Scholarlink

In order to solve the problem of low accuracy of fault diagnosis identification of transmission line based on transient current traveling wave, a method based on the combination of Gramian Angular Field(GAF) transform and ResNeXt-18 deep learning framework is offered to achieve the accurate diagnosis of transmission line fault types. First, the original transient current traveling wave signal captured by the transmission line distributed fault surveillance system is standardized, and then the time series imaging of the current traveling wave signal is realized by the GAF transform; second, the ResNeXt-18 deep learning framework is used to achieve the refined diagnosis of transmission line faults through the training learning of the GASF dataset; Comparison experiments are conducted with ResNet, GoogLeNet, and CNN, and the experimental results show that ResNeXt-18 has better diagnostic accuracy.

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

Transmission Line Fault Diagnosis Based on GAF and Deep Learning Algorithms

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Yongzeng Ji,
  • Wanxing Feng,
  • Jian Li,
  • Yingpu Xie,
  • Haonan Zhu,
  • Jianyu Ge

摘要

In order to solve the problem of low accuracy of fault diagnosis identification of transmission line based on transient current traveling wave, a method based on the combination of Gramian Angular Field(GAF) transform and ResNeXt-18 deep learning framework is offered to achieve the accurate diagnosis of transmission line fault types. First, the original transient current traveling wave signal captured by the transmission line distributed fault surveillance system is standardized, and then the time series imaging of the current traveling wave signal is realized by the GAF transform; second, the ResNeXt-18 deep learning framework is used to achieve the refined diagnosis of transmission line faults through the training learning of the GASF dataset; Comparison experiments are conducted with ResNet, GoogLeNet, and CNN, and the experimental results show that ResNeXt-18 has better diagnostic accuracy.