storage – Cost of BOP – Cost of fixed O&M and replacement costs

Renewable integration, power regulating, generation capacity Bulk energy storage, T&D applications

Yes

No

Review of other sources and manufacturers

Sensitivity of the costs to the size and cycle number

Electric Power Research Institute (EPRI) (2013)

Joint Research Center (JRC) (2011)

Karellas and Tzouganatos (2013)

EPRI/USA

[124]

No

No life cycle costs for CAES

575

and energy) – Fixed O&M costs

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

PHS, CAES, flywheel, lead–acid, NaS, Li-ion, Zn–Br, VRFB

576

Table 2 (continued ) Affiliation/ country

Reference ID

EES technologiesb

Schoenung (2011a)

Sandia National Laboratories (DOE)/USA

[105]

PHS, CAES, flywheel, lead–acid, NaS, Ni–Cd, Liion, Zn–Br, VRFB, SCES

Schoenung (2011b)

Sandia National Laboratories (DOE)/USA

[128]

Sioshansi et al. (2011)

National RE lab., Uni of Ohio/USA

Steward et al. (2009)

Tan et al. (2013)

a b c

Future cost estimation

Applicationc

Sensitivity analysis

Probabilistic model

Method for cost estimation

– Capital cost of PCS, BOP and storage unit – Fixed O&M costs – Replacement costs

No

Bulk energy storage

Yes

No

In combination with Includes cost analysis for variable [104,106,127] provides speed PHS an analytical framework for LCC of energy storage

Hydrogen (also an update for other technologies)

– Capital cost of PCS, BOP and storage unit – Fixed O&M costs

No

Bulk energy Yes storage, distributed generation, and wind integration

No

With review of other Includes cost of storage tank and sources develops a underground storage for hydrogen. framework for analysis of Analysis of wind curtailment costs LCC

[45]

PHS, CAES

No – Total capital cost (per unit of power) – Variable O&M costs – Cost of upgrade

Arbitrage/capacity payment

Yes

No

Authors' model, other sources

Considers the effect of storage on the system prices

National RE Lab. (NERL), DOE/USA

[129]

PHS, CAES, NaS, Ni–Cd, VRFB, hydrogen (fuel cell and gas turbine)

– Capital cost of PCS Yes and storage unit – Fixed O&M costs – Variable O&M costs – Replacement costs

Bulk energy storage

Yes

No

Review of other sources, a holistic analysis of hydrogen-based systems

Includes different gas-to-power systems for hydrogen, namely fuel cells and gas turbines Examines tank and caverns for hydrogen storage

Shandong University/ China

[130]

Flywheel, lead–acid, NaS, SMES, SCES

No – Total capital cost (per unit of energy)

Power quality in microgrids

No

No

Review of other sources

No LCC costs No power rating

Cost items

Those publications that report the cost data of one single EES technology are not included in this table, but reported in each technology's subsection. Those that are among examined technologies in this study. Based on the terminology used in the reference.

Notes

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

Author (publisher), date

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

data, based on the credit of the report and reliability of the estimation method, as well as tools, models, and estimation procedure used in each reference. Those publications that are reviewed in this study for collecting the cost data of different EES systems are listed in Table 2. The main cost items that each publication has covered, as well as their methodology, sensitivity analysis, probability analysis based on the uncertainties, and the inclusion of estimations for the future costs are also compared. The efforts are made to avoid referencing those studies that are merely citing some other available contributions, without adding new information. In some cases that a secondary reference (e.g. a review article) adds more data to the original reference or reports some references that were not accessible, those new additions are individually examined (if possible) and the secondary reference is also listed in Table 2. In general, estimating the cost of EES involves both analysis and judgment. The cost figures that are relatively unreliable, outlier, and outdated are averted. For the sake of brevity, those publications with the cost data of a single project are not listed in Table 2, while taken into account in the analysis process. 3.4. EES technologies and related costs In this section, EES technologies are briefly reviewed with regard to their main techno-economic characteristics relevant for LCC analysis. The reported costs in this study are in their original values in the text body followed by their equivalent euro, if needed. However, the results and values presented in the tables and calculations are converted to euro and inflation-adjusted.2 3.4.1. Mechanical energy storage systems 3.4.1.1. Pumped hydroelectric storage (PHS). With a total installed capacity of over 125 GW, PHS3 represents 3% of the total installed electricity generation capacity in the world and 99% of the electricity storage capacity [110]. PHS is the only commercially-proven, largescale EES with no additional fuel needs. PHS is characterized for large power capacity (100–2000 MW), long lifetime, relatively long discharge time, and high efficiency. This has favored PHS over other alternatives for bulk energy storage, from daily energy time shift to seasonal storage. PHS's duty can be extended to include ancillary services, e.g. frequency regulation, in turbine mode. The use of variable speed pumping may introduce new capabilities and flexibility for ancillary services in charging phase as well (pump mode). In a study by EPRI [132], the profitability of PHS is examined by considering multiple services, including energy arbitrage, frequency regulation, and spinning and non-spinning reserve capacity. While installation of new PHS plants inclined in early 90s due to the environmental concerns and scarcity of favorable sites, new projects are again proposed after recent projections for the future development of RES and free electricity markets. For instance in the EU, there were 7.4 GW proposed projects for the period of 2009– 2018, increasing the total installed capacity of PHS by 20% in the region. In the US, the new project proposals were 30 GW by 2009 [45]. Despite its relatively low generating costs, PHS has shown to be capital intensive. Prior to the required capital investment; siting, environmental impacts, permitting, land demand, project contingency, and long-lead construction time of PHS plant are the main barriers in further adoption of PHS in the power systems. Recent technological advances and new tools might further ease the discovery of potential sites for PHS in the future [133]. 2 Annual inflation rate of 2.45% (based on the average rate of the EU for the period of 2003–2013), 1d ¼ 1.21€, and 1€¼ 1.34$ are applied throughout this study [131]. 3 In this study, PHS mainly refers to the pure PHS (also called off-stream or closed-loop), in which the output power is solely generated by returning the pumped water back to the lower reservoir, river, or sea.

577

To address the above-mentioned challenges, some of new PHS projects are proposed incorporating innovative solutions, e.g. making PHS reservoirs as wastewater treatment storage bodies [134], using piston-floated mechanism in an underground water-filled shaft [135], undersea PHS connected to offshore wind plants [136], water-filled balloon under pressure caused by sand [117], and underground PHS.4 In a study by Pickard [137], the feasibility of underground PHS is examined from technical, economic, and environmental aspects, indicating that excavation costs comprise 82% of the TCC for such systems. Based on [134], 25% of the permitted PHS projects in the US by 2010 are those with at least one underground reservoir. The construction and installation costs of PHS are estimated to be as twice as conventional hydropower plants with similar capacity, while operating costs are almost equal [117]. The cost of PCS may increase by 30–40% by applying variable speed pumping in PHS plants [138,139]. The main cost elements in the construction of a PHS plant are listed in Table 3. TCC is highly depended on topographical and geological characteristics of the examined site. The long lead time of PHS project may result in the intensification of initial cost estimations. For instance, despite the details in project contingency, the capital costs of a 1000 MW upgrade PHS project was announced 810 M€ in 2009 [36], but was modified later to 1700 M€ in 2014 [140]. The project contingency of PHS is deemed to be typically in the range of 10–15% and the accuracy of cost estimations may vary between 20 and þ25% [110]. There is no significant cost reduction forecast for the PCS section as it already consists of mature technologies. Regarding the storage reservoirs, the estimated costs vary more significantly, from 10 $/kWh (7.5 €/kWh) [107] to 169 $/kWh (126 €/kWh) [111]. In addition to the specific features of the site, the cost of storage depends on the plant size, 69 $/kWh (52 €/kWh) for a 14.4 GWh plant while 103 $/kWh (77 €/kWh) for 11.7 GWh storage capacity [111]. The results of this study show the cost of PCS of 513 €/kW and storage cost of 68 €/kWh, on average. More details of the results can be seen in Section 4.1 and Appendix A. 3.4.1.2. Compressed air energy storage (CAES). With two plants in operation, CAES is the second commercially proven, large-scale EES, after PHS. The configuration of CAES can be in the form of diabatic (D-CAES), with the need for additional fuel in the expansion process, or advanced adiabatic5 (AA-CAES). The overall efficiency of D-CAES is approximately 42%, as for Huntorf power plant in Germany (with 320 MW power rating) [115]. The efficiency can be improved by 12%point with adding a recuperator to recover the waste heat from the gas-fired expansion process as for McIntosh CAES plant in Alabama. The recent research focus is mainly dedicated to AA-CAES, in which the efficiency can reach to 70% with eliminating the supplemental gasfiring procedure. This process enhancement increases the cost of AA-CAES by 30–40% compared to the conventional counterparts [115]. Concerning the storage unit for compressed air, underground salt caverns, natural aquifers, and depleted natural gas reservoirs are respectively the most cost-efficient options for capacities up to several hundreds of megawatts (discharge time of 8–26 h) [110]. Aboveground CAES (typically a pressure vessel) may have capacities of 3–15 MW (with 2–4 h discharge time), at higher costs but easier project implementation compared to the underground type. The energy ratio6 of aboveground CAES is 0.79–0.81, which is typically higher than underground plants with 0.68–0.75, implying further need for additional fuel. CAES can be built in power

4 In underground PHS, the lower (or both) reservoir is located deep under the ground level. 5 In AA-CAES, the extra heat of air compression is recovered by a thermal storage unit to heat up the air during the expansion process. 6 Energy ratio is the ratio of energy consumption (from additional fuel like natural gas) to the output electric energy.

578

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

Table 3 Cost elements of pumped hydropower storage (PHS), based on [138]. Cost items Direct costsa

Indirect costsb

Other costs

Civil works (storage section) Power station costs Dams, spillways, water diversion, and embankments Intakes Surface penstocks Vertical shaft Horizontal power tunnels Steel-lined tunnel Electromechanical works (PCS) Transmission works Switchyard

Planning and investigation Environmental studies Licensing and permitting Preliminary and final design Quality assurance Construction management Administration

Transmission interconnections Infrastructure upgrade Initial charging energy (filling) Pumping Life cycle operation and maintenance Time cost of money Escalations Interest during construction Bank fees Depreciation

a b

Project contingency is considered 25% for PCS and civil structure, and 35% for underground works. Indirect costs vary between 15 and 30% of the total direct costs.

Table 4 Estimating the share of each cost item for underground CAES, based on the data from [117,141,142]. Cost item

Cavern leaching Construction and equipment Cushion gas for one cavern Total energy related capital costa Compressor part and related construction Turbine section Total PCS costsb Balance of plantc (BOP) and other costs Total capital costsd (TCC)

Total (%) 4 44 28 75 13 7 20 5 100

a

Based on the figures for a greenfield plant. Assuming equal power rating for turbine and compressor, otherwise the costs should be scaled accordingly. c Including connections, transformers, regulation, and instrument. d The costs are based large scale plant, assuming 4 h discharge time. b

ratings up to 2000 MW, with flexibility in input/output power, depending on the storage capacity. The costs of CAES can be conveniently divided into two main sections: storage- and power-related costs. Storage-related costs may be inexpensive if the cavern already exists. The costs of power trains are generally as for the conventional gas turbine plants, including turbine, compressor, and related ancillary equipment. The share of main capital cost items of an underground D-CAES plant are separately illustrated for power- and energy-related parts in Table 4. The LCC of a CAES plant is however highly depended on the additional fuel costs, related emission costs, and charging electricity prices. The optimal economic operation of CAES in different electricity markets has been subject to research in a broad range of studies [125,143–147]. CAES is capable to provide different services including energy arbitrage, reserve capacity, and wind integration based on the structure of the electricity market [148– 150]. CAES has a typical construction time of three years, 95% availability, and 99% reliability at starting time. The geologically appropriate formations in the service territory of the CAES operator is however one of the project challenges. Yet the conventional CAES relies on fossil fuels, an issue that can be resolved by the implementation of AA-CAES or the use of biomass-

derived gas or hydrogen instead [151]. The cost of storing air in geologic formations is compared with as for hydrogen in Table 7, in Section 3.4.4. Since all the equipment items used in CAES are established technologies, no significant reduction in cost is expected for the near future. As the key components and controls should be verified for the second-generation CAES, a process contingency of 10–15% is expected for such plants [110]. The project contingency related to the site geology of underground CAES plants is estimated to be 10%, which should be considered in project planning and implementation. The review of publications listed in Table 2 shows that the cost of storage may differ from 4 to 48 €/kWh, depending on the site and scale of the plant. Fixed O&M costs are estimated in the range of 14 €/kW‐yr in [117] and [125], while other references have estimations lower than 5 $/kW‐yr (3.7 €/kW‐yr) [110,111,118]. The results for the main cost elements of CAES are summarized in Section 4.1 and Appendix A. The average cost of PCS is in the range of 845 €/kW, while the storage costs varies between 40 for aboveground and 110 €/kWh for underground storage, on average.

