Deep neural networks for understanding and diagnosing partial discharge data

V. M. Catterson, B. Sheng

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

6 Citations (Scopus)

Abstract

Artificial neural networks have been investigated for many years as a technique for automated diagnosis of defects causing partial discharge (PD). While good levels of accuracy have been reported, disadvantages include the difficulty of explaining results, and the need to hand-craft appropriate features for standard two-layer networks. Recent advances in the design and training of deep neural networks, which contain more than two layers of hidden neurons, have resulted in improved results in speech and image recognition tasks. This paper investigates the use of deep neural networks for PD diagnosis. Defect samples constructed in mineral oil were used to generate data for training and testing. The paper demonstrates the improvements in accuracy and visualization of learning which can be gained from deep learning.
LanguageEnglish
Title of host publication2015 IEEE Electrical Insulation Conference (EIC)
Place of PublicationPiscataway, NJ, USA
PublisherIEEE
Pages218-221
Number of pages4
ISBN (Print)9781479973521
DOIs
Publication statusPublished - 7 Jun 2015
EventIEEE Electrical Insulation Conference 2015 - Seattle, United States
Duration: 7 Jun 201510 Jun 2015

Conference

ConferenceIEEE Electrical Insulation Conference 2015
CountryUnited States
CitySeattle
Period7/06/1510/06/15

Fingerprint

Partial discharges
Defects
Image recognition
Network layers
Mineral oils
Speech recognition
Neurons
Visualization
Neural networks
Testing
Deep neural networks
Deep learning

Keywords

  • artificial neural networks
  • deep neural networks
  • partial discharge
  • diagnostics
  • UHF monitoring
  • defects in oil

Cite this

Catterson, V. M., & Sheng, B. (2015). Deep neural networks for understanding and diagnosing partial discharge data. In 2015 IEEE Electrical Insulation Conference (EIC) (pp. 218-221). Piscataway, NJ, USA: IEEE. https://doi.org/10.1109/ICACACT.2014.7223616
Catterson, V. M. ; Sheng, B. / Deep neural networks for understanding and diagnosing partial discharge data. 2015 IEEE Electrical Insulation Conference (EIC). Piscataway, NJ, USA : IEEE, 2015. pp. 218-221
@inproceedings{f56e3b872c9c4968a37af92aa77fe129,
title = "Deep neural networks for understanding and diagnosing partial discharge data",
abstract = "Artificial neural networks have been investigated for many years as a technique for automated diagnosis of defects causing partial discharge (PD). While good levels of accuracy have been reported, disadvantages include the difficulty of explaining results, and the need to hand-craft appropriate features for standard two-layer networks. Recent advances in the design and training of deep neural networks, which contain more than two layers of hidden neurons, have resulted in improved results in speech and image recognition tasks. This paper investigates the use of deep neural networks for PD diagnosis. Defect samples constructed in mineral oil were used to generate data for training and testing. The paper demonstrates the improvements in accuracy and visualization of learning which can be gained from deep learning.",
keywords = "artificial neural networks, deep neural networks, partial discharge, diagnostics, UHF monitoring, defects in oil",
author = "Catterson, {V. M.} and B. Sheng",
year = "2015",
month = "6",
day = "7",
doi = "10.1109/ICACACT.2014.7223616",
language = "English",
isbn = "9781479973521",
pages = "218--221",
booktitle = "2015 IEEE Electrical Insulation Conference (EIC)",
publisher = "IEEE",

}

Catterson, VM & Sheng, B 2015, Deep neural networks for understanding and diagnosing partial discharge data. in 2015 IEEE Electrical Insulation Conference (EIC). IEEE, Piscataway, NJ, USA, pp. 218-221, IEEE Electrical Insulation Conference 2015, Seattle, United States, 7/06/15. https://doi.org/10.1109/ICACACT.2014.7223616

Deep neural networks for understanding and diagnosing partial discharge data. / Catterson, V. M.; Sheng, B.

2015 IEEE Electrical Insulation Conference (EIC). Piscataway, NJ, USA : IEEE, 2015. p. 218-221.

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

TY - GEN

T1 - Deep neural networks for understanding and diagnosing partial discharge data

AU - Catterson, V. M.

AU - Sheng, B.

PY - 2015/6/7

Y1 - 2015/6/7

N2 - Artificial neural networks have been investigated for many years as a technique for automated diagnosis of defects causing partial discharge (PD). While good levels of accuracy have been reported, disadvantages include the difficulty of explaining results, and the need to hand-craft appropriate features for standard two-layer networks. Recent advances in the design and training of deep neural networks, which contain more than two layers of hidden neurons, have resulted in improved results in speech and image recognition tasks. This paper investigates the use of deep neural networks for PD diagnosis. Defect samples constructed in mineral oil were used to generate data for training and testing. The paper demonstrates the improvements in accuracy and visualization of learning which can be gained from deep learning.

AB - Artificial neural networks have been investigated for many years as a technique for automated diagnosis of defects causing partial discharge (PD). While good levels of accuracy have been reported, disadvantages include the difficulty of explaining results, and the need to hand-craft appropriate features for standard two-layer networks. Recent advances in the design and training of deep neural networks, which contain more than two layers of hidden neurons, have resulted in improved results in speech and image recognition tasks. This paper investigates the use of deep neural networks for PD diagnosis. Defect samples constructed in mineral oil were used to generate data for training and testing. The paper demonstrates the improvements in accuracy and visualization of learning which can be gained from deep learning.

KW - artificial neural networks

KW - deep neural networks

KW - partial discharge

KW - diagnostics

KW - UHF monitoring

KW - defects in oil

UR - http://sites.ieee.org/eic/program/

UR - http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=7193737

U2 - 10.1109/ICACACT.2014.7223616

DO - 10.1109/ICACACT.2014.7223616

M3 - Conference contribution book

SN - 9781479973521

SP - 218

EP - 221

BT - 2015 IEEE Electrical Insulation Conference (EIC)

PB - IEEE

CY - Piscataway, NJ, USA

ER -

Catterson VM, Sheng B. Deep neural networks for understanding and diagnosing partial discharge data. In 2015 IEEE Electrical Insulation Conference (EIC). Piscataway, NJ, USA: IEEE. 2015. p. 218-221 https://doi.org/10.1109/ICACACT.2014.7223616