Grid Energy Storage: Big Market, Tough to Tackle

Storing electricity for the grid could boom to a nearly $2.5 billion market by 2015, according to a new GTM Research report. But knowing which storage applications to target will be critical for early movers.

Grid-scale energy storage – it's a tricky market to tackle, but one that offers billions of dollars to those that can get it right.

The grid storage market is set to grow from roughly $365 million today to nearly $2.5 billion by 2015, according to a new research report from GTM Research.

That growth will include a larger role for lithium-ion, flow battery and  advanced lead-acid battery technologies than they now hold in grid storage, as well as growing pressure to expand energy storage to meet the anticipated needs of adding unprecedented amounts of intermittent solar and wind power generation to the grid.

But companies making the first moves into grid storage will have to navigate a complex regulatory and economic environment to put together commercially viable projects, report author John Kluza warned.

That means finding ways to combine a myriad of storage applications, from power-oriented functions like frequency and voltage regulation to load-shifting applications aimed at saving electricity generated at night to power the grid's peak loads on hot afternoons.

And with new battery technologies still quite expensive for more massive energy-oriented storage projects, it's likely that power-oriented storage will grow faster over the short term, Kluza said.

"The markets for power-oriented storage are a lot better developed at this point," he said. Power storage represents a 49 megawatt, or $65 million, market this year, but that should grow to about 479 megawatts, or $500 million, by 2015, he said.

Those include tasks like frequency regulation, or providing short bursts of power to stabilize the grid – a market that some regional grid system operators in the United States have already opened to third parties, Kluza noted.

Flywheels can serve this purpose – that's the goal of flywheel maker Beacon Power, which has landed Department of Energy loan guarantees to assist its plans to build up to 40 megawatts of frequency regulation storage for grid operators New York Independent System Operator and PJM (see Green Light post).

But Kluza said that lithium-ion batteries are likely to become the technology of choice for such power-oriented storage applications in the coming years.

"The market is ready for it from a government regulation perspective, and it's already in place," he said. "Existing manufacturers can make it at a price that allows an economic return."

Kluza named lithium-ion battery makers A123 Systems and Altair Nanotechnologies as two contenders in that emerging grid storage market. Both companies have delivered battery systems to utility AES Corp., he noted (see A123 Systems Inks Battery Deal With Chrysler).

But A123 could jump ahead of AltairNano if utility Southern California Edison succeeds in securing a DOE stimulus grant that would, among other things, allow it to hire A123 to build a 32-megawatt grid storage battery to help manage wind power (see SoCal Edison Wants A123's Biggest Grid Battery Ever).

A123 also has an agreement with AES to deliver up to 16 megawatts of batteries, according to documents A123 has filed with the Securities and Exchange Commission.

But while these short-term, power-oriented storage needs are the most promising in the short term, energy-oriented storage – keeping energy for longer periods of time to meet peak demand times or store solar and wind power – is expected to become an even bigger market, Kluza said.

The year 2009 saw about 147 megawatts of such storage produced at a value of about $300 million, but that's expected to grow to 1,321 megawatts with a value of $1.98 billion by 2015, he said.

However, Kluza warned, "The energy markets are still in flux and being developed, or identified even – especially in the United States, where there are a lot of regulatory uncertainties."

That's not as true elsewhere in the world, where large-scale energy storage projects using sodium sulfur batteries are underway. NGK Insulators of Japan is the sole mass producer of that technology, which is best suited for massive, stationary installations since it runs at very high temperatures (see Top Ten Smart Grid: Energy Storage).

Among the largest customers, Japanese utility Tokyo Electric has installed about 300 megawatts of NGK batteries, France's EDF has bought about 150 megawatts worth, while the Abu Dhabi Water and Electric Authority has bought 50 megawatts, he said.

In the United States, NGK's biggest customer has been American Electric Power, which has 7 megawatts of batteries installed and 4 megawatts more under development, and Xcel Energy with a 1-megawatt battery. AEP is using the batteries to mitigate overloaded power lines, and Xcel is seeking to better manage wind power (see GridPoint to Manage Wind Power Battery Storage).

General Electric is also developing sodium sulfur batteries to power train locomotives as well as for future grid applications, and could become a major contender in that market, Kluza said (see GE Aims At Energy Storage For Trains, Grid).

But sodium sulfur will likely start to see some competition from flow batteries and advanced versions of old-fashioned lead acid batteries in coming years, he said.

Flow batteries, which share some characteristics with fuel cells, are becoming more cost-competitive with sodium sulfur batteries, he noted. Companies such as Deeya Energy, ZBB Energy Corp and Premium Power are making large-scale flow batteries (see ZBB Seeks DOE Dollars to Expand Flow Battery Production).

And advanced lead-acid batteries could do away with the key maintenance concerns that have limited the traditional battery chemistry's attractiveness for bulk energy storage, he noted. C&D Technologies, Axion Power International Inc. and Exide Technologies are among the companies working on grid-scale advanced lead-acid batteries.

