Enhanced Multi-Objective Microgrid Scheduling Through Adaptive BSA With Dynamic Cognitive Learning

This paper proposes an enhanced multi-objective optimization approach called dynamic cognitive learning bird swarm algorithm (DCL-BSA) for microgrid scheduling. It integrates adaptive inertia weights, time-varying cognitive coefficients, and Levy flight patterns to improve exploration and exploitation capabilities. The comprehensive scheduling model incorporates economic costs, environmental impacts, and time-of-use pricing considerations, while a fuzzy decision-making framework effectively balances competing objectives. Extensive simulations on a realistic microgrid system demonstrate DCL-BSA's superior performance compared to existing methods. The algorithm achieves 7.0% lower operational costs, and 15.8% reduced environmental impact compared to baselines while maintaining higher convergence reliability with a 98.5% success rate. Computational efficiency analysis shows 16.1% faster convergence and 25% reduced memory utilization. The results validate DCL-BSA's effectiveness in developing robust scheduling strategies.