Transmission Line Fault Diagnosis Based on Variational Modal Decomposition and Long and Short Term Memory Network
In order to improve the timeliness and reliability of fault early warning,a fault early warning method based on variational mode decomposition(VMD)sample entropy and long short term memory network is proposed.Firstly,VMD is used to decompose the three-phase voltage signal to obtain a series of modal components.Then,sample entropy is used to extract fault features to improve the correlation of fault features.Finally,Bayesian optimization parameter tuning is used to improve the performance of long and short term memory network,optimize the fault diagnosis model of transmission lines,and improve the convergence speed and prediction accuracy of the model.The simulation results show that VMD+BLSTM has the highest fault diagnosis rate of transmission lines,fewer iterations and better noise robustness,compared with Wavelet+BP,Wavelet+SVM,and Wavelet+Whale+ELM.
万方数据
维普
