Floating offshore wind turbines (FOWTs) are widely recognized as the key technology to exploit the abundant wind resource of many coastal areas characterized by water depths greater than 50 meters, where traditional fixed-bottom solutions would result in a not competitive COE. FOWTs are studied and designed mostly relying on numerical simulation tools that must be validated by computations and experiments to ensure their predictiveness with respect to reality. 

In the last decade, the wind energy scientific community put a great deal of effort into the verification of numerical codes, both through code-to-code and numerical-experimental comparisons and all of them shown a great dispersion of results. Uncertainties appear to be significative especially for what concerns the prediction of the interaction between the wind turbine aerodynamic loads and the low-frequency platform modes, the unsteady wind turbine wake and the hydrodynamic viscous loads. From the past testing experience also emerged that a reliable database to study and understand these problems is currently lacking. Prototype testing is affected by the well-known problems typical of full-scale experiments: environmental conditions cannot be controlled neither precisely described and the generality of the resulting data is often not completely asserted; long times and high costs are often involved; the size of recent FOWT prototypes was also not significative with respect the floating systems considered by commercial projects.

Scale model testing can complement under many aspects full-scale experiments, providing low-uncertainty data at reasonable costs, while widening the variety of dynamic conditions that are investigated in a fully-controlled environment. However, FOWTs are interested by different physics phenomena that are difficult to simultaneously reproduce in a scaled environment. At this purpose, different experimental methodologies were developed.

The object of this work is to investigate the main differences, potentialities, drawbacks and consequences of two experimental methodologies recently adopted by the authors to perform scale model experiments about three floating offshore wind turbines. The first one, was used within the Hydralab+ SparBOFWEC research framework. A 1/40 Froude-scaled model of the OC3 spar-buoy floating wind turbine based on the 5MW NREL was realized and used to investigate the overall structure dynamics under combined wind and wave loads at the DHI wave basin. The second one, was developed at Politecnico di Milano (PoliMi) to overcome the limitations imposed by Froude scaling on the correct reproduction of aerodynamic loads. A 6-degrees-of-freedom (DOFs) hardware-in-the-loop (HIL) system is used to reproduce the FOWT dynamics inside the PoliMi wind tunnel. Aerodynamic and control-related loads are generated by a 1/75 physical scale model of the wind turbine, whereas platform rigid-body dynamics and hydrodynamic loads are simulated in real-time from the integration of a numerical model. This methodology was used by the authors within the LIFES50+project to study the dynamics of two innovative floating wind turbines based on the DTU 10MW.

The two methodologies are compared in terms of design requirements for the floating system scale model, fidelity of the reproduced aerodynamic and hydrodynamic loads, uncertainties in the experiment results, especially those concerning the rotor dynamics and platform motions.