Power System Transient Stability Assessment Method Based on Convolutional Neural Network

摘要

A method based on convolutional neural network is proposed for power system transient stability assessment in this paper, which can overcome the shortcomings of traditional evaluation methods and satisfy the requirements with high assessment accuracy of power system transient stability assessment problems. Firstly, the structural characteristic of convolutional neural network is introduced in this paper, then the applicability in transient stability assessment problems is analyzed. Secondly, the training methods of convolutional neural network are optimized according to the characteristics of transient stability assessment problem, and the batch normalization algorithm is added to establish the transient stability assessment model. Finally, the simulation performed on the New England 10-machine 39-node system demonstrates the effectiveness of the proposed method.