Enhancement of hydrodynamics modeling for floating offshore wind turbines using multi-objective genetic algorithm

Abstract

Floating offshore wind turbines (FOWTs) are pivotal for enhancing the U.S. energy supply by harnessing deep-water wind resources with higher capacity factors. Accurate hydrodynamic modeling is essential for predicting FOWT responses to varying sea states, as hydrodynamic coefficients—added mass, damping, and drag—govern platform dynamics. However, these coefficients are highly sensitive to platform geometry and environmental conditions, presenting significant challenges for predictive modeling. This study presents an optimization framework that automates the tuning of hydrodynamic coefficients across varying sea states. The hydrodynamics model is developed based on Morison’s equation and enhanced with second-order wave kinematics, wave stretching, MacCamy-Fuchs corrections, depth-dependent coefficients, and component-wise discretization. A multi-objective Genetic Algorithm (GA) is employed to calibrate coefficients using data from free-decay and irregular wave tests. The framework treats hydrodynamic coefficients as design variables and evaluates fitness based on dynamic responses in both time and frequency domains. To support generalization, regression models are developed to estimate damping coefficients under varying sea states. Validation on two reference platforms—the OC3 Spar-Buoy and VolturnUS-S Semi-Submersible—demonstrates the framework’s adaptability. Results show that incorporating depth- and sea-state-dependent coefficients significantly improves response predictions compared to decay-test-only models, highlighting the benefits of automated hydrodynamic optimization for FOWT modeling.