Energy storage opportunities and trends
Historically, electricity storage has been deployed by power companies, whether network operators, generators or vertically integrated utilities, but more recently, there has been interest from project developers seeking to benefit from a more liberal electricity market.
Author: Anthony Price
Several large-scale battery storage projects have been completed in Great Britain and northern Ireland since 2010, and more are under construction, for applications including reductions in network constraints, integration of renewable energy, provision of frequency response, other ancillary services and improvement of the end user's service. As energy storage systems’ costs decrease, opportunities to use energy storage at all scales become more commercially attractive. Projects in Japan, China and the USA demonstrate that electricity storage technologies can be deployed at large scales, with project sizes of 50 MW or more. Case studies show how energy storage can be used on its own, as a merchant operation, or in conjunction with other assets on the electricity network. Though battery projects have a high profile, are relatively quick to construct and have low operating and maintenance costs, other storage technologies may also be used such as flywheels, thermal cycles and compressed or liquefied air. These different technologies have different operating parameters, which may influence the technology selection for a particular application and careful consideration is required when selecting a location for storage projects, taking into account local conditions. The improvement in performance, and cost reductions indicate increased deployment of electricity storage at different scales and for many applications in the evolving smart grid.
The increased interest in electricity storage can be attributed to several factors including:
- Increased availability of advanced technology for distributed energy storage.
- An increasing proportion of electricity generated from variable renewable resources such as solar PVC and wind, leading to consideration of the use of storage for energy balancing and the provision of ancillary services such as reserve power, frequency regulation and constraint management.
- Increased attention on the role of the smart grid as the electricity network changes from a passive, centralised system into an actively managed distributed system.
Historically, electricity storage has been deployed by power companies, whether network operators, generators or vertically integrated utilities, but more recently, there has been interest from project developers seeking to benefit from a more liberal electricity market. The investment by project finance companies in providing services such as short-term operating reserve (STOR) has expanded into consideration of provision of other services such as frequency response and demand side response. There is now commercial interest in the use of storage technologies across the energy spectrum, both in standalone applications, and in conjunction with other assets such as solar farms and industrial or commercial buildings.
This article considers recent developments and looks at trends in the energy storage industry.
‘Storage’ in the electricity system is the conversion of electricity into a form of energy which can be kept, the keeping of the energy which has been so converted and the reconversion of that energy into electrical energy.
‘Storage facility’ in the electricity system means a facility which converts electricity into a form of energy which can be kept, keeps the energy which has been so converted and reconverts that energy into electrical energy.
Purpose of storage
Electricity is unlike other energy vectors in that production has to be kept in balance with consumption in order that the electricity system can remain stable. Electricity storage is used to introduce a delay between the production of electricity and its final use: the delay may be only short term, or it may extend over hours, days or longer.
Maintaining the balance between supply and demand over short time periods (second by second) is used to maintain frequency and keep the system stable. Over longer time periods, (minutes or hours) storage is used to lower the costs of producing electricity, by transferring production from times of high cost to times of lower cost.
Energy storage technologies used for applications on the electricity power systems can be broadly categorised by the type of storage media . Only capacitors and superconducting electromagnetic energy storage store electricity as electricity, other types use a different form of potential energy as the medium. Mechanical types may use flywheels, with energy stored in a rotating mass, many systems are now based on grouping together banks of flywheels to meet large-scale power requirements. Pumped hydrosystems store energy stored as gravitational potential energy, water pumped to a higher reservoir can drive turbines when the water is released from its elevated height.
Compressed air systems use containers from steel or other high strength materials, or geological formations such as excavated salt caverns or disused mine shafts to contain air at high pressure . The air drives a turbine when it expands, and performance can be enhanced with thermal recovery or by mixing the high-pressure air with fuel in the combustion chamber of a gas turbine. Chemical and electrochemical types include rechargeable batteries and thermal techniques include heat and cold stores. Some manufacturers and developers are considering hybrid systems such as flywheel and compressed air combinations, in order to combine the most attractive benefits of each technology.
An energy storage facility brings together an energy storage medium (such as a battery, flywheel or other storage device) usually with a power conversion system for converting AC input power to DC and vice versa, along with a control and management system.
