Application of time-frequency distribution and neural networks for fault classification in power electronics

Author

Leonowicz, Zbigniew ; Lobos, Tadeusz

Author_Institution

Dept. of Electr. Eng., Wroclaw Univ. of Technol., Poland

fYear

2003

fDate

29-31 July 2003

Firstpage

67

Lastpage

71

Abstract

A new method of fault analysis and detection by signal classification in frequency converters is presented. The Wigner Ville time frequency distribution is used to produce the representation of the signal and the probabilistic neural network as a classifier. The accuracy and robustness of the proposed method is investigated on signals obtained during the different fault mode operations of the industrial frequency converter.

Keywords

Wigner distribution; fault location; frequency convertors; neural nets; power electronics; signal classification; signal representation; time-frequency analysis; Wigner Ville time frequency distribution; classifier; fault analysis; fault classification; fault detection; fault mode operations; frequency converters; industrial frequency converter; neural networks; power electronics; robustness; signal classification; time-frequency distribution; Fault detection; Frequency conversion; Intelligent networks; Neural networks; Noise robustness; Pattern classification; Performance analysis; Signal analysis; Signal resolution; Time frequency analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Computational Intelligence for Measurement Systems and Applications, 2003. CIMSA '03. 2003 IEEE International Symposium on

Print_ISBN

0-7803-7783-4

Type

conf

DOI

10.1109/CIMSA.2003.1227204

Filename

1227204