EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Research on a Fault-Diagnosis Method for Heavy-Duty Bearings of Thin Coal-Seam Shearers

Minghao Li, Shuting Wang, Xiao-Guang Zhang, Xiaoxu Yi, Dongsheng Wu

Symmetry · 19.07.2026 · 18 (7) · 1219

DOI: 10.3390/sym18071219 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Symmetry, online erschienen am 19.07.2026. Verfasst von Minghao Li und 4 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Fundamental Research Funds for the Universities of Liaoning Province, Liaoning Province Science and Technology Plan Joint Program Project 2025.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Aiming at the problems of fault samples being difficult to obtain in fault diagnosis research on heavy-duty bearings of thin-seam shearers and the insufficient diagnostic performance of existing methods under small-sample conditions, this study focuses on dataset construction, sample augmentation and fault diagnosis. First, based on the virtual prototype model of the cutting-unit transmission system of a thin-seam shearer, three-dimensional models of healthy bearings and four typical fault types (inner ring fault, outer ring fault, rolling-element fault and cage fault) of heavy-duty bearings were built using SolidWorks 2024. Vibration signals were collected through ADAMS dynamic simulation and converted into time-frequency images via Continuous Wavelet Transform (CWT), thereby constructing an original fault dataset with five states. Furthermore, a conditional generative adversarial network incorporating VGG perceptual loss (VGG-CGAN) was proposed to achieve targeted sample augmentation for the five bearing states, effectively alleviating the class-imbalance problem. On this basis, an improved ResNet50 fault-diagnosis model was constructed, and Bayesian optimization was used to automatically tune key hyperparameters. Experimental results show that the improved ResNet50 model achieved an accuracy of 87.14% on the self-built thin-seam shearer heavy-duty bearing dataset and 98.28% on the public CWRU dataset. The proposed method exhibits strong diagnostic performance and generalization ability under small-sample and imbalanced data conditions. This study can provide new ideas and useful references for fault diagnosis of heavy-duty bearings in thin-seam shearers.

Förderung

  • National Natural Science Foundation of China
  • Fundamental Research Funds for the Universities of Liaoning Province
  • Liaoning Province Science and Technology Plan Joint Program Project 2025

Zitieren

APA

Li, M., Wang, S., Zhang, X., Yi, X., & Wu, D. (2026). Research on a Fault-Diagnosis Method for Heavy-Duty Bearings of Thin Coal-Seam Shearers. Symmetry, 18(7), 1219. https://doi.org/10.3390/sym18071219

Vancouver

Li M, Wang S, Zhang X, Yi X, Wu D. Research on a Fault-Diagnosis Method for Heavy-Duty Bearings of Thin Coal-Seam Shearers. Symmetry. 2026;18(7):1219. doi:10.3390/sym18071219

BibTeX
@article{Li2026,
  author = {Li, Minghao and Wang, Shuting and Zhang, Xiao-Guang and Yi, Xiaoxu and Wu, Dongsheng},
  title = {Research on a Fault-Diagnosis Method for Heavy-Duty Bearings of Thin Coal-Seam Shearers},
  journal = {Symmetry},
  year = {2026},
  volume = {18},
  number = {7},
  pages = {1219},
  doi = {10.3390/sym18071219}
}
Kostenlos starten →