By Topic

Fast match-based vector quantization partial discharge pulse pattern recognition

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
$31 $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

4 Author(s)
Abdel-Galil, T.K. ; Dept. of Electr. & Comput. Eng., Univ. of Waterloo, Ont., Canada ; Hegazy, Y.G. ; Salama, M.M.A. ; Bartnikas, R.

A novel approach for the classification of cavity size in terms of their apparent charge versus applied voltage (ΔQ-V) partial discharge pattern characteristics is described. The method makes use of the fast match-based vector quantization procedure, wherein a given partial discharge pattern is matched against a set of known partial discharge patterns in a database. The ΔQ-V partial discharge patterns for different cavity sizes are considered as a sequence of events rather than as ΔQ-V curve representations. In the training phase, each cavity size represents a unique class, which emits its own ΔQ-V sequence, and vector quantization (VQ) is used to assign labels for this sequence of events. In the testing phase, a fast match algorithm is proposed to determine the degree of similarity between the labels of the tested phenomena and the prestored labels for different partial discharge patterns previously stored during the training phase. The best-matched model pinpoints the cavity size class. The results demonstrate that while the implementation of such classifier is simple, it achieves high classification rates; this positions the method as a competitive alternative vis-a`-vis other previously proposed classifiers, which suffer from both larger computational burdens and inherently more complicated structures.

Published in:

Instrumentation and Measurement, IEEE Transactions on  (Volume:54 ,  Issue: 1 )