In order to separate the compound faults from rolling bearing, and improve diagnosis accuracy, a method based on morphological filtering (MF) and sparse component analysis (SCA) was proposed to deal with the blind source separation (BSS) problem of rotation machines in the case of nonlinear, noisy source mixing and the number of failure sources are unknown. The morphological filtering was used to extract modulation features embedded in the observed signals and to ensure signals presented in sparse mode, and then SCA was used to separate unknown sources from mixed signals. In the over-completed and underdetermined condition, the method was applied to analyze the faulty rolling bearing acceleration signals. Analysis results show that this method can separate and extract the rolling bearing’s fault characteristic efficiently.
LI Yu-chuan;WU Xing;CHI Yi-lin;LIU Chang.
Blind separation for rolling bearing faults based on morphological filters and sparse component analysis[J]. Journal of Vibration and Shock, 2011, 30(12): 170-174