Experimental and numerical investigation on the dynamic response of a 15 MW semisubmersible floating offshore wind turbine including 1-mooring line failure scenario

A 1:64-scaled physical model test was conducted in a 3D wave tank for dynamic analysis of a 15 MW semisubmersible floating offshore wind turbine (FOWT) under combined wind/wave/current environmental excitations in a parked condition for intact and 1-mooring-line-failed scenarios. The semisubmersible is characterized by non-uniform columns and multiple heave plates, and it hosts the IEA 15 MW reference wind turbine and connected to a 3 × 2 chain-catenary mooring system at water depth of 51.2 m. Aero-hydro-servo-mooring-elastic coupled-dynamics mid-fidelity digital twin model was built in parallel. The static-offset and free-decay test results were compared with experiments to check coupled-system stiffness, damping and natural periods. Under intact conditions in extreme sea, several sensitivity studies were conducted and analyzed: (a) full quadratic transfer function (QTF) versus Newman's approximation, (b) viscous drag on heave plates, (c) linear and nonlinear hydrostatics, (d) turbulent and steady extreme wind models. It was found that the QTF method better estimates low-frequency platform and tension responses. Comparisons between intact and 1-line-failed conditions were also conducted. The line failure resulted in significant increase of platform's offset and neighboring mooring tensions, but transient response due to sudden disconnection did not show noticeable overshoots. Overall comparisons between experiment and simulation showed consistently good agreements. • 1:64-scale model test for global performance of a 15 MW semisubmersible FOWT. • Platform characterized by multiple heave plates and non-uniform columns. • Extreme load cases for intact and 1-line failed conditions compared to simulations. • Sensitivities for QTF/nonlinear hydrostatic/turbulent wind/drag coefficients checked. • Slowly varying responses better agreed using full QTF than Newman approximation.