Experimental Study on the wind turbine wake meandering with the help of a non-rotation simplified model and of a rotating model

Abstract

In order to study the wind turbine wake and its eventual interactions with neighbouring wind turbines, several numerical and physical modelling approaches are used. Some model the wind turbine with the simplest model, that is the actuator disc concept, adding a drag source (i.e. pressure loss) within the surface swept by the blades (numerical, physical,). Some use the Blade Element Momentum Theory, which takes into account the blade rotation effect on the wake and the aerodynamic features of the blades,. Some use Large Eddy Simulation to compute the unsteady flow around the entire rotor. In a wind resource assessment context, the latter one is not mature enough to be used since the computation times are extremely long. The second one has more acceptable computation time but the first one is still the most attractive to model the far wake, according to its simplicity of implementation and short computation time. On the other hand, the issue is that it is difficult to assess the errors induced by the absence of blades and associated rotation momentum on the wake development. Furthermore, the level of turbulence intensity encountered in the atmospheric incoming flow plays a role on this issue : First, the higher the turbulence intensity is, the faster the spectral signature of the blades disappears in the wake and the faster the azimuthal velocity induced by the rotational momentum is overwhelmed in ambiant velocity fluctuations. These intuitive remarks need to be quantified. In this context, a previous study of the present authors compared the wake properties of a model of a 3-blade rotating wind turbine (D = 416mm, TSR = 5.8, CT = 0.50)(Fig. 2) and of a porous disc made of metallic mesh (Fig. 3), generating the same velocity deficit as the wind turbine (Fig. 3). Both models were tested in an atmospheric boundary layer (ABL) wind tunnel. A special care had been supplied to reproduce in the wind tunnel, at a smaller geometric scale, the flow and turbulence properties of an ABL (called later on ”imodelled ABL”?). The spectral content of the turbulence and so, the turbulence length scales of the ABL are then reproduced. The modelled boundary layer had neutral stability conditions and the turbulence intensity at hub height was 13%. The 3D flow properties were measured from x = 0.5D to 3D downstream of the wind turbine. The mean flow and the spectral content were studied and the conclusion was that the additional rotation momentum and the blade signature in flow are no longer visible at the beginning of the far wake, expected at x = 3D). On the other hand, this statement is valid for the tested freestream turbulence intensity (13% at hub height) and cannot be extrapolated to any freestream conditions. One still can expect that the weaker the freestream turbulence intensity is, the farther the tip vortex signature and the additional rotation momentum remain visible within the wake. To complete this results, an experimental study about the unsteady properties of the wake downstream of the rotating wind turbine and of the porous disc located in a modelled ABL has been undertaken. Indeed, it is established that the large turbulence length scales present in the ABL can be responsible of the wake meandering process ). This phenomenom will be studied in the wind tunnel through space-time correlations of velocity signals measured at locations diametrically opposed in the wake, and compared for the two type