Optimal Scheduling Method of Combined Wind–Photovoltaic–Pumped Storage System Based on Improved Bat Algorithm

Abstract

Pumped storage power stations not only serve as a special power load but also store excess electricity from the power system, significantly reducing the curtailment of wind and solar power. This dual function ensures the stable operation of the power grid and enhances its economic benefits. The scheduling optimization problem of a combined wind–solar–pumped storage system is addressed in this study, and an optimization scheduling model is proposed with the objective of maximizing total system revenue. The model is designed to comprehensively account for the generation revenues from wind power, photovoltaic power, thermal power, and pumped storage, as well as the penalty costs associated with pollutant emissions. To address the limitations of traditional algorithms, which are prone to being trapped in local optima and exhibit slow convergence, an improved bat algorithm was developed. The algorithm is enhanced through the use of chaotic mapping to expand the initial solution space, the incorporation of adaptive step-size updates to improve convergence efficiency, and the integration of the Cauchy function to strengthen global search capabilities, thereby effectively avoiding local optima. Simulation results have demonstrated that the improved algorithm achieves significant improvements over traditional bat algorithms and particle swarm optimization (PSO) in terms of optimization efficiency, with total revenue increases of 21.9% and 24.6%, respectively. The optimized scheduling plan is shown to fully utilize the flexible regulation capabilities of pumped storage, mitigating the adverse effects of wind and photovoltaic output fluctuations on grid operations and achieving a balanced trade-off between economic and environmental objectives.