Fault Diagnosis of EMU Rolling Bearing Based on EEMD and SVM

Rolling bearing is an important and fragile component in the EMU. To give a safe condition assessment of rolling bearing, especially for early fault diagnosis, is very necessary and become an urgent thing to the EMU. A fault detection and diagnosis method based on EEMD, sample entropy and SVM is proposed in this paper. Firstly, the investigated signal is decomposed into several IMFs by EEMD. Then, the values of sample entropy of IMFs are extracted as the feature vectors, and finally the fault detection and classification are carried out with feature vectors by using SVM. The method can effectively diagnose the fault. Compared with the results of EMD and SVM combination, this method is more accurate and valid.