Rolling bearing fault diagnosis based on PSO-VMD

To address the problem that the number of decomposition layers K and penalty factor a in variational modal decomposition (VMD) have a great influence on the accuracy of the decomposition results, a method is proposed to determine the optimal values of K and a using particle swarm optimization (PSO)....

Full description

Saved in:
Bibliographic Details
Published in2022 6th International Conference on Wireless Communications and Applications (ICWCAPP) pp. 280 - 284
Main Authors Chen, Lingping, Li, Zhe, Zhang, Long, Liu, Yi
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.08.2022
Subjects
Analytical models
Correlation
Data models
Fault diagnosis
Particle swarm optimization algorithm
Rolling bearing
Rolling bearings
Variational modal decomposition
Vibrations
Wireless communication
Online AccessGet full text

Cover

More Information
Summary:To address the problem that the number of decomposition layers K and penalty factor a in variational modal decomposition (VMD) have a great influence on the accuracy of the decomposition results, a method is proposed to determine the optimal values of K and a using particle swarm optimization (PSO). In addition, a correlation analysis method is used to quantify the correlation between each component decomposed by VMD and the original signal of the fault, which helps to find the IMF component with more fault information among all IMF components. Finally, the simulation signal and experimental data of rolling bearing inner ring faults are analyzed and diagnosed respectively to verify the effectiveness of the PSO-VMD fault diagnosis method proposed in this paper.
DOI:10.1109/ICWCAPP57292.2022.00076