The early electricity networks, typically constructed and operated in the period 1880–1910 and based on direct current systems, were hybrid systems combining both generation and battery energy storage systems. A battery system supported the local network in Colchester in the 1880s . Some battery systems, even at a large megawatt scale, persisted until the 1930s where DC systems remained operational.
The development of hydroelectric power was soon followed by the introduction of pumped hydroelectricity storage. Early systems used separate pumps and turbines for the charging and discharging process, later systems, developed from the 1930s used reversible pumps and turbines. The number and size of pumped hydroelectricity storage facilities has increased, and modern pumped hydrosystems are often based on pump/turbine/alternator sets in the size range of 500 MW. Worldwide, the total installed capacity of pumped hydroelectricity is estimated to be 144 GW . There are four operating pumped storage plants in Great Britain, the most recently constructed plant at Dinorwig in North Wales, commissioned in 1984 has six motor/turbine sets, each rated at 312 MVA for pumping and 330 MVA for generation .
In comparison, the total worldwide capacity of other types of electricity storage, connected to power networks for charging and discharging to an electricity network is still in the region of 1–2 GW in 2016. Nevertheless, recent interest, and falling costs, particularly for battery storage systems, shows that new network connected projects are being installed at a rate approaching 1 GW/year worldwide .
Project development timetables are relatively short for advanced electricity storage. Pre-project work to identify a site, obtain planning permission or other consents and a network connection may take from 3 months to a year, but on site work is now expected to be between 9 and 18 months for most battery projects, with similar timescales for other advanced technologies. Large-scale projects such as compressed air, which require the use of underground geological formations have construction timescales of 2 years or more, and pumped hydro is in the range of 3 years to 10 years or more.
The long lifetimes of mechanically based storage technologies such as compressed air, liquid air and pumped hydro result in lower whole life costs, and with the correct financing arrangements such energy storage projects can be very cost effective.
Electricity storage technology development
Research, development and refinement are active in all of the three key areas of the energy storage: the storage medium, the power conversion system and the control and management system.
Though lead–acid batteries were first developed by Gaston Plante in 1859, they are still in widespread use, not only for vehicle starting lighting and ignition applications, but also for off grid and back-up applications. There is a well-developed and efficient recycling infrastructure which can monetise the scrap value of the lead. Developments such as carbon doping of the electrodes and bipolar electrodes have improved the operational performance and lifetime of the batteries . They provide a relatively low cost and acceptable form of energy storage, and recent large-scale projects include a 36 MW project at Notrees, Goldsmith, Texas (though this will be refurbished with lithium-ion batteries in 2016) and a number of projects in Hawaii.
The first lithium-ion battery electrochemistry was developed in 1991, originally for telephones, cameras and computers, but several manufacturers sought to expand their markets, and after applying the technology to electric vehicles lithium-ion batteries are now used for grid applications. In the period 2014–2015, numerous installations in the 10 MW or greater size range have been completed in America, Europe and Asia . Extensive R&D to develop improved lithium-based batteries is ongoing, and it is difficult to provide an up-to-date review of the current state of the art.
Lithium-based batteries are extensively used in modern electric vehicles. The trend is expanding from using batteries in electric vehicles to other stationary applications, either reusing EV battery packs that are beyond their useful life for traction purposes in stationary storage or to use EV batteries as grid connected storage during recharging. This needs careful consideration as any use of the battery has implications on its degradation and future performance, and this must be balanced against the financial rewards from such a service. Nevertheless, this is an emerging area, and with high investment in research and development in the smart grid arena from governments, agencies, power companies, the battery industry and others, technical and commercial problems are likely to be solved and integration of electric vehicles into a two-way part of the network will be achieved.
Several companies have announced storage suitable for domestic use. Tesla, an EV manufacturer has produced a battery system which can be integrated with a domestic control system, other manufacturers including Moixa and Powervault have developed battery and control systems specifically for the domestic and small commercial market. However, it is not just the battery that is important, the control methodology is also significant, and some manufacturers, for example, Stem Inc., are concentrating on the power conversion system and the system integration including software.
