THz Wave Detection of Gap Defects Based on a Convolutional Neural Network Improved by a Residual Shrinkage Network

Zhonghao Zhang; Guozheng Peng; Yuanpeng Tan; Tianjiao Pu; Liming Wang

doi:10.17775/CSEEJPES.2020.02460

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

THz Wave Detection of Gap Defects Based on a Convolutional Neural Network Improved by a Residual Shrinkage Network

Regular Papers | 更新时间：2026-10-15

- THz Wave Detection of Gap Defects Based on a Convolutional Neural Network Improved by a Residual Shrinkage Network
- CSEE Journal of Power and Energy Systems Vol. 9, Issue 3, Pages: 1078-1089(2023)
- 作者机构：
  
  1. China Electric Power Research Institute,Beijing,China
  2. Tsinghua Shenzhen International Gradual School,Shenzhen,China
- 作者简介：
- 基金信息：
- DOI：10.17775/CSEEJPES.2020.02460
  CLC：
- Published：01 May 2023
- 稿件说明：
移动端阅览
Zhonghao Zhang, Guozheng Peng, Yuanpeng Tan, et al. THz Wave Detection of Gap Defects Based on a Convolutional Neural Network Improved by a Residual Shrinkage Network[J]. CSEE Journal of Power and Energy Systems, 2023, 9(3): 1078-1089.
DOI：

Zhonghao Zhang, Guozheng Peng, Yuanpeng Tan, et al. THz Wave Detection of Gap Defects Based on a Convolutional Neural Network Improved by a Residual Shrinkage Network[J]. CSEE Journal of Power and Energy Systems, 2023, 9(3): 1078-1089. DOI： 10.17775/CSEEJPES.2020.02460.

摘要

Abstract

Internal air gap is a serious type of defect in the insulation equipment

which threatens the safe operation of the power grid. In order to diagnose the position and thickness of the internal air gap

this paper proposes a terahertz wave detection method based on wavelet analysis and a CNN (convolution neural network) model. According to the time-frequency characteristics of the wavelet cluster

the calculation method of air gap depth is proposed. To determine the thickness of the internal air gap

the performances of several classification methods

such as waveform feature analysis

Bayes

MLP (Multi-layer Perceptron)

SVM (Support Vector Machine) and CNN are compared. The results show that the CNN modified by a residual shrinkage network and SVM (CNN-RSN-SVM) has the best performance. By adjusting the parameters

the classification accuracy of the CNN-RSN-SVM model can be elevated to 98.91%. Furthermore

the 3D imaging method of air gap defect based on wavelet analysis and CNN-RSN-SVM classification model is formed.

关键词

Keywords

references

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Terahertz Imaging Reconstruction for Super-Resolution Identification of Millimeter-Scale Defects in XLPE Cable Insulation

Multi-Synchrosqueezing Transform Approach for Forced Oscillation Source Location of Power Systems in Time-Frequency and Frequency Domains

Modeling VSC with Active/Reactive Current Excitation and Internal Voltage Response for Grid Amplitude/Frequency Modulation Dynamics Analysis

Dual Clustering Detection Model of Power Theft Based on User Behavior Analysis

Sub/Super-Synchronous Oscillation Detection Based on Matching Synchroextracting Wavelet Transform

Related Author

Shuaibing Li

Daqi Zhang

Yi Cui

Yongqiang Kang

Haiying Dong

Tao Jiang

Nan Ye

Xue Li

Related Institution

School of New Energy and Power Engineering, Lanzhou Jiaotong University

School of Electrical Engineering and Computer Science, The University of Queensland

Department of Electrical Engineering, Northeast Electric Power University

School of Electrical and Electronic Engineering, Huazhong University of Science and Technology

School of Electrical Engineering, Guangxi University

AI问答

⁰