Collapse analysis of a large plastic pipe using cohesive zone modelling technique is undertaken.
•
Fracture energy release rate Gc is determined experimentally through a single cantilever beam test setup.
•
Experimental findings are implemented in a cohesive zone model.
•
The model is used to determine the competing collapse mechanisms in a large subsea plastic pipe subjected to negative internal pressure.
Abstract
Polypropylene (PP) plastic pipes have recently gained widespread application in non-pressurized gravity pipes used for seawater intake lines in the petrochemical industry. These pipes consist of a solid wall base pipe, on which an outer reinforcement called the omega-profile is spirally winded and hot fusion bonded. The omega-profile is usually filled with grout to provide on-bottom stability for subsea installation. It is of high importance that the bond between the omega-profile and the base pipe has sufficient strength to provide resistance against buckling of the pipeline system. The objective of this study is to investigate the collapse behaviour of such large-diameter PP pipes subjected to a negative internal pressure. The bond is modelled with cohesive zone modelling technique with the aim to determine the failure mode that governs the collapse behaviour of the pipe, e.g. buckling or delamination. Experiments where conducted on single cantilever beam (SCB) specimens cut from the pipe to determine the cohesive bond strength between the omega-profile and base pipe. The findings from the experiments are implemented in a full pipe model, where the surface between the omega-profile and base pipe is assigned bond strength characteristics in accordance with the experimental results. The FEA results of the non-linear collapse analysis of the full pipe model show that for the range of grout stiffness values considered (0 ≤ Eg ≤ 30 GPa), the governing failure mode of the pipe is initiated by buckling and proceeded by delamination. For delamination to govern the failure mode, a grout stiffness greater than 36 GPa in combination with a weaker bond strength than the experimentally measured would be required. The methodology presented in this study gives a rather accurate tool for the design and analysis of this type of structures, and can reliably assess the bond strength level required in view of the governing failure modes, e.g. buckling and delamination.