EUVIMEDEUVIMED entdecken
Crossref

Attention-based spatiotemporal graph models for full-field acceleration reconstruction

Adarsh Gupta, Shahrul Kadri Ayop, Anowarul Habib

Japanese Journal of Applied Physics · 31.03.2026 · 65 (6) · 068003

DOI: 10.35848/1347-4065/ae50eb ↗

Kurz gefasst

Fachartikel in der Zeitschrift Japanese Journal of Applied Physics, online erschienen am 27.03.2026. Verfasst von Adarsh Gupta und 2 weiteren Autor:innen. Die Arbeit nennt 21 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Structural health monitoring (SHM) often suffers from sensor sparsity, making full-field response estimation challenging. We propose a physics-informed spatio-temporal graph network that represents the structure as a complete graph with 34-dimensional node descriptors and 17-dimensional edge attributes. A dual-stream encoder integrates graph attention networks (spatial features) with bidirectional recurrent units (temporal features), followed by cross-attention to propagate information from observed to unobserved nodes. Validated on the IASC-ASCE SHM benchmark with 50% sensor coverage, the model achieves R 2 = 0.52 and correlation ρ = 0.73 in the test split, demonstrating consistent reconstruction of dominant structural dynamics under sparse sensing.

Zitieren

APA

Gupta, A., Ayop, S. K., & Habib, A. (2026). Attention-based spatiotemporal graph models for full-field acceleration reconstruction. Japanese Journal of Applied Physics, 65(6), 068003. https://doi.org/10.35848/1347-4065/ae50eb

Vancouver

Gupta A, Ayop SK, Habib A. Attention-based spatiotemporal graph models for full-field acceleration reconstruction. Japanese Journal of Applied Physics. 2026;65(6):068003. doi:10.35848/1347-4065/ae50eb

BibTeX
@article{Gupta2026,
  author = {Gupta, Adarsh and Ayop, Shahrul Kadri and Habib, Anowarul},
  title = {Attention-based spatiotemporal graph models for full-field acceleration reconstruction},
  journal = {Japanese Journal of Applied Physics},
  year = {2026},
  volume = {65},
  number = {6},
  pages = {068003},
  doi = {10.35848/1347-4065/ae50eb}
}
Kostenlos starten →