Identification of Traveling Wave Current of Cable Fault in Distribution Network

With the increase of distribution network load and the complexity of the operating environment of power equipment, the identification and diagnosis of distribution network cable faults have become particularly important. Cable fault traveling wave signals are key indicators for detecting electrical faults and can provide valuable information for fault location and early warning. However, cable traveling wave signals are usually interfered by noise, resulting in reduced accuracy of fault detection. This paper proposes a distribution network cable fault traveling wave identification method based on adaptive filtering and convolutional neural network (CNN). First, the adaptive filtering method is used to remove noise to improve the quality of the signal; then, CNN is used to extract and classify the features of the denoised traveling wave signal. Experimental results show that this method has significant advantages in cable fault identification, with an accuracy rate of 90.1 %, which can effectively improve the accuracy and reliability of fault identification. Compared with traditional machine learning methods, CNN shows better performance in processing complex cable traveling wave data.