Signal processing of vibrations for condition monitoring of an induction motor

Author

Pöyhönen, Sanna ; Jover, Pedro ; Hyötyniemi, Heikki

Author_Institution

Control Eng. Laboratory, Helsinki Univ. of Technol., Finland

fYear

2004

fDate

2004

Firstpage

499

Lastpage

502

Abstract

Vibration monitoring is studied for fault diagnostics of an induction motor. Several features of vibration signals are compared as indicators of broken rotor bar of a 35 kW induction motor. Regular fast Fourier transform (FFT) based power spectrum density (PSD) estimation is compared to signal processing with higher order spectra (HOS), cepstrum analysis and signal description with autoregressive (AR) modelling. The fault detection routine and feature comparison is carried out with support vector machine (SVM) based classification. The best method for feature extraction seems to be the application of AR coefficients. The result is found out with real measurement data from several motor conditions and load situations.

Keywords

autoregressive processes; condition monitoring; fast Fourier transforms; fault diagnosis; feature extraction; induction motors; signal processing; support vector machines; vibrations; 35 kW; autoregressive modelling; cepstrum analysis; fast Fourier transform; fault detection routine; fault diagnostics; feature extraction; higher order spectra; induction motor; power spectrum density estimation; signal processing; support vector machine; vibration monitoring; Cepstral analysis; Cepstrum; Condition monitoring; Fast Fourier transforms; Induction motors; Rotors; Signal analysis; Signal processing; Support vector machine classification; Support vector machines;

fLanguage

English

Publisher

ieee

Conference_Titel

Control, Communications and Signal Processing, 2004. First International Symposium on

Print_ISBN

0-7803-8379-6

Type

conf

DOI

10.1109/ISCCSP.2004.1296338

Filename

1296338