Multi-objective optimization design for the carcass of marine flexible pipes based on BP neural network and genetic algorithm

Abstract The collapse of the carcass is a prominent failure mode in marine flexible pipes. This paper explores the multi-objective optimization design of the carcass layer, focusing on pivotal design variables such as the thickness of the steel strip and the height-thickness ratio of the profile. The objective functions encompass the unit length weight and critical collapse value. The optimization process integrates the BP neural network with the Non-dominated Sorting Genetic Algorithm-II (NSGA-II). Furthermore, the improved minimum distance selection method is applied to extract optimal results from the Pareto front. The insights obtained from this paper hold significant potential to contribute to engineering applications, particularly in advancing the design methodologies for carcass collapse resistance.