3.4.1.3. Flywheel energy storage. Flywheels are among mature technologies that have been long used for different motorgenerator applications, e.g. as power buffer in electric vehicles. Flywheels are fast-responding, in the scale of milliseconds, with short duration discharge, in the scale of seconds to minutes, which make them suitable for power-related services, including UPS, frequency regulation, and integration of intermittent RES [110]. The most common application is to play role as a ride-through to switch between different sources of power. Having relatively high energy efficiency (typically higher than 85%), long life cycles (hundred thousand discharges and more than 15 yr) regardless the working temperature and depth of discharge (DoD), and lower environmental impacts, favors flywheels over the conventional batteries in similar applications [18]. With established manufacturing technology and suppliers, flywheels can be scaled up to tens of megawatts for grid-scale applications, e.g. a 20 MW frequency regulation plant in Stephentown, New York, by Beacon Power [152]. The use of flywheel in hybrid energy systems, namely wind-diesel, is also examined in the literature [18,153]. A wind-hydrogen plant in Utsira in Norway is equipped with a 200 kW flywheel EES that can store 5 kWh electric energy for a few seconds. Flywheels are also employed in the sites with the need for 24/7 power availability, e.g. data centers, to eliminate the power outages of several seconds or bridging to the back-up systems [154]. Based on the spinning

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

speed, flywheels can be divided to low-speed (less than 6000 rpm) and high-speed configurations, of which the latter is equipped with more advanced material and machinery to increase the overall efficiency. The cost of high-speed flywheels, manufactured with magnetic bearing, can be 5 times higher than low-speed types. As flywheel is targeted for power related applications, it is more accurate to calculate the costs based on the unit of power (€/kW) rather than energy (€/kWh). In a study by DOE [155], the use of flywheel in combination with a valve-regulated lead–acid (VRLA) battery is examined for UPS services. The costs of the two systems are compared and concluded that flywheel has 57% lower LCC compared to VRLA batteries, over a 20-yr lifetime for the same service. Flywheel's TCC (purchasing and installation) is however 42% higher than VRLA batteries. The use of electromechanical flywheels compared to electrochemical batteries entails less environmental hazards and safety issues related to the batteries, including spill containment, detection of hydrogen, eyewash stations, and ventilation requirements [18]. A summary of the main LCC items related to flywheels are presented in Table 5. While the goal is not to directly compare the two estimations, the changes in the results due to technical and financial assumptions seem to be remarkable. It should be noted that these costs vary between different manufacturers and for different applications. More particularly, the levelized costs of stored energy may vary significantly based on the discharge time of the flywheel [106]. Besides the cost of stored energy, O&M costs show wider variations in the available estimations, while cost of PCS stands around 300 €/kW. The replacement costs of high-speed flywheel vary between 85 and 215 €/kW, given the period of 10 yr for the replacement. For further details about the costs, see Section 4.1 and Appendix A.

3.4.2. Electrochemical battery energy storage Rechargeable (secondary) battery energy storage (BES) comprises a wide range of technologies based on the material used in electrodes and electrolytes, and the functioning system. Since the purpose of this study is not to discuss the working principles and pure technical features of each BES technology, further details in this respect can be seen in [9,15,16,158–161]. BES systems covered in this Section are both conventional (e.g. lead–acid, NaS, Ni–Ca) and flow batteries (e.g. VRFB, Zn–Br) with focus on their stationary, utility-scale applications. Table 5 A sample for main cost elements of flywheel energy storage, the range of 100300 kW (based on the data from [156,157]). Cost item

Purchase cost (€/kW) Installation cost (€/kW) Total capital costs (TCC) (€/kW) Bearing replacement costc (€/kW-yr) Vacuum pump replacementd (€/kW-yr) Fixed O&M (€/kW-yr) Variable O&M (€/kW-yr) Stand-by power consumption cost (€/kW-yr) Make-up energy (€/kW-yr) Annualized life cycle costs (ALCC) (€/kW-yr)

Application UPSa

Area regulationb

290 19 309 1-3 0.7 5 – 0.5 – 38

– – 1255 – – 11.6 10.1 – 5.7 257

a Costs are for uninterruptible power supply (UPS) application, 20 yr lifetime, and 6% discount rate. b Costs are for area regulation application, 10 yr lifetime, 10% discount rate, and present worth factor 0.2. c Considering 5 yr replacement period. d Considering 7 yr replacement period.

579

3.4.2.1. Lead–acid battery. Lead–acid batteries are the oldest form of BES with established record in a wide range of applications. Lead–acid batteries have been a common choice in microgrids or isolated power systems, for power quality, UPS, and spinning reserve applications [9]. Their limited life cycles ( 2500), short discharge time, and low energy density ( 50 Wh/kg) make them not favorable choice for energy time-shift purposes. However, large lead–acid batteries with discharge time of hours are in operation, e.g. in Chino project, California, with a power capacity of 10 MW and 4 h discharge time [158]. New advances in lead–acid battery's configuration has offered improved characteristic for the utility scale applications. Advanced VRLA (valve-regulated lead acid) batteries equipped with carbon-featured electrodes can reach 10 times longer life cycles compared to the conventional ones [16]. Lead–acid batteries are among low cost EES systems. While lead prices are directly influential in final prices, the cost varies widely from different suppliers, depending on the configuration design, duty cycles, and design lifetime [110]. Moreover, battery's temperature should be kept in limits specified by the supplier ( 5 to þ40 1C) otherwise it suffers from significant degradation in expected lifetime, entailing extra operating costs [155]. The power related and BOP costs of VRLA are estimated to be in the same range as flooded cell (conventional) lead–acid batteries, but the storage compartment has 25–35% higher costs [162]. The main cost elements of advanced lead– acid batteries are depicted in Section 4.1 and Appendix A. It should be noted that the costs are related to those systems suitable for bulk energy storage or T&D support services, with discharge time of approximately 4 h. The costs for frequency regulation services are mainly similar to those presented in [110]. The range for the estimation of fixed O&M costs is between 3.2 and 13 €/kW‐yr, and PCS costs are expected as of 322–400 €/kW. 3.4.2.2. Sodium–sulfur (NaS) and sodium–nickel–chloride batteries (NaNiCl2). NaS batteries have been developed by NGK Insulators and Tokyo Electric Power since 1987. The batteries are one of the most proven electrochemical storage technologies in MW scale, with projected total installations of 606 MW by 2012 [110]. NaS batteries have shown capabilities in power quality applications and power time shift, with relatively high overall efficiency (75– 85%), 2500–4500 life cycles, expected lifetime of 15 yr, and discharge time up to 7 h [12,163]. The power rating is scalable, promising more utility-scale demonstrations in the future. A largescale, 300 MW project was to be delivered to Abu Dhabi Water & Electricity Authority in UAE [16], but later moderated to 60 MW. As of Jan 2014, the largest NaS project (by NGK Insulators) is a 70 MW (490 MWh) battery ordered by the Italian transmission system operator (TSO), Terna S.p.A [163]. With regard to the initial price of this contract, a total capital cost of 1430 €/kW (204 €/kWh) is estimated for NaS batteries, in projects with tens of megawatts scale. NaS battery projects are reported to bear a project contingency of 1–5% depended on the site conditions [110]. The cost data of NaS batteries show a relatively higher consistency in the literature as they are mainly supplied by one manufacturer. Based on the review performed in this study, the levelized costs of PCS and storage section are on average 366 €/kW and 298 €/kWh, respectively. The main cost elements for NaS batteries are summarized in Section 4.1 and Appendix A. Similar to NaS, sodium–nickel–chloride batteries, known as ZEBRA (Zero Emission Battery Research), are high-temperature batteries (270–350 1C), in which nickel chloride is employed as the cathode instead of sulfur [164]. They have been commercially available since about 1995 and have been successfully employed in several mobile applications. The focus of research is nowadays in developing advanced ZEBRA batteries, with enhanced energy densities for load-leveling and integration of variable RES. The cost of ZEBRA batteries are not well established, but estimated by several vendors for MW-scale applications. For instance, the capital cost may vary

580

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

between 2800 and 4300 $/kW (2100–3200 €/kW) for projected installed power rating of 50 MW scale [111]. A typical lifetime of 8–10 yr and 2600 life cycles are considered for this type of batteries. This domain of lifetime favors ZEBRA batteries for distributed power applications. The main cost items of ZEBRA batteries are presented in Section 4.1 and Appendix A (PCS cost of 470 €/kW and storage cost of 510 €/kWh, on average). In general, more research is needed to address the energy density and environmental issues of Na-ion batteries for their large-scale adoption in the grid-scale services [165,166].

3.4.2.3. Nickel–cadmium battery (Ni–Cd). Ni–Cd7 batteries are among the oldest BES technologies that are further developed since 1990s. They offer relatively high energy density (55–75 Wh/kg), low maintenance need, and life cycles between 2000 and 2500. The life cycle is highly depended on DoD so that it can reach 50,000 cycles in 10% DoD [12]. Ni–Cd batteries have served in different applications from power quality and emergency reserve to telecommunication and portable services. The world's largest Ni–Cd battery, and the US largest BES, has been in operation since 2003 in Fairbanks, Alaska (USA), with power rating of 27 MW (15 min discharge time) capable to boost to 40 MW (7 min) [16]. The major drawbacks of Ni–Cd are relatively high capital costs (see Section 4.1 and Appendix A) and the problems in disposal handling associated with the toxicity of the heavy metals (Ni and Cd) [7]. It is also reported that the memory effect, susceptibility to overcharging, and relatively low efficiency can be other limiting barriers subject to more improvements in the future [160].

3.4.2.4. Lithium-ion battery (Li‐ion). The first commercial Li-ion batteries were produced in early 1990s. They were first targeted for portable applications but were employed in grid-scale, stationary applications as well. High energy density ( 200 Wh/kg), long lifetime ( 10,000 cycles), and relatively high efficiency (0.85–0.90) have offered sufficient motives for the development of these batteries [28]. The largest Li-ion EES serves at the Laurel Mountain Wind Farm, in Moraine, Ohio supplied by AES Energy Storage [167]. The project is an advanced Li-ion storage with the power rating of 32 MW (8 MWh storage), targeted for enhancing the wind power plant's capability in providing capacity services and grid stability for the PJM electricity market. In general, the future perspective seems to be promising for Li-ion batteries in grid-scale applications as the final price is declining and the functionality is ever improving by optimizing manufacturing costs, extending the lifetime, using new materials, and improving the safety parameters [168]. It is estimated that the share of Li-ion batteries in the market reaches 35 GWh by 2015, providing frequency regulation and power quality services [110]. The results of cost analysis show relatively consistent figures for PCS costs with 463 €/kW, including BOP cost of 80 €/kW on average (see Section 4.1 and Appendix A for full details).

3.4.2.5. Flow batteries. Flow batteries store energy in the electrolyte solutions, opposite to the conventional BES in which the electrodes are responsible for this task. Hence, the ratings of power and energy can be designed independently: energy capacity is determined by the quantity of electrolyte stored in external tanks while power rating is designed based on the active area of the cell compartment. It makes flow batteries favorable for both energy and power related storage 7 Ni–Cd batteries are equipped with nickel hydroxide and cadmium hydroxide as positive and negative electrode plates, respectively, and an alkaline-based electrolyte [9].

applications, maintaining a high rate of discharge time up to 10 h [9,168]. With relatively low energy density (10–75 Wh/kg) [109], limited operating temperature range (10–35 1C) [168], and high capital costs, VRFB8 are yet to be commercialized for grid-scale applications. However, their flexibility in discharge time, power rating, and energy capacity in addition to their long lifetime (þ13,000) [117], motivates further research for developing VRFBs. The largest reported VRFB is a 3 MW (16 min discharge time) unit at Sumitomo's Densetsu Office, in Osaka, Japan, targeted for peak shaving [16]. The breakdown of the cost elements of VRFB systems is presented in Table 6, indicating en equal share for PCS and storage costs. For MW-scale projects, a process contingency of 5–8% and project contingency of 10–15% may increase the TCC. However, it is estimated that the increase in the cost of flow batteries would be less size-dependent compared to the conventional counterparts. The R&D on new configurations and materials has attracted a wide attention and plays a key role in the cost reduction and performance enhancement of the flow batteries [169–171]. In [172], the feasibility and economic features of deep eutectic solvents is compared to liquid-ion electrolytes. The results show that while deep eutectic solvents imply higher capital costs, they are rather environmentally benign and biodegradable, offering a widely available raw material with marginal environmental impacts. The results of this study show that the cost of PCS is in the range of 424–527 €/kW for VRFB systems. The cost of other types of flow batteries are also examined in this study, namely zinc–bromine (Zn–Br), iron–chromium (Fe–Cr) and polysulfide–bromide (PSB). Despite having relatively low efficiency (0.6–0.65), Zn–Br batteries offer higher DoD, approximately fully discharged [110]. For Zn–Br batteries the recent estimations show the cost of PCS in the range of 151–595 €/kW, with the average of 444 €/kW. The storage cost and replacement costs (after 15 yr) are approximately 195 €/kWh, for bulk energy storage and T&D applications with 365–500 cycles per year. Fe–Cr flow batteries are in the first stages of R&D. In [111], the cost of Fe–Cr batteries is estimated based on the data provided by one manufacturer. The estimations imply relatively lower capital costs for this configuration compared to other types of flow batteries, approximately 1100–1360 €/kW for MW-scale applications. The cost of PCS for this system is 360 €/kW which is lower than other flow batteries (more details in Section 4.1 and Appendix A). Flow batteries may have lower costs in larger scales [174] and in long discharge times (several hours) [175], compared to other battery types. Flow batteries have also shown to have the minimum carbonequivalent emissions during their life cycle, compared to lead–acid, flywheel, and superconductors [176]. For further details about historical trends in research and development of different flow battery technologies; chemistry of their electrolyte, electrode and membrane components; and different materials, configurations, and applications of flow batteries refer to [175,177,178]. 3.4.3. Electric and magnetic energy storage 3.4.3.1. Capacitors and supercapacitors. Capacitors9 are the most direct method to store electricity, offering fast-response with life cycles of tens of thousands and very high efficiency. The low energy density of conventional capacitors has led the research on SCES (supercapacitor energy storage), with electrochemical double layer capacitors (DLC) and pseudocapacitors as the main configurations [179]. The main drawback of SCES relies on their short storage duration, low energy density, and high self-discharge loss. They are 8 In VRFB, vanadium redox couples, V2 þ /V3 þ in the negative and V4 þ /V5 þ in the positive half-cells, produce power by exchanging H þ through a hydrogen-ion permeable polymer membrane [160]. 9 A capacitor has two metal plates separated by a dielectric, which is itself a non-conducting medium. If one of the plates is charged with DC electricity the other plate is induced a charge with an opposite sign [9].

