International Journal of Advanced Information and Communication Technology


Transformer Fault Classification Using Support Vector Machine Method

J. Eslin Fathima, A. Venkatasami, Einstein College of Engineering, Tirunelveli, Tamilnadu, India.

DOI : 01.0401/ijaict.2014.01.33

International Journal of Advanced Information and Communication Technology

Received On : September 13, 2014

Revised On : October 20, 2014

Accepted On : November 12, 2014

Published On : December 05, 2014

Volume 01, Issue 08

Pages : 168-172

Abstract


Power transformers are important equipments in power system. Smooth functioning is the key to ensure hossle-free operation. Dissolved Gas Analysis (DGA) is a well known technique to analyze faults in Transformers. Rogers ratio method was attempted for transformer fault diagnosis and the same is reported. To improve the diagnosis accuracy soft computing techniques are generally used. There are several soft computing Techniques available for diagnosis. This work proposes a new method of DGA diagnosis based on Support Vector Machine (SVM) method. SVM can change a non-linear learning problem into a linear learning problem to reduce the algorithm complexity. Experimental data from TC 10 database is used to illustrate the performance of the SVM method.

Keywords


componentDGA diagnosis, IEC TC 10, gas ratio, new method

Cite this article


J. Eslin Fathima, A. Venkatasami, “Transformer Fault Classification Using Support Vector Machine Method” INTERNATIONAL JOURNAL OF ADVANCED INFORMATION AND COMMUNICATION TECHNOLOGY, pp.168-172, December 05, 2014.

Copyright


© 2014 J. Eslin Fathima, A. Venkatasami. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.