Grid Scale Energy Storage: Technologies and Forecasts Through 2015

by John Kluza

Distributed Grid Scale Energy Storage

The development of the electrical grid is one of the most impressive accomplishments of the 20th century. Its scale and ubiquity speak volumes about how important it has become to modern life. The modern grid, however, is still largely based on the original design that Westinghouse and Edison debated in the late 1800s, and it is not designed for modern electrical loads, distributed energy sources or optimal efficiency. Smart grid technology is now beginning to be rolled out to improve the system. One of the major pieces required to make the new smart grid effective is a buffer in the system that can store energy to balance the whole grid system.

The development of distributed grid scale energy storage technology offers great potential to improve the architecture and operation of the electrical grid. A number of interrelated factors are driving the adoption of distributed grid scale energy storage including:

  • Development of renewable and/or distributed energy sources. For instance, wind power requires approximately 3 MW to 5 MW of additional frequency regulation electric ancillary service for every 100 MW wind power installed and would also benefit from load shifting energy storage.
  • Utilities’ desire for more efficient use of generation, transmission and distribution assets
  • Public’s desire for carbon reduction and more efficient use of fuel resources
  • Increasing power quality/reliability requirements from end users


There are many ways that energy storage can help balance electric flow on the grid. This report will discuss distributed storage with discharge times of minutes or longer in order to focus in on particular sectors of the market that are undergoing rapid development and offer massive opportunities for growth. There are a number of classes of technologies with different properties that make them more suitable for specific market applications.

Power and Energy Positioning of Energy Storage Options

Batteries of many different classes are used in these grid scale energy storage systems. One of the most familiar technologies in use is the lithium-ion battery. While the handful of installations already on the grid do not use the same exact chemistry commonly found in laptops and cell phones, they share some of the same components, packaging, manufacturing processes and properties. Lithium-ion batteries for grid use are more similar to those for electric vehicles and are designed to perform well on the metrics needed in power-oriented grid applications, including cycle life, cost and power capacity.

The advanced lead acid battery, based on the familiar traditional lead acid battery, is in very early stages of being used on the grid for energy-oriented applications. While considerably different from traditional car starter type lead acid batteries, they share much of the same chemistry and packaging. However, these new ones have new parts integrated into them, such as ultracapacitor material or other alternate electrode material which greatly improves their cycle life and power delivery, while allowing them to still be built in a low cost fashion on equipment similar to that used for traditional lead acid batteries.

The type of battery which currently is most used on the grid is from the sodium family of batteries, the sodium sulfur battery. It is conceptually similar to a traditional battery in many ways, however its implementation uses liquid sodium maintained at 300°C, so its design, construction and enclosure are quite different from traditional batteries. The various configurations of these batteries have acceptable power, energy and life cycle levels for many grid applications, both power and energy-oriented, but their cost, even in current mass production, makes them only suitable for very high value applications.

The technology for the flow battery has been explored for decades but is only now becoming a viable alternative for cost effective energy-oriented grid installations. Flow batteries differ from traditional batteries in that they pump their active electrochemical material through a reaction chamber, instead of containing it all inside the cell. This way, their discharge time (energy) is related to size of the tanks of liquid active material and their discharge rate (power) is independently related to the size of the reaction chamber. There are a number of variations on this chemistry and some have recently achieved low enough costs and high enough life cycles to be installed in pilot grid applications.

The final technology examined by this report is flywheels. These systems connect a motor/generator to a spinning rotor and spin it faster with the motor to charge the system then slow it down through the generator to discharge it. There are a handful of flywheels used in power-oriented installations on the grid already. Though they are expensive initially, they have a very long life, which could eventually make the upfront cost worthwhile.


Markets and Revenues

One of the most interesting and complicated aspects of grid scale energy storage is that there are many applications that create value on the grid. This report describes 14 applications identified and researched by New York State Energy Research and Development Authority (NYSERDA) which cover the range of possible methods to create value with grid scale energy storage. These applications include power-oriented (fast) and energy-oriented (load shifting or LS) options for utilities/grid operators, end-users and renewable power. Each of these applications creates a benefit stream of cash and non-cash benefits. Many of these benefit streams are independent of each other and some can be accomplished at the same time by one energy storage installation.

By analyzing estimates of benefit value created by each application and the cost of the energy storage systems, it was found that building an installation that only accomplishes one of these applications does not typically create enough value to offer an attractive payback on the project. Players need to find ways to create enough value by bundling applications together into what this report will refer to as “implementations” and identify ways to capture enough of that created value themselves to make the installation economically attractive.

The applications were examined to identify which benefits could be accomplished at the same time and how they might be bundled together to create more value for the energy storage system owner. These possible implementations were compared to existing and proposed installations, and a set of six primary implementation types was developed. This set characterizes each of the grid scale installations identified and denotes the types of value each installation seeks to create and capture.

The best case scenario for value generated by each implementation was estimated and compared to the cost of the energy storage system built with each technology. This was used to develop a matrix of likely economic outcomes for each of the technologies in each of the implementations and identify which combinations make economic sense. For each implementation, existing installations with available economic data were also highlighted for comparison to the estimated numbers and these results were collected in their own economic outcomes matrix to show which technologies have been economical in practice.

