Transmission Line Fault Diagnosis Based on Wavelet Packet Analysis and Convolutional Neural Network

Abstract

Accurate classification of fault types for highvoltage transmission lines is a prerequisite for fault location and fault recovery. To this end, this paper proposes a fault classification method based on wavelet packet analysis and convolutional neural network. Firstly, according to the difference between the three-phase voltage and current wavelet energy vectors in the line fault, the samples representing different fault categories are constructed, then, the sample of the fault category is reconstructed into a grayscale image by the wavelet energy probability vector. The sample is trained by the convolutional neural network algorithm to obtain a CNN model that identifies different fault types. The simulation results show that the method has fast recognition speed and fault recognition accuracy is not affected by parameters such as fault resistance, fault location and system operation mode. It has strong practicability and reliability.