None of the battery-based technologies can compete on a dollar-for-dollar basis with such old-fashioned energy storage methods as pumped hydro.

Furthermore, utilities can be challenged to fit storage into the regulatory framework that constrains how they spend and make money, Kluza said.

That's a point that Ed Cazelet, a former board member of the California Independent System Operator and founder of would-be grid storage startup Megawatt Storage Farms, has made in describing the challenges companies like his will face (see Green Light post).

Still, he and others predict a huge future demand for energy storage to help solar, wind and other renewable (and unpredictable) energy generation sources fit into a grid made to deliver predictable amounts of power.


Interact with smart grid industry visionaries from North American utilities, innovative hardware and software vendors and leading industry consortiums at The Networked Grid on November 4 in San Francisco.

Comments [15]

  • Beacon Power 09/14/09 8:19 PM

    Beacon Power is a very niche game changer for frequency regulation.
    Goldman Sachs has been buying.

    Reply
  • Charles R. Toca 09/21/09 1:42 PM

    One flow battery technology a bit under the radar since the publicly traded VRB Power was acquired by Prudent Energy is the vanadium redox system - the VRB-ESS.  More information is at http://www.Utility-Savings.com.

    Reply
  • FDDoty 09/28/09 9:49 PM

    It’s important to be a bit more careful with units.  Energy storage is measured in kWhr or MWhr, not kW or MW.  The later refer to peak power capacity.  The prices of peak capacity, which you’ve mentioned, might matter the most for 3-15 minute control and regulation problems that are outside the ramp rates of coal power plants, but it’s energy storage that matters when it comes to bringing down peak rates. 

    The cost of energy storage in batteries ranges from $100 to $1500/kWhr.  The cost of energy storage in flywheels in higher.
    The tank-component cost of storing energy in a liquid fuel such as diesel is $0.02/kWhr.
    The only practical method of storing the enormous amounts of energy needed to have a significant effect on the price of peak energy is in the chemical bonds of stable liquid fuels.  Advanced processes are being developed that will allow gasoline, diesel, jet fuel, etc. to be synthesized very efficiently from CO2 and H2O using off-peak grid energy.  The development path is relatively straightforward.  When you have three to four orders of magnitude working in favor of synthesized carbon-neutral fuels for grid-scale energy storage, nothing else has a chance in the long term.

    Four recent peer-reviewed publications on recent technical advances in synthesizing fuels from CO2 and off-peak grid energy are available for download at the windfuels website.

    Reply
      • Peter A 09/29/09 1:44 PM

        While I agree with you on principle about the potential for fuel synthesis, I think the reliance on (presumably fresh) water is their main weakness. We all know that fresh water will become increasingly scarce over time. Now if the process could use seawater ...

        By the way, thanks for confirming the order(s) of magnitude difference in energy density and cost per kwh between electrical storage and hydrocarbons. Unless battery technology exhibits “Moore’s Law” type of effects, it has zero chance of substituting, let alone displacing liquid fuels. That’s why I no longer see the EV as anything more than a niche product.

      • Victor Babbitt 10/1/09 12:34 PM

        Excellent point on the common misunderstanding about MW/MWh.  However, your analysis is a great simplification.  Electrons come directly out of the batteries, they don’t come directly out of a tank used to store deisel.  With that logic, we could say that the electrolyte tanks used in a flow battery (where the activated electrolyte is actually stored), has a ‘tank component’ cost in the pennies per KWh as well.

  • Charles R. Toca 09/29/09 11:42 AM

    FDDoty: Good point on MW v. MWhrs.  Important to maintain distinction.  For example, the 32 “MW” lithium ion battery mentioned above for SCE is actually an 8 MW battery with 4 hours of storage = 32 MWHrs.  In reality, if SCE used it for 4 hours of cycling, it would only last 500 plus charge/discharge cycles, so only about 2 years.  It’s more likely they will use if for 15 minute partial charge/discharge cycles, like the AES application noted above.  So now you are looking at a a 2 MWHr battery in actual use (8 MW x 15 minutes)  So how do you value the system?  Not as a 32 “MW” system as mentioned above, and probably not as a 32 “MWHr” system if it’s practical use will only be for 15 minute cycling.

    Reply
      • Victor Babbitt 10/1/09 12:30 PM

        Charles: A123 is quoting quite amazing cyclability, and in fact seems to have the highest cyclability of any large scale plate battery, better even than Sodium Sulfur.  The bigger question is cost: SCE is stating they are looking for $25 million from the DOE for this project.. add the 50% cost share and we get a $50 million project.. $1.56 million / MWh?

