Research on Frequency Regulation Market Application Strategy of Integrated Fire-Storage System Based on Q -Reinforcement Learning

With the gradual establishment of market-based trading mode of power frequency regulation auxiliary service, new power supply formats such as integrated fire-storage system are developing rapidly. At the same time, the bidding strategy of integrated fire-storage system in frequency regulation market has become the focus of research in this field. In order to meet the demand of high frequency and high precision market transaction of the integrated fire storage system, a frequency regulation market declaration strategy based on Q-reinforcement learning is proposed. Q-reinforcement learning is a model-free reinforcement learning algorithm. Considering the bidding characteristics of the current frequency regulation market, the environment variables, action library and return function of the frequency regulation market declaration agent of the integrated fire storage system are constructed, and the implementation process of the algorithm is proposed. The effectiveness of the proposed method is verified by constructing an example based on the actual grid data of a province. Example results show that the results of this algorithm are basically consistent with the current target declaration strategy, and its computational efficiency is significantly higher than the existing algorithm, which is more in line with the trading needs of operators of integrated fire storage system in frequency regulation market.