Department of Artillery Engineering, Army Engineering University, He Ping Road, Shijiazhuang China
In the fault diagnosis of automaton, the vibration signal presents non-stationary and non-periodic, which make it difficult to extract the fault features. To solve this problem, an automaton fault diagnosis method based on morphological component analysis (MCA) and ensemble empirical mode decomposition (EEMD) was proposed. Based on the advantages of the morphological component analysis method in the signal separation, using the morphological difference of the components in the automatic vibration signal, different sparse dictionaries were constructed to separate the components, eliminates the noise components and extracted the effective fault characteristic component, the extracted impact components are decomposed by EEMD and the energy feature of each IMF component is calculated as the fault features, then put the fault features into SVM (Support Vector Machine) and identify the faults. Through the construction simulation example and the typical fault simulation test of automatic machine, it showed that the morphological component analysis method had better noise reduction and signal separation effect. Compared with the traditional EEMD method, the feature extraction method based on the MCA-EEMD can distinguish automaton fault types more effectively.