Research on numerical simulation and fault feature extraction methods for non-healthy helical gear transmission

Abstract

To analyze the meshing contact characteristics of faulty helical gears, finite element models of gear transmission systems under both healthy and faulty conditions were established. Acceleration signals of the gear transmission in both conditions were obtained, and the number of spectral peaks around the meshing frequency in the acceleration signal spectrum was counted for different fault cases. A gear fault experiment was carried out, and the trend in the number of spectral peaks in the measured acceleration signal spectrum was found to be consistent with the simulation results, which verifies the effectiveness of the finite element model and simulation results proposed in this study. The Subtraction Average-Based Optimizer (SABO) was used to optimize the key parameters of Variational Mode Decomposition (VMD) and Maximum Correlated Kurtosis Deconvolution (MCKD). The envelope spectrum of gear fault signals under strong noise interference was analyzed, confirming that appropriate selection of key parameters for VMD and MCKD allows for effective extraction of gear fault characteristic frequencies under noisy conditions.