Day-ahead self-scheduling strategy for energy storage based on deep reinforcement learning algorithm with hindsight experience replay

With the development of renewable power systems, energy storage will participate in the power market as a major power source, so it is necessary to study the self-scheduling strategy for energy storage to participate in the day-ahead power market. At present, the intelligent quotation simulation technology of energy storage based on deep reinforcement learning algorithm has been developed rapidly, while most of the traditional reinforcement learning algorithms are aimed at optimizing the offer strategy of energy storage for a single time period, and it is difficult to optimize the one-day strategy of energy storage as a whole. Thus, this paper adopts the Soft Actor Critic algorithm with Hindsight experience replay to simulate the self-scheduling behavior of energy storage, and the simulation results show the effectiveness of the algorithm.