Ultimate Capacity of Subsea CO2 Transport Pipelines with Semi-Elliptical Corrosion Defects under Combined Internal Pressure and Bending Moment: A Finite Element Study

Abstract

This study investigates the ultimate bending capacity of API 5L X52 subsea CO₂ transport pipelines as part of environmentally sound technology within Sustainable Development Goal (SDG) 17, incorporating idealized semi-elliptical corrosion defects under combined internal pressure and bending moment. A nonlinear finite element model was benchmarked against published four-point bending tests on corroded pipes with rectangular defects for validation purposes. The validated modeling procedure was then applied to a deterministic parametric study involving 33 selected cases, with corrosion depth ratios of = 0.1 – 0.8, defect length ratios of = 0.2 – 1.2, and internal pressures ranging from 0 to 20 MPa. The results show that corrosion depth is the primary parameter controlling bending-capacity degradation because it reduces the residual wall thickness, local section stiffness, and shell stability at the remaining ligament. Increasing defect length reduces the bending capacity up to a saturation region, after which the response is mainly governed by the weakest local cross-section. For the investigated semi-elliptical defect configuration, internal pressure reduces the ultimate bending capacity rather than improving it. The pressure-induced circumferential stress interacts with the bending-induced axial stress, promoting earlier yielding and local buckling at the corroded ligament. The normalized ultimate bending capacity at  and  decreased from 0.95 to 0.86 as the pressure increased from 0 to 20 MPa. These findings indicate that the residual bending capacity of corroded CO₂ pipelines should be assessed by considering the coupled effects of metal-loss geometry, pressure-induced membrane stress, and local shell instability. The results are deterministic estimates and are limited to the investigated pipe geometry, idealized single-defect morphology, material model, and loading range. The benchmark validation supports the FE framework but does not directly validate the semi-elliptical defect morphology.