李彦君,刘友波,冉金周,刘俊勇.深度学习驱动的电网无功-电压优化控制策略模型[J].电测与仪表,2023,60(11):165-173. LI Yanjun,LIU Youbo,RAN Jinzhou,LIU Junyong.Optimal Control Strategy Of Reactive Power And Voltage In Power Grid Driven By Deep Learning[J].Electrical Measurement & Instrumentation,2023,60(11):165-173.
深度学习驱动的电网无功-电压优化控制策略模型
Optimal Control Strategy Of Reactive Power And Voltage In Power Grid Driven By Deep Learning
On the basis of applying deep neural network, this paper proposes a methodology to estimate and optimize the network losses for a designated power grid. This method is seeking the relationship among the network losses, operation state of the reactive compensation equipments and transformers with the deep neural network, and then to be optimized by improved genetic algorithm, find out the running state of the system control equipments under the condition of optimal network losses, which is convenient for real-time on-line regulation and control. The method gets rid of the convergence problem in solving nonlinear mixed variable programming model in the traditional power flow calculation and optimization process. This paper using the node voltages data replace of the power flow distribution data to achieve dimension reduction while ensuring global response information in power system,optimizing the training effect of deep neural network with good point set samples and improving traditional genetic algorithms to enhance the global and convergence speed of search. In the last, the proposed scheme is tested on the IEEE 30-bus system and the effectiveness of the methodology is proved to be effective.
