EUVIMEDEUVIMED entdecken
Crossref

Image-Quality-Guided Consistency-Alignment Network for Fault Diagnosis

Zhuowei Li, Jianxu Mao, Wenbin He, Yaonan Wang, Junfei Yi, Caiping Liu, Hui Zhang

IEEE Transactions on Reliability · 2026 · 75 · 2117-2130

DOI: 10.1109/tr.2026.3702064 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Reliability, 2026. Verfasst von Zhuowei Li und 6 weiteren Autor:innen. Die Arbeit nennt 39 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Hunan Province, Postdoctoral Fellowship Program and the China Postdoctoral Science Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Signal-to-image transformation has been widely used in mechanical fault diagnosis to provide unified visual representations for downstream diagnostic models. However, the resulting diagnostic images can exhibit substantial quality variations caused by measurement noise, sensor degradation, and partial signal loss. Most existing methods either discard low-quality samples or implicitly assume equal reliability across samples, leading to information loss or degraded robustness. This article proposes an image-quality-guided consistency-alignment network that explicitly estimates sample quality and leverages degraded data during training. First, vibration signals are decomposed into sub-bands, and energy-ratio criteria are used to select informative components for reconstruction. The reconstructed signals are subsequently fused into fixed-layout two-dimensional images via a sector-allocation strategy. Next, a diagnosis-aware quality assessment module assigns a quality score to each image to guide training. Low-quality samples are further regularized via cross-quality feature alignment using a quality-weighted supervised pairwise loss, encouraging them to align with high-quality counterparts from the same class. Finally, a Swin transformer backbone performs classification. Experiments on the proprietary robotic curved-surface grinding fault dataset dataset and the public Southeast University Bearing Fault Dataset dataset demonstrate consistent gains under controlled mixed-quality settings, with accuracies of up to 99.4% and 98.7%, respectively, while additional evaluations under reproducible degradations further confirm the robustness of the proposed quality-aware learning mechanism.

Förderung

  • National Natural Science Foundation of China
  • Natural Science Foundation of Hunan Province
  • Postdoctoral Fellowship Program and the China Postdoctoral Science Foundation
  • Open Fund of State Key Laboratory of Mechanical Transmission for Advanced Equipment
  • Open Research Fund of State Key Laboratory of Precision Manufacturing for Extreme
  • National Science and Technology Major Project

Zitieren

APA

Li, Z., Mao, J., He, W., Wang, Y., Yi, J., Liu, C., & Zhang, H. (2026). Image-Quality-Guided Consistency-Alignment Network for Fault Diagnosis. IEEE Transactions on Reliability, 75, 2117-2130. https://doi.org/10.1109/tr.2026.3702064

Vancouver

Li Z, Mao J, He W, Wang Y, Yi J, Liu C, et al. Image-Quality-Guided Consistency-Alignment Network for Fault Diagnosis. IEEE Transactions on Reliability. 2026;75:2117-2130. doi:10.1109/tr.2026.3702064

BibTeX
@article{Li2026,
  author = {Li, Zhuowei and Mao, Jianxu and He, Wenbin and Wang, Yaonan and Yi, Junfei and Liu, Caiping and Zhang, Hui},
  title = {Image-Quality-Guided Consistency-Alignment Network for Fault Diagnosis},
  journal = {IEEE Transactions on Reliability},
  year = {2026},
  volume = {75},
  pages = {2117--2130},
  doi = {10.1109/tr.2026.3702064}
}
Kostenlos starten →