The energy storage participation in the optimal operation of the microgrid can effectively solve the problem of system security and stability caused by large-scale grid connection of renewable energy. Based on bi-level programming theory, an energy storage optimization configuration model with load fluctuation, system cost and energy storage SOC deviation is established. An improved multi-objective particle swarm optimization algorithm is proposed to solve the model. According to the optimal similarity, the value of inertia weight is guided, and the cross-mutation operation is introduced in time to improve the convergence of the algorithm and improve the ability to jump out of the local optimal solution. To ensure the globality and uniformity of the Pareto solution set, a strategy for dynamically updating the set based on multiple iteration directions is proposed. Finally, the weight is established based on information entropy, and the optimal scheme is selected by the TOPSIS method. The modified IEEE-33 node system is used to analyze the effectiveness and superiority of the proposed algorithm in solving the optimal configuration of microgrid energy storage. This work is supported by National Natural Science Foundation of China (No. 51707089) and Science and Technology Project of State Grid Corporation of China “Key Technologies and Empirical Research on Electricity-cored Multi-energy Complementary Integrated Power Conversion Equipment”.