Pricing and Charging Scheduling for Cooperative Electric Vehicle Charging Stations via Deep Reinforcement Learning

Abstract

The rapid adoption of electric vehicles (EVs) stimulates the proliferation of charging stations (CSs), motivating the cooperative management of growing CSs. However, cooperative CS management still remains an open problem, due to the uncertain user behavior and heterogeneous service capabilities. To capture the CS dynamics caused by uncertain user behavior, we propose a deep reinforcement learning (DRL)-based cooperative method for multiple CSs, towards maximizing the total profit. The proposed method determines pricing and charging scheduling decisions for CSs, considering stochastic CSs selection and its impact on CSs energy supply. In order to reduce the computational burden of dimensions caused by the time-varying decisions, we design a discretization strategy for action space, based on the current market rule of tiered pricing and CS types. The simulations using real data demonstrate that our proposed method can obtain higher profit than the independent operation and benchmark cooperative algorithms such as Q-learning.