EUVIMEDEUVIMED entdecken
Crossref

DSMA-Net: A Lightweight Time–Frequency Dual-Stream Multiattention Network for Rolling Bearing Fault Diagnosis With Limited Data

Jinglun Xu, Xu Wang, Qingle Zhang, Zhiqiang Li, Jigen Xia, Dibo Hou, Pingjie Huang

IEEE Sensors Journal · 15.05.2026 · 26 (10) · 15899-15911

DOI: 10.1109/jsen.2026.3680735 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Sensors Journal, erschienen am 15.05.2026. Verfasst von Jinglun Xu und 6 weiteren Autor:innen. Gefördert von: Ningbo Science and Technology Plan Project, Shaoxing Science and Technology Plan Project, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate rolling bearing fault diagnosis is essential for ensuring the reliability of industrial systems; however, practical applications are often constrained by small-sample training where discriminative fault information is insufficiently exposed (e.g., weak and sparsely distributed fault-related components and ambiguous spectral cues), as well as strong noise interference and restricted deployment resources. To address these challenges, a lightweight time–frequency dual-stream multiattention network (DSMA-Net) is proposed. A discrete cosine transform-driven channel attention (DCT-CA) module with lightweight nonlinear mapping is designed to capture multicomponent intrachannel energy descriptors from time-domain signals. To emphasize fault-relevant spectral regions, a frequency-domain attention mechanism is developed, including element-wise channel sensing and channel-specific spatial attention (SA). In addition, a bidirectional cross-attention fusion mechanism is employed to align mid- and high-level temporal and spectral features. Experiments on two bearing datasets collected from different test rigs under small-sample conditions demonstrate the effectiveness of DSMA-Net. It achieves a test accuracy of 99.12% with only 78.63K parameters and 35.51M FLOPs, outperforming existing lightweight state-of-the-art (SOTA) methods. Additional evaluations under noise conditions and extremely low training sample sizes further confirm its robustness and consistent performance in challenging scenarios. The proposed method provides an efficient and deployment-friendly solution for noisy and resource-constrained industrial scenarios.

Förderung

  • Ningbo Science and Technology Plan Project
  • Shaoxing Science and Technology Plan Project
  • National Natural Science Foundation of China

Zitieren

APA

Xu, J., Wang, X., Zhang, Q., Li, Z., Xia, J., Hou, D., & Huang, P. (2026). DSMA-Net: A Lightweight Time–Frequency Dual-Stream Multiattention Network for Rolling Bearing Fault Diagnosis With Limited Data. IEEE Sensors Journal, 26(10), 15899-15911. https://doi.org/10.1109/jsen.2026.3680735

Vancouver

Xu J, Wang X, Zhang Q, Li Z, Xia J, Hou D, et al. DSMA-Net: A Lightweight Time–Frequency Dual-Stream Multiattention Network for Rolling Bearing Fault Diagnosis With Limited Data. IEEE Sensors Journal. 2026;26(10):15899-15911. doi:10.1109/jsen.2026.3680735

BibTeX
@article{Xu2026,
  author = {Xu, Jinglun and Wang, Xu and Zhang, Qingle and Li, Zhiqiang and Xia, Jigen and Hou, Dibo and Huang, Pingjie},
  title = {DSMA-Net: A Lightweight Time–Frequency Dual-Stream Multiattention Network for Rolling Bearing Fault Diagnosis With Limited Data},
  journal = {IEEE Sensors Journal},
  year = {2026},
  volume = {26},
  number = {10},
  pages = {15899--15911},
  doi = {10.1109/jsen.2026.3680735}
}
Kostenlos starten →