Physics-Guided Graph Convolutional Network for Ship Structural Failure Mode Classification
Mathematics · 21.05.2026 · 14 (10) · 1768
DOI: 10.3390/math14101768 ↗Kurz gefasst
Fachartikel in der Zeitschrift Mathematics, online erschienen am 21.05.2026. Verfasst von Shengpeng Li und 5 weiteren Autor:innen. Die Arbeit nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Project of Virtual-Real Fusion Analysis Technology for Ship Structure Fatigue Performance, Project of Virtual-Real Fusion Testing Technology for Ship Structure Performance.
Abstract (Originalsprache)
Ship structural failure mode classification still relies heavily on subjective expert judgment, which is time-consuming and may introduce uncertainty in safety assessment. Although deep learning provides a promising avenue for automation, many existing learning approaches rely on 2D image representations and may therefore suffer from geometric occlusion and information loss when projecting complex 3D stiffened structures. To address these challenges, we propose a Physics-Guided Graph Convolutional Network (PGGCN) for failure mode classification. Specifically, our method models finite-element (FE) meshes directly as graphs, preserving the holistic topology and displacement-field fidelity without viewpoint dependency. We further incorporate domain knowledge through a hybrid strategy: a Deep Graph Convolutional Network (DeepGCN) first detects local component buckling states such as plate or web buckling, and a logic matrix derived from classical failure definitions subsequently determines panel-level failure modes. To enable systematic evaluation, we construct a dataset spanning diverse stiffened-panel geometries via Latin Hypercube Sampling. Progressive analysis states from each loading case are organized into task-specific graph samples for supervised learning. Experiments on the test set achieve accuracies of 95.48% and 91.42% for plate- and web-buckling classification, respectively, and 89.56% for panel-level failure mode discrimination. These results demonstrate that the proposed method provides an interpretable framework for automated failure mode classification from FE meshes in ship stiffened panels.
Förderung
- Project of Virtual-Real Fusion Analysis Technology for Ship Structure Fatigue Performance
- Project of Virtual-Real Fusion Testing Technology for Ship Structure Performance
Zitieren
APA
Li, S., Xu, Y., Cao, H., Wei, P., Zhang, R., & Zhu, Z. (2026). Physics-Guided Graph Convolutional Network for Ship Structural Failure Mode Classification. Mathematics, 14(10), 1768. https://doi.org/10.3390/math14101768
Vancouver
Li S, Xu Y, Cao H, Wei P, Zhang R, Zhu Z. Physics-Guided Graph Convolutional Network for Ship Structural Failure Mode Classification. Mathematics. 2026;14(10):1768. doi:10.3390/math14101768
BibTeX
@article{Li2026,
author = {Li, Shengpeng and Xu, Yi and Cao, Hanxi and Wei, Pengyu and Zhang, Ruonan and Zhu, Zhikui},
title = {Physics-Guided Graph Convolutional Network for Ship Structural Failure Mode Classification},
journal = {Mathematics},
year = {2026},
volume = {14},
number = {10},
pages = {1768},
doi = {10.3390/math14101768}
}
EUVIMED