Optimal Placement and Sizing of Battery Energy Storage System for Loss Reduction in A Power Distribution System Using Chaotic Coyote Optimization Algorithm

Abstract

The growing global electricity demand and increasing integration of renewable energy have intensified the need for efficient and resilient power distribution systems. Power losses in distribution networks remain a persistent challenge, increasing operational costs and reducing efficiency. This study investigates the optimal placement and sizing of battery energy storage systems to minimize losses in a 148-bus distribution network. A Chaotic Coyote Optimization Algorithm is proposed, which enhances the conventional Coyote Optimization Algorithm by incorporating chaotic maps to improve exploration, prevent local optima trapping, and accelerate convergence. Two scenarios were evaluated: (i) a grid-connected system and (ii) a grid-connected system with solar photovoltaic integration. For each scenario, single-unit and dual-unit battery energy storage configurations were tested using simultaneous and sequential optimization strategies. The Chaotic Coyote Optimization Algorithm consistently outperformed the Coyote Optimization Algorithm, Whale Optimization Algorithm, and Particle Swarm Optimization, achieving faster convergence and greater solution accuracy. Results showed that dual-unit battery energy storage configurations significantly reduced losses compared to single-unit setups, with the best case (dual-unit plus solar photovoltaic integration) achieving a 51.4 percent reduction in active power losses and an improvement in minimum bus voltage from 0.927 per unit to 0.975 per unit. The study demonstrates that combining battery energy storage with renewable integration not only reduces distribution losses but also enhances voltage stability. The proposed Chaotic Coyote Optimization Algorithm framework provides a robust methodology for energy storage optimization, offering valuable insights for utilities in planning sustainable, cost-effective, and resilient smart grids.