Cost-optimal sizing of battery energy storage systems in microgrids using artificial Rabbits optimization

Abstract

This paper presents a cost-optimal sizing framework for Battery Energy Storage Systems (BESS) in grid-connected microgrids using the Artificial Rabbits Optimization (ARO) algorithm. The main objective is to minimize the total operational cost of the microgrid by optimally determining the size of the BESS under real-world constraints, including dynamic pricing, varying load, and renewable energy availability. The proposed model incorporates technical and economic considerations, including depth-of-discharge limits, initial battery state-of-charge (SOC), and different wind turbine models. Three operational scenarios are evaluated: without BESS (Case A), and with BESS initialized at 20 %, and 100 % SOC (Cases B, and C). ARO is benchmarked against Particle Swarm Optimization (PSO), Artificial Bee Colony (ABC), and Firefly Algorithm (FA). For example, in Case C, ARO achieved the lowest operational cost of $778.81/day, compared to $793.86/day of PSO, $901.78/day of ABC, and $786.18/day of FA​. Additionally, in Case A, where no BESS is included, the total cost was $1069.10/day, while the introduction of optimally sized BESS in Case C reduced the cost to $778.81/day, demonstrating a significant economic benefit. Sensitivity analysis further confirms the robustness of the approach to changes in PV and WT generation, load demand, and battery efficiency. The results validate the effectiveness and computational efficiency of ARO for realistic and flexible microgrid energy management.