Transformer dissolved gas analysis using least square support vector machine and bootstrap

Tang Wenhu, Shintemirov Almas, Q. H. Wu

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

16 Citations (Scopus)

Abstract

This paper presents a least square support vector machine (LS-SVM) approach to dissolved gas analysis (DGA) problems for power transformers. Two methods are employed to improve the diagnosis accuracy for DGA analysis. Firstly, bootstrap preprocessing is utilised to equalise the sample numbers for different fault types. Then, the preprocessed samples are inputted to a classier for fault classification. For comparison purposes, four classifiers are utilised, i.e. artificial neural network (ANN), K-nearest neighbour (ANN), simple SVM and LS-SVM. The classification accuracy of LS-SVM is then compared with the ones of ANN, KNN and a simple SVM. The results indicate that the LS-SVM approach can significantly improve the diagnosis accuracies for transformer fault classification.

Original languageEnglish
Title of host publicationProceedings of the 26th Chinese Control Conference, CCC 2007
Pages482-486
Number of pages5
DOIs
Publication statusPublished - 2007
Externally publishedYes
Event26th Chinese Control Conference, CCC 2007 - Zhangjiajie, China
Duration: Jul 26 2007Jul 31 2007

Other

Other26th Chinese Control Conference, CCC 2007
CountryChina
CityZhangjiajie
Period7/26/077/31/07

Fingerprint

Least Squares Support Vector Machine
Gas fuel analysis
Transformer
Bootstrap
Support vector machines
Artificial Neural Network
Fault
Neural networks
Power Transformer
Power transformers
Preprocessing
Nearest Neighbor
Classifiers
Classifier
Gas

Keywords

  • Bootstrap
  • Dissolved gas analysis
  • Least square support vector machine
  • Transformer

ASJC Scopus subject areas

  • Computer Science Applications
  • Control and Systems Engineering
  • Applied Mathematics
  • Modelling and Simulation

Cite this

Wenhu, T., Almas, S., & Wu, Q. H. (2007). Transformer dissolved gas analysis using least square support vector machine and bootstrap. In Proceedings of the 26th Chinese Control Conference, CCC 2007 (pp. 482-486). [4347139] https://doi.org/10.1109/CHICC.2006.4347139

Transformer dissolved gas analysis using least square support vector machine and bootstrap. / Wenhu, Tang; Almas, Shintemirov; Wu, Q. H.

Proceedings of the 26th Chinese Control Conference, CCC 2007. 2007. p. 482-486 4347139.

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

Wenhu, T, Almas, S & Wu, QH 2007, Transformer dissolved gas analysis using least square support vector machine and bootstrap. in Proceedings of the 26th Chinese Control Conference, CCC 2007., 4347139, pp. 482-486, 26th Chinese Control Conference, CCC 2007, Zhangjiajie, China, 7/26/07. https://doi.org/10.1109/CHICC.2006.4347139
Wenhu T, Almas S, Wu QH. Transformer dissolved gas analysis using least square support vector machine and bootstrap. In Proceedings of the 26th Chinese Control Conference, CCC 2007. 2007. p. 482-486. 4347139 https://doi.org/10.1109/CHICC.2006.4347139
Wenhu, Tang ; Almas, Shintemirov ; Wu, Q. H. / Transformer dissolved gas analysis using least square support vector machine and bootstrap. Proceedings of the 26th Chinese Control Conference, CCC 2007. 2007. pp. 482-486
@inproceedings{a77f443417764cf99a992d7f84c8eabf,
title = "Transformer dissolved gas analysis using least square support vector machine and bootstrap",
abstract = "This paper presents a least square support vector machine (LS-SVM) approach to dissolved gas analysis (DGA) problems for power transformers. Two methods are employed to improve the diagnosis accuracy for DGA analysis. Firstly, bootstrap preprocessing is utilised to equalise the sample numbers for different fault types. Then, the preprocessed samples are inputted to a classier for fault classification. For comparison purposes, four classifiers are utilised, i.e. artificial neural network (ANN), K-nearest neighbour (ANN), simple SVM and LS-SVM. The classification accuracy of LS-SVM is then compared with the ones of ANN, KNN and a simple SVM. The results indicate that the LS-SVM approach can significantly improve the diagnosis accuracies for transformer fault classification.",
keywords = "Bootstrap, Dissolved gas analysis, Least square support vector machine, Transformer",
author = "Tang Wenhu and Shintemirov Almas and Wu, {Q. H.}",
year = "2007",
doi = "10.1109/CHICC.2006.4347139",
language = "English",
isbn = "7900719229",
pages = "482--486",
booktitle = "Proceedings of the 26th Chinese Control Conference, CCC 2007",

}

TY - GEN

T1 - Transformer dissolved gas analysis using least square support vector machine and bootstrap

AU - Wenhu, Tang

AU - Almas, Shintemirov

AU - Wu, Q. H.

PY - 2007

Y1 - 2007

N2 - This paper presents a least square support vector machine (LS-SVM) approach to dissolved gas analysis (DGA) problems for power transformers. Two methods are employed to improve the diagnosis accuracy for DGA analysis. Firstly, bootstrap preprocessing is utilised to equalise the sample numbers for different fault types. Then, the preprocessed samples are inputted to a classier for fault classification. For comparison purposes, four classifiers are utilised, i.e. artificial neural network (ANN), K-nearest neighbour (ANN), simple SVM and LS-SVM. The classification accuracy of LS-SVM is then compared with the ones of ANN, KNN and a simple SVM. The results indicate that the LS-SVM approach can significantly improve the diagnosis accuracies for transformer fault classification.

AB - This paper presents a least square support vector machine (LS-SVM) approach to dissolved gas analysis (DGA) problems for power transformers. Two methods are employed to improve the diagnosis accuracy for DGA analysis. Firstly, bootstrap preprocessing is utilised to equalise the sample numbers for different fault types. Then, the preprocessed samples are inputted to a classier for fault classification. For comparison purposes, four classifiers are utilised, i.e. artificial neural network (ANN), K-nearest neighbour (ANN), simple SVM and LS-SVM. The classification accuracy of LS-SVM is then compared with the ones of ANN, KNN and a simple SVM. The results indicate that the LS-SVM approach can significantly improve the diagnosis accuracies for transformer fault classification.

KW - Bootstrap

KW - Dissolved gas analysis

KW - Least square support vector machine

KW - Transformer

UR - http://www.scopus.com/inward/record.url?scp=37749014535&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=37749014535&partnerID=8YFLogxK

U2 - 10.1109/CHICC.2006.4347139

DO - 10.1109/CHICC.2006.4347139

M3 - Conference contribution

SN - 7900719229

SN - 9787900719225

SP - 482

EP - 486

BT - Proceedings of the 26th Chinese Control Conference, CCC 2007

ER -