An efficient hydrodynamic-structure unified state-space model for the coupled dynamics during floating wind turbine wet-towing construction

Wet-towing is a critical installation stage for a large floating offshore wind turbine (FOWT), yet the tower is still commonly treated as a rigid component in all existing towing simulations. This simplification may compromise the prediction of hydroelastic load transfer as turbine capacity and tower height continue to increase. To address this limitation, this study develops an efficient hydrodynamic-structural coupled time-domain model for the wet-towing of the International Energy Agency 15 MW FOWT for the first time. A static condensation method is introduced to incorporate tower elasticity into the global six degrees of freedom floating body formulation, while a state-space model is established to replace the radiation memory convolution in the Cummins equation. The further derived hydrodynamic-structural unified state-space model enables rapid time-domain simulation of the towing system under forward speed effects. The proposed model is first validated against Newman's classical double-barge benchmark and free decay simulations. Frequency-domain results reveal that forward speed mainly governs the evolution of surge, heave, and pitch response amplitude operators, while tower elasticity significantly reshapes the hydroelastic resonance characteristics. The tower base loads are concentrated mainly in the frequency ranges of 0.8–1.0 and 1.2–1.5 rad/s, whereas the most critical tower top stress is controlled by a low-frequency surge-pitch coupled regime near 0.19–0.22 rad/s rather than by heave resonance. Time-domain comparisons with a validated OpenFAST and AQWA coupled framework show that the proposed model accurately reproduces the dominant pitch motion, nacelle acceleration, tower shear force, and stress evolution under both regular and irregular wave conditions. The results further indicate that stronger turbulent wind, longer wave periods, and higher towing speed intensify fore-aft bending. Additionally, the proposed model reduces computational time by approximately 80% in regular waves and more than 54% in irregular waves, which demonstrates strong potential for efficient engineering assessment of FOWT towing safety.