Optimal Configuration of Energy Storage Capacity in Low-voltage DC Microgrid under Different Operating Modes

A multi-scenario-based capacity configuration method for low-voltage DC microgrids is used to manage the issues of high uncertainty in renewable energy output and high light rejection rate in low-voltage DC microgrid systems. Firstly, using the K-means clustering algorithm to cluster annual photovoltaic output data, obtaining four typical days of photovoltaic power and their corresponding probabilities. Then, based on different output scenarios, considering the different operating states of grid-connected converters, the system operation modes are divided. Set constraints such as energy storage state of charge and load loss rate to minimize the annual total investment cost. Finally, the artificial bee colony optimization algorithm was used to solve the capacity configuration results of each unit, and the optimal capacity configuration results were obtained under different operating modes. The research results indicate that the proposed method can effectively lower the light rejection rate, and the annual investment cost is the lowest in the bidirectional grid connection mode; At the same time, selecting an appropriate load loss rate can significantly reduce the annual investment cost.