Bearing Health State Detection Based on Informer and CNN + Swin Transformer
Machines · 04.07.2024 · 12 (7) · 456
DOI: 10.3390/machines12070456 ↗Kurz gefasst
Fachartikel in der Zeitschrift Machines, online erschienen am 04.07.2024. Verfasst von Chunyang Liu und 7 weiteren Autor:innen. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key R & D Program of China, Major Science and Technology Projects of Longmen Laboratory.
Abstract (Originalsprache)
In response to the challenge of timely fault identification in the spindle bearings of machine tools operating in complex environments, this study proposes a method based on a combination of infrared imaging with an Informer and a CNN + Swin Transformer. The aim is to achieve real-time monitoring of bearing faults, precise fault localization, and classification of fault severity. To accomplish this, an angular contact ball bearing was chosen as the research subject. Initially, an infrared image dataset was constructed, encompassing various fault positions and degrees, by simulating different forms of bearing faults. Subsequently, an Informer-based bearing temperature prediction model was established to select faulty bearing data. Lastly, the faulty data were input into the CNN + Swin Transformer model for bearing fault recognition and classification. The results demonstrate that the Informer model accurately identifies abnormal temperature rises during bearing operation, effectively screening out faulty bearings. Under steady-state conditions, the model achieves a classification accuracy of 97.8%. Furthermore, after employing the Informer screening process, the proposed model exhibits a recognition precision of 98.9%, surpassing other models such as CNN, SVM, and Swin Transformer, which are mentioned in this paper.
Förderung
- National Key R & D Program of China
- Major Science and Technology Projects of Longmen Laboratory
Zitieren
APA
Liu, C., Zou, W., Hu, Z., Li, H., Sui, X., Ma, X., Yang, F., & Guo, N. (2024). Bearing Health State Detection Based on Informer and CNN + Swin Transformer. Machines, 12(7), 456. https://doi.org/10.3390/machines12070456
Vancouver
Liu C, Zou W, Hu Z, Li H, Sui X, Ma X, et al. Bearing Health State Detection Based on Informer and CNN + Swin Transformer. Machines. 2024;12(7):456. doi:10.3390/machines12070456
BibTeX
@article{Liu2024,
author = {Liu, Chunyang and Zou, Weiwei and Hu, Zhilei and Li, Hongyu and Sui, Xin and Ma, Xiqiang and Yang, Fang and Guo, Nan},
title = {Bearing Health State Detection Based on Informer and CNN + Swin Transformer},
journal = {Machines},
year = {2024},
volume = {12},
number = {7},
pages = {456},
doi = {10.3390/machines12070456}
}
EUVIMED