Fault diagnosis of planetary gearbox with multi-channels vibration data based on a novel feature extraction method and LSSVM

The vibration signal of the planetary gearbox is complex and mixed with high noise owing to the variable operating conditions and complex operating environments. However, it is difficult to obtain accurate diagnostic results. To improve the efficiency of feature extraction and fault diagnosis, a fault diagnosis method based on improved multi-resolution singular value decomposition (IMRSVD), variational mode decomposition (VMD), dispersion entropy (DE), and LSSVM with multi-channel vibration data is proposed. Firstly, an index based on signal energy is proposed to improve the self-adaption of MRSVD in decomposition times, and IMRSVD is applied to eliminate noise in the vibration signal. Secondly, a novel composite feature extraction method is proposed based on the VMD-DE and multi-channel vibration data. Finally, the fault diagnosis of the planetary gearbox was achieved using the LSSVM with composite features. The proposed method is validated on actual planetary gearbox datasets, and the results indicate that the proposed method can achieve higher fault diagnosis accuracy.