Fault diagnosis method of rolling bearing based on MSSA-VMD algorithm

Abstract

Abstract Early fault signals of rolling bearings are affected by environmental noise, which makes it challenging to extract fault characteristics. In this paper, a bearing fault diagnosis method based on Sparrow Search Algorithm (SSA) and Variable Modal Decomposition (VMD) is proposed to achieve adaptive signal decomposition and feature extraction. First, the fusion shock index is constructed based on the envelope spectrum pulse factor and correlation coefficient to measure the shock fault components. Then, the VMD parameters are adaptively optimized by the improved sparrow search algorithm MSSA, and the fusion shock index is established as the optimal modal component selection method of the fitness function. Finally, the envelope spectrum of the signal after noise reduction is analyzed to extract the fault characteristic frequency. By analyzing the bearing outer ring fault signal and inner ring fault signal under strong noise, the results show that this method can adaptively determine the mode number and penalty factor of VMD, effectively extract the fault signal characteristics, and realize rolling bearing fault diagnosis.