Wavelet-LDA-neural network based short circuit occurrence detection in induction motor winding

D. A. Asfani*, Syafaruddin, M. H. Purnomo, T. Hiyama

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Citations (Scopus)

Abstract

The paper proposes the short circuit identification method for induction motor winding. Four states of motor operation are defined as normal operation, starting of short circuit, steady state short circuit and ending of short circuit. The neural network based detection system is utilized to distinguish these defined operation states. Motor current is processed using discrete wavelet transformation to extract energy component of high frequency signal, which is latterly used for variable detection. Three different wavelet types varied by five levels of transformation are evaluated using linear discriminant analysis (LDA) in order to obtain the most appropriate wavelet filter for detection task. A laboratory experiment is performed to validate the accuracy of the proposed method.

Original languageEnglish
Title of host publicationSDEMPED 2011 - 8th IEEE Symposium on Diagnostics for Electrical Machines, Power Electronics and Drives
Pages330-336
Number of pages7
DOIs
Publication statusPublished - 2011
Event8th IEEE Symposium on Diagnostics for Electrical Machines, Power Electronics and Drives, SDEMPED 2011 - Bologna, Italy
Duration: 5 Sept 20118 Sept 2011

Publication series

NameSDEMPED 2011 - 8th IEEE Symposium on Diagnostics for Electrical Machines, Power Electronics and Drives

Conference

Conference8th IEEE Symposium on Diagnostics for Electrical Machines, Power Electronics and Drives, SDEMPED 2011
Country/TerritoryItaly
CityBologna
Period5/09/118/09/11

Keywords

  • Fault detection
  • Induction motor winding
  • Linear discriminant analysis
  • Neural networks
  • Wavelet transforms

Fingerprint

Dive into the research topics of 'Wavelet-LDA-neural network based short circuit occurrence detection in induction motor winding'. Together they form a unique fingerprint.

Cite this