Circuit Fault Diagnosis Based on Wavelet Packet and Neural Network

Konferenz: ISTET 2009 - VXV International Symposium on Theoretical Engineering
22.06.2009 - 24.06.2009 in Lübeck, Germany

Tagungsband: ISTET 2009

Seiten: 4Sprache: EnglischTyp: PDF

Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt

Kuczynski, Andrzej; Ossowski, Marek (Technical University of Lodz)

In this paper, neural network algorithms of fault diagnosis for analog circuit containing MOS transistors are presented. Measurement of dynamic supply current is utilized for detecting the catastrophic faults. A discrete wavelet transform is used as preprocessor in order to reduce the nodes in input layer and hidden layer of BP neural network. The illustrative numerical examples are presented.