An Effective Envelope Analysis Using Gaussian Windows for Evaluation of Fault Severity in Bearing

Abstract

This paper proposes an efficient method for evaluation of the fault severity in bearing using the discrete wavelet packet transform (DWPT) and the envelope analysis. The acoustic emission (AE) signals for each defect are first decomposed to the sub-band signals. The envelope power spectrum analysis is performed on each sub-band to detect the frequency periodic impulses showing the abnormal symptoms of bearing defects. It is essential to select an optimal sub-band for reliable assessment of the fault severity in bearing. A ratio of defect spectral component to residual spectral component (RDR) is calculated from their envelope power spectrum using the Gaussian window for an optimal sub-band selection which shows clearly information about failures. As a result, the severe degree of bearing defects is assessed based on the RDR calculation. The effectiveness of the proposed scheme is validated through experimental results of evaluating the different fault conditions under variable crack size in bearing.