Classification of partial discharge signals by combining adaptive local iterative filtering and entropy features

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

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Electromagnetic Interference (EMI) is a technique for capturing Partial Discharge (PD) signals in high voltage power plant apparatus. EMI signals can be non-stationary which makes their analysis difficult, particularly for pattern recognition applications. This paper, extends upon a previously developed software condition-monitoring model for improved EMI events classification based on time-frequency signal decomposition and entropy features. The idea of the proposed method is to map multiple discharge source signals captured by EMI and labelled by experts, including PD, from the time domain to a feature space which aids in interpretation of subsequent fault information. Here, instead of using only one permutation entropy measure a more robust measure, called Dispersion Entropy (DE), is added to the feature vector. Multi-Class Support Vector Machine (MCSVM) methods are utilized for classification of the different discharge sources. Results show an improved classification accuracy compared to previously proposed methods. This yields to a successful development of an experts knowledge-based intelligent system. Since this method is demonstrated to be successful with real field data, it brings the benefit of possible real world application for EMI condition monitoring.
LanguageEnglish
Article number406
Number of pages14
JournalSensors
Volume18
Issue number2
DOIs
Publication statusPublished - 31 Jan 2018

Fingerprint

electromagnetic interference
Electromagnetic Phenomena
Partial discharges
Entropy
Signal interference
entropy
Condition monitoring
Power Plants
permutations
Intelligent systems
power plants
pattern recognition
Pattern recognition
Support vector machines
high voltages
Power plants
Software
Decomposition
computer programs
decomposition

Keywords

  • EMI Method
  • partial discharge
  • permutation entropy
  • dispersion entropy
  • classification
  • expert system
  • EMI events (discharge sources)

Cite this

Mitiche, Imene ; Morison, Gordon ; Nesbitt, Alan ; Hughes-Narborough, Michael ; Stewart, Brian G. ; Boreham, Philip. / Classification of partial discharge signals by combining adaptive local iterative filtering and entropy features. In: Sensors. 2018 ; Vol. 18, No. 2.
@article{b097cbca608b41e8920421ff062162c3,
title = "Classification of partial discharge signals by combining adaptive local iterative filtering and entropy features",
abstract = "Electromagnetic Interference (EMI) is a technique for capturing Partial Discharge (PD) signals in high voltage power plant apparatus. EMI signals can be non-stationary which makes their analysis difficult, particularly for pattern recognition applications. This paper, extends upon a previously developed software condition-monitoring model for improved EMI events classification based on time-frequency signal decomposition and entropy features. The idea of the proposed method is to map multiple discharge source signals captured by EMI and labelled by experts, including PD, from the time domain to a feature space which aids in interpretation of subsequent fault information. Here, instead of using only one permutation entropy measure a more robust measure, called Dispersion Entropy (DE), is added to the feature vector. Multi-Class Support Vector Machine (MCSVM) methods are utilized for classification of the different discharge sources. Results show an improved classification accuracy compared to previously proposed methods. This yields to a successful development of an experts knowledge-based intelligent system. Since this method is demonstrated to be successful with real field data, it brings the benefit of possible real world application for EMI condition monitoring.",
keywords = "EMI Method, partial discharge, permutation entropy, dispersion entropy, classification, expert system, EMI events (discharge sources)",
author = "Imene Mitiche and Gordon Morison and Alan Nesbitt and Michael Hughes-Narborough and Stewart, {Brian G.} and Philip Boreham",
year = "2018",
month = "1",
day = "31",
doi = "10.3390/s18020406",
language = "English",
volume = "18",
journal = "Sensors",
issn = "1424-8220",
number = "2",

}

Classification of partial discharge signals by combining adaptive local iterative filtering and entropy features. / Mitiche, Imene; Morison, Gordon; Nesbitt, Alan; Hughes-Narborough, Michael; Stewart, Brian G.; Boreham, Philip.

In: Sensors, Vol. 18, No. 2, 406, 31.01.2018.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Classification of partial discharge signals by combining adaptive local iterative filtering and entropy features

AU - Mitiche, Imene

AU - Morison, Gordon

AU - Nesbitt, Alan

AU - Hughes-Narborough, Michael

AU - Stewart, Brian G.

AU - Boreham, Philip

PY - 2018/1/31

Y1 - 2018/1/31

N2 - Electromagnetic Interference (EMI) is a technique for capturing Partial Discharge (PD) signals in high voltage power plant apparatus. EMI signals can be non-stationary which makes their analysis difficult, particularly for pattern recognition applications. This paper, extends upon a previously developed software condition-monitoring model for improved EMI events classification based on time-frequency signal decomposition and entropy features. The idea of the proposed method is to map multiple discharge source signals captured by EMI and labelled by experts, including PD, from the time domain to a feature space which aids in interpretation of subsequent fault information. Here, instead of using only one permutation entropy measure a more robust measure, called Dispersion Entropy (DE), is added to the feature vector. Multi-Class Support Vector Machine (MCSVM) methods are utilized for classification of the different discharge sources. Results show an improved classification accuracy compared to previously proposed methods. This yields to a successful development of an experts knowledge-based intelligent system. Since this method is demonstrated to be successful with real field data, it brings the benefit of possible real world application for EMI condition monitoring.

AB - Electromagnetic Interference (EMI) is a technique for capturing Partial Discharge (PD) signals in high voltage power plant apparatus. EMI signals can be non-stationary which makes their analysis difficult, particularly for pattern recognition applications. This paper, extends upon a previously developed software condition-monitoring model for improved EMI events classification based on time-frequency signal decomposition and entropy features. The idea of the proposed method is to map multiple discharge source signals captured by EMI and labelled by experts, including PD, from the time domain to a feature space which aids in interpretation of subsequent fault information. Here, instead of using only one permutation entropy measure a more robust measure, called Dispersion Entropy (DE), is added to the feature vector. Multi-Class Support Vector Machine (MCSVM) methods are utilized for classification of the different discharge sources. Results show an improved classification accuracy compared to previously proposed methods. This yields to a successful development of an experts knowledge-based intelligent system. Since this method is demonstrated to be successful with real field data, it brings the benefit of possible real world application for EMI condition monitoring.

KW - EMI Method

KW - partial discharge

KW - permutation entropy

KW - dispersion entropy

KW - classification

KW - expert system

KW - EMI events (discharge sources)

UR - http://www.mdpi.com/1424-8220/18/2/406

U2 - 10.3390/s18020406

DO - 10.3390/s18020406

M3 - Article

VL - 18

JO - Sensors

T2 - Sensors

JF - Sensors

SN - 1424-8220

IS - 2

M1 - 406

ER -