EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Research on a Multi-Dimensional Information Fusion Mechanical Wear Fault-Diagnosis Algorithm Based on Data Regeneration

Qifan Zhou, Bosong Chai iD, Kunwen Ran, Yingqing Guo iD, Shan Zhou, Wangyu Wu iD, Kun Wang, Yao Ni iD

Sensors · 15.06.2025 · 25 (12) · 3745

DOI: 10.3390/s25123745 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 15.06.2025. Verfasst von Qifan Zhou und 7 weiteren Autor:innen. Beteiligte Einrichtung: School of Power and Energy, Northwestern Polytechnical University, Xi'an 710129, China. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 68 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science and Technology Major Program, Guangdong Basic and Applied Basic Research Foundation, Science and Technology Projects in Guangzhou.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Under laboratory conditions for recording a small amount of data, the characteristics of the phenomena distribution become a limitation of machine learning and advanced deep learning concepts for the diagnosis and localization of mechanical wear faults. In this paper, we adopt the combination of the diffusion model and TTT (test-time training), based on the sample distribution of feature data under the laboratory conditions, and we use the pre-trained decoder to decode the data into a continuous potential representation of natural language for sampling, to achieve data regeneration. Subsequently, the TTT algorithm becomes a model with weights in the hidden state itself. The gradient step on the self-supervised loss is selected as the update rule, which is trained synchronously during the testing time, adhering to the concept of migration learning, to construct a high-dimensional mapping relationship between the feature parameters and the failure modes of the mechanical wear. The final validation results show that the diagnosis accuracy reaches more than 95% for six types of typical aero-engine mechanical wear faults.

Schlagwörter

Stichwörter der Autor:innen

aero-enginediffusion modelfault diagnosis algorithmsmechanical weartest-time training

Autor:innen und Institutionen

  • School of Power and Energy, Northwestern Polytechnical University, Xi'an 710129, China.
  • School of Computer Science and Technology, Zhejiang University, Hangzhou 310027, China.
  • School of Computer Science, University of Liverpool, Liverpool L69 3DR, UK.
  • Institute for Aero Engine, Tsinghua University, Beijing 100084, China.
  • School of Integrated Circuits, Guangdong University of Technology, Guangzhou 510006, China.

ORCID-Profile: Bosong Chai, Yingqing Guo, Wangyu Wu, Yao Ni

Förderung

  • National Science and Technology Major Program
  • Guangdong Basic and Applied Basic Research Foundation
  • Science and Technology Projects in Guangzhou
  • Key Science and Technology Program of Henan Province

Zitieren

APA

Zhou, Q., Chai, B., Ran, K., Guo, Y., Zhou, S., Wu, W., Wang, K., & Ni, Y. (2025). Research on a Multi-Dimensional Information Fusion Mechanical Wear Fault-Diagnosis Algorithm Based on Data Regeneration. Sensors, 25(12), 3745. https://doi.org/10.3390/s25123745

Vancouver

Zhou Q, Chai B, Ran K, Guo Y, Zhou S, Wu W, et al. Research on a Multi-Dimensional Information Fusion Mechanical Wear Fault-Diagnosis Algorithm Based on Data Regeneration. Sensors. 2025;25(12):3745. doi:10.3390/s25123745

BibTeX
@article{Zhou2025,
  author = {Zhou, Qifan and Chai, Bosong and Ran, Kunwen and Guo, Yingqing and Zhou, Shan and Wu, Wangyu and Wang, Kun and Ni, Yao},
  title = {Research on a Multi-Dimensional Information Fusion Mechanical Wear Fault-Diagnosis Algorithm Based on Data Regeneration},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {12},
  pages = {3745},
  doi = {10.3390/s25123745}
}
Kostenlos starten →