Field-Validated Digital Twin–Enabled Structural Health Monitoring for Offshore High-Pile Wharves
Future Cities and Environment · 08.05.2026 · 12 · 22
DOI: 10.70917/fce-2026-010 ↗Kurz gefasst
Fachartikel in der Zeitschrift Future Cities and Environment, online erschienen am 08.05.2026. Verfasst von Seyed Reza Samaei und 1 weiteren Autor:innen. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Offshore high-pile wharves operate under harsh, variable loads that can mask the earliest signs of deterioration. We present a closed-loop structural health monitoring (SHM) framework that couples multimodal machine/deep learning with a physics-based digital twin (DT). A year-long field deployment at an operating high-pile wharf in Western Asia instrumented six piles with triaxial accelerometers and strain gauges (GPS-synchronized, 100 Hz). After quality control and environmental/operational variability normalization, an unsupervised VAE produces an anomaly score that is fused with a CNN–GNN classifier to yield window-level damage probabilities and pile-by-elevation localization. The DT assimilates these diagnostics through weighted least squares and an EKF, with measurement covariance derived from model uncertainty; “what-if” simulations quantify the structural consequence of candidate repairs. Under chronological splits with leave-one-pile-out, the framework achieves F1 = 0.95 and AUC-ROC = 0.98, with false-alarm rate ≈ 2–5 % and median detection latency ≈ 45 s. Spatial heatmaps consistently pinpoint the affected elevation, while online updating reduces mean frequency bias from ≈ 4.8 % to ≈ 0.9 % and increases mode-shape MAC from ≈ 0.81 to ≥ 0.92. First-mode shifts of ≈ 3–4 % observed in the field are reproduced by DT “stiffness-restoration” scenarios, enabling risk-informed maintenance planning. Compared on identical normalized windows, the proposed method outperforms classical vibration-based indices in both discrimination and false-alarm control. The results demonstrate a scalable, explainable pathway to predictive asset management for critical maritime infrastructure. Beyond structural safety, the proposed framework contributes to sustainable port operation by enabling condition-based maintenance, reducing unnecessary inspections, extending asset service life, and lowering the embodied carbon associated with premature repair or replacement.
Zitieren
APA
Samaei, S. R., & Riffat, J. (2026). Field-Validated Digital Twin–Enabled Structural Health Monitoring for Offshore High-Pile Wharves. Future Cities and Environment, 12, 22. https://doi.org/10.70917/fce-2026-010
Vancouver
Samaei SR, Riffat J. Field-Validated Digital Twin–Enabled Structural Health Monitoring for Offshore High-Pile Wharves. Future Cities and Environment. 2026;12:22. doi:10.70917/fce-2026-010
BibTeX
@article{Samaei2026,
author = {Samaei, Seyed Reza and Riffat, James},
title = {Field-Validated Digital Twin–Enabled Structural Health Monitoring for Offshore High-Pile Wharves},
journal = {Future Cities and Environment},
year = {2026},
volume = {12},
pages = {22},
doi = {10.70917/fce-2026-010}
}
EUVIMED