  • Dr. Utsawa K. Chaturvedi 10/17/09 8:52 AM

    Despite less price, the storage of electrical energy in chemical form and getting it back in electrical form, is not as efficient as it is in rechargeable batteries. The pollution while changing chemical energy to electric is another issue.
    Yes, we have to develop batteries / super-capacitors many order of magnitude before it is used in gird energy storage and stabilization.

    Reply
  • JLB 11/18/09 7:54 PM

    Eamon—Thanks for reviewing those market forecasts.  I see a fundamental confusion in them and in virtually all articles on grid storage, too: sale of technology is confused with sale of technology-based service.  This distinction is crucial in the one grid-storage market that’s ready today (“power applications,” esp. ancillary services). Here the significant question is which technology will enable its posessor to be low bidder (in day-ahead and real-time markets) and to be the most profitable operator of an ancillary service business, not which technology will have the most grid-connected MWs.  And given how much lower O&M and replacement costs are for flywheels than big batteries and how much longer they will last, it sure seems that flywheel plants will get the lion’s share of the FR revenue.  That’s what matters.  Beacon Power isn’t even actively trying to sell the flywheels it makes; it plans to use them to provide a flywheel-based frequency regulation (FR) service, using sale-of-service accounting.  So when these market forecasters project sales of flywheels to be negligible compared to lithium (or other) batteries, they miss the point, at least for investors.  Beacon is a good investment precisely because it is NOT FORECAST TO SELL tons and tons of flywheels to competitors, but will will use them to make itself the most profitable operator of a frequency regulation service. 

    There may soon be more MWs of lithium-ion batteries sold for the purpose, but that doesn’t mean they’ll beat out Beacon’s plants for FR business.  Utilities are interested in scaled-up lithium batteries mainly because they expect (questionably, in my view!) millions and millions of scaled-down lithium batteries to be out there one day in plug-in cars.  They’re looking forward to the day when all these batteries will be aggregated together (big and little bateries) into one enormous virtual battery, managed by the smart grid and giving the utilities a heck of a lot more electrons to sell.  Getting into big lithium batteries now for whatever piece of the FR business they can get (and there is enough to go around) will be a dress rehersal for that day.  All of which may (or may not!) come to pass.  But meanwhile Beacon is already serving a real market today is about to build two 20 MW FR plants (one starts next Monday).

    Reply
  • StevePluvia 11/18/09 10:57 PM

    Victor, what amazing cycling is a123 claiming?  Nickle hydrogen batteries can cycle 20,30,40,000 times, they have impressive power/weight ratios and costs well below lithium.

    Reply
  • JoeJoe 11/18/09 11:22 PM

    My money is on load side technologies vs. supply side technologies. From an dollars and sense perspective enabling loads to respond to prices and frequency should be much cheaper that batteries, flywheels, synchronous condensers etc. From a performance perspective, loads can respond to changes in frequency faster than supply side technologies (generators). Responsive loads offer a one two punch that’s hard to beat. Batteries/flywheels are great for stand-alone applications (hospitals, communication hubs) but out of place in generic interconnected applications.

    Reply
      • JLB 11/19/09 8:44 AM

        Yes.  But the things about flywheels or any storage really, is that they are load and supply—they can absorb as well as inject energy.  Today’s approach—ramping standby generators up or down—can only adjust supply.  They can reduce supply but not reverse it.

  • StevePluvia 11/19/09 9:30 AM

    Talk to me baby.  Give me some JoeJoe examples.  Here’s the Nickle-Hydrogen battery—tested to 3000 cycles with zero degradation; target cost slightly higher than lead acid:

    http://www.ergenics.com/page15.htm

    They also have a prototype solar engine that can be driven via solar hot water or waste heat; claimed cost $2.50ish/watt installed…

    http://www.ergenics.com/page27.htm

    Very interesting technology…

    Reply
  • JoeJoe 11/19/09 2:52 PM

    JLB… Hydro and thermal units act as loads in synchronous condense mode. The technique works but why turn a generator into a motor when you’ve got plenty of motors on the system already?

    Steve… Anything that cycles on and off and isn’t overly time sensitive can be looked at as a price sensitive load - water pumping, HVAC, water heaters, dishwashers, refrigerators, washer/dryer. Which applications make sense will depend on the profile of daily electricity prices.

    -Dishwashers and washing machines can be programmed to run in the middle of the night. Refrigerators can be programmed to run the defrost cycles and ice making at night.
    -It’s technically possible to have a water heater turn itself on at 4 o’clock in the morning (during the low priced trough) and heat itself up from from 140 to 175 - that’s 4 kWh of displaced energy use. Would the more durable construction and extra insulation make this idea a no-go? I don’t know… It depends on prices.
    -It’s technically possible to do the same thing in reverse with your refrigerators and HVAC system using ice batteries. I’m not endorsing these technologies just throwing them out there.

    You could argue that batteries could still make sense. Why not charge the battery in the middle of the night and help power your home/business during the day? Sure… that’s technically possible but the Benjamins offer a strong counter argument.

    Reply
.