科技创新与应用2025，Vol.15Issue(19)：44-47,4.DOI:10.19981/j.CN23-1581/G3.2025.19.010
基于人工智能的高压电缆接地故障诊断方法研究
于金亮 1冯俊博 1齐福东 1郭怀龙1
作者信息
- 1. 国网冀北电力有限公司张家口供电公司,河北 张家口 075000
- 折叠
摘要
Abstract
This paper proposes a fault diagnosis method based on deep learning to solve the problems of low positioning accuracy and slow diagnosis efficiency in high-voltage cable grounding fault diagnosis.The method first preprocesses and extracts features on the collected cable fault signals through wavelet transform,and builds a feature data set containing multiple ground fault types;then designs an improved convolutional neural network model that integrates the attention mechanism and residual connection structure,realizing adaptive learning and classification of fault characteristics;Finally,a real-time fault diagnosis system is developed to achieve rapid fault location and identification.Experimental results show that in the fault diagnosis of 10～35 kV high-voltage cables,the fault location accuracy of this method reaches 98.5%,which is 15%higher than the traditional method,and the average diagnosis time is shortened to 2.3 seconds,and it still maintains stable diagnosis performance in complex noise environments.This method has been applied in transmission line fault diagnosis of a provincial power company,providing an effective guarantee for the safe and stable operation of the power system.关键词
高压电缆/接地故障诊断/深度学习/故障特征提取/智能诊断系统Key words
high-voltage cable/ground fault diagnosis/deep learning/fault feature extraction/intelligent diagnosis system分类
信息技术与安全科学引用本文复制引用
于金亮,冯俊博,齐福东,郭怀龙..基于人工智能的高压电缆接地故障诊断方法研究[J].科技创新与应用,2025,15(19):44-47,4.
