A Peak Regulation Ancillary Service Optimal Dispatch Method of Virtual Power Plant Based on Reinforcement Learning

With the development of power market reform in China, the market trading mechanism has been improved. Auxiliary service market has become an important part in current market transaction reform. As an effective form of user side participating in power grid market transaction, virtual power plant(VPP) is expected to become an important auxiliary service provider. This paper proposes the basic structure of VPP under energy Internet and analyzes the response characteristics of distributed energy resource. A peak regulation auxiliary service optimization dispatch method of VPP based on reinforcement learning algorithm is proposed to solve the operation optimization problem of VPP participating in the peak regulation auxiliary service market. Based on the strong adaptability of reinforcement learning, this method can meet the operation control requirements of different scenarios and different types of VPPs. Finally, a case study is constructed based on the actual data of a VPP demonstration project in Northern Hebei of China, which verifies the effectiveness of the proposed method.