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

Table 6 Breakdown of the internal cost of a sample VRFB system, data extracted from [173]. VRFB's cost item

Share of total (%)a

V2O5 (solute) Electrolyte manufacture Tanks Total storage costs Activated carbon‐felt electrode Bipolar current collector Frame and associated components Ion‐exchange membrane Electrolyte storage tanks (  2) Pumps (  2) Control system Total flow cell costs (equivalent to PCS) Total internal capital costs

28 10 7 45 7 2 19 2 8 7 11 55 100

a The values are given for a system with 1.75 m2/kW electrode area, and 6.0 kg/ kWh required amount of V2O5.

mainly employed in power quality services, including ride-through and bridging [9]. Ongoing research on materials, e.g. nanostructured materials [180], can promote SCES in grid-scale applications. The capital cost of capacitors is reported to be in the range of 1100–1500 €/kW [121]. Supercapacitor are also good candidates to smooth the short-term high frequency fluctuations caused by swell effects in the marine current systems, especially for one generator plants [181]. More details about the performance and LCC of asymmetric lead– carbon capacitors can be seen in [127]. 3.4.3.2. Superconducting magnetic energy storage (SMES). A SMES10 system is capable to store energy in a magnetic field so that it can be instantaneously discharged back, offering electricity storage in a pure electrical format. SMES systems are characterized for a very high energy storage efficiency (  97%), fast response (few milliseconds), and long life cycles (100,000) [12]. These features make SMES system capable in providing power quality services for industrial consumers, carryover energy during the voltage sags and momentary power outages, and frequency regulation [59,182]. The typical power rating is in kW to several MW but research is carried out to improve the power rating. The main challenges in the utilization of SMES rely on high capital cost and environmental considerations related to the strong magnetic fields. Typical capital cost of SMES is reported in the range of 150–250 €/kW of power rating for power quality applications [9]. Based on [121], the present value of the LCC of a 100 MW scale SMES can be approximately 1500 €/kW of installed capacity, assuming 30 yr lifetime and grid stabilization services. In [183], the energy costs of two different configurations are compared (solenoid and toroid), concluding that the cost of superconductors may reduce by 85% with increasing the storage capacity from kWh to MWh scale. 3.4.4. Power to gas energy storage technologies It is expected that with the increase in the share of intermittent RES in power systems, the need for long-term EES systems becomes more urgent. Gas storage systems, in the form of hydrogen or synthesized methane, have high energy density and marginal losses in long-term storage. Gas storage systems are also compatible with the existing infrastructure for natural gas storage and transmission, as well as conversion technologies. They can be delivered in different types at consumption terminals, including power, heat, and transportation fuel. In principle, hydrogen can be 10 A SMES system is a device that stores energy in the magnetic field, by converting AC to DC current flowing through a superconducting wire in a large magnet [20].

581

produced from water and then further converted to synthetic methane by reacting with CO2. The process of power-to-gas conversion, energy storage, and final energy utilization by means of gas storage systems is illustrated in Fig. 2. Gas storage systems offer the possibility for integrating the process of carbon capture and storage (CCS) in an efficient energy storage and power production system. In addition to power-to-gas storage systems based on electrolysis, biogas production and storage can be considered as a measure to increase both the flexibility of the power system and share of bioenergy [184]. 3.4.4.1. Hydrogen storage. Hydrogen energy storage is the process of production, storage, and re-electrification of hydrogen gas. Hydrogen is usually produced by electrolysis and can be stored in underground caverns, tanks, and gas pipelines. Hydrogen can be stored in the form of pressurized gas, liquefied hydrogen in cryogenic tanks, metal hydride or in chemical compounds (ammonia, methanol, etc.) [117]. The existing natural gas networks are capable to store additional hydrogen up to 5% of their capacity, without significant degradation in the performance [185]. This way, energy can be transmitted and delivered in higher capacities (4.5 times more than high-voltage transmission lines) while lower transmission losses (1% in gas pipelines while 4% in power transmission lines) [115]. Energy density of hydrogen (can be pressurized and stored in 200 bar) is as high as Li-ion batteries, which implies the need for significantly smaller storage reservoirs compared to PHS and CAES. The stored hydrogen can be converted back to the electricity by fuel cells (compatible for mobile applications), gas-fired turbines, or gas-fired engines [186]. Today, the relatively low overall efficiency and huge capital costs are two major barriers in commercial implementation of hydrogen-based storage in gridscale applications. Since the cost of each power production method varies along with their advantages and requirements, it is not an easy task to establish consistent cost estimation for hydrogen-based systems. The capital cost of electrolysis itself varies among different configurations, projected 590 €/kW for solid-oxide electrolysis plants in 2020 [117]. These plants have power-to-hydrogen efficiency of 98% and net electrolysis efficiency of 83%, due to their heat demand. For alkaline electrolysis, the capital costs are in the range of 1400 €/kW while maintaining 43–66% power-to-hydrogen efficiency. Polymer electrolyte membrane (PEM) electrolyzer cell offers the power-to-hydrogen efficiency of 68–72% and net efficiency of 88% due to heat production. The cost of storage part heavily depends on the use of available infrastructure, for example gas employing caverns or gas pipelines, or building new facilities. In general, it is estimated that the cost of aboveground storage section would be around 15 $/kWh (11 €/kWh) [128], while for the underground caverns ranging from 0.002 to 49 $/kWh (0.002–0.41 €/kWh) [129]. Table 7 compares the cost of geologic storage caverns for CAES and hydrogen per unit of delivered energy. In [117], the cost of a MW-scale hydrogen plant, comprising cavern storage and gas internal combustion engine, is estimated as of 3055 €/kW with 35% overall efficiency (AC-to-AC). In [14], the capital costs, O&M costs, and replacement cost of hydrogen systems including electrolyzer (700 kW), storage tank, and PEM fuel cells (500 kW), is compared with a CAES plant. It is concluded that hydrogen plant has higher O&M costs but lower capital costs, 2.97 M€ compared to 5.5 M€ for CAES (500 kWh storage capacity). The integration of hydrogen storage in different power systems and the associated advantages and drawbacks has also been subject for different studies [187]. In [188], the total electricity generation cost using hydrogen storage (solid and gaseous) is estimated to be lower than that for the system without storage back up. However, the economic feasibility of the use of hydrogen

582

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

Fig. 2. The pathway for power to gas energy storage (hydrogen and methane), conversion, and final utilization in different sectors [119].

Table 7 The cost of storing air (in CAES) and hydrogen in different underground formations (data from [128], adjusted to 2014-€).

which an overview of different EES technologies and comparison between their main technical characteristics are maintained.

Cavern formation type

Air storage (€/kWh)

Hydrogen storage (€/kWh)

4. Results and discussion

Natural porous rock formations from depleted gas or oil sites Solution-mined salt caverns Dry-mined salt caverns Abandoned limestone or coal mines Geologic storage of hydrogen Rock caverns from excavation of impervious rock formations

0.10

0.002

4.1. Results of the review for individual cost items

1.01 9.71 9.71 N/A 29.55

0.02 0.14 0.14 0.25 0.41

This Section reports the main individual cost items of the EES technologies comparatively. While the ultimate goal is not to prioritize the EES systems based on their associated costs, this comparison provides the reader with the scale of magnitude and variation in cost of different technologies. The data are collected and examined from the references listed in Table 2, as well as other individual projects presented under corresponding subsections for each EES system in Section 3. The results will be presented by statistical methods to account for the variability and disparity of the cost estimations in the literature, in addition to the magnitude of the most likely values. Due to the limited number of reported data and relatively high disparity, the use of parametric methods, e.g. probability density functions, might entail relatively significant parametric assumptions and unrepresentative values affected by outliers. The results usually show a deviation from normal distributions and the data are usually skewed to the near end of each range. Hence, the fivepoint descriptive (non-parametric) method is employed to report the results. The average values are the median of each range and the interquartile range (IQR), or so-called middle-fifty range, represents the spectrum that contains 50% of the reported cost data. This way, the reported average (median) is unaffected by extreme outliers [194]. The outliers are either lower than three IQR minus the first quartile or higher than three IOR plus the third quartile, and excluded from the reported ranges. The cost of PCS for different EES technologies can be compared in Fig. 3. The range of power rating for each EES system is shown in Appendix B, while the more detailed costs can be found in each technology's corresponding table in the Appendix A. As the results reveals, there is a wide range of variability in the PCS cost of some battery technologies (e.g. ZEBRA) in the literature. More importantly, the cost of power electronics of commercial and mature technologies (PHS and CAES) is also rather inconsistent in the reviewed publications. This cost variability intensifies the uncertainty in investigation of LCC, even for mature and wellestablished EES systems. CAES systems show the most expensive PCS systems averaging 845 €/kW. The lower variation in the cost data of PCS for aboveground CAES might be the result of fewer sources of estimation for this configuration in the literature. The cost of storage compartment of MW-scale EES technologies applicable for energy-related applications is illustrated in Fig. 4.

storage is highly depended on other value streams applicable in the future [189].

3.4.4.2. Methane synthesis and storage. Hydrogen can be further converted to methane by reacting with CO2, increasing the stability and energy density of the stored media, from 360 to 1200 kWh/m3 at 200 bar [115]. The existing natural gas infrastructure can conveniently store, transmit, and convert the synthesized methane back to the final form of energy use. With the losses of 18–25% during the methane production process, it is estimated that the AC-to-AC efficiency of this storage process would be only 33–40% with today's technologies [115]. The cost of methane synthesis from hydrogen is around 1000 €/kW [190].

3.4.5. Other electricity storage technologies There are other EES systems under R&D that are not studied in this contribution due to the lack of information about their costs and functionality, including nano-supercapacitors, hydrogen–bromine flow batteries, advanced Li-ion batteries, novel mechanical energy storage systems (based on gravity forces). While not examining the costs in details, the capital cost of metal–air batteries (Zn–air) are, however, covered in this study. The technical features that are important and influential in cost analysis are presented in the Appendix B. It should be noted that the goal of this study is not to analyze the technical features of EES as there are adequate contributions in this regard in the existing literature. However, to provide consistency between the cost data and the corresponding technical data, the same references that are examined in this study for the cost data of EES systems are reviewed to extract the technical characteristics of each EES technology. For further details about each technology, readers can refer to, for example [4,6–9,12,110,111,115,117,191–193], in

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

583

Fig. 3. Cost of power electronics for EES technologies, including BOP costs where applicable, based on the review of the references listed in Table 2 (the average is illustrated above each range). More details including PCS costs of hydrogen-based systems are shown in the Appendix A.

Fig. 4. Cost of storage part (e.g. tank, reservoir, or electrolyte compartment) for different EES systems after reviewing the publications presented in Table 2 (the average is shown above each bar). Note: costs are given for the typical size of each technology (see Appendix A and B for further details).

The calculated levelized costs are for EES systems with different discharge times and should be considered in comparisons. The results are reported for typical size of each technology, e.g. 8 h for PHS while 6–7 h for NaS batteries. For further details about the discharge time of each EES system, see the Appendix B. The results reveals a relatively high variabilty in the cost estimations of the storage part in battery technologies. These cost sparity even dominate the price gap among different technologies. The variability of the storage costs is, however, considerably lower for the mechanical EES systems, intrducing underground CAES as the cheapest one (40 €/kWh). Considering the fact that the cost of storage reservoirs of PHS and underground CAES is highly depended on the geography and geology of the site, yet the uncertainty in the associated costs is less than batteries. This can be attributed to the limited experience in the production and

deployment of large-scale batteries for utility-level applications, resulting in scattered and inconsistent cost data for these technologies. Fe–Cr battery as an exception has more consistent results, which can be the result of acquistion of the costs from one available source. The cost of storage section for those EES technologies that are employed in power-related applications, namely flywheel, SCES, and SMES, is not compared in Fig. 4. Fixed O&M costs of EES systems are illustrated in Fig. 5. The costs are mostly for energy-related and T&D support applications (except for flywheel). As these applications demand a working regime of one full cycle per day, the costs migth be higher for other applications with more frequent chrge/discharge requirements, e. g. frequency regulation services. For further details for numeric values, see the corresponding subsection for each technoogy in the Appendix A. The variable O&M costs of different EES systems are

584

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

Fig. 5. Cost of fixed O&M for EES technologies based on the review of the references listed in Table 2 (the average is illustrated above each range). For further details and also for variable O&M costs see Appendix A.

Fig. 6. Replacement costs of battery storage systems per unit of stored energy at rated DoD based on the references presented in Table 2 (average values are illustrated above each range).

illustrated in Appendix A. The costs of purchasing power in charging phase is not included in the estimations, as it directly depends on the market and application of the asset. The range of natural gas prices in the examined literature varies between 8 and 20 €/MWh, while the emission costs were between 18 and 22 €/ton CO2. Since the costs are stacked in some literatures, it was impossible to extract the fuel and emission costs separately. The replacement costs of batteries are compared in Fig. 6. The lower range of replacement costs for Fe–Cr batteries is attributed to the one supplier price range. It is very important to notice the replacement time of each battery, reported in the corresponding subsection in the Appendix A. For instance, while the replacement cost of ZEBRA batteries is on average 182 €/kWh compared to 195

€/kWh for Zn–Br batteries, the replacement time is around 8 and 15 yr, respectively.

