Configuration Optimization of Lazy-Wave Dynamic Umbilicals Using Random Forest Surrogates and NSGA-II

Dynamic umbilicals, as critical components connecting offshore platforms to subsea production systems, can effectively decouple platform motions through a lazy-wave configuration, thereby reducing top tension and fatigue damage. To address the engineering challenges of numerous configuration design variables and time-consuming dynamic analyses for dynamic umbilicals, an efficient design optimization framework based on surrogate modeling and multi-objective optimization is proposed. An integrated finite-element model of a lazy-wave dynamic umbilical–offshore platform system is developed in OrcaFlex, incorporating environmental loads, material properties, and geometric parameters. The arrangement parameters of clump weights and buoyancy modules are selected as design variables, and the dynamic responses and parameter sensitivities of multiple configurations are investigated. Using simulation data, surrogate models for predicting tension and curvature are constructed via random forest regression, achieving coefficients of determination (R2) of 0.9948 and 0.9121 on the test set, respectively. Based on the surrogate predictors, the Non-dominated Sorting Genetic Algorithm II (NSGA-II) is employed to solve a multi-objective optimization problem that minimizes the maximum tension and curvature, yielding a set of Pareto-optimal solutions. The proposed approach effectively improves the stability and reliability of the dynamic umbilical system under complex sea states.