This is the authors' abstract. We don't add key points for this paper.
The high slenderness ratio inherent in offshore pipelines renders them susceptible to buckling under various stresses encountered during the operational period. The predominant mode of buckling for buried offshore pipelines tends to be upward. Hence, this study employs finite-element analysis and artificial neural networks to probe the upheaval displacement behavior of offshore pipelines and evaluate the impact of diverse soil, pipe, and installation parameters. Through finite-element analysis, the study inspects the variation in uplift resistance of pipeline segments by altering factors such as the effective soil shear strength (Su), embedment depth (H), effective soil unit weight (γ′), interface adhesion factor (Rinter), and interface tensile capacity (T), with graphical presentations of the results. Validation of numerical models against laboratory experiments and prior numerical investigations fortifies the study’s credibility. Furthermore, the results of the numerical analysis are used for the construction of a neural network aimed at predicting the nondimensional uplift capacity factor using four nondimensional key input parameters: the interface adhesion factor (Rinter), embedment depth ratio (H/D), incremental shear strength factor (k/γ′), and interface tensile capacity factor (T/Su0). The weights of the input parameters and hidden neurons are utilized to propose a correlation equation linking the input parameters to the uplift capacity factor (Nu). Sensitivity analysis is then employed to examine the influence of these parameters, with results presented both in tabular format and through neural interpretation diagrams. Comparative analysis with existing literature and numerical studies demonstrates satisfactory alignment with the correlation equation derived from the neural network. The numerical analysis reveals an increased uplift capacity factor with rising H/D and T/Su0. Notably, for lower T/Su0, the uplift capacity factor demonstrates a decreasing trend with increasing k/γ′, which reverses for higher T/Su0 values (e.g., T/Su0=60). Moreover, both the numerical and sensitivity analyses indicated a minimal impact of the Rinter on the resistance of the pipe segment against its uplift. Thus, the findings from the current study revealed significant influences of several influencing parameters, as validated against past studies, indicating a promising avenue for predictive modeling and risk assessment in offshore pipeline engineering.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Seth et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: