A Fault Diagnosis Scheme for Gearbox Based on Improved Entropy and Optimized Regularized Extreme Learning Machine

Wei Zhang, Hong Lu, Yongquan Zhang, Zhangjie Li (), Yongjing Wang, Jun Zhou, Jiangnuo Mei and Yuzhan Wei
Additional contact information
Wei Zhang: School of Mechanical and Electronic Engineering, Wuhan University of Technology, Wuhan 430070, China
Hong Lu: School of Mechanical and Electronic Engineering, Wuhan University of Technology, Wuhan 430070, China
Yongquan Zhang: School of Mechanical and Electronic Engineering, Wuhan University of Technology, Wuhan 430070, China
Zhangjie Li: School of Mechanical and Electronic Engineering, Wuhan University of Technology, Wuhan 430070, China
Yongjing Wang: Department of Mechanical Engineering, University of Birmingham, Birmingham B15 2TT, UK
Jun Zhou: School of Mechanical and Electronic Engineering, Wuhan University of Technology, Wuhan 430070, China
Jiangnuo Mei: School of Mechanical and Electronic Engineering, Wuhan University of Technology, Wuhan 430070, China
Yuzhan Wei: School of Mechanical and Electronic Engineering, Wuhan University of Technology, Wuhan 430070, China

Mathematics, 2022, vol. 10, issue 23, 1-28

Abstract: The performance of a gearbox is sensitive to failures, especially in the long-term high speed and heavy load field. However, the multi-fault diagnosis in gearboxes is a challenging problem because of the complex and non-stationary measured signal. To obtain fault information more fully and improve the accuracy of gearbox fault diagnosis, this paper proposes a feature extraction method, hierarchical refined composite multiscale fluctuation dispersion entropy (HRCMFDE) to extract the fault features of rolling bearing and the gear vibration signals at different layers and scales. On this basis, a novel fault diagnosis scheme for the gearbox based on HRCMFDE, ReliefF and grey wolf optimizer regularized extreme learning machine is proposed. Firstly, HRCMFDE is employed to extract the original features, the multi-frequency time information can be evaluated simultaneously, and the fault feature information can be extracted more fully. After that, ReliefF is used to screen the sensitive features from the high-dimensional fault features. Finally, the sensitive features are inputted into the optimized regularized extreme learning machine to identify the fault states of the gearbox. Through three different types of gearbox experiments, the experimental results confirm that the proposed method has better diagnostic performance and generalization, which can effectively and accurately identify the different fault categories of the gearbox and outperforms other contrastive methods.

Keywords: hierarchical refined composite multiscale fluctuation dispersion entropy (HRCMFDE); fault diagnosis; gearbox; regularized extreme learning machine; ReliefF; grey wolf (search for similar items in EconPapers)
JEL-codes: C (search for similar items in EconPapers)
Date: 2022
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2227-7390/10/23/4585/pdf (application/pdf)
https://www.mdpi.com/2227-7390/10/23/4585/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jmathe:v:10:y:2022:i:23:p:4585-:d:992693

Access Statistics for this article

Mathematics is currently edited by Ms. Emma He

More articles in Mathematics from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().