4.2. Total capital cost (TCC) of EES systems In this section, the TCC of EES systems are comparatively presented per unit of power rating and storage capacity. For estimating the TCC, a wider range of studies were available, compared to the segregated cost elements discussed previously. In other words, TCC can be calculated whether from the costs of PCS, BOP, and storage part by using Eq. (2)[105, 107], or be directly obtained from the manufacturer/literature, per unit of installed

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

585

Fig. 7. Total capital cost (TCC) of large-scale EES systems per unit of nominal power rating based on the review of the references listed in Table 2 and applying Eq. (2) (the average values above each bar and further details including storage size in Appendix A and C).

power [112], stored energy [12], or both [9,110]. The TCC of different grid-scale EES systems per unit of power rating are illustrated in Fig. 7. The costs are presented for the typical storage size of each system, which is not primarily identical among different technologies. For example, the TCC of PHS and CAES is equivalent to 8-hour discharge time, while the TCC of lead–acid and VRFB are for 4 h discharge. The number of technologies examined in this Section is more than previous analyses, as SMES, SCES, and Zn–air are also included. The results indicate that underground CAES offers the lowest capital costs (893 €/kW) for bulk energy storage systems, followed by Ni–Cd and Fe–Cr batteries, 1092 and 1130 €/kW, respectively. For power quality applications, SCES and SMES show the lower costs, 229 and 218 €/kW, respectively. However, it should be noted that the range of available literature for the latter two technologies is rather limited and cannot be conveniently generalized to gridscale applications. ZEBRA batteries are the most expensive EES systems with a TCC of 3370 €/kW. The range of the TCC of batteries is rather wide, which implies more uncertainty in the estimation of the costs for utility-scale systems. Comparing the TCC of hydrogen based technologies show that hydrogen gas turbines (GT) would cost rather half the hydrogen-fuel cells (FC) systems. The TCC of ESS systems based on the unit of storage capacity is also presented in Appendix C. 4.3. Life cycle costs (LCC) of EES systems and uncertainty analysis The LCC of EES technologies can be determined by applying the framework presented in Section 3.2.2, having TCC, fixed and variable O&M costs, replacement costs, and disposal/recycling costs11, if applicable. The LCC of EES systems is directly depended 11 Due to the lack of data for disposal and recycling costs, they are not examined in this study.

on the characteristics of the service (e.g. number of cycles per year), the power market (e.g. interest rate and price of power), and technological features (e.g. DoD and replacement time for batteries). The LCC of EES systems can be presented in different ways, including discounted cost items to the corresponding net present value (€/kW) [129]; annualized costs throughout the lifetime of the application (€/kW‐yr)[105]; LCOE discharged by EES (€/kWh) [111]; or LCOS (€/kWh) [107]. LCC analysis is conducted for three main applications: bulk energy storage (energy arbitrage), T&D support, and frequency regulation. This way, in addition to exemplifying the framework and cost data reviewed in this study, the effect of uncertainties in LCC analysis of EES systems are also examined. The main features of three application categories are summarized in Table 8 and economic assumptions are shown in Table 9. Different studies have investigated the LCC of EES systems in different markets and for various applications [128,162]. The sensitivity of the LCC of EES systems to the discharge time, capital cost, overall efficiency, and discount rate are also discussed in details [118,120,127,129]. However, the systematic analysis of the uncertainties in input data and assumptions related to LCC calculations is rather rare in the literature. Neglecting or overlooking the uncertainties in input parameters decreases the level of the accuracy of the results of LCC analysis [195]. Among the examined references, Battke et al. [113] has considered the uncertainties in input parameters, including cost of storage part, overall efficiency, lifetime, and life cycle numbers in the analysis of LCC of four battery systems. In addition to more input parameters for uncertainty analysis, this study extends the number of examined technologies to include mechanical storage (PHS, CAES, and flywheel), hydrogen-based, and more electrochemical BES systems. The review of the cost data of EES systems reveals that the uncertainty in PCS costs, O&M costs, and replacement costs are also considerable for some EES technologies (see Section 4.1).

586

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

Table 8 Three common applications of EES systems and their requirements, data from [110,127,156]. Application type (usage frequency)

Example of application

Power rating

Response time

Discharge time

Cycles per year

Application lifetime

EES systems

Long-duration, frequent

Bulk energy storage, energy arbitrage Capacity credit, spinning reserve, T&D support Frequency regulation, RES integration, power quality

þ10 MW

min

4–8 h

250–300

20

1–10 MW

s–min

0.5–2 h

300–400

15

o 0.25 h

þ 1000

10

PHS, CAES, lead–acid, NaS, Ni–Cd, VRFB, Fe–Cr CAES (aboveground), lead–acid, NaS, ZEBRA, Li-ion, VRFB, Zn–Br, Fe–Cr, Ni–Cd, hydrogen Flywheel, lead–acid, Li-ion

Medium duration, fast response Short duration, highly frequent

0.1–2 MW ms–s

Table 9 Economic parameters and assumptions for the analysis of LCC of EES systems. Parameter

Value

Note

Average yearly inflation ratea Discount rateb Charging electricity priceb,c Power price escalation rate Fuel costs (natural gas)d,e Fuel cost escalation rate Carbon emission costse

2.5% 8% 50 €/MWh 0% 20–25 €/MWh 0% 8–22 €/ton CO2

[131] Sensitivity performed Sensitivity performed Assumption Uncertain parameter Assumption Uncertain parameter

a

Average of the EU inflation rate for the period of 2003–2013 [131]. Otherwise mentioned in the sensitivity analyses. Based on average of wholesale prices in EU [131]. d Based on wholesale prices in EU (2007–2012) [131]. e Notice that these values are used in this study for uncertainty analysis and they are not necessarily consistent with those reported in reviewed publications. b c

Fig. 8. Comparing the results of cost calculations with and without uncertainty analysis for VRFB (vanadium-redox) systems.

Therefore, this study examines all the cost segments in uncertainty analysis, as well as overall efficiency, fuel and emission costs for CAES, lifetime and life cycle numbers, and replacement period (11 uncertain input parameters). Employing the Monte Carlo method and MATLAB simulation tool, the uncertainty analysis in LCC is conducted by assigning stochastic values for the uncertain input parameters, from the corresponding ranges. This method is recommended for the modeling of systems with varying several deterministic input parameters simultaneously [196]. For the cost items, the uncertainty in input data is examined for the corresponding interquartile ranges (IQR) presented in Section 4.1, which represent the middle 50% threshold of the probable values. For the remaining parameters, the uncertainty in input data is stochastically selected from those ranges presented in Tables 9 and B.1 of Appendix B.

In Fig. 8, the distribution of the results of uncertainty analysis for VRFB systems is illustrated, after 10,000 simulation runs. The results of uncertainty analysis can be properly fitted by a normal probability distribution, indicating a mean value of 706 €/kW‐yr for this technology. However, without uncertainty analysis, the ALCC of VRFB system is 663 €/kW‐yr, which shows 6.5% error and it even does not lie in the one standard deviation (SD) threshold of the corresponding probabilistic distribution after considering uncertainties. In general, conducting uncertainty analysis in LCC calculations shows changes from  17% to þ39% compared to the results obtained without considering the effect of uncertainties. For instance, in calculating ALCC for lead–acid batteries, the results with uncertainty analysis indicate differences of  17% in bulk energy storage,  11% in T&D support, and þ13% in frequency

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

587

Fig. 9. The annualized life cycle costs (ALCC) of EES systems in bulk energy storage and related uncertainties, considering 250 cycles per year, 8% interest rate and 8 h discharge time (see Tables 8 and 9 for other input parameters). The average values are shown above each bar.

Fig. 10. The annualized life cycle costs (ALCC) of EES systems and related uncertainty for T&D support applications, with 400 cycles per year, 8% interest rate and 2 h discharge time (see Tables 8 and 9 for other input parameters).

regulation, compared to the results for the same technology without uncertainty analysis. For each of three applications mentioned in Table 8, the results are first presented and discussed for ALCC (€/kW‐yr). The share of capital cost, charging electricity costs, and O&M costs (including replacement costs) are separately shown for each technology. These results are presented by boxplots, indicating the range of 95% threshold, 68% likelihood or one standard deviation band (1SD), as well as the average values (arithmetic mean). The annualized ALCC of EES systems applicable for bulk energy storage (energy arbitrage) are illustrated in Fig. 9. PHS offers the minimum costs (240 €/kW‐yr) for this service, with relatively low inconsistency. The costs of fuel and emissions decrease the profitability of CAES, while it is the cheapest technology in terms of capital costs. The uncertainty in LCC of lead–acid batteries is rather high, due to more dispersed input cost elements and diverse suppliers. The ALCC of EES systems applicable for T&D support services are shown in Fig. 10. While aboveground CAES surpasses other technologies for this service with lower costs (160 €/kW‐yr), Li-ion batteries show rather the highest costs (493 €/kW‐yr). The widest uncertainty in LCC is however related to hydrogen–FC systems. The replacement costs of batteries may impose extra costs that make them unfavorable for highly frequent charge/discharge

Fig. 11. The annualized life cycle costs (ALCC) and its uncertainty for EES systems applicable in frequency regulation and power quality services, 1000 cycles per year, 8% interest rate and max 15 min discharge time (see Tables 8 and 9 for other input parameters).

services. Comparing those EES systems applicable for both examined applications, NaS and lead–acid show lower costs for T&D support compared to bulk energy storage. This can be attributed to

588

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

the lower discharge time requirements, shorter application lifetime, and consequently less operating cycles. The ALCC of fast-responding EES systems applicable for frequency regulation and similar services are depicted in Fig. 11. The results indicate that flywheel offers lower LCC while having the highest capital costs. Those technologies with higher number of life cycles are more favorable for this service to avert the costs of replacement.

4.3.1. Sensitivity analysis In this section, the dependency of LCC on interest rate, electricity prices, and discharge time is examined. The results are presented in levelized cost of electricity (LCOE) delivered by the EES technology. This way, the grid operators and decision makers are able to compare the power prices discharged by EES with those of other alternatives relevant to the market in question. The electricity prices are varied in the range of 0-100 €/MWh and interest rate from 6% to 10%. Zero charging power prices are chosen to quantify the net internal costs added by storing and discharging power through EES systems. The authors suggested the term levelized cost of storage (LCOS) to account

for the net internal costs of EES systems without including the influence of price of charging electricity (see Eq. (10)). The results for the bulk energy storage are illustrated in Fig. 12, showing the LCOS for different EES systems at the left end of each red bar (upper sensitivity bar). The results indicate that the LCC of those EES systems that are subject to significant replacement costs during the lifetime are more sensitive to the interest rate, e.g. Ni–Cd, VRFB and lead–acid. The LCOE delivered by PHS, as the most cost-efficient technology is in the range of 120 €/MWh, with charging prices of 50 €/MWh. The LCOE discharged by EES systems in T&D support services and the corresponding sensitivity analyses are shown in Fig. 13. LCC shows wider sensitivity to the electricity prices for those EES systems with relatively low efficiencies. For example, the LCOE for hydrogen–FC increases from 480 to 600 €/MWh, if power prices rise from 50 to 100 €/MWh. Aboveground CAES offers the most costefficient option with LCOE totaling 202 €/MWh for this application. The sensitivity of LCC of EES systems for power quality and frequency regulation applications is comparatively lower to power prices due to the lower power consumption (Fig. 14). Since these technologies are

Fig. 12. Levelized cost of electricity (LCOE) delivered by EES systems in bulk energy storage and the sensitivity of LCOE to interest rate and electricity prices.

Fig. 13. Levelized cost of electricity (LCOE) delivered by EES systems in T&D support and similar services, and the sensitivity of LCOE to interest rate and electricity prices.

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

not relevant for energy-related service, the cost figures presented in the sensitivity analysis are not in the form of LCOE. The levelized costs (LCOE and LCOS) of EES systems are also sensitive to the discharge time. The LCOS of different EES systems applicable for bulk energy storage are illustrated in Fig. 15, for different discharge times up to 8 h. The price of charging power is not included in the costs to show the pure value of each EES systems. The results indicate that the rank order of the batteries is different based on the discharge time. For example, Ni–Cd shows the lower costs compared to lead–acid and VRFB in one-hour discharge time, while it is the most expensive EES system in discharge times higher than 3 h. PHS offers the minimum costs with the LCOS of 54 €/MWh. The LCOE of each EES system can be produced by adding the cost of charging power to the LCOS values shown in Fig. 15, subject to EES inefficiencies (see Eq. (10)).

589

The future cost of EES systems is also provided in some references, predicting cost reductions due to further commercialization [112,117]. However, the wide usage of EES systems raises questions regarding their environmental aspects [198], material intensity [199–201], and other societal impacts [40,202]. The extensive employment of EES systems can affect the electricity prices by smoothing the high peaks and troughs, deteriorating the expected revenues. According to Awad et al. [207], the clearing market price of electricity can be furthermore influenced by the location (in power transmission versus distribution level) and size (distributed versus central) of the EES systems. Therefore, any holistic approach should consider the impact of large-scale EES systems on the energy system to evaluate the associated costs and potential benefits [203–206]. As the integration of EES systems is linked with the large-scale penetration of variable RES, the associated policy and market barriers should be addressed by considering all the direct and indirect, system-level impacts of EES technologies [208–210].

4.4. Discussion 5. Conclusions While the cost of discharged power is not the only criterion for the selection of EES systems, it compares different alternatives in fulfilling the target services in a market, based on the associated benefits [197]. The cost of pilot plants can be higher due to the administrative expenses, as well as the lack of economy of scale.

Fig. 14. Sensitivity of ALCC to interest rate and electricity prices for EES systems in frequency regulation and similar services.

Fig. 15. The levelized cost added by storage (LCOS) to the price of charging power, in different discharge times per one cycle (bulk energy storage with 250 cycles per year, interest rate 8%).

