Fault detection method for gear condition based on LMD method and principle component analysis

According to the non-stationary characteristics of gear fault vibration signals, a fault detection method for gear condition base on LMD method and principle component analysis is proposed. The LMD method is used to deal with the non-stationary signals, which can decompose the vibration signals into a finite number of product functions (PFs). Then the PFs containing the main fault information are selected, and the energy and the time domain statistics feature parameters can be extracted from the PFs, from which the initial feature vector matrixes can be formed. By applying the principle component analysis technique to the initial feature vector matrixes, the principle components are obtained. The euclidean distance is used for classification of gear working condition. The analysis results show that the proposed method can effectively identify the gear condition and fault pattern.