Whether or not storage is needed depends on the aspirations for wind power. Although there is a large potential for wind power expansion in serving non-base load markets, this should be regarded as a near term opportunity for expanding wind generation from its current very small base production level. If wind energy is to become a truly large player in electricity markets globally it must be able to compete with base load power, which accounts for most electricity generation. Notably, 70% of US electricity and 60% of global electricity are currently provided by coal and nuclear power, mostly via base load power plants.
Moreover, and ultimately more importantly, if wind power is to make a major impact in climate change mitigation, wind power would have to displace primarily base load coal power, which accounts for more than 80% of US CO2 emissions for power generation and 70% of CO2 emissions from power generation worldwide.
Base load wind power systems can be either systems using thermal power, such as natural gas combined cycle (NGCC) backup, or wind power with storage. With the former, wind can provide no more than 40% of base load power, but with storage over 80% of base load power can be provided by wind. Moreover, the NGCC backup option would require a substantial increase in power system dependence on natural gas -- with attendant concerns about high and volatile prices and supply security.
Although the potential of wind with storage to displace conventional base load power is not resource constrained, it is constrained by the remoteness of good wind resources and the added costs of transmitting remote supplies to major electricity markets. We have investigated wind coupled to compressed air energy storage (CAES), which would often be the most cost-effective storage option, with long-distance transmission and have found that remote wind/CAES systems would not only be able to provide base load power but would be able to compete with fossil power systems in a wide range of circumstances. The dispatch (short run marginal) cost for the wind/CAES option would typically be significantly less than for fossil power systems, so the plant would be dispatched first. This means that under a wide range of competitive energy market conditions the wind/CAES system could sustain a high capacity factor and force fossil power plants to operate at a lower capacity factor and thus at higher total generation cost. Please visit http://www.princeton.edu/~ssuccar/wpm.html for evidence to support this claim. We have found the wind/CAES option would be especially attractive under a climate-change mitigation policy, in which case wind/CAES could provide less costly power than coal plants with CO2 capture and storage technologies at a comparable greenhouse gas emission rate.
Thus we contend that energy storage has the potential to provide an economically viable pathway for wind to enter base load markets while also providing a number of ancillary benefits such as enhanced transmission utilisation, improved energy security, reduced system-wide emission levels and greater realisable penetration levels for wind.
Our point is not that wind does not need to compete in base load markets, but that holding generation in reserve is cheaper than storing and releasing power. "Reserve" here refers specifically to the small addition to overall system reserve that is introduced to deal with the uncertainties associated with wind generation. That addition is modest -- 5% of the wind capacity, say, if wind is providing 15% of the electricity. Using storage instead of additional reserve to "level the output" will almost certainly need a much higher capacity, to cater for low wind periods.
Base load capacity on most electricity systems is actually only about 40% of peak demand. The other 60% of generating facilities must respond to the variations in consumer demand. When wind is mixed in with these, the additional variability is barely noticeable at first and rises only slowly. The indication is that the most promising markets for storage are in "flattening" the variations of the system as a whole, irrespective of the amount of wind on the system. Any storage system dedicated solely to wind is likely to have a tough time, as the additional value of firm power over non-firm power is quite small.
We are certainly not suggesting that storage will never be appropriate. With increasing volumes of wind energy, more additional reserve will be needed. The extra cost of that reserve rises slowly and levels out at about $5/MWh of wind generation, even at wind penetrations of around 50%. If storage can meet the tough cost targets and be cheaper than reserve, that is fine.