Identification of partial discharge location using probabilistic neural networks and the fuzzy c-means clustering approach

Ju Chu Hsieh, Cheng-Chi Tai, Ming Shou Su, Yu Hsun Lin

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

This article proposes an approach for location of partial discharge sources in the power cable and gas-insulated load break switches using a probabilistic neural networks and the fuzzy C-means clustering approach. Three different defect positions are designed in the power cable and gas-insulated load break switches. The three different defect positions of partial discharge occurrence are located by the proposed method. Discrete wavelet transform is employed to suppress noises of measured signals by the high-frequency current transformer. The proposed method can assist electrical engineers in making accurate statistical judgments. To accurately discover the different defect positions, the proposed method uses feature extraction and statistical analysis of the measured signals. Finally, experimental results validate that the proposed approach can effectively determine the location of partial discharge sources in practical partial discharge measurement.

Original languageEnglish
Pages (from-to)60-69
Number of pages10
JournalElectric Power Components and Systems
Volume42
Issue number1
DOIs
Publication statusPublished - 2014 Jan 2

Fingerprint

Partial discharges
Neural networks
Defects
Cables
Switches
Electric instrument transformers
Discrete wavelet transforms
Gases
Feature extraction
Statistical methods
Engineers

All Science Journal Classification (ASJC) codes

  • Energy Engineering and Power Technology
  • Mechanical Engineering
  • Electrical and Electronic Engineering

Cite this

@article{187796fc2bdc4662b67a5f5022f0a3df,
title = "Identification of partial discharge location using probabilistic neural networks and the fuzzy c-means clustering approach",
abstract = "This article proposes an approach for location of partial discharge sources in the power cable and gas-insulated load break switches using a probabilistic neural networks and the fuzzy C-means clustering approach. Three different defect positions are designed in the power cable and gas-insulated load break switches. The three different defect positions of partial discharge occurrence are located by the proposed method. Discrete wavelet transform is employed to suppress noises of measured signals by the high-frequency current transformer. The proposed method can assist electrical engineers in making accurate statistical judgments. To accurately discover the different defect positions, the proposed method uses feature extraction and statistical analysis of the measured signals. Finally, experimental results validate that the proposed approach can effectively determine the location of partial discharge sources in practical partial discharge measurement.",
author = "Hsieh, {Ju Chu} and Cheng-Chi Tai and Su, {Ming Shou} and Lin, {Yu Hsun}",
year = "2014",
month = "1",
day = "2",
doi = "10.1080/15325008.2013.844210",
language = "English",
volume = "42",
pages = "60--69",
journal = "Electric Power Components and Systems",
issn = "1532-5008",
publisher = "Taylor and Francis Ltd.",
number = "1",

}

Identification of partial discharge location using probabilistic neural networks and the fuzzy c-means clustering approach. / Hsieh, Ju Chu; Tai, Cheng-Chi; Su, Ming Shou; Lin, Yu Hsun.

In: Electric Power Components and Systems, Vol. 42, No. 1, 02.01.2014, p. 60-69.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Identification of partial discharge location using probabilistic neural networks and the fuzzy c-means clustering approach

AU - Hsieh, Ju Chu

AU - Tai, Cheng-Chi

AU - Su, Ming Shou

AU - Lin, Yu Hsun

PY - 2014/1/2

Y1 - 2014/1/2

N2 - This article proposes an approach for location of partial discharge sources in the power cable and gas-insulated load break switches using a probabilistic neural networks and the fuzzy C-means clustering approach. Three different defect positions are designed in the power cable and gas-insulated load break switches. The three different defect positions of partial discharge occurrence are located by the proposed method. Discrete wavelet transform is employed to suppress noises of measured signals by the high-frequency current transformer. The proposed method can assist electrical engineers in making accurate statistical judgments. To accurately discover the different defect positions, the proposed method uses feature extraction and statistical analysis of the measured signals. Finally, experimental results validate that the proposed approach can effectively determine the location of partial discharge sources in practical partial discharge measurement.

AB - This article proposes an approach for location of partial discharge sources in the power cable and gas-insulated load break switches using a probabilistic neural networks and the fuzzy C-means clustering approach. Three different defect positions are designed in the power cable and gas-insulated load break switches. The three different defect positions of partial discharge occurrence are located by the proposed method. Discrete wavelet transform is employed to suppress noises of measured signals by the high-frequency current transformer. The proposed method can assist electrical engineers in making accurate statistical judgments. To accurately discover the different defect positions, the proposed method uses feature extraction and statistical analysis of the measured signals. Finally, experimental results validate that the proposed approach can effectively determine the location of partial discharge sources in practical partial discharge measurement.

UR - http://www.scopus.com/inward/record.url?scp=84890361351&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84890361351&partnerID=8YFLogxK

U2 - 10.1080/15325008.2013.844210

DO - 10.1080/15325008.2013.844210

M3 - Article

VL - 42

SP - 60

EP - 69

JO - Electric Power Components and Systems

JF - Electric Power Components and Systems

SN - 1532-5008

IS - 1

ER -