In 2011, the DeepCwind Consortium performed 1/50th-scale model tests on three offshore floating wind platforms at the Maritime Research Institute Netherlands (MARIN) using a geometrically scaled model of the National Renewable Energy Laboratory (NREL) 5 MW reference turbine. However, due to the severe mismatch in Reynolds number between full scale and model scale, the strictly Froude-scaled, geometrically-similar (geo-sim) wind turbine underperformed greatly, which required significant modification of test wind speeds to match key wind turbine aerodynamic loads, such as thrust. The conclusion from these prior efforts was to abandon a geometrically similar model turbine and use a performance-matched turbine model in its place, keeping mass and inertia properties properly scaled, but utilizing modified blade geometries to achieve required performance at the lower Reynolds numbers of the Froude scaled model. To this end, the University of Maine and MARIN worked in parallel to develop performance-matched turbines designed to emulate the full scale performance of the NREL 5 MW reference turbine at model scale conditions. An overview of this performance-matched wind turbine design methodology is presented and examples of performance-matched turbines are provided.
The DeepCwind semi-submersible platform was retested at MARIN in 2013 using the MARIN Stock Wind Turbine (MSWT), which was designed to closely emulate the performance of the original NREL 5 MW turbine. This work compares the wind turbine performance of the MSWT to the previously used geometrically scaled NREL 5 MW turbine. Additionally, turbine performance testing of the 1/50th-scale MSWT was completed at MARIN and a 1/130th-scale model was tested at the University of Maine under Reynolds numbers corresponding to the Froude-scaled model test conditions. Results from these tests are provided to demonstrate effects on model test fidelity. Comparisons of the performance response of the geometrically matched turbine to the performance-matched turbines are also presented to illustrate the performance-matched turbine methodology. Lastly, examples of the fully dynamic floating system performance using the original geometrically scaled NREL 5 MW turbine and the MSWT are investigated to illustrate the implementation of the model test procedure as well as the effects of turbine performance on floater response. Using the procedures employed for the MARIN tests as a guide, the results of this work support the development of protocols for properly designing scale model wind turbines that emulate the full scale design for Froude-scale wind/wave basin tests of floating offshore wind turbines.