An optimal design model for sandwich pipe considering structural and thermal insulation requirements

Abstract

Sandwich pipe (SP) is a promising solution in ultra-deepwater applications. Due to the vast choices of materials and thicknesses for each layer, the design process can be tailored for a specific scenario. The purpose of the thesis is to investigate an optimal design method for sandwich pipes based on structural and thermal insulation requirements. Experimental test results of SPs subjected to external pressure are correlated with those from a nonlinear finite element model. A parametric study is then conducted for a wide range of variables to analyze the collapse behavior of fullscale SP with strain-hardening cementitious composite (SHCC) core. An equation to predict the SP collapse pressure is proposed based on the numerical results using supervised learning techniques and stochastic algorithms. The software OLGA is employed to analyze the insulation performance of the SP with the SHCC core. A mathematical model for the SP heat transfer is developed under both steady-state and shut-in conditions. A cost estimate model for SPs, including material cost, fabrication cost, and welding cost, is also considered. A mixed-integer nonlinear programming model, incorporating structural and thermal insulation requirements, is established for the optimal design of the SP. The performance of the proposed model is evaluated in a case study. Finally, a parametric study is conducted to investigate the optimal SP configurations under different working conditions.