YOLO-MVCP: A Lightweight Fault Diagnosis Method of Rolling Bearing Based on STFT Time–frequency Graph and MobileViT Network Pruning

Yu Wang (), Yue Li () and Shufeng Zhang ()
Additional contact information
Yu Wang: National University of Defense Technology, National Key Laboratory of Equipment State Sensing and Smart Support
Yue Li: National University of Defense Technology, National Key Laboratory of Equipment State Sensing and Smart Support
Shufeng Zhang: National University of Defense Technology, National Key Laboratory of Equipment State Sensing and Smart Support

A chapter in Data-Driven Methods for Reliability and Safety Engineering: Applications in Industrial Systems, 2026, pp 333-348 from Springer

Abstract: Abstract To address the limitation that existing rolling bearing fault diagnosis methods are unable to simultaneously achieve high diagnostic accuracy and lightweight model deployment, a compact fault diagnosis approach is proposed based on short-time Fourier transform time–frequency representations and MobileViT network pruning. The initial step involves processing the raw vibration signal with a short-time Fourier transform to generate a 2D time–frequency image. Subsequently, a YOLO-MobileViT network is constructed to train the diagnostic model, and redundant secondary channels are eliminated through network pruning, enabling lightweight compression of the high-precision model following fine-tuning. Experimental validation on the CWRU dataset demonstrates that the YOLO-MVCP achieves a fault diagnosis accuracy of 100 percent, while maintaining a model size of only 1.24 MB. This balance between diagnostic accuracy and model compactness confirms its practical applicability and engineering value.

Keywords: Fault diagnosis; Short-time Fourier transform; MobileViT; Network pruning; Light weight (search for similar items in EconPapers)
Date: 2026
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

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:spr:ssrchp:978-3-032-22873-4_24

Ordering information: This item can be ordered from
http://www.springer.com/9783032228734

DOI: 10.1007/978-3-032-22873-4_24

Access Statistics for this chapter

More chapters in Springer Series in Reliability Engineering from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().