Wind tunnel

The model in the wind tunnel was to the scale of 1:150. Reynolds scaling technique was used.To achieve the right properties at the expected Reynolds number the geometry of the blade was modified.
The purpose of the test was to investigate the aerodynamic interaction in-between the turbines.

The tests were performed for one wind velocity and 5 different directions: -10°, -5°, 0°, 5°, 10°.
Movie from our own wind tunnel shows the test upset and the model.


The wind properties in the tunnel were modelled as close as possible to an offshore wind: the velocity variation was represented as well as the turbulence percentage of about 9%.
The aft turbine working alone gave 30% more power than the forward turbines alone. This is due to the wind velocity profile and the higher hub position of the aft turbine.
When all three turbines were started together the setting had to be adjusted to reach an optimum.
The power of the aft turbine was then reduced to 89% of the average of port and starboard turbine power. The influence the forward turbines have on the aft turbine is likely to be more pronounced in full scale than in model scale.
The forward turbines were only marginally affected by one another or by the aft turbine.
The working conditions of the aft turbine, with its wings moving from high undisturbed wind into the slowed down very turbulent wake of the leading turbines, have not been studied explicitly in the present project, but should be considered, in particular with respect to fatigue life of wings and other components of the turbine.This is addressed in the Risø report (See system Engineering).