Energy Management System for Microgrids based on Deep Reinforcement Learning

The increasing use of distributed and renewable energy resources presents a challenge for traditional control methods due to the higher complexity and uncertainty brought by these new technologies. To address these challenges, rein-forcement learning algorithms are used to design and implement an energy management system (EMS) for different microgrids configurations. Reinforcement Learning (RL) approach seeks to train an agent from their interaction with the environment rather than from direct data such as in supervised learning. With this in mind, the problem of energy management is be posed as a Markov decision process and it is solved using different state-of-the-art Deep Reinforcement Learning (DRL) algorithms, such as Deep Q-Networks (DQN), Proximal Policy Optimization (PPO) and Twin Delayed Deep Deterministic Policy Gradient (TD3). Additionally, these results are compared with traditional EMS implementations such as Rule-Based and Model Predictive Control (MPC) used as benchmarks. Simulations are run with the novel Pymgrid module build for this purpose. Preliminary results show that RL agents have comparable results to the classical implementations with some possible benefits for generic and specific use cases.