High-temperature batteries, principally based on either the sodium sulphur chemistry, or the sodium nickel halide chemistry have continued to be deployed in numerous applications where high power and high energy are required. These systems are based on the use of a solid beta alumina electrolyte and liquid molten sodium and sulphur electrodes. The battery packs are thermally insulated in order to keep the battery cells at their operating temperature. The batteries have a high-energy density and high efficiency. Reference (battery briefing): A large battery system of 50 MW output and 300 MWh capacity began operation in March 2016 at Busen in Japan after a 9 month construction programme. The batteries, supplied by NGK Insulators Ltd., are installed in 252 shipping containers on a 14,000 m 2 site. The battery is used by the local power company to balance supply and demand.
Conventional batteries such as lithium-ion, lead–acid or the high-temperature types are limited in the energy storage capacity by the amount of active materials contained within the battery cells. An alternative battery type combines the attributes of a fuel cell with a rechargeable battery. Instead of the electrolytes being contained entirely within the cell, the electrolytes are stored in external tanks and pumped through the system. This is a cost-effective method to increase the energy storage capacity of the battery. Three are several manufacturers with commercial products as well as a number of companies at an early stage of development. Many systems are based on the all-vanadium system or the zinc bromine system, other combinations of electrolytes are being developed based on iron chrome, zinc chlorine and hydrogen bromine. Further improvements such as using air as an electrolyte could bring further cost reductions. A modular vanadium flow battery, the Cellstrom FB10/100 supplied by Gildemeister AG has been in continuous operation on a dairy farm in Veirakker, the Netherlands since 2009. This 15 kW/100 kWh battery, operating in conjunction with a 50 kWp solar panel has been used to maximise solar power self-consumption and to reduce power flows to and from the local network . Large-scale flow battery installations include a 5 MW/10 MWh all-vanadium system supplied by Rongke Power in Dalian, China, and a 60 MWh system with a nominal power output of 15 MW and maximum 30 MW, installed in January 2016 in the Minami-Hayakita substation in Japan and supplied by Sumitomo Electric Industries. A similar installation is shown in Fig 1.
As well as mature technologies such as nickel cadmium, nickel iron, other new types of batteries are constantly being developed including zinc air, sodium ion and ionic liquids. In general terms, battery development is a lengthy process, and several years’ development and testing are required before a new battery technology can be considered commercial.
Environmental and safety issues
Most energy storage systems operate as sealed systems, with little or no emissions during normal service. One of the benefits of integrating electricity storage into a power system is that the overall efficiency of system is improved, which brings environmental improvements such as reductions in emissions through reducing the need for fossil fired peaking plant, for example. However, there is concern that safety, and the environmental implications of an unexpected event are considered, especially as incidents such as battery fires, or catastrophic failures of flywheels can cause widespread damage.
A full risk assessment, which considers the local environmental impact should be carried out at the design stage, as this will influence the choice of technology and location for the specific application.
The recycling and disposal of batteries is regulated by European and national legislation and information on compliance should be available from suppliers.
Network connected applications
A major use for battery and flywheel energy storage has been and continues to be uninterruptible power supplies (UPSs) and back-up power. UPS are available in various configurations and sizes from watts to multi-MW systems. At the small end, integrated UPS systems are able to operate in line with a domestic computer, and these provide protection against both transients (momentary variations in current, voltage or frequency), on the network, or interruptions to the power supply. The system needs sufficient ride-through power to cover short faults, and if necessary to also provide reserve power for an hour or even more. The UPS may send a signal to the PC instructing it to begin a safe shutdown if a power system fault occurs and the reserve battery power falls below a safe level.
Similar systems, but at a much larger scale, are in operation in data centres, commercial properties and industry worldwide. Some systems are hybrid, perhaps using a flywheel (or flywheels) to maintain the load for seconds, before handing over the duty to a battery or to a standby generator or both.
These UPS applications are valued against the cost of lost load, and in many cases it is preferred to maintain the security of the UPS by not assigning any other use to the storage plant.
Customer side of meter applications
Energy storage can be used as a means of storing electrical energy drawn from the network for discharge back to the network at a later time, when the electrical energy has more value, either technically or commercially. Customer side renewable generation such as solar PV or from charging from the network at a low tariff rate offers.
Battery manufacturers and electric vehicle manufacturers are able to work collaboratively to develop products that cross-over between transport applications and stationary applications. Some demonstration projects are in place which illustrate how batteries, previously used in electric vehicles, can be reused as standalone storage, or in applications where, still connected in the electric vehicle, the battery provides services either to the consumer or to the network.
