Title: Rolling bearing fault diagnosis based on VMD reconstruction and DCS demodulation

Authors: Dong Zhen; Dongkai Li; Guojin Feng; Hao Zhang; Fengshou Gu

Addresses: School of Mechanical Engineering, Hebei University of Technology, Tianjin 300401, China ' School of Mechanical Engineering, Hebei University of Technology, Tianjin 300401, China ' School of Mechanical Engineering, Hebei University of Technology, Tianjin 300401, China ' School of Mechanical Engineering, Hebei University of Technology, Tianjin 300401, China ' Centre for Efficiency and Performance Engineering, University of Huddersfield, UK

Abstract: As a major component of rotating machinery, rolling bearings are prone to failure because they usually work in harsh environment and are subjected to heavy cyclic loads. Meanwhile, the fault characteristics of bearings are easily submerged by noise and difficult to extract. To solve this problem, a fault diagnosis method based on variational mode decomposition (VMD) and degree of cyclostationarity (DCS) demodulation is proposed. First, the sparsity-based reconstruction factor can distinguish the sensitivity of VMD modes, and it is used to reconstruct all VMD modes to denoise the signal. Secondly, taking the advantage that DCS demodulation analysis can obtain more useful information, it is applied to the reconstructed signal to extract the fault characteristic frequencies. Finally, simulation studies show the effectiveness of combining VMD and DCS in fault diagnosis, and the advantages of the proposed method are verified through experiments with rolling bearing inner race, outer race and compound faults.

Keywords: sparsity; DCS demodulation; rolling bearing; fault diagnosis; variational mode decomposition; VMD.

DOI: 10.1504/IJHM.2022.125092

International Journal of Hydromechatronics, 2022 Vol.5 No.3, pp.205 - 225

Received: 19 Jan 2022
Accepted: 22 Mar 2022
Published online: 25 Aug 2022 *