Present storage media are steam and molten salts. At 1,200 degrees C, new materials flowing through the system, perhaps molten glass or nanoparticles in salts, may make it possible to use compressed air as the secondary storage medium.
Presently, the cost for molten salts to provide six hours of storage for a 150-megawatt solar power plant is approximately $75 million, according to Raade. As research advances, storage can be developed to operate at 700 degrees C (which Raade said will be “soon”) and the cost will fall to $20 million. The cost at 1,200 degrees, he said, will be $8 million.
That translates, Raade said, to $25 per kilowatt-hour, compared to present costs of $100 per kilowatt-hour and today’s utility-scale battery cost of $400 per kilowatt-hour.
Solar power plant thermal energy storage (TES) is already more reliable than batteries, which remain vulnerable to heating inefficiencies and overheating malfunctions. Lead acid batteries are capable, Raade said, of 2,000 charge-discharge cycles. Sodium sulfur batteries can get 4,500 cycles or about eleven years of on-and-off service. Electric vehicle makers expect today’s lithium-ion battery packs to serve seven to ten years. TES, Raade said, can be counted on for 11,000 cycles, which translates to 30 years of after-sun service.
It can also, Raade said, return as much as 95 percent to 98 percent of the heat it captures, a greater round trip efficiency than competitors. According to Raade, flow batteries come in at around 75 percent, pumped hydro at 82 percent and lead acid batteries at 87 percent.
There are good reasons why, despite advantages and promise, thsolar power plants with TES continue to struggle for support, said E3 Senior Consultant Andy Taylor. He listed the sharp drop in the cost of photovoltaic (PV) solar, the recessionary fall in electricity demand, and the drop in natural gas and wind energy prices.
Perhaps more importantly, Taylor noted, the “increased penetration of intermittent resources into power systems requires a shift in economic valuation.” There is a need for, instead of the familiar levelized cost of electricity (LCOE) analysis, a “more sophisticated assessment of operational and reliability impacts and associated value to the utility.”
Because it is less amenable to large-scale storage, PV solar is significantly more challenging for grid operators than CSP with TES, which is both dispatchable and more reliable.
Taylor proposed a new valuation formula. Energy value plus capacity value, less the cost of grid integration, should be what utilities use, because “LCOE does not address the integration cost,” Taylor noted.
As higher levels of PV solar are integrated onto the grid, the capacity factor of solar energy falls because its variability becomes a larger factor. CSP with TES extends the capability of solar to meet the grid’s needs, extending solar’s capacity factor.
“Storage can shift solar energy to periods of peak demand,” Taylor added, as well as providing “firm capacity and ancillary services” and reducing “renewables integration challenges.”
In Taylor’s equation, this translates into a higher “capacity value” and results in a higher value to utilities.
“Thermal storage is not new,” said Torresol Energy Development Senior Vice President Felicia Bellows in describing her company’s solar power tower projects with storage in Spain and Chile. From Torresol’s experience with its 120-megawatt Gemasolar facility, which has fifteen hours of TES, Bellows said, the company’s strategy has become “get on-line and run, run, run.”
Its TES-enabled capacity factor of 75 percent, Bellows said, allows Gemasolar to “push solar out into the other hours and make as much money as we do at noon.”
During the Spanish winter, demand peaks at 10 a.m. and 8 p.m. Gemasolar meets the first surge in demand from solar radiation and the second from storage. During the Spanish summer, demand peaks at noon and 10 p.m. Again, Gemasolar responds to the first from radiation and the second from storage.
Depending on how Gemasolar’s turbine is used, the stored capacity can be extended in time or output volume.
Bellows said Torresol’s leaders have experienced a “shift in mindset.” They do not think of Gemasolar and CSP with TES as renewable energy but rather as “a typical thermal plant with a renewable source.”
Because the technology, she concluded, is proven and reduces operational risk, “it is financable, it is exceeding expectations, and so it is commercial.”
National Renewable Energy Laboratory (NREL) CSP Program Manager Mark Mehos described extensive modeling studies, particularly those undertaken by NREL’s Paul Denholm, that confirm Torresol’s real-world experience as well as Taylor’s theories.
“Grid flexibility is a key factor influencing future penetration of renewable generation,” according to Mehos, and TES increases grid flexibility because it “allows shifting of the solar resource to periods of reduced solar output with relatively high efficiency.” Therefore, he noted, more use of CSP with TES “could result in greater use of non-dispatchable PV and wind, especially at higher penetrations.”
Herman K. Trabish, www.greentechmedia.com