The LCC of different grid-scale EES technologies were analyzed by conducting an extensive review of the existing literature, considering uncertainties in cost data and technical parameters. The results reveal that the cost estimations/projections of the EES systems are rather dispersed and inconsistent among different references. The cost estimations rely on assumptions and scaling the size, the case for most of battery systems, which reduces the consistency among different sources of data. Most of the EES systems are in formative stages of commercialization and those commercial plants are mainly site-specific resulting in more inconsistency in the cost data. Hence, a robust LCC analysis should account for the uncertainties. The cost items of EES systems were first separately analyzed. CAES has the highest costs for PCS (845 €/kW) while NiCd batteries offer the minimum power interface costs (240 €/kW). However, electrochemical batteries show higher costs for storage compartment (up to 800 €/kWh for Li-ion). Hydrogen-based and underground CAES have lowest costs of storage, 4 and 40 €/kWh, respectively. More details of the cost elements are presented in Appendix A for each technology. In terms of TCC (total capital cost), underground CAES (with 890 €/kW) offers the most economical alternative for bulk energy storage, while SMES and SCES are the cheapest options in power quality applications. However, the cost data for these electro-magnetic EES systems are rather limited and for small-scale applications. The TCC of hydrogen-based systems indicate a large difference between gas turbine (1570 €/kW) and fuel cell systems (3240 €/kW). TCC of different EES systems are illustrated in more details in Appendix C. In the calculation of LCC, the effect of uncertainties is different and can affect the results by 5–17% in most of the examined cases. The results indicated that mechanical energy storage systems, namely PHS and CAES, are still the most cost-efficient options for bulk energy storage. PHS and CAES approximately add 54 and 71 €/MWh respectively, to the cost of charging power. The project's environmental permitting costs and contingency may increase the costs, however. The uncertainty in LCC of CAES may increase if the fuel and emission costs could not be consistently established for the application's lifetime. Among the commercialized batteries, NaS offers relatively lower LCC in both energy arbitrage and T&D support applications. However, the uncertainties in the cost of batteries are rather wide, even larger than difference in costs between different technologies. Hence, the most optimal option should be selected based on other technical and project-specific characteristics. Since replacement costs comprise a major and distinctive part of LCC of the batteries, optimal cycle numbers that can make the highest revenues should be defined, based on the service requirements and the regime of charge and

590

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

discharge (DoD). Flywheels offer the most cost-efficient option in power quality and frequency regulation applications (discharge time up to a few minutes), with lower operational costs. Hydrogen-based storage and other EES systems with relatively low efficiency demonstrate higher sensitivity to the electricity prices, indicating the need for more R&D to become economically competitive. This study aims at providing a milestone for future researches that examine the integration of EES systems by contributing in costbenefit analysis. The analysis can be improved in the future by realization of more demonstration plants and establishment of the costs at different stages throughout the plant's lifetime. Considering more parameters in LCC of the batteries can improve the practicality of the results, e.g. dependencies between DoD and life cycle numbers or optimal life cycle numbers based on the service lifetime.

electricity is not included in variable O&M costs in the following tables, as it depends on the market in which the EES is adopted. See Tables A1–A12

Table A3 Main cost items of flywheel energy storage systems. Cost item

Average

Middle fifty range, IQR

Range

PCSa (€/kW) Storage sectionb(€/kWh) Fixed O&M (€/kW -yr) Variable O&M (€/MWh) Replacement costsc (€/kW)

287 2815 5.2 2.0 151

284–356 1030–18159 4.8–5.6 1.1–2.9 118–184

263–470 865–47764 4.3–6.0 0.2–3.8 85–216

a

Including BOP costs. As flywheel systems are typically employed for power quality applications with discharge time of seconds up to 30 min, the direct use of storage cost may entail ambiguity. c Every four years, given based on the unit of power rating. b

Acknowledgments The authors would like to present their sincere appreciation for the supports provided by STEEM project (Sustainable Transition of European Energy Markets) and the Aalto Energy Efficiency Research Program.

Appendix A. Cost elements of different EES systems The main cost items of each EES systems are depicted in separate tables to compare the values and variability of the cost items in the literature for each individual technology. The results are calculated based on the detailed review of the references reported in Table 2. The average is the median of each range, and outliers are not included in the ranges. It should be noted that the price of purchasing

Table A4 Main cost items of lead–acid battery systems. Cost item

Average

Middle fifty range, IQR

Range

PCS (€/kW) BOP (€/kW) Storage sectiona (€/kWh) Fixed O&M (€/kW -yr) Variable O&M (€/MWh) Replacement costsb(€/kW)

378 87 618 3.4 0.37 172

322–440 65–108 264–661 3.3–6.1 0.35–0.49 157–264

195–594 43–130 184–847 3.2–13.0 0.15–0.52 50–560

a Mainly for MW-scale systems with rated DoD of 80%, used for bulk energy storage and T&D support (discharge time 4 h). b Every 8 yr for the mentioned application (365 cycles per year).

Table A1 Main cost items of PHS systems. Cost item

Average

Middle fifty range, IQR

Range

PCS (€/kW) BOP (€/kW) Storage sectiona (€/kWh) Fixed O&Mb (€/kW-yr) Variable O&M (€/MWh)

513 15 68 4.6 0.22

410–805 9–22 41–115 3.9–7.7 0.20–0.79

373–941 3–28 8–126 2.0–9.2 0.19–0.84

a

Mainly for storage sizes with 8 h discharge time. Major fixed O&M is expected every 20 yr totaling 84 €/kW of installed capacity. b

Table A5 Main cost items of NaS (sodium–sulfur) battery systems. Cost item a

PCS (€/kW) Storage sectionb (€/kWh) Fixed O&M (€/kW-yr) Variable O&M (€/MWh) Replacement costsc (€/kW)

Average

Middle fifty range, IQR

Range

366 298 3.6 1.8 180

314–553 277–358 3.3–16.5 0.3–4.6 180–307

241–865 180–563 2.0–17.3 0.3–5.6 180–443

a

Including BOP costs, which is estimated in the range of 80 €/kW. Mainly for MW-scale systems with rated DoD of 80%, used for bulk energy storage and T&D support (discharge time 6–7.2 h). c Every 8 yr for the mentioned application (365 cycles per year). b

Table A2 Main cost items of first-generation CAES systems. Cost item

PCS (€/kW) a

Storage section (€/kWh) Fixed O&Mb (€/kW -yr) c

Variable O&M (€/MWh)

a

Type of CAES

Average

Middle fifty range, IQR

Range

Aboveground Underground Aboveground Underground Aboveground Underground Aboveground Underground

846 843 109 40 2.2 3.9 2.2 3.1

825–866 696–928 97–120 30–47 2.2–3.0 2.6–4.0 2.1–2.6 2.6–3.6

804–887 549–1014 86–131 4–64 2.2–3.7 2.0–4.2 1.9–3.0 2.2–2.5

Mainly for storage sizes with 8 h discharge time. Major fixed O&M is expected every 5 yr totaling 67 €/kW of installed capacity. c As natural gas prices are not equal in different studies, the variable O&M costs entails more uncertainty. On average, the fuel costs are in the range of 8–20 €/MWh and emission cost is 18–22 €/ton CO2, for example see [110,111].

Table A6 Main cost items of Ni–Cd (nickel–cadmium) battery systems. Cost item a

PCS (€/kW) Storage sectionb (€/kWh) Fixed O&M (€/kW -yr) Replacement costsc (€/kW)

Average

Middle fifty range, IQR

Range

239 780 11 525

213–279 571–1020 5–19 502–549

206–329 564–1120 4–24 478–573

b

a

Including BOP costs. Mainly for MW-scale systems with rated DoD of 80%, used for bulk energy storage and T&D support (discharge time 2–4 h). c Every 10 yr for the mentioned application (365 cycles per year). b

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

591

Table A7 Main cost items of sodium–nickel chloride (NaNiCl2) battery systems, known as ZEBRA. Cost item

Average

Middle fifty range, IQR

Range

PCSa (€/kW) Storage sectionb (€/kWh) Fixed O&M (€/kW -yr) Variable O&M (€/MWh) Replacement costsc (€/kW)

472 509 5.5 0.6 182

379–611 410–723 3.7–7.1 0.41–1.0 148–202

335–638 366–778 3.3–7.2 0.38–2.1 107–202

a b c

Including BOP costs. Mainly for MW-scale systems with rated DoD of 80%, used for bulk energy storage and T&D support (discharge time 5 h). Every 8 yr for the mentioned application (365 cycles per year).

Table A8 Main cost items of Li-ion battery systems. Cost item a

PCS (€/kW) Storage sectionb (€/kWh) Fixed O&M (€/kW-yr) Variable O&M (€/MWh) Replacement costsc (€/kW) a b c

Average

Middle fifty range, IQR

Range

463 795 6.9 2.1 369

398–530 676–1144 4.9–11.2 0.99–3.6 284–505

241–581 470–1249 2.0–13.7 0.4–5.6 187–543

Including BOP costs of 80 €/kW. Mainly for MW-scale systems with rated DoD of 80%, used for bulk energy storage and T&D support (discharge time 0.5–2 h). Every 5 yr for the mentioned application (365–500 cycles per year).

Table A9 Main cost items of VRFB (vanadium-redox) battery systems. Cost item a

PCS (€/kW) Storage sectionb (€/kWh) Fixed O&M (€/kW-yr) Variable O&M (€/MWh) Replacement costsc (€/kW) a b c

Average

Middle fifty range, IQR

Range

490 467 8.5 0.9 130

478–518 440–536 4.3–16.1 0.5–1.2 114–165

472–527 433–640 3.4–17.3 0.2–2.8 111–192

Including BOP costs approximately 25 €/kW. Mainly for MW-scale systems with rated DoD of 80%, used for bulk energy storage and T&D support (discharge time 4 h). Every 8 years for the mentioned application (365–500 cycles per year).

Table A10 Main cost items of zinc–bromine (Zn–Br) battery systems. Cost item a

PCS (€/kW) Storage sectionb (€/kWh) Fixed O&M (€/kW-yr) Variable O&M (€/MWh) Replacement costsc (€/kW) a b c

Average

Middle fifty range, IQR

Range

444 195 4.3 0.6 195

343–470 178–314 3.6–5.4 0.4–1.0 148–198

151–595 178–530 3.2–6.9 0.3–2.0 101–201

Including BOP costs approximately 25 €/kW. Mainly for MW-scale systems with rated DoD of 80%, used for bulk energy storage and T&D support (discharge time 2–5 h). Every 15 yr for the mentioned application (365 cycles per year).

Table A11 Main cost items of iron–chrome (Fe–Cr) battery systems. Cost item

Average

Middle fifty range, IQR

Range

PCSa (€/kW) Storage sectionb (€/kWh) Fixed O&M (€/kW -yr) Variable O&M (€/MWh) Replacement costsc (€/kW)

362 145 3.3 0.4 29

333–393 126–152 2.8–4.0 0.2–0.6 24–33

326–523 64–156 2.7–6.9 0.1–1.0 14–38

a b c

Including BOP costs approximately 25 €/kW. Mainly for MW-scale systems with rated DoD of 80%, used for bulk energy storage and T&D support (discharge time 2–5 h). Every 15 yr for the mentioned application (365 cycles per year).

592

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

Table A12 Main cost items of hydrogen-based EES systems. Cost item

Configuration

Average

Middle fifty range, IQR

Range

PCSa (€/kW)

Hydrogen-FCb Hydrogen-GTc

2465 1548

1630–3884 1359–2673

1383–4453 1102–3362

Storage sectionb (€/kWh)

Aboveground Underground

130 3.7

128–132 0.2–11.6

125–134 0.02–12.4

O&M (€/kW -yr)

Hydrogen-FCb hydrogen-GTc

25 35

24–39 25–45

16–44 23–48

a b c

Including BOP costs approximately 25 €/kW. Electrolysis and fuel cell. Electrolysis and small-to medium scale gas turbine.