Network side of meter applications
Energy storage can be applied on the network side of the meter, either to benefit the network, as in the example given above for Leighton Buzzard, or to benefit a sole user such as a storage plant operator or a generator. Further uses are under consideration such as provision of storage to local community energy projects.
As well as being used to ‘trade energy’, such devices can also be used to supply ancillary services to the network, and also to supply support services to the local network operator. This range of services provides numerous opportunities and an increased level of complexity in planning and designing storage plants.
Currently, National Grid Electricity Transmission a subsidiary of National Grid plc., purchase a number of services from market participants. Some of these are shown in Table 1 .
|SERVICE||RESPONSE TIME||DURATION||MINIMUM POWER||RELEVANCE FOR STORAGE|
|enhanced frequency response||<1 s||15 min||1 MW||suitable for fast acting storage|
|primary response||<10 s||30 s||3 MW||suitable for storage|
|secondary response||<30 s||30 min||3 MW||suitable for storage|
|high response||<10 s||unlimited||3 MW||possible, for long duration types|
|fast reserve||<2 min||15 min||50 MW||possible|
|STOR||<4 h||2 h||1 MW||possible|
|black start||–||varies||varies||location specific|
Table 1: Ancillary services in Great Britain
Many of these services can be provided by storage of many types including pumped hydro, compressed air and batteries. Though many more projects now use storage to provide fast acting ancillary services such as frequency response, practical examples include a 17 MW, 14.4 MWh lead–acid battery in Steglitz, West Berlin, which provided frequency response up to 1994 and a 10 MW lithium-ion battery installed in 2016 by AES at Kilroot Power Station in northern Ireland.
Project developers will need to consider many factors including:
- The application: This will determine the power and energy rating of the energy storage facility and have a significant bearing on the location.
- Future applications.
- Details of the electrical connection to the network including connection voltage, real and reactive power requirements.
- Preferences for project management: If there is a requirement to develop using an EPC contract, the selection of the contractor may itself determine the choice of the battery or other energy storage medium supplier, the power conversion system and the choice of the system integrator.
Containerised or building based
Many energy storage suppliers now supply systems (especially batteries) packaged in ISO shipping containers, which can be delivered in an advanced state of preparation to the operating site. These have several advantages: they are relatively simple to deploy and on site work may be reduced significantly, but they are not a universally suitable solution.
Shipping containers may not be visually attractive and acceptable in certain areas, they require to be sited on adequate foundations (which could represent a sizable proportion of a new building cost), and they may involve higher operation and maintenance costs. Many battery types require temperature control and this is easier in a single large building rather than in individual containers. Access to shipping containers for repair or replacement of components such as individual battery racks is often restricted, which leads to higher maintenance costs. It is good practice to include plans for expansion or modification in the initial design as reconfiguring the layout of a containerised system can be expensive and be more complicated than work inside a large building.
Specification and sizing of the energy storage facility
The correct specification and sizing of the energy storage facility will depend on an accurate assessment of the intended and future applications.
The application will determine the power rating for the charge and discharge cycles (which may be different) and energy storage capacity can be determined by integrating the time required for the storage facility to operate at each power level. The actual energy storage capacity will need to take into account specific characteristics of the storage medium technology. For example, some battery types such as lithium ion can be cycled between 5 and 95% state of charge, whereas other types such as lead–acid may be restricted to cycling between 45 and 100% state of charge. It is important to specify the usable energy storage capacity and not the theoretical capacity.
The use of flywheel systems outside the established UPS application sector is growing, and a number of installations, particularly in the USA illustrate fast response time, and low maintenance costs. Typical applications include the provision of frequency response (often known as frequency regulation in the USA) .
Electricity storage facilities will be connected to the network in accordance with the standard codes such as the distribution code or the grid code in a similar way to other devices. Devices connected to the distribution network will need to comply with specific requirements concerning disconnection; these requirements may be counter to the intended application for the device. For example, a distribution network operator (DNO) may require a device to disconnect during periods when the frequency is outside limits, but in order to provide frequency response services to the transmission system operator (TSO) such disconnection is inappropriate. The scale of storage devices, in comparison with other resources on the network, becomes important. For example, connecting a 10 MW storage device on a 11 kV distribution feeder may be perfectly satisfactory if the storage device is moved from charge to discharge and vice versa at low ramp rates, but if it is switched within fractions of a second, for example, to provide frequency response to the TSO, such high ramp rates may be unacceptable as they would lead to excessive voltage swings on the distribution feeder. It is recommended that the DNO is consulted at an early stage when considering the location of an energy storage project.
