Based on the combination of improved Local Mean Decomposition (LMD), Multi-scale Permutation Entropy (MPE) and Hidden Markov Model (HMM), the fault types of bearings are diagnosed. Improved LMD is proposed based on the self-similarity of roller bearing vibration signal by extending the right and left side of the original signal to suppress its edge effect. First, the vibration signals of the rolling bearing are decomposed into several product function (PF) components by improved LMD respectively. Then, the phase space reconstruction of the PF1 is carried out by using the mutual information (MI) method and the false nearest neighbor (FNN) method to calculate the delay time and the embedding dimension, and then the scale is set to obtain the MPE of PF1. After that, the MPE features of rolling bearings are extracted. Finally, the features of MPE are used as HMM training and diagnosis. The experimental results show that the proposed method can effectively identify the different faults of the rolling bearing.

A novel bearing vibration signal fault feature extraction and recognition method based on the improved local mean decomposition (LMD), permutation entropy (PE) and the optimized K-means clustering algorithm is put forward in this paper. The improved LMD is proposed based on the self-similarity of roller bearing vibration signal extending the right and left side of the original signal to suppress its edge effect. After decomposing the extended signal into a set of product functions (PFs), the PE is utilized to display the complexity of the PF component and extract the fault feature meanwhile. Then, the optimized K-means algorithm is used to cluster analysis as a new pattern recognition approach, which uses the probability density distribution (PDD) to identify the initial centroid selection and has the priority of recognition accuracy compared with the classic one. Finally, the experiment results show the proposed method is effectively to fault extraction and recognition for roller bearing.