Bearing Multi-faults Diagnosis Based on Improved Morphological Component Analysis
Li Hui1; Zheng Hai-qi2; Tang Li-wei2
1 Department of Electromechanical Engineering, Shijiazhuang Institute of Railway Technology, Shijiazhuang 050041;2 First Department, Ordnance Engineering College, Shijiazhuang 050003
Abstract:Morphological component analysis (MCA) is a signal or image processing method based on signal morphological diversity and sparse representation. MCA takes advantage of the sparse representation of the analyzed data in over-complete dictionaries to separate features in the data based on their morphology. According to the shortcomings of traditional morphological component analysis which the dictionary need to be selected manually and threshold selection, an improved approach to MCA combining adaptive dictionary and hard threshold MOM strategy is proposed. The simulative and experimental results show that not only the component of morphological diversity is separated, but also the signal noise ratio of separated signal is improved, the multi-faults of the bearing of a gearbox can be effectively detected.