Multi-agent deep reinforcement learning–based approach for optimization in microgrid clusters with renewable energy

Abstract

To the microgrid clusters with renewable energy, a microgrid clusters optimization method based on multi-agent deep reinforcement learning is proposed. First of all, in order to make full use of renewable energy, the electric-heat dynamic distribution factor is introduced into the microgrid clusters system model. Secondly, each microgrid is divided into an agent, so that the optimization problem is transformed into a reinforcement learning model in a multi-agent environment. The optimization of the microgrid clusters is completed with the goal of economy. The results of numerical examples show that the agent trains and learns from environmental information such as load, renewable energy and electricity price. The trained model can generate optimization strategy in real time, which can effectively deal with the uncertainty of renewable energy. Compared with the single agent algorithm, the training process of the model is easier to converge stably.