Metamodel approach for reliability-based design optimization of a steel catenary riser

A reliability-based design optimization (RBDO) methodology is presented for the design of a steel catenary riser (SCR) under dynamic environmental loads. The purpose of this work is to optimize the cost of products subjected to probabilistic constraints. Searching for the optimal design of the riser in a wide range of design variables is computationally very expensive if time-consuming codes for dynamic analysis are necessary in the iteration process. In this study, the effectiveness of the proposed RBDO using a metamodel is firstly studied and validated through a beam test, then applied to the industrial dynamic optimization problem. The design variables of structures are assumed to be uncertain, and some other parameters such as loading and material properties are considered random. The performance function is approximated using metamodels to avoid time-consuming finite-element analysis during the optimization iteration. A single-loop method is used to decouple the double-loop RBDO problem. The reliability is finally confirmed through Monte Carlo simulations. According to the analysis, the presented methodology is more rational and realistic compared with deterministic optimization.