Study on Wear Fault Diagnosis of Planetary Gearbox Based on STOA-VMD Combined with 1.5-Dimensional Envelope Spectrum

Aiming at the problems of obvious nonlinear characteristics and difficult identification of fault characteristics of planetary gearbox vibration signals, a fault diagnosis method based on Sooty Tern Optimization Algorithm (STOA) optimized Variational Modal Decomposition (VMD) and 1.5-dimensional envelope spectrum is proposed. Firstly, the STOA is used to optimize the parameters of variational modal decomposition; Secondly, the signal is decomposed by variational mode method to obtain multiple eigenmode components; Then, based on the correlation kurtosis, the vibration signal is reconstructed with corresponding coefficients of eigenmode components; Finally, the reconstructed signal is analyzed using 1.5-dimensional envelope spectrum. The effectiveness of STOA-VMD is verified by simulation signals. The planetary gearbox test-bed is built, the whole life cycle data of the tooth surface wear fault is collected, and the wear fault is diagnosed using STOA-VMD combined with 1.5-dimensional envelope spectrum, and compared with other methods. The results show that the proposed method is effective and accurate for fault feature extraction.