Appendix B. Summary of technical characteristics of EES systems See Table B1 Table B1 Technical characteristics of electrical energy storage (EES) systems, based on the review of the references in Table 2. EES technology

Power range (MW)

Discharge time (ms–h)

Overall efficiency

PHS CAES (underground) CAES (aboveground) Flywheel Lead–acid NaS NaNiCl2 (ZEBRA) Ni–Cd Li-ion VRFB Zn–Br Fe–Cr PSB SMES Capacitors SCES Hydrogen (fuel cell)

10–5000 5–400 3–15 Up to 0.25 Up to 20 0.05–8 50 Up to 40 up to 0.01 0.03–3 0.05–2 1–100 15 0.1–10 Up to 0.05 Up to 0.3 0.3–50

1–24 h 1–24 h 2–4 h ms–15 m s–h s–h 2–5 h s–h m–h s–10 h s–10 h 4–8 h s–10 h ms–8 s ms–60 m ms–60 m s–24 h

0.70–0.82 0.7–0.89 0.70–0.90 0.93–0.95 0.70–0.90 0.75–0.90 0.86–0.88 0.60–0.73 0.85–0.95 0.65–0.85 0.60–0.70 0.72–0.75 0.65–0.85 0.95–0.98 0.60–0.65 0.85–0.95 0.33–0.42

Power density (W/kg)

Energy density (Wh/kg)

Storage durability

Self-discharge (per day)

Lifetime (yr)

Life cycles (cycles)

0.5–1.5 30–60

Negligible Small Small 100% 0.1–0.3% 20% 15% 0.2–0.6% 0.1–0.3% Small Small Small 10–15% 40% 20–40% Negligible

50–60 20–40 20–40 15–20 5–15 10–15 15 10–20 5–15 5–10 5–10 10–15 10–15 15–20 5–8 10–20 15–20

20000–50000 413,000 413,000 20,000–100,000 2000–4500 2500–4500 2500–3000 2000–2500 1500–4500 10,000–13,000 5000–10,000 410,000 2000–2500 4100,000 50,000 4100,000 20,000

1000 75–300 150–230 150–200 50–1000 50–2000 166 45

5–100 30–50 150–250 100–140 15–300 150–350 10–35 30–85

h–months h–months h–days s–min min–days s–h s–h min–days min–days h–months h–months

500-2000 100,000 800–23,500 500

0.5–5 0.05–5 2.5–50 100–10,000

h–months min–h s–h s–h h–months

Appendix C. Total capital cost of different EES systems See Table C1

Table C1 Total capital cost (TCC) of grid-scale EES systems based on the review of the sources listed in Table 2. EES technology

PHS CAES Flywheel Lead–acid NaS Ni–Cd ZEBRA Li-ion VRFB Zn–Br PSB Fe–Cr Zn–air Supercapacitors

Configuration

Conventional Aboveground Underground High-speed Advanced – – – – – – – – – Double-layer

Total capital costa (TCC), per unit of power rating €/kW

Total capital costa (TCC), per unit of storage capacityb €/kWh

Min

Average

Max

Min

Average

Max

1030 774 1286 590 1388 1863 2279 874 2109 1277 1099 927 1376 1313 214

1406 893 1315 867 2140 2254 3376 1160 2512 1360 1132 1093 1400 1364 229

1675 914 1388 1446 3254 2361 4182 1786 2746 1649 1358 1308 1425 1415 247

96 48 210 1850 346 328 596 973 459 257 170 1071 527 262 691

137 92 263 4791 437 343 699 1095 546 307 220 1147 569 271 765

181 106 278 25049 721 398 808 1211 560 433 281 1153 611 417 856

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

593

Table C1 (continued ) EES technology

SMES Hydrogen

Configuration

– Fuel cellc (FC) Gas turbined (GT)

Total capital costa (TCC), per unit of power rating €/kW

Total capital costa (TCC), per unit of storage capacityb €/kWh

Min

Average

Max

Min

Average

212 2395 1360

218 3243 1570

568 4674 2743

5310 399 227

6090 540 262

Max 6870 779 457

a It should be noted that the capital costs are calculated based on typical discharge time (storage size) for each technology, which is not necessarily the same among different EES systems (for typical discharge time, see the corresponding table for each technology in Appendix A). Minimum and maximum values are the bands of interquartile range (middle-fifty likelihood) and the average value is the median of whole sample, excluding outliers. It should be noted that the costs of grid interconnections and infrastructure requirements are not included in this estimation. b For the batteries, the storage capacity is equivalent to the rated DoD. c Electrolysis and fuel cell with steel tank storage system. d Electrolysis and small-to medium scale gas turbine with underground storage.

References [1] International Energy Agency (IEA). World energy outlook 2013. Paris: OECD/ IEA; 2013. [2] European Commission. Strategic energy technologies [online]. Available: 〈http://setis.ec.europa.eu/technologies〉; 2013. [3] Sandia National Laboratories. Energy storage systems program [online]. Available: 〈http://www.sandia.gov/ess/〉; 2013. [4] Evans A, Strezov V, Evans TJ. Assessment of utility energy storage options for increased renewable energy penetration. Renew Sustain Energy Rev 2012;16 (6):4141–7. [5] Electricity Storage Association (ESA). Electricity storage technology comparison [online]. Available: 〈http://www.electricitystorage.org/〉; 2013. [6] Hall PJ, Bain EJ. Energy-storage technologies and electricity generation. Energy Policy 2008;36(12):4352–5. [7] Hadjipaschalis I, Poullikkas A, Efthimiou V. Overview of current and future energy storage technologies for electric power applications. Renew Sustain Energy Rev 2009;13(6–7):1513–22. [8] Ibrahim H, Ilinca A, Perron J. Energy storage systems-characteristics and comparisons. Renew Sustain Energy Rev 2008;12(5):1221–50. [9] Chen H, Cong TN, Yang W, Tan C, Li Y, Ding Y. Progress in electrical energy storage system: A critical review. Progr Nat Sci 2009;19(3):291–312. [10] Ter-Gazarian AG, editor. Energy storage for power systems. 2nd ed.. London, UK: The Institution of Engineering and Technology; 2011. [11] Baker J. New technology and possible advances in energy storage. Energy Policy 2008;36(12):4368–73. [12] Díaz-González F, Sumper A, Gomis-Bellmunt O, Villafáfila-Robles R. A review of energy storage technologies for wind power applications. Renew Sustain Energy Rev 2012;5(16):2154–71 (4). [13] Punys P, Baublys R, Kasiulis E, Vaisvila A, Pelikan B, Steller J. Assessment of renewable electricity generation by pumped storage power plants in EU member states. Renew Sustain Energy Rev 2013;10(26):190–200. [14] Karellas S, Tzouganatos N. Comparison of the performance of compressed-air and hydrogen energy storage systems: Karpathos island case study. Renew Sustain Energy Rev 2014;29(0):865–82. [15] Dunn B, Kamath H, Tarascon J. Electrical energy storage for the grid: a battery of choices. Science 2011;334(6058):928–35. [16] Poullikkas A. A comparative overview of large-scale battery systems for electricity storage. Renew Sustain Energy Rev 2013;27:778–88. [17] Alotto P, Guarnieri M, Moro F. Redox flow batteries for the storage of renewable energy: a review. Renew Sustain Energy Rev 2014;29(0):325–35. [18] Sebastián R, Peña Alzola R. Flywheel energy storage systems: review and simulation for an isolated wind power system. Renew Sustain Energy Rev 2012;16(9):6803–13. [19] Bolund B, Bernhoff H, Leijon M. Flywheel energy and power storage systems. Renew Sustain Energy Rev 2007;11(2):235–58. [20] Ali MH, Wu B, Dougal RA. An overview of SMES applications in power and energy systems. IEEE Trans Sustain Energy 2010;1(1):38–47. [21] Noriega JR, Iyore OD, Budime C, Gnade B, Vasselli J. Characterization system for research on energy storage capacitors. Rev Sci Instrum 2013;84:5. [22] Steffen B, Weber C. Efficient storage capacity in power systems with thermal and renewable generation. Energy Econ 2013;36(0):556–67 (3). [23] Rugolo J, Aziz MJ. Electricity storage for intermittent renewable sources. Energy Environ Sci 2012;5(5):7151–60. [24] Di Silvestre ML, Riva Sanseverino E. Modelling energy storage systems using Fourier analysis: an application for smart grids optimal management. Appl Soft Comput J 2013. [25] Makarov YV, Du P, Kintner-Meyer MCW, Jin C, Illian HF. Sizing energy storage to accommodate high penetration of variable energy resources. IEEE Trans Sustain Energy 2012;3(1):34–40.

[26] Barbour E, Wilson IAG, Bryden IG, McGregor PG, Mulheran PA, Hall PJ. Towards an objective method to compare energy storage technologies: development and validation of a model to determine the upper boundary of revenue available from electrical price arbitrage. Energy Environ Sci 2012;5(1):5425–36. [27] Zhu D, Wang Y, Yue S, Xie Q, Pedram M, Chang N. Maximizing return on investment of a grid-connected hybrid electrical energy storage system. In: Proceedings of the Asia and South Pacific Design Automation Conference, ASP-DAC; 2013. [28] Evans L, Guthrie G, Lu A. The role of storage in a competitive electricity market and the effects of climate change. Energy Econ 2013;36:405–18. [29] Brekken TKA, Yokochi A, Von Jouanne A, Yen ZZ, Hapke HM, Halamay DA. Optimal energy storage sizing and control for wind power applications. IEEE Trans Sustain Energy 2011;2(1):69–77. [30] Yuan Y, Li Q, Wang W. Optimal operation strategy of energy storage unit in wind power integration based on stochastic programming. IET Renew Power Gener 2011;5(2):194–201. [31] Fertig E, Apt J. Economics of compressed air energy storage to integrate wind power: a case study in ERCOT. Energy Policy 2011;39(5):2330–42. [32] Bradbury K, Pratson L, Patiño-Echeverri D. Economic viability of energy storage systems based on price arbitrage potential in real-time U.S. electricity markets. Appl Energy 2014;114:512–9. [33] Wade NS, Taylor PC, Lang PD, Jones PR. Evaluating the benefits of an electrical energy storage system in a future smart grid. Energy Policy 2010;38(11):7180–8. [34] He X, Delarue E, D'haeseleer W, Glachant J-. A novel business model for aggregating the values of electricity storage. Energy Policy 2011;39(3): 1575–1585. [35] DOE. Grid energy storage [online]. Available: 〈http://energy.gov/oe/down loads/grid-energy-storage-december-2013〉; 2013. [36] Deane JP, Ó Gallachóir BP, McKeogh EJ. Techno-economic review of existing and new pumped hydro energy storage plant. Renew Sustain Energy Rev 2010;14(4):1293–302. [37] Denholm P, King JC, Kutcher CF, Wilson PPH. Decarbonizing the electric sector: combining renewable and nuclear energy using thermal storage. Energy Policy 2012;44:301–11. [38] Li Y, Cao H, Wang S, Jin Y, Li D, Wang X, et al. Load shifting of nuclear power plants using cryogenic energy storage technology. Appl Energy 2014;113: 1710–1716. [39] Electric Power Research Institute (EPRI). EPRI–DOE handbook of energy storage for transmission and distribution applications. Palo Alto, California: EPRI and U.S. Department of Energy (DOE); 2003. [40] Sioshansi R, Denholm P, Jenkin T, Weiss J. Estimating the value of electricity storage in PJM: arbitrage and some welfare effects. Energy Econ 2009;31 (2):269–77. [41] Eyer J. Electric utility transmission and distribution upgrade deferral benefits from modular electricity storage. New Mexico, California: Sandia National Laboratories; 2009. [42] Walawalkar R, Apt J, Mancini R. Economics of electric energy storage for energy arbitrage and regulation in New York. Energy Policy 2007;35(4): 2558–2568. [43] Ekman CK, Jensen SH. Prospects for large scale electricity storage in Denmark. Energy Convers Manag 2010;51(6):1140–7. [44] Connolly D, Lund H, Finn P, Mathiesen BV, Leahy M. Practical operation strategies for pumped hydroelectric energy storage (PHES) utilising electricity price arbitrage. Energy Policy 2011;39(7):4189–96. [45] Sioshansi R, Denholm P, Jenkin T. A comparative analysis of the value of pure and hybrid electricity storage. Energy Econ 2011;33(1):56–66. [46] Zakeri B, Syri S. Economy of electricity storage in the Nordic electricity market: the case for Finland. In: Proceedings of the 11th international conference on the European Energy Market, EEM14, Krakow, Poland; 28–30 May 2014. [47] Muche T. Optimal operation and forecasting policy for pump storage plants in day-ahead markets. Appl Energy. 2014;113:1089–99.

