Scholarlink

High-voltage cable lines in cities ensure the efficient transportation and distribution of electricity. Ensuring the safe and reliable operation of these cable lines under various environmental conditions is of vital importance to the entire city’s power system. Approximately 80% of cable faults are caused by insulation defects. Partial discharge is a significant indicator of insulation deterioration. Therefore, real-time monitoring of partial discharge signals in high-voltage cables and precise identification of their types are of great significance for preventing faults and extending the service life of equipment. At present, the identification of partial discharge in cables mainly relies on intelligent algorithms, which have problems such as large computational load. However, expert systems can make rapid judgments but cannot accurately diagnose the types of faults. This article will respectively introduce expert systems and convolutional neural networks, and combine expert systems and intelligent algorithms to construct a high-voltage cable condition assessment system. This enables accurate determination of fault types while reducing the amount of computation. Finally, pattern recognition was carried out on the partial discharge of faulty cables identified by the expert system, verifying the effectiveness of the method.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Partial Discharge Pattern Recognition and Early Warning of XLPE High-voltage Cables Based on CNN Deep Learning

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Ji Wu,
  • Xiaokang Lei,
  • Yinge Li,
  • Tian Zeng,
  • Xin Yu,
  • Xiaosheng Peng

摘要

High-voltage cable lines in cities ensure the efficient transportation and distribution of electricity. Ensuring the safe and reliable operation of these cable lines under various environmental conditions is of vital importance to the entire city’s power system. Approximately 80% of cable faults are caused by insulation defects. Partial discharge is a significant indicator of insulation deterioration. Therefore, real-time monitoring of partial discharge signals in high-voltage cables and precise identification of their types are of great significance for preventing faults and extending the service life of equipment. At present, the identification of partial discharge in cables mainly relies on intelligent algorithms, which have problems such as large computational load. However, expert systems can make rapid judgments but cannot accurately diagnose the types of faults. This article will respectively introduce expert systems and convolutional neural networks, and combine expert systems and intelligent algorithms to construct a high-voltage cable condition assessment system. This enables accurate determination of fault types while reducing the amount of computation. Finally, pattern recognition was carried out on the partial discharge of faulty cables identified by the expert system, verifying the effectiveness of the method.