Development of floating offshore wind turbine technology is in need of more scaled model tests to investigate its dynamical characteristics. This paper presents the model testing investigations of a spar-buoy and semi-submersible floating wind turbines, and conducts a series of comparisons on the experimental results of the two floating wind turbines. It is found that the spar-buoy floating wind turbine is more sensitive to wind loading and has larger amplitudes of platform motion in wind only cases, especially for the yaw motion. By contrast, the semi-submersible floating wind turbine is more sensitive to wave loading, especially for the second order difference-frequency wave loading. In addition, the dynamic responses of the two floating wind turbines are both influenced by the current flow. Compared to the semi-submersible floating wind turbine, the more regular “figure-eight” surge-sway trajectory is observed in the spar-buoy floating wind turbines, showing the influence from vortex-induced-motion. This paper provides more valuable experimental details and results to the scientific community with regard to the spar-buoy and semi-submersible floating wind turbines. Moreover, the comparison of the dynamic behaviours of the two floating wind turbines in various conditions helps to understand the characteristics and mechanism about the floating wind turbine technology.

There are different approaches utilized to establish dynamic models for floating offshore wind turbines, but their distinction and applicability are often confusing. This paper presents the theories about four different dynamic models developed for simulating floating wind turbines, including a single-rigid-body model, a corrected single-rigid-body model, a multi-rigid-body model, and a multi-rigid-flexible-body model. Subsequently, a series of comparisons between the four dynamic models are conducted to assess their differences and applicability for simulating floating offshore wind turbines. The results show that the multi-rigid-flexible-body model is most accurate but most time-consuming. In most cases, the multi-rigid-body model and the multi-rigid-flexible-body model can be performed, except for the aeroelastic effects and global dynamic responses caused by the vibration of the tower and blades. In contrast, the single-rigid-body model has simpler dynamical methodologies and less computational cost. However, it has some internal limitations, for instance, limited degrees of freedom, neglecting aeroelastic effects, neglecting gyroscopic effects, etc. Hence, the corrected single-rigid-body model is proposed and performs better than the original single-rigid-body model at the cost of a little time-consumption.