压缩空气储能
储能
抽蓄发电
可再生能源
压缩空气
环境科学
发电站
电网储能
间歇式能源
工艺工程
电力
工程类
功率(物理)
分布式发电
电气工程
机械工程
物理
量子力学
作者
Firdovsi Gasanzade,Francesco Witte,Ilja Tuschy,Sebastian Bauer
标识
DOI:10.1016/j.enconman.2022.116643
摘要
Compressed air energy storage in geological porous formations, also known as porous medium compressed air energy storage (PM-CAES), presents one option for balancing the fluctuations in energy supply systems dominated by renewable energy sources. The strong coupling between the subsurface storage facility and the surface power plant via the pressure of the compressed air, which directly determines the amount of energy stored and the power rates achievable, requires the consideration of the fluctuating supply and demand of electric power, the specific technical design of the compressed air energy storage plant and the subsurface storage processes to determine achievable power rates, storage capacities and overall performance. In this paper, we present subsurface storage designs using a set of future energy system scenarios with different fractions of renewable energy supply and technical options for the power plant. Our findings indicate that the PM-CAES systems can supply 115 MW of electric power and between 12.1 GWh and 49.9 GWh of electric energy for up to 429 h, thus offering grid-scale power storage capacity. The storage design is robust against variations in future energy system scenarios and different power plant configurations, with efficiencies between 0.54 and 0.67 and energy densities between 0.12 and 0.28 kWh per kilogram of stored air. The storage design can be improved further by using horizontal instead of vertical wells, which also reduces induced pressure increases in the storage formation. This study for the first time provides a complete framework for assessing achievable storage rates and capacities for PM-CAES based on detailed forecasts of future energy systems, the geological and geotechnical setting as well as engineering aspects of the compressed air energy power plant.
科研通智能强力驱动
Strongly Powered by AbleSci AI