Research on Separation and Clustering Method of Multi-source PD Signals for Metal-clad Switchgear

LI Ruipeng; HUANG Chao; ZHANG Wei; ZHANG Long; YANG Yuqi; LI Hongjie

doi:10.13296/j.1001-1609.hva.2015.10.017

您当前的位置：

首页 >

文章列表页 >

Research on Separation and Clustering Method of Multi-source PD Signals for Metal-clad Switchgear

更新时间：2025-04-26

- Research on Separation and Clustering Method of Multi-source PD Signals for Metal-clad Switchgear
- High VoltageApparatus Vol. 51, Issue 10, Pages: 110-116(2015)
- 作者机构：
  
  1. 西安高压电器研究院有限责任公司
  2. 深圳供电局有限公司
  3. 西安交通大学
- 作者简介：
- 基金信息：
- DOI：10.13296/j.1001-1609.hva.2015.10.017
  CLC： TM591
- Published：2015
- 稿件说明：
移动端阅览
[1]李锐鹏,黄超,张炜,张龙,杨宇琦,李洪杰.开关柜多源局部放电信号分离及聚类方法研究[J].高压电器,2015,51(10):110-116.

LI Ruipeng, HUANG Chao, ZHANG Wei, et al. Research on Separation and Clustering Method of Multi-source PD Signals for Metal-clad Switchgear[J]. High voltageapparatus, 2015, 51(10): 110-116.
[1]李锐鹏,黄超,张炜,张龙,杨宇琦,李洪杰.开关柜多源局部放电信号分离及聚类方法研究[J].高压电器,2015,51(10):110-116. DOI： 10.13296/j.1001-1609.hva.2015.10.017.

LI Ruipeng, HUANG Chao, ZHANG Wei, et al. Research on Separation and Clustering Method of Multi-source PD Signals for Metal-clad Switchgear[J]. High voltageapparatus, 2015, 51(10): 110-116. DOI： 10.13296/j.1001-1609.hva.2015.10.017.

摘要

文中采用暂态对地电压（transient earth voltage

TEV）法对10 kV高压开关柜局部放电进行检测

并利用短时傅里叶变换（short-time Fourier transform

STFT）分析了开关柜不同类型放电下局放TEV信号的时间频率特性

结果表明不同放电TEV信号的时频特性不同。基于此

又对信号STFT时频分析结果提取了信号的时间中心tc、频率中心fc以及中心矩μc3个特征参数。试验表明多源局部放电TEV信号能够在tc

c三维特征空间内实现信号分离。进一步利用GK模糊聚类方法在tc

c三维空间内实现了对多源放电TEV信号的智能分离与聚类。试验及分析结果表明与基于傅里叶变换的传统方法比较

文中方法具有更优的信号分离效果

且实现了多源信号的智能聚类。

Abstract

Transient earth voltage（TEV） method is applied to detect the partial discharge in 10 kV metal-clad switchgear. Time-frequency characteristics of different PD TEV signals in metal-clad switchgear are analyzed employing short-time Fourier transform（STFT）method and tests show that different PD TEV signals have different time-frequency performance. A new separation method of multi-source PD TEV signals for switchgear is introduced by extracting time center tc

frequency center fcand center moment μcfrom time-frequency result of TEV signals.Multi-source PD signals can be gathered and separated in tc

μcspace using the proposed method and Gustafson Kessel（GK）clustering algorithm is applied to cluster the signals intelligently. Tests and analysis prove that the propose method for multi-source PD separation has better performance compared with traditional method based on Fourier transform and it can also cluster multi-source PD signals automatically.

关键词

Keywords

references

Views

194

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Time-frequency Characteristics of Partial Discharge Pulse with Wigner-ville Distributions

Bispectrum Estimation and Time-frequency Analysis of Electromagnetic Signal Radiated from Corona Discharge

Fault Diagnosis of Power Transformer On-load Tap Changer Based on EMD-HT Analysis of Vibration Signal

Research on Location Method and Parameter Estimation of PD Signal for Metal-clad Switchgear

Comparative Study of Partial Discharge Detection Technologies Based on the Switchgear Defect Models

Related Author

LI Yan-ming

LUO Yong-fen

LI Yan-mu

LI Jun-hao

SI Wen-rong

YUAN Peng

HU Xiao-feng

LIU Shang-he

Related Institution

School of Electrical Engineering,Xi’an Jiaotong University

National Key Laboratory of Science and Technology on Strong Electromagnetic Environment Simulation and Protection,Ordnance Engineering College

Education and Evaluation Center of Guangdong Power Grid Corporation

College of Energy and Electrical Engineering,Hehai University

Nanjing Power Supply Company,Jiangsu Electric Power Company

⁰