An Intelligent Power Flow Violation Adjustment Method Based on Adversarial Process
Shiqi Duan, Juan Yu, Zhifang Yang, Maosheng Gao, Tao Chen, Guangde Dong · 2023
Adjusting the power flow mode is vital for daily power system operation. However, with the increase in power system scale and uncertainty, it is tougher to ensure the N-l critics when adjusting the power flow mode. To this end, this paper proposes an intelligent power flow violation adjusting method based on adversarial process to meet N-l verification. While any branch power flow or voltage magnitude violation occurs in the N-1 verification, it can intelligently generate a new power flow mode to satisfy the N-l verification by changing node injection power, e.g., adjusting the generation of generators or renewable resources, shedding loads, and transfer loads. Firstly, the functional relationship between the power flow input and output under the N-l multiple topologies is constructed based on the fully connected neural network (FCN). Then, the power flow violation adjusting method is constructed based on the adversarial process of the power flow calculation neural networks. Finally, through simulations of the IEEE 30-bus system, it is verified that the proposed intelligent power flow violation adjusting method can quickly generate a power flow mode that meets N-l verification.