EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Knowledge Graph-Driven Graph Neural Networks for Equipment Fault Prediction in Maglev Train Systems

Chunlong Yu, Yi Peng, Kunyan Li, Jianyu Guo, Yi Wang, JingJing Chen

Applied Sciences · 19.06.2026 · 16 (12) · 6205

DOI: 10.3390/app16126205 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 19.06.2026. Verfasst von Chunlong Yu und 5 weiteren Autor:innen. Die Arbeit nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the National 14th Five-Year Science Key Research and Development Program of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Equipment fault prediction in maglev train systems poses substantial challenges: fault events are inherently rare, class distributions are severely imbalanced, and individual equipment units are subject to complex spatial and functional couplings that single-device statistical approaches fundamentally cannot capture. To address these challenges, this study proposes a Knowledge Graph-driven Graph Neural Network (KG-GNN) framework. A fault knowledge graph encompassing equipment, fault, temporal, and environmental entities is constructed to unify multi-source maintenance data. Graph connectivity is established via three spatial relation types (co-location, co-zone, and co-level), with edge weights derived from Laplacian-smoothed Lift scores quantifying fault co-occurrence strength. A two-layer GATv2Conv-based graph attention network is designed: the first layer employs four-head attention with explicit edge-weight integration to capture heterogeneous neighborhood influences, while the second layer produces compact node embeddings via single-head attention. A Top-20 sparsification strategy suppresses weak-association noise, and training under severe class imbalance is stabilized through Focal Loss and F2-Score-guided early stopping. On the test set, the proposed method achieves an F2-Score of 0.5703, Recall of 0.6825, and AUC-ROC of 0.9329 (single-run evaluation); multi-seed evaluation (5 seeds) yields F2 = 0.5645 ± 0.0035, Recall = 0.6789 ± 0.0095, and AUC-ROC = 0.9298 ± 0.0026, outperforming the MLP baseline by 18.3% in F2-Score and substantially exceeding GCN (F2 = 0.1476 ± 0.0176) and GATConv (F2 = 0.4284 ± 0.0097). Ablation studies confirm the individual contributions of authentic graph topology, precise edge weighting, and graph sparsification to overall performance.

Förderung

  • the National 14th Five-Year Science Key Research and Development Program of China

Zitieren

APA

Yu, C., Peng, Y., Li, K., Guo, J., Wang, Y., & Chen, J. (2026). Knowledge Graph-Driven Graph Neural Networks for Equipment Fault Prediction in Maglev Train Systems. Applied Sciences, 16(12), 6205. https://doi.org/10.3390/app16126205

Vancouver

Yu C, Peng Y, Li K, Guo J, Wang Y, Chen J. Knowledge Graph-Driven Graph Neural Networks for Equipment Fault Prediction in Maglev Train Systems. Applied Sciences. 2026;16(12):6205. doi:10.3390/app16126205

BibTeX
@article{Yu2026,
  author = {Yu, Chunlong and Peng, Yi and Li, Kunyan and Guo, Jianyu and Wang, Yi and Chen, JingJing},
  title = {Knowledge Graph-Driven Graph Neural Networks for Equipment Fault Prediction in Maglev Train Systems},
  journal = {Applied Sciences},
  year = {2026},
  volume = {16},
  number = {12},
  pages = {6205},
  doi = {10.3390/app16126205}
}
Kostenlos starten →