Electric Vehicle Charging and Discharging Algorithm based on Reinforcement Learning with Data-driven Approach in Dynamic Pricing Scheme

Abstract

In the smart grid environment, penetration of electric vehicles (EV) is increasing and dynamic pricing and V2G are being introduced. In this situation, automatic charging and discharging scheduling responding to the electricity prices that change over time is required in order to reduce charging cost oa to sample variables related to charger usage pattern, so that the variables can be casted in the training process of a reinforcement learning agent. By doing this, the agent can optimally learn the characteristics of the target charger. We analyze the effectiveness of our proposed algorithm from two perspectives, i.e., charging cost and the load shifting effect. Simulation results show that the proposed method outperforms the benchmarks that simply model usage pattern through general assumptions in terms of charging cost and the load shifting effect.