Fault Diagnosis for Power Transformer Based on SVM Information Fusion

DOI

10.2991/emeit.2012.447How to use a DOI?

Keywords

transformer,SVM,information fusion,fault diagnosis

Abstract

A kind of fault diagnosis method for power transformer based on multi-SVM information fusion is proposed in this paper. Based on the "one against one" multi-classification SVM and pairwise coupling method, the multi-classification SVM probability outputs are obtained and regarded as the basic probability function of information fusion to diagnose. It also makes full use of complementary information from gases dissolved in transformer oil analysis data and electrical test data. From the example analysis results, it shows better performance than a single feature SVM method in the aspect of accuracy and generalization.

Copyright

© 2012, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Li-ping Sima
AU  - Xing-zhi Su
AU  - Bo Wang
AU  - Peng Dou
AU  - Gen-cai Liu
PY  - 2012/09
DA  - 2012/09
TI  - Fault Diagnosis for Power Transformer Based on SVM Information Fusion
BT  - Proceedings of the 2nd International Conference on Electronic & Mechanical Engineering and Information Technology (EMEIT 2012)
PB  - Atlantis Press
SP  - 2016
EP  - 2019
SN  - 1951-6851
UR  - https://doi.org/10.2991/emeit.2012.447
DO  - 10.2991/emeit.2012.447
ID  - Sima2012/09
ER  -