Bi-level optimal configuration of hybrid shared energy storage capacity in wind farms considering prediction error

In wind farms, hybrid energy storage (HES) can effectively mitigate the fluctuation and intermittency of wind power output and effectively compensate for the prediction errors of wind power. However, the high cost of HES has prevented its large-scale adoption. Inspired by the sharing economy, this paper introduces the concept of hybrid shared energy storage (HSES) in wind farms. A rolling optimization (RO) strategy is formulated to obtain the reference power curve of the HES, and a bi-level optimal configuration model of HSES is established to minimize the wind farm's daily operating cost and optimize the compensation's performance for the prediction errors. The upper-level model aims to obtain the optimal rental capacity of HSES, and the lower level is an energy trading mechanism to optimize the trading power of the wind farm in real-time. The simulation results reveal the excellent performance and economy of the proposed HSES model for wind farms.