Prediction of local buckling behavior of reeled pipeline with probabilistic approach using finite element analysis / Sheralia Ufairah Abdullah

Sheralia Ufairah, Abdullah (2019) Prediction of local buckling behavior of reeled pipeline with probabilistic approach using finite element analysis / Sheralia Ufairah Abdullah. Masters thesis, University of Malaya.

	PDF (The Candidate’s Agreement) Restricted to Repository staff only Download (598Kb)
Preview	PDF (Thesis M.A) Download (2418Kb) | Preview

Abstract

Reel-lay is a fast and cost-effective method for the installation of infield flowlines and smaller export lines up to 20”. Pipelines installed by this method undergo large plastic deformation during reeling. The natural variation of yield strength and wall thickness in the steel pipe dictates the potential for bending strength mismatches between adjacent pipes. These mismatches can cause a localized peak in strain and drive gross deformation of the pipe, which may result in a buckle if not addressed at the engineering stage. DNVGL-ST-F101, the industrial standard widely used in the oil and gas industry for subsea pipelines, provides a formula to analyze local buckling using the displacement controlled condition. However, equations derived in the code are fixed with safety factors, limiting the minimum acceptable wall thickness. The standard also considered the presented requirements using results from pure bending. Based on successful track records that use wall thickness less than the standard has specified, Technip has published a paper addressing the discrepancy. The standard and the refined assessment procedure by Technip are both studied in this research, and then used to establish the probabilistic method. This research outlines the probabilistic method which firstly defines the probability of failure in calculating the mismatch. Finite element model is then developed and analyzed to verify that the level of safety associated with the method is met. The reeling studies carried out in this research has shown that the probabilistic method requires far less analyses to be done, while still meeting DNVGL’s requirement.

Actions (For repository staff only : Login required)