Fault Diagnosis Of Locomotive Bearing Based On Variational Mode Decomposition

Variational mode decomposition is a new signal decomposition algorithm.This paper compares the performance of variational mode decomposition with common signal adaptive decomposition algorithm and its application in fault diagnosis,and obtains the following results:The characteristics and shortcomings of empirical mode decomposition algorithm are studied,and the mode mixing problem in empirical mode decomposition is discussed.The causes of mode mixing are analyzed,and modal mixing mode is summarized.The simulation signal analysis shows that the variational mode decomposition can effectively suppress the mode mixsing of various modes in the empirical mode decomposition.Compared with empirical mode decomposition,variational mode decomposition is based on a mathematical framework rather than an iterative algorithm,and a complex signal is synchronously decomposed into a plurality of finite bandwidth,approximately orthogonal intrinsic modal functions.The computer-generated and engineering actual signal analysis shows that the variational mode decomposition is superior to the empirical mode decomposition and the ensemble empirical mode decomposition in terms of noise robustness and decomposition capability.Aiming at the problem that the penalty factor and modal number parameter set in the variational mode decomposition method have great influence on the decomposition results,a reasonable parameter setting scheme is put forward and verified effectively by studying the filtering characteristics and the decomposition time.Based on the above research results,a fault diagnosis method for locomotive bearing based on variational mode decomposition is proposed.The operation test of the HXD3 locomotive proved that the above method can accurately diagnose the bearing failure of the axle box and the traction motor.