Wind Resource Assessment over a complex Terrain by Wind Tunnel Modeling, Uncertainty Quantification

The wind resource assessment in complex terrain is a demanding task with lots of parameters to consider, often, simple numerical models are not sufficient to solve it. Wind tunnel testing is one of the possibilities to evaluate the wind on a complex topography like a mountain, a cliff or a forest. Additionally to measurement errors that can be estimated from classical derivation of measurement chain equations and from calibration, the wind tunnel simulation, as any simulation, also encounters uncertainties due to the assumptions. The uncertainty quantification aims at estimating the scatter obtained when the conditions differ from the original assumption made.

The goal of this work is to quantify the sensitivity of three main parameters driving the physical modeling: the Reynolds number, the flow angle and the inlet wind profile.

The test case chosen for the study is the Bolund island (Denmark) because of the important database including field measurements, numerical simulations and wind tunnel data. A 1/450 model is tested at the von Karman Institute in a 0.35m x 0.35m x 2m test section adapted for atmospheric flow modeling. Two-component PIV and hot-wire anemometry measurement techniques are performed for different configurations. The Reynolds number can be modified by a factor of 5, inlet wind conditions are changed from Cat. I (offshore conditions) to Cat. III (rough conditions) and the wind direction is modified in a range from -15 to +15 degrees of angle.

Results are discussed and compared with CFD and field measurements. The influence of each parameter is discussed. As an example, the inflow conditions has a major influence on the turbulent level downstream of the hill, a low turbulence in the inlet is leading to a very important increase of the turbulence downstream as the eddies will take more time to dissipate in a low turbulence flow.