A Reinforcement Learning Approach For Ev Charging Station Dynamic Pricing And Scheduling Control

Electric vehicles (EVs) are regarded as one of the most effective ways to reduce the oil demands and gas emissions. A charging station with well-designed scheduling and pricing strategy can benefit both EV users and the electricity distribution system. This paper proposes a profit-maximizing joint charging scheduling and pricing scheme for a public charging station. We first show that the profit-maximizing strategy is the solution to a Markov decision process (MDP). To solve the problem, we propose a reinforcement learning (RL) approach that does not require any non-causal information or distributional information. To ensure convergence and a low computational complexity of the RL approach, we propose a state linear approximation scheme. Through simulating with real-world data, we show that the proposed RL algorithm achieves on average 138.5% higher profit than representative benchmark algorithms.