594

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

[48] Aggarwal SK, Saini LM, Kumar A. Electricity price forecasting in deregulated markets: a review and evaluation. Int J Electr Power Energy Syst 2009;31 (1):13–22. [49] Aslani A, Naaranoja M, Wong K-V. Strategic analysis of diffusion of renewable energy in the Nordic countries. Renew Sustain Energy Rev 2013;22:497–505. [50] Wong J, Lim YS, Tang JH, Morris E. Grid-connected photovoltaic system in Malaysia: a review on voltage issues. Renew Sustain Energy Rev 2014;29 (0):535–45. [51] Gutiérrez-Martín F, Da Silva-Álvarez RA, Montoro-Pintado P. Effects of wind intermittency on reduction of CO2 emissions: the case of the Spanish power system. Energy 2013;61:108–17. [52] Keatley P, Shibli A, Hewitt NJ. Estimating power plant start costs in cyclic operation. Appl Energy 2013;111:550–7. [53] Energy Research Partnership (ERP). The future role for energy storage in the UK. London, UK: ERP; 2011. [54] Steinke F, Wolfrum P, Hoffmann C. Grid vs. storage in a 100% renewable Europe. Renew Energy 2013;50:826–32. [55] Aboumahboub T, Schaber K, Tzscheutschler P, Hamacher T. Optimal configuration of a renewable-based electricity supply sector. WSEAS Trans Power Syst 2010;5(2):120–9. [56] Benitez LE, Benitez PC, van Kooten GC. The economics of wind power with energy storage. Energy Econ 2008;30(4):1973–89. [57] Howlader AM, Urasaki N, Yona A, Senjyu T, Saber AY. A review of output power smoothing methods for wind energy conversion systems. Renew Sustain Energy Rev 2013;26:135–46. [58] Sundararagavan S, Baker E. Evaluating energy storage technologies for wind power integration. Sol Energy 2012;86(9):2707–17. [59] Hasan NS, Hassan MY, Majid MS, Rahman HA. Review of storage schemes for wind energy systems. Renew Sustain Energy Rev 2013;21:237–47. [60] Fares RL, Meyers JP, Webber ME. A dynamic model-based estimate of the value of a vanadium redox flow battery for frequency regulation in Texas. Appl Energy 2014;1(113):189–98 (0). [61] Loisel R. Power system flexibility with electricity storage technologies: a technical-economic assessment of a large-scale storage facility. Int J Electr Power Energy Syst 2012;42(1):542–52. [62] Weis TM, Ilinca A. The utility of energy storage to improve the economics of wind-diesel power plants in Canada. Renew Energy 2008;33(7):1544–57. [63] Arabali A, Ghofrani M, Etezadi-Amoli M. Cost analysis of a power system using probabilistic optimal power flow with energy storage integration and wind generation. Int J Electr Power Energy Syst 2013;53:832–41. [64] Chowdhury MM, Haque ME, Aktarujjaman M, Negnevitsky M, Gargoom A. Grid integration impacts and energy storage systems for wind energy applications – a review. 2011 IEEE power energy society general meeting; 2011. [65] Østergaard PA. Comparing electricity, heat and biogas storages' impacts on renewable energy integration. Energy 2012;37(1):255–62. [66] Al-Karaghouli A, Kazmerski LL. Energy consumption and water production cost of conventional and renewable-energy-powered desalination processes. Renew Sustain Energy Rev 2013;8(24):343–56. [67] Alanne K, Saari A. Distributed energy generation and sustainable development. Renew Sustain Energy Rev 2006;10(6):539–58. [68] Niemi R, Mikkola J, Lund PD. Urban energy systems with smart multi-carrier energy networks and renewable energy generation. Renew Energy 2012;48:524–36. [69] Grünewald PH, Cockerill TT, Contestabile M, Pearson PJG. The socio-technical transition of distributed electricity storage into future networks-system value and stakeholder views. Energy Policy 2012;50:449–57. [70] Toledo OM, Oliveira Filho D, Diniz ASAC. Distributed photovoltaic generation and energy storage systems: a review. Renew Sustain Energy Rev 2010;14 (1):506–11. [71] Hill CA, Such MC, Chen D, Gonzalez J, Grady WM. Battery energy storage for enabling integration of distributed solar power generation. IEEE Trans Smart Grid 2012;3(2):850–7. [72] Kaldellis JK, Zafirakis D. Optimum energy storage techniques for the improvement of renewable energy sources-based electricity generation economic efficiency. Energy 2007;32(12):2295–305. [73] Zafirakis D, Chalvatzis KJ. Wind energy and natural gas-based energy storage to promote energy security and lower emissions in island regions. Fuel 2014;115:203–19. [74] Darby S, Strömbäck J, Wilks M. Potential carbon impacts of smart grid development in six European countries. Energy Eff 2013;6(4):725–39. [75] Hashmi M, Hänninen S, Mäki K. Developing smart grid concepts, architectures and technological demonstrations worldwide – a literature survey. Int Rev Electr Eng 2013;8(1):236–52. [76] Welsch M, Howells M, Bazilian M, DeCarolis JF, Hermann S, Rogner HH. Modelling elements of smart grids – enhancing the OSeMOSYS (open source energy modelling system) code. Energy 2012;46(1):337–50. [77] Niemi R, Lund PD. Alternative ways for voltage control in smart grids with distributed electricity generation. Int J Energy Res 2012;36(10):1032–43. [78] Finn P, Fitzpatrick C. Demand side management of industrial electricity consumption: Promoting the use of renewable energy through real-time pricing. Appl Energy 2014;113:11–21. [79] Warren P. A review of demand-side management policy in the UK. Renew Sustain Energy Rev 2014;29:941–51. [80] Hedegaard K, Mathiesen BV, Lund H, Heiselberg P. Wind power integration using individual heat pumps – analysis of different heat storage options. Energy 2012;47(1):284–93.

[81] Amoroso FA, Cappuccino G. Advantages of efficiency-aware smart charging strategies for PEVs. Energy Convers Manag 2012;54(1):1–6. [82] Lindgren J, Niemi R, Lund PD. Effectiveness of smart charging of electric vehicles under power limitations. Int J Energy Res 2013. [83] Mets K, D'Hulst R, Develder C. Comparison of intelligent charging algorithms for electric vehicles to reduce peak load and demand variability in a distribution grid. J Commun Netw 2012;14(6):672–81. [84] Mu Y, Wu J, Jenkins N, Jia H, Wang C. A spatial-temporal model for grid impact analysis of plug-in electric vehicles. Appl Energy 2014;114:456–65. [85] Sundström O, Binding C. Flexible charging optimization for electric vehicles considering distribution grid constraints. IEEE Trans Smart Grid 2012;3 (1):26–37. [86] Tikka V, Lassila J, Makkonen H, Partanen J. Case study of the load demand of electric vehicle charging and optimal charging schemes in an urban area; 2012. [87] Järventausta P, Repo S, Rautiainen A, Partanen J. Smart grid power system control in distributed generation environment. Annu Rev Control 2010;34 (2):277–86. [88] Kádár P. Application of optimization techniques in the power system control. Acta Polytech Hung 2013;10(5):221–36. [89] Pradeep Y, Seshuraju P, Khaparde SA, Joshi RK. Flexible open architecture design for power system control centers. Int J Electr Power Energy Syst 2011;33(4):976–82. [90] Shuto T, Nagata M, Yoshimura K, Sugiuchi T, Takeshita M, Yonei K. A study on power system control considering both transient stability and voltage stability. IEEJ Trans Power Energy 2013;133(10):740–5. [91] Westermann D, Kratz M. A real-time development platform for the next generation of power system control functions. IEEE Trans Ind Electron 2010;57(4):1159–66. [92] Azadeh A, Babazadeh R, Asadzadeh SM. Optimum estimation and forecasting of renewable energy consumption by artificial neural networks. Renew Sustain Energy Rev 2013;27:605–12. [93] Hossain R, Maung Than OoA, Shawkat Ali ABM. Historical weather data supported hybrid renewable energy forecasting using artificial neural network (ANN). Energy Proc 2012;14:1035–40. [94] Kontu K, Fang T-, Lahdelma R. Forecasting district heating consumption based on customer measurements. Euroheat Power (Engl. Ed) 2013;10 (3):16–20. [95] Salonen K, Niemelä S, Fortelius C. Application of radar wind observations for low-level NWP wind forecast validation. J Appl Meteorol Climatol 2011;50 (6):1362–71. [96] Ulbricht R, Fischer U, Lehner W, Donker H. Optimized renewable energy forecasting in local distribution networks. ACM Int Conf Proc Ser 2013:262–6. [97] Holmgren M, Haarla L, Matilainen J, Holttinen H. Power regulation resources required by wind power in Finland and regulation characteristics of power plants. 2009 CIGRE/EEE PES joint symposium: integration of wide-scale renewable resources into the power delivery system; 2009. [98] Østergaard PA. Regulation strategies of cogeneration of heat and power (CHP) plants and electricity transit in Denmark. Energy 2010;35(5):2194–202. [99] Lund P. Large-scale urban renewable electricity schemes – integration and interfacing aspects. Energy Convers Manag 2012;63:162–72. [100] Lund H, Andersen AN, Østergaard PA, Mathiesen BV, Connolly D. From electricity smart grids to smart energy systems – a market operation based approach and understanding. Energy 2012;42(1):96–102. [101] Connolly D, Lund H, Mathiesen BV, Pican E, Leahy M. The technical and economic implications of integrating fluctuating renewable energy using energy storage. Renew Energy 2012;43:47–60. [102] Rinne S, Syri S. Heat pumps versus combined heat and power production as CO2 reduction measures in Finland. Energy 2013;57:308–18. [103] Kaldellis JK, Zafirakis D, Kavadias K. Techno-economic comparison of energy storage systems for island autonomous electrical networks. Renew Sustain Energy Rev 2009;13(2):378–92. [104] Schoenung SM. Characteristics and technologies for long- vs. short-term energy storage. New Mexico, California: Sandia National Laboratories; 2001. [105] Schoenung S. Energy storage systems cost update. New Mexico, California: Sandia National Laboratories; 2011. [106] Schoenung SM, Hassenzahl WV. Long- vs. short-term energy storage technologies analysis: A life-cycle cost study. New Mexico, California: Sandia National Laboratories; 2003. [107] Poonpun P, Jewell WT. Analysis of the cost per kilowatt hour to store electricity. IEEE Trans Energy Convers 2008;23(2):529–34. [108] Brealey RA, Myers SC, Allen F. Principles of corporate finance. New York, NY: McGraw-Hill/Irwin; 2011. [109] Ferreira HL, Garde R, Fulli G, Kling W, Lopes JP. Characterisation of electrical energy storage technologies. Energy 2013;53:288–98. [110] Electric Power Research Institute (EPRI). Electric energy storage technology options: a white paper primer on applications, costs, and benefits. Palo Alto, California: EPRI; 2010. [111] Akhil AA, Huff G, Currier AB, Kaun BC, Rastler DM, Chen SB, et al. DOE/EPRI 2013 electricity Storage handbook in collaboration with NRECA. New Mexico, California: Sandia National Laboratories; 2013. [112] Kintner-Meyer M, Balducci PJ, Jin C, Nguyen TB, Elizondo MA, Viswanathan VV, et al. Energy storage for power systems applications: a regional assessment for the northwest power pool (NWPP). Richland, WA (US): Pacific Northwest National Laboratory (PNNL); 2010.

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

[113] Battke B, Schmidt TS, Grosspietsch D, Hoffmann VH. A review and probabilistic model of lifecycle costs of stationary batteries in multiple applications. Renew Sustain Energy Rev 2013;25:240–50. [114] Abrams A, Fioravanti R, Harrison J, Katzenstein W, Kleinberg M, Lahiri S, et al. Energy storage cost-effectiveness methodology and preliminary results. California, USA: DNV KEMA Energy and Sustainability, California Energy Commission; 2013. [115] Auer J, Keil J. State-of-the-art electricity storage systems: Indispensable elements of the energy revolution. Frankfurt am Main, Germany: Deutsche Bank AG; 2012. [116] Connolly D. A review of energy storage technologies for the integration of fluctuating renewable energy. Limerick, Ireland: University of Limerick; 2010. [117] Danish Energy Agency. Generation of electricity and district heating, energy storage and energy carrier generation and conversion: technology data for energy plants. Denmark: Energi Styrelse; 2012. [118] Electric Power Research Institute (EPRI). Cost-effectiveness of energy storage in California: application of the energy storage valuation tool to inform the California public utility commission proceeding R, 10-12-007. Palo Alto, California: EPRI; 2013. [119] EASE/EERA. Joint EASE/EERA recommendations for a European energy storage technology development roadmap towards 2030. Brussels: The European Association for Storage of Energy (EASE) and the European Energy Research Alliance (EERA); 2013. [120] Hittinger E, Whitacre JF, Apt J. What properties of grid energy storage are most valuable. J Power Sources 2012;5/15(206):436–49. [121] Inage S. Prospects for large-scale energy storage in decarbonised power grids. Paris: OECD/IEA; 2009. [122] Joint Research Centre (JRC). Technology map of the European strategic energy technology plan (SET-plan). Petten, The Netherlands: JRC European Commission/Institute for Energy and Transport; 2011. [123] Kintner-Meyer M, Jin C, Balducci P, Elizondo M, Guo X, Nguyen T, et al. Energy storage for variable renewable energy resource integration – a regional assessment for the Northwest Power Pool (NWPP). 2011 IEEE/PES power systems conference and exposition, PSCE 2011. 2011. [124] Lund H, Salgi G. The role of compressed air energy storage (CAES) in future sustainable energy systems. Energy Convers Manag 2009;50(5):1172–9. [125] Lund H, Salgi G, Elmegaard B, Andersen AN. Optimal operation strategies of compressed air energy storage (CAES) on electricity spot markets with fluctuating prices. Appl Therm Eng 2009;29(5–6):799–806. [126] Salgi G, Lund H. System behavior of compressed-air energy-storage in Denmark with a high penetration of renewable energy sources. Appl Energy. 2008;85(4):182–9. [127] Schoenung SM, Hassenzahl W. Long- vs. short-term energy storage: Sensitivity analysis. New Mexico, California: Sandia National Laboratories; 2007. [128] Schoenung S. Economic analysis of large-scale hydrogen storage for renewable utility applications. Albuquerque: Sandia National Laboratory, (NM) (2011 Aug) Report No.: SAND20114845. Contract No.: DEAC0494AL85000; 2011. [129] Steward D, Saur G, Penev M, Ramsden T. Lifecycle cost analysis of hydrogen versus other technologies for electrical energy storage. US: National Renewable Energy Laboratory (NREL); 2009. [130] Tan X, Li Q, Wang H. Advances and trends of energy storage technology in microgrid. Int J Electr Power Energy Syst 2013;44(1):179–91. [131] European Commission, “Eurostat” [online]. Available: 〈http://epp.eurostat.ec. europa.eu/portal/page/portal/eurostat/home〉; 2014. [132] Electric Power Research Institute (EPRI). Quantifying the value of hydropower in the electric grid: modeling results for future scenarios. Palo Alto, California: EPRI; 2012. [133] Connolly D, MacLaughlin S, Leahy M. Development of a computer program to locate potential sites for pumped hydroelectric energy storage. Energy 2009;35(1):375–81. [134] Yang C-, Jackson RB. Opportunities and barriers to pumped-hydro energy storage in the united states. Renew Sustain Energy Rev 2011;15(1):839–44. [135] Gravity Power LCC. GPM system overview and operation [online]. Available: 〈http://www.gravitypower.net/Technology.aspx〉; 2014. [136] Slocum AH, Fennell GE, Dündar G, Hodder BG, Meredith JDC, Sager MA. Ocean renewable energy storage (ORES) system: analysis of an undersea energy storage concept. Proc IEEE 2013;101(4):906–24. [137] Pickard WF. The history, present state, and future prospects of underground pumped hydro for massive energy storage. Proc IEEE 2012;100(2):473–83. [138] Electric Power Research Institute (EPRI). Quantifying the value of Hydropower in the electric grid: plant cost elements. Palo Alto, California: EPRI; 2011. [139] Ardizzon G, Cavazzini G, Pavesi G. A new generation of small hydro and pumped-hydro power plants: advances and future challenges. Renew Sustain Energy Rev 2014;3(31):746–61. [140] Axpo Holding AG. Linthal 2015 project [online]. Available: 〈http://www.axpo. com/axpo/ch/en/axpo-erleben/linthal-2015.html〉; 2013. [141] Danish Energy Agency (energi styrelse). Underground storage of gas [online]. Available: 〈http://www.ens.dk/en〉; 2013. [142] Madlener R, Latz J. Economics of centralized and decentralized compressed air energy storage for enhanced grid integration of wind power. Appl Energy 2013;101:299–309. [143] Drury E, Denholm P, Sioshansi R. The value of compressed air energy storage in energy and reserve markets. Energy 2011;36(8):4959–73. [144] Wolf D, Kanngießer A, Budt M, Doetsch C. Adiabatic compressed air energy storage co-located with wind energy-multifunctional storage commitment

