CRACK ANALYSIS ON A SUBSEA PIPELINE DUE TO ANCHOR DRAGGING USING EXTENDED FINITE ELEMENT METHOD (XFEM)

• Main Author: Filius Santoso, Jason
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/70163
• Institution: Institut Teknologi Bandung

Subsea pipeline remains to be the preferred solution for economically transferring hydrocarbon products between offshore facilities or towards land facilities for export. Strict standards and safety requirements are provided to regulate subsea pipeline design processes to produce an optimal and safe design, mitigating various modes of failure from the operation-based design itself. Nevertheless, pipelines are still susceptible to external damages. One of the major causes of pipeline accidents is due to ship anchoring. The probability of this incident is low, but the resulting damage may be fatal. During its operation, damages on the pipeline structure may appear due to lack of welding fusion or penetration during fabrication, sharply localized corrosion during service, a combination of corrosion, residual stress, and poor microstructure of steel, or even third party impacts such as anchoring or vessel collision. Damages on pipelines due to fabrication flaws or operational damages often manifest themselves as cracks. Once a crack is identified on an operating pipeline, the conventional ratio of operational stress to the yield strength of the pipe cannot be the only integrity acceptance criterion, but fracture mechanics needs to be taken into consideration. Most pipeline systems are under cyclic loads and other external loads during its operational period. These loads may cause fractures such as cracks to initiate and propagate, causing fatal failure of the pipe. One dangerous pipeline failure scenario is a crack that grows continuously, damaging a significant length of the pipe, ending in rapid leakage of pipe contents in huge amounts. Therefore, this study focuses on modeling anchor drag on a cracked subsea pipeline. This research complements previous similar studies of anchor drag modeling on a subsea pipeline with an inclusion of an existing crack. The objective of this study is to perform crack analysis and model crack propagation on a subsea pipeline, to determine the influence of crack on a subsea pipeline dragged by a ship anchor, and to determine the effect of initial crack dimensions to the crack propagation on a dragged subsea pipeline. The approach taken in this study is mathematical modeling, specifically numerical analysis using finite element method. Modeling is done by utilizing Abaqus, especially the extended finite element method (XFEM) feature. XFEM is used to perform crack analysis and crack growth modeling on a pipe model while eliminating the usage of a complex mesh and remeshing in propagation modeling as required in the conventional finite element method. The general procedure of this study include the review of relevant literatures, data collection, calculation of stress intensity factor (SIF) under operational loads, validation of crack model by comparing the SIF from modeling results and the theoretical SIF calculated using API 579-1, global modeling of pipe and anchor interaction to determine the stresses at maximum displacement, local modeling of pipe and anchor interaction to study crack behaviour, and analysis of all modeling results. This study results in a finite element model capable of assessing SIF values of cracks on a subsea pipeline. These models are validated using theoretical SIF values calculated using API 579-1, with a maximum deviation of 4.61% throughout all nine variations of initial cracks (a = 4, 5, 6 mm; 2c = 20, 40, 60 mm). It is concluded that all cases of initial crack will not propagate deeper into the pipe based on Abaqus SIF evaluation for operating conditions. The largest variation of the studied crack dimensions yields an SIF value of only 9.703 ksi?in which is significantly lower than the pipe fracture toughness, 153.713 ksi?in, therefore the pipe is still fit for service under normal operating conditions. The studied pipeline suffers a maximum displacement of 214.565 meters due to the anchor drag of the vessel used in this study. At this point, all variations of initial crack have propagated and caused pipeline leakage. The smallest crack variation causes the pipe to leak at an anchor drag distance of 67.839 meters, while it only takes 33.389 meters of anchor drag distance to cause the pipeline with the largest crack variation to leak. Each millimeter of initial crack depth is more influential than initial crack length to both the pipe strength in withholding leakage due to crack propagation under anchor drag loads and the extension of crack length on the exterior surface of the pipe.