When the single-ended method is used to range traveling wave fault of transmission line,there will be confusion be-tween the opposite bus reflection wave head and the fault point reflection wave head in the first three wave heads.To solve this prob-lem,a single-ended fault ranging method with TT transformation and back propagation neural network algorithm is proposed.Firstly,TT transformation is used to analyze the single-ended fault traveling wave signal,and time values and polarity of the first three travel-ing wave heads are collected as the sample data set.Then the sample data set is learned with the strong fitting ability of BP neural net-work.Finally,a trained neural network model is obtained,which is used to range the single-ended fault of the transmission line.In ad-dition,genetic algorithm is introduced to optimize the BP neural network,so as to improve the ranging accuracy of the neural network.The simulation results of Matlab/Simulink show that the ranging error of this method meets the requirement of practical application,and the ranging accuracy is not affected by fault type and transition resistance,so this method is reliable.
维普
万方数据