Research on Reinforcement Learning-based Optimization Algorithm for Electric Vehicle Charging Scheduling

With the rapid growth in the number of electric vehicles, how to effectively manage their charging demand has become an important challenge for power systems. In this paper, we propose an EV charging scheduling optimization algorithm that combines power big data analysis and deep reinforcement learning (DQN). The algorithm aims to optimize the charging scheduling to balance the grid load, reduce the charging cost and enhance the satisfaction rate of users’ charging demand. Through simulation experiments, the effectiveness of the DQN optimization algorithm in reducing the peak hour grid load, lowering the charging cost and enhancing user satisfaction is verified. The experimental results show that the optimization scheme significantly reduces the peak grid load, reduces the charging cost by about 10.29%, and improves the user demand satisfaction rate by 5.2%.