Stochastic optimal sizing of hybrid energy storage system considering uncertain AGC regulation requirements

Integrating energy storage systems into thermal power plants offers an effective solution for enhancing automatic generation control (AGC) performance. However, existing research has primarily focused on deterministic AGC regulation commands, which fail to accommodate the inherent uncertainties in frequency regulation requirements. To address this gap, this paper proposes a stochastic optimal sizing scheme for hybrid energy storage systems (HESSs) that accounts for uncertain AGC regulation requirements. First, the conditional generative adversarial network is employed to generate a large set of AGC regulation commands, followed by the hierarchical clustering method to extract typical scenarios. Then, a bi-level stochastic optimization model integrating HESS optimal sizing and frequency regulation strategies is developed. The outer level formulates a multi-objective optimization model that maximizes both the economic benefits and frequency regulation performance to determine the optimal HESS capacity. Based on model predictive control, the inner level optimizes the power allocation among thermal power unit, battery energy storage system (BESS), and supercapacitor energy storage system (SCESS) while incorporating AGC tracking errors and BESS cycle life. Furthermore, the multi-objective particle swarm optimization method is used to solve the outer-level sizing model, and the interior point method is employed to solve the inner-level quadratic programming model. The final optimal solution can be obtained through the alternate iteration between the two levels. Case studies demonstrate that the proposed scheme enhances AGC performance, extends BESS service life by 56.1% and increases the net present value by 7.4%, thereby validating its effectiveness.