WAVELET TIME-FREQUENCY ANALYSIS WITH DAUBECHIES FILTERS AND DIMENSION REDUCTION METHODS FOR FAULT IDENTIFICATION INDUCTION MACHINE IN STATIONARY OPERATIONS

Marcus Varanis¹; Robson Pederiva²

¹ Federal University of Grande Dourados (UFGD), Brazil; ² University of Campinas (UNICAMP)

doi:10.20906/CPS/COB-2015-2601

Resumo

This work is a contribution to the study of Signal Processing Techniques based on the Wavelet Transform for extracting Energy and Entropy parameters from vibration signals for detecting faults in stationary induction motors. Together with the Wavelet Transform, dimension reduction methods are used for reducing the dimension of the dataset. The use of an experimental bench brings high-precision results.

Palavras-chave: Wavelet Packet; Daubechies Filters; Vibration; Principal Component Analysis; Induction Machine