Research on Classification Method of Urban Underground Cable Partial Discharge Defects

After the urban underground power transmission cable is put into use, the cable section is prone to metal residue, semiconductor residue, insulation moisture and other defects when the cable joint is connected, resulting in partial discharge phenomenon of the cable. This paper analyzes the partial discharge of cable, proposes a convolutional neural network to identify the local discharge signal, and improves the network to obtain better results. This topic has a wide range of application scenarios and is suitable for the identification and analysis of cable discharge signals generated during cable operation.