Preprint: A Comparison of different RL Algorithms for the energy management for bidirectional charging of EVs

Abstract

This study compares different Reinforcement Learning (RL) Algorithms to optimize the bidirectional electric vehicle (EV) charging in a microgrid with dynamic pricing and renewable energy. The environment includes an EV, photovoltaic, stationary battery, household loads, and grid connection. The RL agents have to learn to maximize the profit of the microgrid under the constraint to reach the desired departure SOC’s of the EV battery. The different RL algorithms are compared by their test results.