Deep Reinforcement Learning in Smart Grid: Progress and Prospects

Abstract

The combination of reinforcement learning and deep learning has shown some remarkable results in many scientific fields. Deep reinforcement learning algorithms are particularly good at understanding and modeling adaptive decision-making in dynamic environments. In recent years, this concept has been successfully applied to smart grids. In this paper, we provide a brief introduction to the concepts of reinforcement and deep reinforcement learning to the power system engineers and present research progress and prospects in the field. Additionally, we identify smart grid engineering domains that need extensive pattern-based modeling as being particularly suitable for deep reinforcement learning.