Identification method of cascading failure in high-proportion renewable energy systems based on deep learning
Yuhong Zhu, Xiaoming Liu, Bo Chen, Donglei Sun, Dong Liu, Yongzhi Zhou
Abstract
A high proportion of new energy sources are connected to the grid, which not only alleviates the energy shortage, but also brings hidden dangers to the safe and stable operation of the grid. Aiming at the inadequate consideration of uncertainty caused by new energy sources such as wind power being connected to the grid, a method for identifying voltage-dominant cascading fault types is proposed. A voltage-dominant fault analysis model in the sense of probability is established. Based on the fault data, the nonlinear mapping relationship between the initial fault and the fault propagation characteristics is established through the neural network. Finally, the effectiveness of the method proposed in this paper is verified in the IEEE39-bus system, and the simulation results show that the method proposed in this paper can effectively identify the fault types of high-proportion new energy systems.