EUVIMEDEUVIMED entdecken
Crossref

A Domain Generalization Framework for Online Aero-Engine Rub-Impact Detection Using Adversarial Learning and Multisource Label Aggregation

Haotian Peng, Jie Gao, Yafeng Shen, Wei Wang, Chunyan Hu, Jinsong Du

IEEE Sensors Journal · 01.07.2026 · 26 (13) · 20153-20166

DOI: 10.1109/jsen.2026.3695046 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Sensors Journal, erschienen am 01.07.2026. Verfasst von Haotian Peng und 5 weiteren Autor:innen. Die Arbeit nennt 56 Literaturangaben. Gefördert von: Fundamental Research Project of Shenyang Institute of Automation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Online monitoring of rub-impact faults is essential for aero-engine safety, yet existing methods struggle with asynchronous multimodal data, diverse sensor configurations, and the scarcity of reliable fault annotations. To address these challenges, this article proposes RubNet, a domain generalization (DG) framework that first employs a physics-guided temporal alignment strategy to precisely align high-frequency vibration signals with low-frequency state parameters based on spectral centroid correlations. A multiscale channel attention convolutional neural network (MSCA-CNN) is subsequently utilized to extract robust features, leveraging adversarial DG to learn representations that remain invariant to shifts in sensor position and orientation. Furthermore, to mitigate label noise during temporal fine-tuning, the framework integrates an uncertainty-aware soft label aggregation mechanism that fuses conflicting annotations from human experts, vision language models (VLMs), and pretrained model predictions using the Dawid–Skene (DS) algorithm. The experimental validation on real-world ground test data demonstrates that RubNet achieves state-of-the-art performance with an $F$ 1-score of 93.26% and superior generalization capabilities across unseen sensor domains while maintaining low computational complexity suitable for real-time deployment.

Förderung

  • Fundamental Research Project of Shenyang Institute of Automation

Zitieren

APA

Peng, H., Gao, J., Shen, Y., Wang, W., Hu, C., & Du, J. (2026). A Domain Generalization Framework for Online Aero-Engine Rub-Impact Detection Using Adversarial Learning and Multisource Label Aggregation. IEEE Sensors Journal, 26(13), 20153-20166. https://doi.org/10.1109/jsen.2026.3695046

Vancouver

Peng H, Gao J, Shen Y, Wang W, Hu C, Du J. A Domain Generalization Framework for Online Aero-Engine Rub-Impact Detection Using Adversarial Learning and Multisource Label Aggregation. IEEE Sensors Journal. 2026;26(13):20153-20166. doi:10.1109/jsen.2026.3695046

BibTeX
@article{Peng2026,
  author = {Peng, Haotian and Gao, Jie and Shen, Yafeng and Wang, Wei and Hu, Chunyan and Du, Jinsong},
  title = {A Domain Generalization Framework for Online Aero-Engine Rub-Impact Detection Using Adversarial Learning and Multisource Label Aggregation},
  journal = {IEEE Sensors Journal},
  year = {2026},
  volume = {26},
  number = {13},
  pages = {20153--20166},
  doi = {10.1109/jsen.2026.3695046}
}
Kostenlos starten →