Relevance to active network management and the smart grid
The network of the future will comprise many more active devices than at present. The smart grid uses real-time data to switch moveable demands to balance variable generation and so relies on a mix of hardware and software. The ability of electricity storage, in a variety of forms to buffer electricity will be invaluable as a means of balancing supply with demand. Electricity storage devices could be deployed to improve the technical operation of the network; however, the technology is now in advance of the commercial and regulatory position, as it is unclear how the business model for such systems would operate. It is unrealistic to expect an enduring income from difference in value between purchases and sales of electricity through the store, as the purpose of balancing is to even supply and demand, and hence reduce the price spikes and sags. Remuneration based on throughput such as a tolling arrangement, may be considered as an alternative.
In a similar way that small-scale and distributed generation (including renewable energy) influenced the development of new business models for the electricity market, presented new challenges to the DNOs, particularly with regard to the provision of connections to the network and provided opportunities for greater adoption of sustainable power sources, energy storage will be important at all scales and across a wide variety of locations.
At the technology level:
- (a) The cost of some technologies, especially lithium-ion batteries continue to fall. While the energy storage medium is only part of the total cost of the project, these dramatic cost reductions have opened the market for storage.
- (b) The scale of advanced storage is increasing. Battery projects of up to 10 MW were considered large until 2012, but battery projects of 20 MW or more are now considered routine. Projects up to 50 MW, especially for ancillary services and integration of renewables are expected before the end of 2017.
- (c) Pumped hydro can be competitive, even at relatively small scale (∼100 MW) due to reduced manufacturing costs for motors and turbines and improved site selection.
- (d) Control technologies, particularly associated with smart meters and remote controls are increasing the viability and acceptability of domestic and behind the meter storage.
Other key trends are:
- (a) The opportunity for domestic storage, and storage operated behind the meter by small industrial and commercial businesses is growing, driven by an incentive to maximise the value of self-generated electricity, often from rooftop PV. The storage medium is likely to be batteries, though there is a strong motivation to use a hybrid system, with electrical heating for domestic hot water, which is used inside the house for domestic use.
- (b) Keen commercial interest from renewable generation developers, wishing to add storage to large-scale solar installations, to overcome existing network constraints, provide time shifting of exported power or to offer additional services to other parties including ancillary services to the network.
- (c) Community energy groups have a developing business case to install shared energy storage, owned by a community project or a co-operative. These installations can be sized to match the power and energy requirements of the community, not the individual users, and so be of increased value.
- (d) Distribution companies see technical advantages in owning and operating storage, though the current regulatory regime does not necessarily provide a good business model.
- (e) The use of storage to provide ancillary services is an attractive high value application. Battery energy storage can be fast acting and suitable for providing services such as enhanced frequency response, primary, secondary and high response.
- (f) Long duration energy storage such as flow batteries, liquid air and compressed air are suitable for both energy management, peak shaving as well as supporting the capacity market and demand reduction programmes.
- (g) Local network constraints can be mitigated by the use of well-located energy storage. These services will need to be combined with other services to become financially attractive.
Energy storage is being used in an increased number of applications on the power network, from small scale behind the meter applications to larger installations at scales between 1 and 100 MW. As well as battery technologies, other types of storage are able to provide valuable services to end users of electricity, network operators, supply companies and generators as well as independent storage operators.
- Ter-Gazarian A. G.: ‘.
- Techno-economic performance evaluation of compressed air energy storage in the pacific northwest, Pacific Northwest National Laboratory PNNL-22235, 2013.
- Hedges K. W.: ‘Central station electric lighting’. Proc. of the Institution of Civil Engineers, 1888, Paper 2217.
- International Hydro Power Agency: ‘
- ‘The electric mountain’. Available at http://www.electricmountain.co.uk, accessed April 2016 .
- US Department of Energy: ‘Energy Storage Database’. Available at http://www.energystorageexchange.org, accessed April 2016.
- UK power networks, design and planning considerations for large scale distributed energy storage, SNS 1.2, UKPower Networks 2014.
