Electric vehicle charging scheduling for multi-microgrids load balancing using LSTM load forecasting

Abstract As the ownership of electric vehicles grows rapidly, multi-microgrids are facing challenges such as sharp increases in charging load and unbalanced resource allocation in charging stations. In order to solve these problems, we propose an optimal scheduling strategy to solve the delay load of waiting vehicles using load forecasting. Firstly, using the historical load data and the LSTM time series prediction method, the future load of the charging stations is predicted. Secondly, considering the delay load characteristics of queuing vehicles, a mathematical model of the charging schedule is constructed to dynamically coordinate the queuing delay load with the real-time charging requests. The simulation results demonstrate that the proposed scheduling strategy can effectively reduce the load difference between multi-microgrids, improve the load coordination ability between multi-microgrids, and ensure the safe and stable operation of multi-microgrids. It provides theoretical research support for the efficient coordinated operation of multi-microgrids and electric transportation.