Optimal Design Procedure for Offshore Pipelines based on Computational Simulation of Pipe Forming Process

Abstract

A pipeline is made of segmented steel pipes connected to form a continuous pipe system for the transportation of oil or gas over a long distance. During the past several decades, UOE and JCO pipes have gained increasing application to produce at lower cost offshore pipelines with diameter larger than 16 inches instead of the conventional seamless pipe. In UOE and JCO pipe forming processes, the steel plate is subject to a series of plastic forming including pressing and spring back and a final welding stage to form the circular pipe. However, the complicated histories of the plastic forming processes executed in the UOE and JCO pipe forming methods involve the following problems. First, the formed pipes develop material properties differing significantly from those of the raw plate. The repeated loading and unloading cycles conducted throughout the forming processes alter the yield strength and curved shape of the stress-strain response of the pipe due to the Bauschinger effect and work hardening. Apart from having critical effect on the structural performance of the steel pipes, these material properties are also representative indicators of their quality. Therefore, the accurate prediction of the material properties will result in non-negligible economy in terms of the cost and time spent for the repeated inspection, design and production performed to secure the strength and structural performance of the formed pipe. The second problem relates to the geometric imperfection and residual stress inherent to the repetition of plastic forming and elastic spring back experienced throughout the UOE and JCO forming processes. Along with the material properties, the ovality and residual stress of the pipe are dominant parameters determining its collapse pressure or bending capacity but occur in such an unpredictable manner that they increase the design uncertainty. The consideration of all these interrelated parameters by means of coefficients as well as the high material and geometrical nonlinearities in the design limits the accuracy of the prediction. This loss of accuracy itself results in excessively conservative design that does not guarantee the pipe to provide consistent quality and satisfactory structural performance. Such situation stresses the pressing need for a method enabling to predict accurately the material properties and structural performance of UOE and JCO steel pipes. This thesis presents an optimal design procedure for offshore pipelines manufactured by UOE and JCO forming processes. The proposed procedure involves (1) the computational simulation of UOE and JCO pipe forming processes by finite element analysis to provide accurate prediction of the parameters of the formed pipe including its material properties, geometrical imperfections, and residual stress

(2) the structural analysis of the steel pipe using the results of the simulation

and, (3) the maximization of the collapse pressure of the formed steel pipe known to be the main structural performance, while ensuring its producibility and quality. To improve the accuracy of the simulation of the UOE and JCO pipe forming processes, nonlinear combined hardening model is applied to describe the plastic characteristics including yield plateau and evolution of Youngs modulus as well as work hardening and Bauschinger effect. The strain-stress response is obtained by tension-compression cyclic test on the raw material, and the genetic algorithm and RMS method are combined to derive fifteen material parameters. Finite element simulation of the UOE, UOC, JCOE, and JCOC forming processes is performed and the corresponding configuration, material properties, geometric imperfections, and residual stresses are derived for each of the processes. From these results, the yield strength can be predicted directly and the producibility of the steel pipe can be checked by monitoring the shape change of the plate and the reaction force applied to the forming tools. The validity of the numerical simulation of the forming process as well as the derived results are verified by the tensile test conducted on a sample cut from a steel pipe produced by UOE forming. Numerical analyses are then performed to estimate the collapse pressure and bending capacity of steel pipe based on the simulation outputs. Here also, the results are in good agreement with the experimental results of previous studies. Parametric analysis is performed to investigate the effect of the pipe expansion and compression on its material properties and structural performance. It is found that larger pipe expansion increases the tensile yield strength but degrades the collapse performance. Therefore, executing compression instead of expansion can increase significantly the collapse performance but with some loss of the tensile yield strength. However, neither compression nor expression appears to affect relevantly the bending capacity. Finally, the optimal design procedure for UOE and JCO pipes is proposed considering the trade-off effect of the design variables on the yield strength and collapse pressure. The proposed procedure is seen to improve the design consistency and efficiency compared to conventional methods and to achieve maximized collapse pressure while securing the producibility and quality of the UOE and JCO pipes.