Adaptive multiscale morphology analysis and its application in fault diagnosis of bearings

In order to solve the problem of impulsive features extraction from strong noise background, an adaptive multiscale morphology analysis (AMMA) algorithm was proposed. Corresponding to the analysis signal, the length scale and height scale were defined separately to select structuring elements for multiscale morphology analysis. An adaptive algorithm based on the information of local peaks of the signal was discussed. Numerical simulation experiments show that the proposed AMMA algorithm is better than the single-scale morphology analysis algorithm for extracting morphological features, and avoids the drawbacks of the ambiguity of selecting structuring elements and the dependence of empirical rules. The proposed AMMA algorithm is also examined in morphology analysis of the experimental signal measured from a bearing with faults. The results confirm that the proposed AMMA algorithm is able to extract various features clearly.