The significant technological and legal challenges and opportunities of each implementation were also investigated and discussed for each implementation. Concerns about regulatory barriers to accruing the multiple benefits of an implementation to one entity and the challenges of recovering value from non-cash benefits like improved customer service were highlighted for each implementation.

The report continues on to an investigation into the current status of the market and projections for the future. The installed base of the total market by manufacturer and sub-market was determined. Total market size for each submarket was estimated both for the US and globally based on expected requirements for each application expected to be profitably served by energy storage. Currently addressable U.S. market size was also calculated for the one application which had sufficient data: frequency regulation ancillary services. Production for each energy storage manufacturer was then estimated to 2015 and summed into a bottom up company-based total annual production estimate for each submarket, including production by technology. This collection of data on estimates of cost and benefits, technical capabilities, market requirements, production levels and implementation strategies was based on interviews with dozens of vendors, utilities, regulators and industry experts, results of a vendor survey and an extensive review of news and research.



In addition to the previously mentioned technology and application investigations, this research also includes profiles of the manufacturers in the industry with a listing of VC transactions for 2008 and 2009H1. The synthesis of this information led to conclusions in three main veins about the future of distributed grid scale energy storage.

  • Power-oriented (fast) energy storage will grow quickly in the near to mid term but will be constrained in the long term by a modest total market size.

    Power-oriented (fast) energy storage is poised for strong near- to mid-term growth. Its most significant component, the frequency regulation market, has recently been opened up for direct entry by energy storage in some ISO regions of the U.S. with additional ISOs anticipated. This means that energy storage can secure contracts for grid frequency regulation on the open market and the owner of the system will get compensated in cash. This capture-able, all cash benefit stream makes obtaining compensation for an energy storage system much less complicated than many other implementations. New highly robust, moderate cost lithium-ion batteries are able to provide this service cost effectively and are beginning to be deployed successfully in a few regions of the U.S. and in Chile. This trend is expected to continue and accelerate with the addition of new renewable resources on the grid and further decreases in the cost of lithium-ion batteries. Production of fast energy storage in 2009 is estimated at 49 MW and is expected to grow to 479 MW or $500 million in 2015. The total market size for fast energy storage is estimated at about 7,137 MW total for the U.S. and about 37,828 MW for the world.

  • Energy-oriented (load shifting) energy storage has a massive total market size, however it is only beginning to be ready to be exploited.

    Energy-oriented (load shifting) energy storage offers a number of potentially lucrative opportunities for implementations that strategically combine applications. While wholesale load shifting is sometimes discussed, it does not create enough value to be cost effective on its own in most situations right now. There are many existing strategic load shifting implementations that are or are close to becoming cost effective. The challenge with these implementations is that some of the benefits are generated as non-cash benefits which can be difficult to monetize. Additionally, the benefits come from bundling different value streams, which are feasible technically but challenging to accrue to one entity for regulatory reasons. In some parts of the world, like Japan, where the value created from single applications is higher or utilities are more easily able to accrue value from the multiple benefits generated, sodium sulfur (NaS) load shifting energy storage has already gained a good foothold and has recently gained favor in other countries like France and the UAE, though only small pilot installations exist in the U.S. New flow battery technology, particularly zinc-bromide, has recently become more cost effective than NaS for many implementations and is expected to grow to surpass NaS installations by 2015. Advanced lead acid batteries are also expected to show impressive growth due to further cost reduction. Additionally, repurposing existing traditional lead acid manufacturing capacity will further enable growth of this technology. The currently variable government regulatory climate for LS storage is expected to somewhat limit near-term growth, but more amenable regulation and mass production cost reduction for LS storage are expected to drive strong mid to long term growth. In 2009, an estimated 147 MW were produced, but in 2015, 1,321 MW are expected to be produced with revenues of $1,978 million. The total market size is estimated at 85,000 MW in the U.S. and 450,000 MW in the world.

    Projected Worldwide Production of Energy Storage in 2009

  • Government regulation will play a large role in determining the rate of the roll out of all energy storage.

    Government regulation is a critical driver of or inhibitor to energy storage technology penetrating the market. Recent history and current trends in government regulation are favorable for energy storage, especially for frequency regulation in the U.S. Recent political action in favor of renewable power, including energy storage tax incentives in bill S.1091 recently proposed in the U.S. Congress, is paving the way for a more favorable environment for LS in the U.S. However, compensation for utilities will also have to change in order to fully spur progress for LS energy storage, including areas such as allowable inclusions in the rate base, risk compensation and ownership of assets useful for both T&D and generation. Other countries with differing structures for utility compensation and greater LS storage penetration, like Japan, France and the UAE, may offer ideas for how to encourage its penetration in the U.S.

John Kluza Analyst at Large

John Kluza holds a Master of Science in Management and Engineering from MIT. While at MIT, John interned at A123 Systems and did competitive analysis and due diligence in order to prepare A123 for an IPO filing. Prior to attending MIT, John worked for 5 years as an engineering consultant on the development of software for control systems and data analysis.

image description

Contact Us