Optimizing economic and operational performance in AC microgrids: An intelligent energy management strategy for BESS using the Generalized Normal Distribution Optimizer

This work addresses the problem of efficient management of battery energy storage systems (BESS) in alternating current microgrids (AC MGs), seeking to optimize technical (energy losses), economic (operating costs), and environmental (emissions) indicators in on-grid and off-grid operation scenarios. Previous studies often lack comprehensive comparisons with multiple optimization methodologies, thorough statistical validation, and adaptability to variable generation-demand scenarios in both grid-connected and isolated modes. To address this gap, the present study applies the Generalized Normal Distribution Optimization (GNDO) algorithm to the energy management of BESS in AC MGs. Its performance is validated against three established methods: a Continuous Genetic Algorithm (CGA), Particle Swarm Optimization (PSO), and the JAYA algorithm. The approach is evaluated on two test systems: 33-node on-grid MG and 27-node off-grid MG under photovoltaic generation and variable demand. Performance metrics include solution quality, standard deviation, and processing time, based on 100 independent runs per method. Results show that GNDO consistently outperforms the benchmark algorithms, with average improvements of 1.4349% in costs, 4.4426% in losses, and 0.1833% in emissions. The findings confirm GNDO's robustness and efficiency in constrained energy management environments. • Proposal of a novel optimization algorithm, GNDO, for managing BESS in AC microgrids. • Validation of GNDO using two representative test systems: 33-node ongrid and 27-node off-grid microgrids. • Demonstrated superior performance of GNDO compared to CGA, PSO, and JAYA in reducing operating costs, energy losses, and C O 2 emissions. • Robust statistical analysis of GNDO across multiple runs, ensuring solution reliability. • Highlighted practical implications for variable-cost and off-grid microgrid scenarios. • Future directions include GNDO integration in multi-objective optimization and systems with wind turbines.