A Two-Timescale Operation Strategy for Battery Storage in Joint Frequency and Energy Markets

The growing penetration of renewable energy in modern power systems requires energy storage to take on more responsibilities in multiple regulation services. Battery energy storage system (BESS) possesses fast response capability and is suitable to shave peak demand and provide frequency support. This article studies coordinated bidding strategies of BESS in frequency regulation and energy markets. Challenge arises from the fact that frequency control and energy arbitrage actions are taken in different timescales, and the capacity used in either market affects the available capacity and revenue in the other one. This article proposes a two-timescale decision framework, offering the hourly base-power bid in the energy market and capacity bid in the frequency regulation market, as well as real-time responses to the automatic generation control (AGC) signal every few seconds. In the fine timescale, we employ a threshold policy to generate AGC response accounting for battery lifespan. In the coarse timescale, we establish a stochastic dynamic programming model and optimize the bidding policy without exact forecasts of market prices. To solve the stochastic dynamic programming model online, a simulation-based policy improvement method is developed to approximate the state-action value function using a heuristic base policy. The performance improvement property brought by simulation is theoretically proven. We carry out comprehensive case studies to validate the effectiveness of the proposed method and analyze the economic impact of electricity prices and battery $E/P$ ratio. Empirical tests show that with an $E/P$ ratio of 3 $\sim$ 5, the BESS gains a higher net revenue across the lifespan.