Classification of EMI discharge sources using time–frequency features and multi-class support vector machine

Imene Mitiche, Gordon Morison, Alan Nesbitt, Michael Hughes-Narborough, Brian G. Stewart, Philip Boreham

Research output: Contribution to journalArticle

4 Downloads (Pure)

Abstract

This paper introduces the first application of feature extraction and machine learning to electromagnetic interference (EMI) signals for discharge sources classification in high voltage power generating plants. This work presents an investigation on signals that represent different discharge sources, which are measured using EMI techniques from operating electrical machines within power plant. The analysis involves time–frequency image calculation of EMI signals using general linear chirplet analysis (GLCT) which reveals both time and frequency varying characteristics. Histograms of uniform local binary patterns (LBP) are implemented as a feature reduction and extraction technique for the classification of discharge sources using multi-class support vector machine (MCSVM). The novelty that this paper introduces is the combination of GLCT and LBP applications to develop a new feature extraction algorithm applied to EMI signals classification. The proposed algorithm is demonstrated to be successful with excellent classification accuracy being achieved. For the first time, this work transfers expert's knowledge on EMI faults to an intelligent system which could potentially be exploited to develop an automatic condition monitoring system.
Original languageEnglish
Pages (from-to)261-269
JournalElectric Power Systems Research
Volume163
Issue numberPart A
Early online date5 Jul 2018
DOIs
Publication statusPublished - 1 Oct 2018

Fingerprint

Signal interference
Support vector machines
Feature extraction
Power plants
Condition monitoring
Intelligent systems
Learning systems
Electric potential

Keywords

  • EMI
  • partial discharge
  • GLCT
  • Uniform LBP
  • Multi-class support vector machine

Cite this

@article{1f1f1331af5a4b99a641922720185458,
title = "Classification of EMI discharge sources using time–frequency features and multi-class support vector machine",
abstract = "This paper introduces the first application of feature extraction and machine learning to electromagnetic interference (EMI) signals for discharge sources classification in high voltage power generating plants. This work presents an investigation on signals that represent different discharge sources, which are measured using EMI techniques from operating electrical machines within power plant. The analysis involves time–frequency image calculation of EMI signals using general linear chirplet analysis (GLCT) which reveals both time and frequency varying characteristics. Histograms of uniform local binary patterns (LBP) are implemented as a feature reduction and extraction technique for the classification of discharge sources using multi-class support vector machine (MCSVM). The novelty that this paper introduces is the combination of GLCT and LBP applications to develop a new feature extraction algorithm applied to EMI signals classification. The proposed algorithm is demonstrated to be successful with excellent classification accuracy being achieved. For the first time, this work transfers expert's knowledge on EMI faults to an intelligent system which could potentially be exploited to develop an automatic condition monitoring system.",
keywords = "EMI, partial discharge, GLCT, Uniform LBP, Multi-class support vector machine",
author = "Imene Mitiche and Gordon Morison and Alan Nesbitt and Michael Hughes-Narborough and Stewart, {Brian G.} and Philip Boreham",
note = "Acceptance from webpage AAM provided by Strathclyde 12m embargo",
year = "2018",
month = "10",
day = "1",
doi = "10.1016/j.epsr.2018.06.016",
language = "English",
volume = "163",
pages = "261--269",
journal = "Electric Power Systems Research",
issn = "0378-7796",
publisher = "Elsevier B.V.",
number = "Part A",

}

Classification of EMI discharge sources using time–frequency features and multi-class support vector machine. / Mitiche, Imene; Morison, Gordon; Nesbitt, Alan; Hughes-Narborough, Michael; Stewart, Brian G.; Boreham, Philip .

In: Electric Power Systems Research, Vol. 163, No. Part A, 01.10.2018, p. 261-269.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Classification of EMI discharge sources using time–frequency features and multi-class support vector machine

AU - Mitiche, Imene

AU - Morison, Gordon

AU - Nesbitt, Alan

AU - Hughes-Narborough, Michael

AU - Stewart, Brian G.

AU - Boreham, Philip

N1 - Acceptance from webpage AAM provided by Strathclyde 12m embargo

PY - 2018/10/1

Y1 - 2018/10/1

N2 - This paper introduces the first application of feature extraction and machine learning to electromagnetic interference (EMI) signals for discharge sources classification in high voltage power generating plants. This work presents an investigation on signals that represent different discharge sources, which are measured using EMI techniques from operating electrical machines within power plant. The analysis involves time–frequency image calculation of EMI signals using general linear chirplet analysis (GLCT) which reveals both time and frequency varying characteristics. Histograms of uniform local binary patterns (LBP) are implemented as a feature reduction and extraction technique for the classification of discharge sources using multi-class support vector machine (MCSVM). The novelty that this paper introduces is the combination of GLCT and LBP applications to develop a new feature extraction algorithm applied to EMI signals classification. The proposed algorithm is demonstrated to be successful with excellent classification accuracy being achieved. For the first time, this work transfers expert's knowledge on EMI faults to an intelligent system which could potentially be exploited to develop an automatic condition monitoring system.

AB - This paper introduces the first application of feature extraction and machine learning to electromagnetic interference (EMI) signals for discharge sources classification in high voltage power generating plants. This work presents an investigation on signals that represent different discharge sources, which are measured using EMI techniques from operating electrical machines within power plant. The analysis involves time–frequency image calculation of EMI signals using general linear chirplet analysis (GLCT) which reveals both time and frequency varying characteristics. Histograms of uniform local binary patterns (LBP) are implemented as a feature reduction and extraction technique for the classification of discharge sources using multi-class support vector machine (MCSVM). The novelty that this paper introduces is the combination of GLCT and LBP applications to develop a new feature extraction algorithm applied to EMI signals classification. The proposed algorithm is demonstrated to be successful with excellent classification accuracy being achieved. For the first time, this work transfers expert's knowledge on EMI faults to an intelligent system which could potentially be exploited to develop an automatic condition monitoring system.

KW - EMI

KW - partial discharge

KW - GLCT

KW - Uniform LBP

KW - Multi-class support vector machine

U2 - 10.1016/j.epsr.2018.06.016

DO - 10.1016/j.epsr.2018.06.016

M3 - Article

VL - 163

SP - 261

EP - 269

JO - Electric Power Systems Research

JF - Electric Power Systems Research

SN - 0378-7796

IS - Part A

ER -