Early Fault Diagnosis of Gearbox Based on VWPSO-VMD

Abstract

Abstract Aiming at the problem that the characteristic of the early weak fault of the gearbox is submerged by the background noise under the low-speed operating condition, this paper proposes a gearbox fault diagnosis method based on the variable weighted particle swarm optimization variational mode decomposition (VWPSO-VMD). First, a variable weight particle swarm optimization algorithm (VWPSO) is proposed, and the effectiveness of the VWPSO is verified by the optimization calculation of test functions. Secondly, using fuzzy entropy as the fitness function, the proposed VWPSO is used to optimize the modal number K and the penalty factor α of the variational mode decomposition (VMD). Then, based on the grey relational analysis method (GRA), the optimal intrinsic mode function (IMF) is selected for envelope demodulation analysis. Finally, VWPSO-VMD and empirical mode decomposition (EMD) are used to diagnose the early crack and broken tooth fault signals of the gearbox. The results show that, compared with EMD, VWPSO-VMD can more accurately extract the early fault characteristic frequency of the gearbox. This provides a new method for the extraction of early fault features of the gearbox.