- Highview power storage, 5 MW liquid air energy storage pre commercial demonstrator storage plant, Highview Power Enterprises Limited 2015.
- Price A.: ‘Briefing electrical energy storage options, energy’. The Proc. of the Institution of Civil Engineers, February 2014, 130010.
- de Veth J.: ‘Evaluating 6 years’ of VFB operational experience in Vierakker, Netherlands (2010–2015)’. Proc. of the Int. Flow Battery Forum 2016, 2016, in print.
- Baxter R.: ‘.
- Moseley P. Garche J.: ‘.
- Akhil A. et al.: ‘.
- Sioshansi R. Cabeza L. F. Yan J.: ‘.
- Cotterell M.: ‘.
Case study: Battery
UK Power Networks, one of the distribution network operating companies in Great Britain, was examining alternatives to conventional upgrades to its electricity distribution system. The town of Leighton Buzzard, in Bedfordshire, is supplied by electricity from a substation which is fed at 33 kV from two overhead lines. As electricity demand rises, the supply to the substation becomes close to its limit, and reinforcement is necessary. A conventional approach is to install an additional overhead line. In this instance, UKPN decided to trial the use of a large battery located at the substation. The battery is charged, usually at times of low power prices, and the battery is then able to be discharged at times of peak demand, so reducing the requirement to import power into the town.
In addition, the battery could also be used for other applications such as providing reserve services to the TSO. The battery was rated at 6 MW, 10 MWh, construction started on site in 2013, the plant came into operation in 2014 and is illustrated in Figs 1–3.
Fig 2: Demonstration liquid air energy storage plant. Photograph by permission of Highview Power Limited
Fig 3: Containerised flow battery installation in Yokohama, Japan (Photograph, Swanbarton)
UKPN was able to consider the experience of other DNOs who had constructed large-scale batteries in the Shetland Isles, Orkney Isles and in the North of England. UKPN considered a number of alternative battery types and the final choice was made as a result of a tender process and technical appraisal. Site specific requirements meant that the battery was installed inside a new building of ∼20 × 40 m, which also accommodated the power conversion system. The power conversion system was supplied by S&C Electric Europe Limited, the lithium-ion battery was supplied by Samsung SDI Co. Ltd. and battery integration software supplied by Younicos AG. Advanced software and control systems supplied by AMT Sybex were included to ensure that the battery was able to meet its primary objective of reducing peak load at times of maximum demand. The battery electricity storage facility is capable of automatic control using an interface supplied by AMT Sybex and battery .
Case study: Liquid air electricity storage
Highview Power, and its partners Viridor, an energy and waste management company, are constructing a 5 MW liquid air energy storage plant which is due to be commissioned in 2016. The plant uses the liquid air as a form of energy storage. Electricity is used to liquefy air, the air is cleaned and cooled until it liquefies, 700 l of air at normal temperature and pressure can be stored as 1 l of liquid air. The air is stored at low pressure in insulated tanks, similar to tanks which are in widespread industrial use. To recover energy from the system, liquid air is withdrawn from the tanks and pumped to high pressure. Stored heat from heat exchangers vapourises the liquid air into high-pressure gas which is used to drive a turbine, which can power an alternator or generator [ 8 ].
Case study: Flow battery for a community
The Scottish Island of Gigha has three wind turbines of 775 kW capacity with a fourth turbine planned of 330 kW capacity. However, because of the constraints of the network connection to the mainland, the additional turbine will be operated at a reduced power factor and reduced power output in order to avoid local voltage rise. REDT Energy plc., a flow battery manufacturer, proposed installing a flow battery of ∼100 kW and 1.25 MWh.
The first proposal was to construct the battery system inside a purpose built building, containing the flow battery modules and large electrolyte storage tanks. During the design and planning stage, the concept was changed to pre-assembled batteries and tanks installed in 20 ft ISO shipping containers. Seven containers will each hold one 15 kW, 240 kWh battery and electrolyte tanks, and one container will accommodate the power conversion system. The system will be connected to the 11 kV local network.
As well as considering the technical requirements for this installation, the project is noteworthy for highlighting practical arrangements for the purchase and sale of electricity as well as the involvement of the local community, as the island and its wind turbines are owned by a community company, Gigha Green Power Limited. The project is due to be operational by the end of 2016 .