Multi-Objective Optimization Scheduling Strategy for Energy Storage Systems Based on Improved NSGA-III

This paper proposes a multi-objective optimization scheduling strategy for energy storage power systems based on an improved NSGA-III algorithm, aiming to address the grid stability challenges posed by distributed photovoltaic (PV) generation in China’s power system. With the rapid growth of renewable energy, particularly solar power, grid instability has become increasingly severe. As an effective solution, energy storage systems (ESS) can enhance grid stability and reliability. To optimize the configuration and scheduling of ESS, this paper introduces an improved NSGA-III algorithm incorporating Levy path-based search optimization, adaptive weight adjustment, dynamic constraint handling, and multi-scale optimization strategies. The proposed improvements effectively mitigate conflicts in multi-objective optimization and provide more efficient solutions across multiple dimensions. To further enhance optimization performance, a bi-level optimization model is employed. The outer layer focuses on optimizing the configuration of ESS, while the inner layer manages the charge-discharge scheduling. The synergy between these two layers ensures both the accuracy and flexibility of ESS deployment and operation. Experimental results demonstrate that the improved NSGA-III algorithm outperforms the conventional NSGA-III in terms of Pareto solution set diversity, convergence, and computational efficiency. The proposed strategy significantly enhances the economic efficiency of ESS and improves grid stability, making it highly valuable for practical applications.