Partial discharge pattern classification for an oil-pressboard interface

A. Abubakar Mas'ud, B.G. Stewart, S. G. McMeekin, A. Nesbitt

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

4 Citations (Scopus)

Abstract

This paper compares the ability of a Single Neural Network (SNN) and an Ensemble Neural Network (ENN) in classifying and discriminating oil-pressboard interface partial discharge (PD) degredation. Discharges were sustained for 15 hours and PD patterns captured, evaluated and correlated with the tracking damage on the pressboard surface. For the same experimental arrangement two samples were stressed, one at 18.5kV (rms) and the other at 30kV (rms). Training data for both the SNN and the ENN comprised statistical parameters obtained from the Φ-q-n discharge patterns. Data sets for application were were split into periods of the first 9 hours and last 6 hours, as these time periods appeared to show most variabilty and stabilty of the statisical paremeters respectively. The results show that both the ENN and the SNN are able to discriminate the Φ-q-n patterns over these periods. It is also shown that the ENN always provides a higher recognition rate for unseen trained data while the SNN actually appears to show a higher ability to discriminate the patterns. 

LanguageEnglish
Title of host publicationConference Record of the 2012 IEEE International Symposium on Electrical Insulation
Pages122-126
Number of pages5
DOIs
Publication statusPublished - 2012
Event2012 19th IEEE International Symposium on Electrical Insulation, ISEI 2012 - San Juan, PR, United States
Duration: 10 Jun 201213 Jun 2012

Conference

Conference2012 19th IEEE International Symposium on Electrical Insulation, ISEI 2012
CountryUnited States
CitySan Juan, PR
Period10/06/1213/06/12

Fingerprint

Partial discharges
Interfaces (computer)
Pattern recognition
Neural networks
Oils

Keywords

  • ensemble neural networks
  • high voltage
  • surface tracking

Cite this

Abubakar Mas'ud, A., Stewart, B. G., McMeekin, S. G., & Nesbitt, A. (2012). Partial discharge pattern classification for an oil-pressboard interface. In Conference Record of the 2012 IEEE International Symposium on Electrical Insulation (pp. 122-126). [6251440] https://doi.org/10.1109/ELINSL.2012.6251440
Abubakar Mas'ud, A. ; Stewart, B.G. ; McMeekin, S. G. ; Nesbitt, A. / Partial discharge pattern classification for an oil-pressboard interface. Conference Record of the 2012 IEEE International Symposium on Electrical Insulation. 2012. pp. 122-126
@inproceedings{3ea046f9a5f34a879560d1d546e563a9,
title = "Partial discharge pattern classification for an oil-pressboard interface",
abstract = "This paper compares the ability of a Single Neural Network (SNN) and an Ensemble Neural Network (ENN) in classifying and discriminating oil-pressboard interface partial discharge (PD) degredation. Discharges were sustained for 15 hours and PD patterns captured, evaluated and correlated with the tracking damage on the pressboard surface. For the same experimental arrangement two samples were stressed, one at 18.5kV (rms) and the other at 30kV (rms). Training data for both the SNN and the ENN comprised statistical parameters obtained from the Φ-q-n discharge patterns. Data sets for application were were split into periods of the first 9 hours and last 6 hours, as these time periods appeared to show most variabilty and stabilty of the statisical paremeters respectively. The results show that both the ENN and the SNN are able to discriminate the Φ-q-n patterns over these periods. It is also shown that the ENN always provides a higher recognition rate for unseen trained data while the SNN actually appears to show a higher ability to discriminate the patterns. ",
keywords = "ensemble neural networks, high voltage, surface tracking",
author = "{Abubakar Mas'ud}, A. and B.G. Stewart and McMeekin, {S. G.} and A. Nesbitt",
year = "2012",
doi = "10.1109/ELINSL.2012.6251440",
language = "English",
isbn = "9781467304887",
pages = "122--126",
booktitle = "Conference Record of the 2012 IEEE International Symposium on Electrical Insulation",

}

Abubakar Mas'ud, A, Stewart, BG, McMeekin, SG & Nesbitt, A 2012, Partial discharge pattern classification for an oil-pressboard interface. in Conference Record of the 2012 IEEE International Symposium on Electrical Insulation., 6251440, pp. 122-126, 2012 19th IEEE International Symposium on Electrical Insulation, ISEI 2012, San Juan, PR, United States, 10/06/12. https://doi.org/10.1109/ELINSL.2012.6251440

Partial discharge pattern classification for an oil-pressboard interface. / Abubakar Mas'ud, A.; Stewart, B.G.; McMeekin, S. G.; Nesbitt, A.

Conference Record of the 2012 IEEE International Symposium on Electrical Insulation. 2012. p. 122-126 6251440.

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

TY - GEN

T1 - Partial discharge pattern classification for an oil-pressboard interface

AU - Abubakar Mas'ud, A.

AU - Stewart, B.G.

AU - McMeekin, S. G.

AU - Nesbitt, A.

PY - 2012

Y1 - 2012

N2 - This paper compares the ability of a Single Neural Network (SNN) and an Ensemble Neural Network (ENN) in classifying and discriminating oil-pressboard interface partial discharge (PD) degredation. Discharges were sustained for 15 hours and PD patterns captured, evaluated and correlated with the tracking damage on the pressboard surface. For the same experimental arrangement two samples were stressed, one at 18.5kV (rms) and the other at 30kV (rms). Training data for both the SNN and the ENN comprised statistical parameters obtained from the Φ-q-n discharge patterns. Data sets for application were were split into periods of the first 9 hours and last 6 hours, as these time periods appeared to show most variabilty and stabilty of the statisical paremeters respectively. The results show that both the ENN and the SNN are able to discriminate the Φ-q-n patterns over these periods. It is also shown that the ENN always provides a higher recognition rate for unseen trained data while the SNN actually appears to show a higher ability to discriminate the patterns. 

AB - This paper compares the ability of a Single Neural Network (SNN) and an Ensemble Neural Network (ENN) in classifying and discriminating oil-pressboard interface partial discharge (PD) degredation. Discharges were sustained for 15 hours and PD patterns captured, evaluated and correlated with the tracking damage on the pressboard surface. For the same experimental arrangement two samples were stressed, one at 18.5kV (rms) and the other at 30kV (rms). Training data for both the SNN and the ENN comprised statistical parameters obtained from the Φ-q-n discharge patterns. Data sets for application were were split into periods of the first 9 hours and last 6 hours, as these time periods appeared to show most variabilty and stabilty of the statisical paremeters respectively. The results show that both the ENN and the SNN are able to discriminate the Φ-q-n patterns over these periods. It is also shown that the ENN always provides a higher recognition rate for unseen trained data while the SNN actually appears to show a higher ability to discriminate the patterns. 

KW - ensemble neural networks

KW - high voltage

KW - surface tracking

UR - http://www.scopus.com/inward/record.url?scp=84866911145&partnerID=8YFLogxK

U2 - 10.1109/ELINSL.2012.6251440

DO - 10.1109/ELINSL.2012.6251440

M3 - Conference contribution book

SN - 9781467304887

SP - 122

EP - 126

BT - Conference Record of the 2012 IEEE International Symposium on Electrical Insulation

ER -

Abubakar Mas'ud A, Stewart BG, McMeekin SG, Nesbitt A. Partial discharge pattern classification for an oil-pressboard interface. In Conference Record of the 2012 IEEE International Symposium on Electrical Insulation. 2012. p. 122-126. 6251440 https://doi.org/10.1109/ELINSL.2012.6251440