This paper presents a hybrid signal processing approach combining the wavelet transform and spectral analysis. A new approach to selecting a mother wavelet for decomposition of vibration signals was demonstrated. Subsequently, the bestsuited decomposition scale was selected based on the energy content of the wavelet coefficients. Subsequently, the envelope spectrum of the wavelet coefficients at the best-suited decomposition scale was used to identify the defect and its location. Experimental analysis of signals from a defect-seeded bearing has shown that the developed approach is more efficient than the conventional spectral analysis techniques.

This content is only available via PDF.
You do not currently have access to this content.