Research on Surface Discharge Image Classification of High-Voltage Cables Based on ResNet Network

This paper proposes an image classification method based on ResNet for detecting surface discharge faults in high-voltage cables. By collecting and labeling a dataset of high-voltage cable surface discharge images, the ResNet deep convolutional neural network model was applied for training. The experimental results show that the model can accurately identify surface discharge and non-discharge states. This method enhances the automation of cable discharge fault detection, contributing to early fault warning and maintenance in power systems, thereby ensuring the safety and reliability of grid operation. The research results indicate that fault detection technology based on deep learning has broad application prospects in smart grids.