EUVIMEDEUVIMED entdecken
Crossref

Probabilistic Digital Twin Framework for Climate-Resilient Offshore Infrastructure under Coupled Corrosion–Fatigue Degradation and Environmental Uncertainty

Seyed Reza Samaei, Mohammad Asadian Ghahfarokhi, Nader Bavanpouri, Mohamad Amein Amjadi Kelishemi, Sajjad Pendashteh

Net Zero · 16.10.2026

DOI: 10.66967/netzero.2026.v2i105 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Net Zero, online erschienen am 16.10.2026. Verfasst von Seyed Reza Samaei und 4 weiteren Autor:innen. Die Arbeit nennt 53 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Offshore assets deteriorate through mechanisms that reinforce one another: corrosion removes section and raises the local stress, the elevated stress accelerates fatigue, and fatigue damage exposes fresh metal to the marine environment. Deterministic integrity assessment cannot represent that feedback or the environmental variability that drives it. This paper presents a physics-informed probabilistic digital twin for offshore infrastructure in which coupled corrosion–fatigue degradation modelling, Monte Carlo uncertainty propagation, sequential Bayesian data assimilation and a physics-constrained neural surrogate are combined within a single lifecycle assessment loop. The degradation state is a bounded composite index of corrosion penetration, Miner fatigue damage, cumulative environmental exposure and an interaction component, with the stress-amplification coupling regularised so that no governing relation becomes singular as failure is approached. Structural performance is tracked through the second-moment reliability index, and the Infrastructure Degradation Instability Index (IDII) is defined as a bounded, dimensionless ratio between an instability driver, the product of normalised degradation rate and dispersion, and the normalised reliability margin. For a 32 mm-wall tubular reference component configured deliberately as an aggressive 30-year demonstration case, the reliability index falls from 3.48 to 0.70 and the target of 1.5 is reached at 17.8 years; the empirical failure probability counted from the 30,000 Monte Carlo limit-state samples differs from Φ(−β) by at most 2.2 × 10−4, so the two-moment approximation is verified rather than assumed. Global variance decomposition attributes 87 % of the variance in end-of-life degradation to environmental severity. Raising the environment-driven variability multiplier from 0.5 to 2.0 at constant mean severity advances the reliability-target year from 21.5 to 10.8 even though the mean degradation trajectory falls, so variability acts through dispersion rather than through the mean. Sequential importance resampling of annual thickness and strain measurements reduces the predictive dispersion of the 30-year degradation state by 81.5%. IDII thresholds of 0.15, 0.35, 0.50 and 0.65 are mapped to reliability-index levels of 3.06, 2.26, 1.71 and 1.05, and a 21-case sensitivity study shows that the composite-index weights move the reliability-target year between 15.9 and 19.3 years while leaving the regime ordering and the warning lead time (6.6–8.1 years) essentially unchanged. The surrogate reproduces held-out trajectories with RMSE 0.0078 and R2 0.999 but is unreliable beyond its training horizon, which is why long-horizon inference is carried by the Bayesian filter. All parameters, distributions and hyperparameters are reported; the framework is presented as probabilistic decision support for a demonstration configuration, with field validation identified as the prerequisite for operational use.

Zitieren

APA

Samaei, S. R., Ghahfarokhi, M. A., Bavanpouri, N., Kelishemi, M. A. A., & Pendashteh, S. (2026). Probabilistic Digital Twin Framework for Climate-Resilient Offshore Infrastructure under Coupled Corrosion–Fatigue Degradation and Environmental Uncertainty. Net Zero. https://doi.org/10.66967/netzero.2026.v2i105

Vancouver

Samaei SR, Ghahfarokhi MA, Bavanpouri N, Kelishemi MAA, Pendashteh S. Probabilistic Digital Twin Framework for Climate-Resilient Offshore Infrastructure under Coupled Corrosion–Fatigue Degradation and Environmental Uncertainty. Net Zero. 2026. doi:10.66967/netzero.2026.v2i105

BibTeX
@article{Samaei2026,
  author = {Samaei, Seyed Reza and Ghahfarokhi, Mohammad Asadian and Bavanpouri, Nader and Kelishemi, Mohamad Amein Amjadi and Pendashteh, Sajjad},
  title = {Probabilistic Digital Twin Framework for Climate-Resilient Offshore Infrastructure under Coupled Corrosion–Fatigue Degradation and Environmental Uncertainty},
  journal = {Net Zero},
  year = {2026},
  doi = {10.66967/netzero.2026.v2i105}
}
Kostenlos starten →