By Topic

A new approach to digital protection of power transformer using support vector machine

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

2 Author(s)
A. M. Shah ; Dept. of Electr. Eng., A.D. Patel Inst. of Technol. (ADIT), Anand, India ; B. R. Bhalja

This paper presents a new classification scheme for power transformer which effectively discriminates between internal faults (phase to ground, phase to phase & phase to phase to ground) and non-internal faults (magnetising inrush, external fault & normal condition). The proposed scheme utilizes CT secondary current signals from both sides (LV and HV) of transformer windings as an input. Various signals during internal and external faults, different types of inrush (initial inrush, residual inrush, recovery inrush and sympathetic inrush) and normal/healthy operating states have been generated by modeling the existing 315 MVA power transformer using PSCAD / EMTDC software package. The simulated data set of more than 4600 operating states of transformer has been used in MATLAB for implementation and validation of the proposed scheme. To illustrate the effectiveness of the proposed scheme in terms of classification accuracy, two types of kernel functions of Support Vector Machine (SVM) classifier have been used by the authors. Using optimum kernel function of SVM, the overall fault classification accuracy has been obtained in which only 33% of total data (4650) has been used for training, whereas the remaining data (67%) has been applied for testing.

Published in:

Electrical and Computer Engineering (CCECE), 2011 24th Canadian Conference on

Date of Conference:

8-11 May 2011