EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

MATE-ViT: A multi-channel contrast-limited adaptive time-frequency enhancement and vision transformer framework for bearing fault diagnosis

Jiangdong Zhao, Jiao Yu, Wenming Wang, Ji Huang, Xiaolu Ma

AIP Advances · 01.06.2025 · 15 (6)

DOI: 10.1063/5.0279740 ↗

Kurz gefasst

Fachartikel in der Zeitschrift AIP Advances, online erschienen am 09.06.2025. Verfasst von Jiangdong Zhao und 4 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: National Natural Science Foundation of China, Horizontal Class II Scientific Research Project of West Anhui University, University-Industry Collaborative Education Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

To address the issues of low contrast, unclear local details, and the difficulty of existing deep learning models in capturing long-range dependencies in time-frequency images generated by current time-frequency construction methods, this paper proposes a bearing fault diagnosis framework (MATE-ViT) based on multi-channel contrast-limited adaptive time-frequency enhancement (CLAHE) and vision transformer (ViT). First, an improved CLAHE algorithm is used to independently enhance the multi-channel time-frequency images, effectively improving the local contrast and detail expression of the images, thereby enhancing the recognizability of fault features. Next, a perception-driven image segmentation and rearrangement strategy is applied, dividing the enhanced image into local feature patches and rearranging them to maximize the complementarity between local features and global information, thus providing optimized input feature representations for subsequent global modeling by the transformer. Finally, the ViT model is applied to extract global features from the enhanced images, achieving accurate classification of bearing fault types. Experimental results show that the proposed MATE-ViT framework outperforms mainstream deep learning models in terms of diagnostic accuracy and robustness under various noise levels and operating conditions, offering a more effective and reliable approach for bearing fault diagnosis.

Förderung

  • National Natural Science Foundation of China
  • Horizontal Class II Scientific Research Project of West Anhui University
  • University-Industry Collaborative Education Program
  • Anhui Province Colleges and Universities Science and Engineering Teachers Practice Plan for Enterprises
  • Natural Science Key Project of West Anhui University
  • High level Talent Research Initiation Fund Project of West Anhui University
  • Open Program of MOE Key Laboratory of Fundamental Physical Quantities Measurements

Zitieren

APA

Zhao, J., Yu, J., Wang, W., Huang, J., & Ma, X. (2025). MATE-ViT: A multi-channel contrast-limited adaptive time-frequency enhancement and vision transformer framework for bearing fault diagnosis. AIP Advances, 15(6). https://doi.org/10.1063/5.0279740

Vancouver

Zhao J, Yu J, Wang W, Huang J, Ma X. MATE-ViT: A multi-channel contrast-limited adaptive time-frequency enhancement and vision transformer framework for bearing fault diagnosis. AIP Advances. 2025;15(6). doi:10.1063/5.0279740

BibTeX
@article{Zhao2025,
  author = {Zhao, Jiangdong and Yu, Jiao and Wang, Wenming and Huang, Ji and Ma, Xiaolu},
  title = {MATE-ViT: A multi-channel contrast-limited adaptive time-frequency enhancement and vision transformer framework for bearing fault diagnosis},
  journal = {AIP Advances},
  year = {2025},
  volume = {15},
  number = {6},
  doi = {10.1063/5.0279740}
}
Kostenlos starten →