Cohesive zone model based reliability analysis for a sandwich pipe. (15th February 2023)

• Record Type: Journal Article
• Authors: Qu, Yangyang; Chen, Nian-Zhong; Yuan, Lin
• Is Part Of: Ocean engineering. Volume 270(2023)
• Publication Date: 2023-02-15
• Subjects: Sandwich pipe (SP) -- Collapse -- Burst -- Cohesive zone model -- Reliability analysis -- Response-surface method; Ocean engineering -- Periodicals; Ocean engineering; Periodicals; 620.4162
• DOI: 10.1016/j.oceaneng.2022.113550
• Languages: English
• ISSNs: 0029-8018

Abstract

Abstract: A cohesive zone model based reliability analysis for a sandwich pipe (SP) is presented in this paper. The collapse or burst strength of SP is predicted by a cohesive zone model based nonlinear finite element (FE) analysis. The interlayer adhesion is simulated by cohesive zone model in accordance with the maximum shear strength. The bilinear traction separation law is employed in the cohesive zone model. A quadratic stress criterion and an energy-based model are used for simulating the damage initiation and evolution. The validity of the FE analysis is then validated by the comparison between model prediction and experimental results. A first-order reliability method coupled with a response-surface method is then used for reliability estimate dealing with implicit limit state functions corresponding to failure modes of collapse and burst. A sensitivity analysis to investigate the influence of design variables on the reliability of SP is further conducted and the results show that working pressure, yield strength of material, and modelling uncertainties are relatively important design variables in predicting the collapse and burst strength of SP, while initial ovalities of internal and external pipes could be regarded as constants in the reliability analysis for improving the computational efficiency. Highlights: Reliability assessment is conducted for a sandwich pipe in terms of collapse and burst. Interlayer adhesion of a sandwich pipe is simulated by the cohesive zone model. Implicit limit state functions are solved by a response surface method. A sensitivity analysis is further carried out for critical parameters.