Scholarlink

In order to solve the problem that it is difficult to locate the grounding fault position of electrical equipment in substation, the grounding fault location model of electrical equipment in substation is constructed by using the current data of equipment in substation monitored by wide-area synchronous current sensing equipment and improving GoogLeNet. Firstly, in order to overcome the problem of noise interference in monitoring signals, time-frequency domain features of current transients are both extracted, and the regional positioning features in the station are constructed together with the monitoring current direction; Secondly, the output layer of GoogLeNet is improved, that is, the support vector machine improved by particle swarm optimization algorithm is replaced, and the classification ability of GoogLeNet is enhanced to adapt to the small sample of faults in substation. Finally, the 35 kV substation fault simulation model is used to obtain sample data after simulating different working conditions, and the proposed fault location model is verified. The results show that the proposed model is effective.

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

Grounding Fault Location Model in Substations Based on Current Sensor Information and Improved-GoogLeNet

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Long He,
  • Yongming Zhu,
  • Jincai Ma,
  • Rong Liu

摘要

In order to solve the problem that it is difficult to locate the grounding fault position of electrical equipment in substation, the grounding fault location model of electrical equipment in substation is constructed by using the current data of equipment in substation monitored by wide-area synchronous current sensing equipment and improving GoogLeNet. Firstly, in order to overcome the problem of noise interference in monitoring signals, time-frequency domain features of current transients are both extracted, and the regional positioning features in the station are constructed together with the monitoring current direction; Secondly, the output layer of GoogLeNet is improved, that is, the support vector machine improved by particle swarm optimization algorithm is replaced, and the classification ability of GoogLeNet is enhanced to adapt to the small sample of faults in substation. Finally, the 35 kV substation fault simulation model is used to obtain sample data after simulating different working conditions, and the proposed fault location model is verified. The results show that the proposed model is effective.