By Glenn Fleishman
A not-so-new notion is gaining traction for storing power generated at nonpeak times: compress regular air into underground chambers, then retrieve it later to spin turbines.
Wind power can be generated any time the wind is blowing at the same cost day and night. Because there's no efficient way to store power when it's generated but not needed, utilities and wind-power farms around the world are already having to slough off power as wind-based generation scales to something beyond scattered projects.
The New York Times blogs about a variety of efforts focused on using the excess electricity from some wind systems to compress air into sealed underground chambers, such as those left behind from various kinds of pumping and mining operations. The compressed air has potential energy that can be released later.
The current generation of compressed air energy storage (CAES) systems have to burn natural gas to heat the compressed air before the air can be used to turn turbines and recapture a good fraction of the energy used in compression. Future CAES plants are planned that skip the natural-gas input, shunting waste head from compression into the decompression process.
Certain parts of the world are better suited to using CAES for energy storage. In Ontario, the Toronto Star reported a few days ago that there are 50,000 wells in the province of which just 2,000 are still in use. Some of these wells are used for a different kind of stored energy: compressed natural gas, pumped and held until demand requires its release. Others could be used to store compressed air.
The comments on the Times blog entry are particularly interesting, with the author of a significant paper on the technology chiming in, along with a wind industry representative named Michael Goggin. Goggin wrote that storage is unnecessary because other types of generation can be shut down on demand in favor of wind—water can be held behind a dam for later release or natural gas held in pipes for later burning.
But that's surprisingly idealistic. In the real world, the cheapest power is used first. If wind power is generated during nonpeak times, less money is paid for it, even with the subsidies in effect in many countries to encourage wind generation. Goggin's scenario works only if the costs are the same among different forms of generation, or a single utility owns the various forms of generation and chooses a more-expensive method to obtain carbon credits or meet greenhouse gas emission goals.
This view also requires that transmission systems are capable of moving wind power at nonpeak times precisely to where it's best needed. As Sandia National Laboratories researcher Georgianne Peek said (in a press release about an Iowa CAES project) in June 2008, "The wind blows in some areas when electricity is not needed or where the transmission system can't accept all of the energy."
If wind power can be offset from nonpeak to peak times, then it becomes more viable, and thus sees greater use. This could balance green-power principles (more wind generation) with market motivations (lowest cost).
While batteries can also be used to store energy, they are expensive to make, use hazardous and toxic metals and compounds, and can't hold energy for very long. They're useful in specific situations, like home storage and backup with solar systems. Peak shifting, in which power generation is used during off hours to be reclaimed in some form during more expensive daytime uses, involves everything from next-generation flywheels to making ice power air conditioning during the day to providing incentives and for future electric-car owners to charge their cars primarily overnight