A bearing fault diagnosis technique based on singular values of EEMD spatial condition matrix and Gath-Geva clustering

This paper employs a combined ensemble empirical mode decomposition (EEMD) and singular value decomposition (SVD) technique to extract useful fault features from the condition monitoring data of rolling element bearings. The fault features is then classified by a Fuzzy clustering method, Gath-Geva (GG) clustering, to obtain the cluster center and membership matrix of each bearing condition for pattern recognition. The bearing fault recognition is realized by calculating the hamming approach degree between the test samples and the known fault clustering centers from the GG clustering. The proposed algorithm is then evaluated first on several sets of simulated bearing defect data with different signal to noise ratios (SNRs) to represent bearing defects with various degrees of severities. Satisfying diagnosis outcome can be obtained from this set of simulation when the SNR is greater than 1. The algorithm is further evaluated using a set of experimental data from a bearing fault test rig. It is found that the proposed algorithm can diagnose all bearing operation conditions accurately based on the experimental data.