[145] [146] [147] [148]

[149]

[150]

[151]

[152]

[153]

[154]

[155]

[156]

[157]

[158] [159] [160]

[161]

[162]

[163] [164] [165]

[166] [167] [168] [169]

[170]

[171]

[172]

[173]

[174]

595

optimization for the German market using GOMES. Energy Syst 2012;3 (2):181–208. Gu Y, McCalley J, Ni M, Bo R. Economic modeling of compressed air energy storage. Energies 2013;6(4):2221–41. Yucekaya A. The operational economics of compressed air energy storage systems under uncertainty. Renew Sustain Energy Rev 2013;22:298–305. Safaei H, Keith DW. Compressed air energy storage with waste heat export: an Alberta case study. Energy Convers Manag 2014;2(78):114–24. Ibrahim H, Younès R, Ilinca A, Dimitrova M, Perron J. Study and design of a hybrid wind-diesel-compressed air energy storage system for remote areas. Appl Energy 2010;87(5):1749–62. Swider DJ. Compressed air energy storage in an electricity system with significant wind power generation. IEEE Trans Energy Convers 2007;22 (1):95–102. Townsend AK, Webber ME. Optimization of technical and operational characteristics of a CAES facility in West Texas to balance intermittent wind power. In: Proceedings of the ASME 2011 5th International Conference on Energy Sustainability, ES 2011; 2011. Denholm P. Improving the technical, environmental and social performance of wind energy systems using biomass-based energy storage. Renew Energy 2006;31(9):1355–70. Beacon Power. Smart energy matrix, 20 MW frequency regulation plant [online]. Available: 〈http://www.beaconpower.com/files/SEM_20MW_2010. pdf〉; 2011. Sebastian R, Pena-Alzola R, Quesada J, Colmenar A. Sizing and simulation of a low cost flywheel based energy storage system for wind diesel hybrid systems. In: Proceedings of the 2012 IEEE International Energy Conference and Exhibition. ENERGYCON 2012. 2012. p. 495–500. Wang D, Ren C, Sivasubramaniam A, Urgaonkar B, Fathy H. Energy storage in datacenters: what, where, and how much? Perform Eval Rev 2012;40:187–98 (1 SPEC. ISS.). Pacific Northwest National Laboratory (US Department of Energy, DOE). Flywheel energy storage: an alternative to batteries for uninterruptible power supply systems. Washington, DC: Federal Energy Management Program/ DOE; 2003. Abele A, Elkind E, Intrator J, Washom B, et al. 2020 strategic analysis of energy storage in California. Sacramento, California: California Energy Commission; 2011. Eyer J. Benefits from flywheel energy storage for area regulation in California – demonstration results. New Mexico, California: Sandia National Laboratories; 2009. Divya KC, Østergaard J. Battery energy storage technology for power systems – an overview. Electr Power Syst Res 2009;79(4):511–20. Liu C, Li F, Lai-Peng M, Cheng H-. Advanced materials for energy storage. Adv Mater 2010;22(8):E28–62. Yang Z, Zhang J, Kintner-Meyer MCW, Lu X, Choi D, Lemmon JP, et al. Electrochemical energy storage for green grid. Chem Rev 2011;111 (5):3577–613. Koohi-Kamali S, Tyagi VV, Rahim NA, Panwar NL, Mokhlis H. Emergence of energy storage technologies as the solution for reliable operation of smart power systems: a review. Renew Sustain Energy Rev 2013;25:135–65. Schoenung SM, Eyer J. Benefit/cost framework for evaluating modular energy storage: a study for the DOE energy storage systems program. New Mexico, California: Sandia National Laboratories; 2008. NGK Insulators LTD. NAS batteries, [online]. Available: 〈http://www.ngk.co. jp/english/products/power/nas/〉; 2014. International Electrotechnical Commission (IEC). Electrical energy storagewhite paper. Geneva, Switzerland: IEC; 2011. Palomares V, Serras P, Villaluenga I, Hueso KB, Carretero-González J, Rojo T. Na-ion batteries, recent advances and present challenges to become low cost energy storage systems. Energy Environ Sci 2012;5(3):5884–901. Ellis BL, Nazar LF. Sodium and sodium-ion energy storage batteries. Current Opin Solid State Mater Sci 2012;16(4):168–77. AES Energy Storage. AES energy storage projects [online]. Available: 〈http:// www.aesenergystorage.com/〉; 2013. Leadbetter J, Swan LG. Selection of battery technology to support gridintegrated renewable electricity. J Power Sources 2012;216:376–86. Lloyd D, Vainikka T, Kontturi K. The development of an all copper hybrid redox flow battery using deep eutectic solvents. Electrochim Acta 2013;6/30 (100):18–23. Lloyd D, Vainikka T, Ronkainen M, Kontturi K. Characterisation and application of the fe(II)/fe(III) redox reaction in an ionic liquid analogue. Electrochim Acta 2013;109:843–51. Peljo P, Lloyd D, Doan N, Majaneva M, Kontturi K. Towards a thermally regenerative all-copper redox flow battery. Phys Chem Chem Phys 2014;16 (7):2831–5. Chakrabarti MH, Mjalli FS, AlNashef IM, Hashim MA, Hussain MA, Bahadori L, et al. Prospects of applying ionic liquids and deep eutectic solvents for renewable energy storage by means of redox flow batteries. Renew Sustain Energy Rev 2014;2(30):254–70. Kear G, Shah AA, Walsh FC. Development of the all-vanadium redox flow battery for energy storage: a review of technological, financial and policy aspects. Int J Energy Res 2012;36(11):1105–20. Jossen A, Sauer D. Advances in redox-flow batteries. In: Proceedings of the first international renewable energy storage conference (IRES I) – the case of energy autonomy: storing renewable energies. Gelsenkirchen, Germany;

596

[175]

[176]

[177] [178]

[179]

[180] [181]

[182]

[183]

[184]

[185]

[186]

[187]

[188]

[189]

[190]

[191]

B. Zakeri, S. Syri / Renewable and Sustainable Energy Reviews 42 (2015) 569–596

Bonn, Germany: EUROSOLAR and the World Council for Renewable Energy (WCRE); 30–31 October 2006. Skyllas-Kazacos M, Chakrabarti MH, Hajimolana SA, Mjalli FS, Saleem M. Progress in flow battery research and development. J Electrochem Soc 2011;158(8):R55–79. Hartikainen T, Mikkonen R, Lehtonen J. Environmental advantages of superconducting devices in distributed electricity-generation. Appl Energy 2007;1 (84):29–38. Wang W, Luo Q, Li B, Wei X, Li L, Yang Z. Recent progress in redox flow battery research and development. Adv Funct Mater 2013;23(8):970–86. Leung P, Li X, Ponce De León C, Berlouis L, Low CTJ, Walsh FC. Progress in redox flow batteries, remaining challenges and their applications in energy storage. RSC Adv 2012;2(27):10125–56. Hall PJ, Mirzaeian M, Fletcher SI, Sillars FB, Rennie AJR, Shitta-Bey GO, et al. Energy storage in electrochemical capacitors: designing functional materials to improve performance. Energy Environ Sci 2010;3(9):1238–51. Liu S, Sun S, You X-. Inorganic nanostructured materials for high performance electrochemical supercapacitors. Nanoscale 2014;6(4):2037–45. Zhou Z, Benbouzid M, Frédéric Charpentier J, Scuiller F, Tang T. A review of energy storage technologies for marine current energy systems. Renew Sustain Energy Rev 2013;2(18):390–400. Pandey SK, Mohanty SR, Kishor N. A literature survey on load–frequency control for conventional and distribution generation power systems. Renew Sustain Energy Rev 2013;9(25):318–34. Zhu J, Zhang H, Yuan W, Zhang M, Lai X. Design and cost estimation of superconducting magnetic energy storage (SMES) systems for power grids. IEEE power and energy society general meeting; 2013. Hahn H, Krautkremer B, Hartmann K, Wachendorf M. Review of concepts for a demand-driven biogas supply for flexible power generation. Renew Sustain Energy Rev 2014;29:383–93. Siemens AG. The most versatile fuel [online]. Available: 〈http://www.sie mens.com/innovation/apps/pof_microsite/_pof-spring-2012/_html_en/elec trolysis.html〉; 2012. Pickard WF, Abbott D. Addressing the intermittency challenge: massive energy storage in a sustainable future [scanning the issue]. Proc IEEE 2012;100(2):317–21. Aguado M, Ayerbe E, Azcárate C, Blanco R, Garde R, Mallor F, et al. Economical assessment of a wind-hydrogen energy system using WindHyGens software. Int J Hydrogen Energy 2009;34(7):2845–54. Lohner T, D'Aveni A, Dehouche Z, Johnson P. Integration of large-scale hydrogen storages in a low-carbon electricity generation system. Int J Hydrogen Energy 2013;38(34):14638–53. Kaldellis JK, Kavadias K, Zafirakis D. The role of hydrogen-based energy storage in the support of large-scale wind energy integration in island grids. Int J Hydrogen Energy 2013. Kloess M. Electric storage technologies for the future power system – an economic assessment. In: Proceedings of the 9th international conference on the European Energy Market (EEM) vol. 12; 2012. Beaudin M, Zareipour H, Schellenberglabe A, Rosehart W. Energy storage for mitigating the variability of renewable electricity sources: an updated review. Energy Sustain Dev 2010;14(4):302–14.

[192] Mahlia TMI, Saktisahdan TJ, Jannifar A, Hasan MH, Matseelar HSC. A review of available methods and development on energy storage; technology update. Renew Sustain Energy Rev 2014;33:532–45. [193] Vazquez S, Lukic SM, Galvan E, Franquelo LG, Carrasco JM. Energy storage systems for transport and grid applications. IEEE Trans Ind Electron 2010;57 (12):3881–95. [194] Introduction to probability and statistics: Principles and applications for engineering and the computing sciences. In: Milton JS, Arnold JC, editors. 4th ed.. Boston, the USA: McGraw-Hill; 2003. [195] Goh YM, Newnes L, McMahon C, Mileham A, Paredis CJJ. A framework for considering uncertainty in quantitative life cycle cost estimation. In: Proceedings of the ASME design engineering technical conference. 2009;8(Parts A and B): p. 3–13. [196] Pfenninger S, Hawkes A, Keirstead J. Energy systems modeling for twentyfirst century energy challenges. Renew Sustain Energy Rev 2014;5(33):74–86. [197] Black M, Strbac G. Value of bulk energy storage for managing wind power fluctuations. IEEE Trans Energy Convers 2007;22(1):197–205. [198] Denholm P, Kulcinski GL. Life cycle energy requirements and greenhouse gas emissions from large scale energy storage systems. Energy Convers Manag 2004;45(13-14):2153–72. [199] Pickard WF. A nation-sized battery? Energy Policy 2012;45:263–7. [200] Barnhart CJ, Benson SM. On the importance of reducing the energetic and material demands of electrical energy storage. Energy Environ Sci 2013;6:1083–92. http://dx.doi.org/10.1039/C3EE24040A. [201] Chatzivasileiadi A, Ampatzi E, Knight I. Characteristics of electrical energy storage technologies and their applications in buildings. Renew Sustain Energy Rev 2013;25:814–30. [202] Carson RT, Novan K. The private and social economics of bulk electricity storage. J Environ Econ Manag 2013;66(3):404–23. [203] Hu Z, Jewell WT. Optimal power flow analysis of energy storage for congestion relief, emissions reduction, and cost savings. 2011 In: Proceedings of the IEEE/PES power systems conference and exposition, PSCE 2011; 2011. [204] Denholm P, Jorgenson J, Hummon M, Jenkin T, Palchak D, Kirby B, et al. The value of energy storage for grid applications. Contract 2013;303:275–3000. [205] Eyer J, Corey G. Energy storage for the electricity grid: benefits and market potential assessment guide. Sandia National Laboratories; Albuquerque, New Mexico, Livermore, California; 2010. [206] Foley A, Díaz Lobera I. Impacts of compressed air energy storage plant on an electricity market with a large renewable energy portfolio. Energy 2013;8/1 (57):85–94. [207] Awad ASA, Fuller JD, EL-Fouly THM, Salama MMA. Impact of energy storage systems on electricity market equilibrium. IEEE Trans Sustain Energy 2014;5 (3):875–85. [208] Sioshansi R, Denholm P, Jenkin T. Market and policy barriers to deployment of energy storage 6. Econ Energy Environ Policy J 2012;1(2):47. [209] Bhatnagar D, Currier A, Hernandez J, Ma O, Kirby B. Market and policy barriers to energy storage deployment. Albuquerque, New Mexico, Livermore, California: Sandia National Laboratories; 2013. [210] Wasowicz B, Koopmann S, Dederichs T, Schnettler A, Spaetling U. Evaluating regulatory and market frameworks for energy storage deployment in electricity grids with high renewable energy penetration. In: Proceedings of the 9th international conference on the european energy market (EEM); vol. 12: 2012.