Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Comparison of 3D transitional CFD simulations for rotating wind turbine wings with measurements

: Schaffarczyk, A.P.; Boisard, R.; Boorsma, K.; Dose, Bastian; Lienard, C.; Lutz, T.; Madsen, H.; Rahimi, Hamid; Reichstein, T.; Schepers, G.; Sørensen, N.; Stoevesandt, B.; Weihing, P.

Fulltext ()

The Science of Making Torque from Wind, TORQUE 2018 : 20-22 June 2018, Milan, Italy
Bristol: IOP Publishing, 2018 (Journal of physics. Conference series 1037)
Art. 022012, 10 pp.
International Conference "The Science of Making Torque from Wind" (TORQUE) <2018, Milan>
Conference Paper, Electronic Publication
Fraunhofer IWES ()

Since the investigation of van Ingen et al., attempts were undertaken to search for laminar parts within the boundary layer of wind turbines operating in the lower atmosphere with much higher turbulence levels than seen in wind tunnels or at higher altitudes where airplanes usually fly. Based on the results of the DAN-Aero experiment and the Aerodynamic Glove project, a special work package Boundary Layer Transition was embedded in IAEwind Task 29 MexNext 3rd phase (MN3). Here, we report on the results of the application of various CFD tools to predict transition on the MEXICO blade. In addition, recent results from a comparison of thermographic pictures (aimed at detecting transition) with 3D transitional CFD are included as well. The MEXICO (2006) and NEW MEXICO (2014) wind tunnel experiments on a turbine equipped with three 2.5 m blades have been described extensively in the literature. In addition, during MN3, high-frequency Kulite data from experiments were used to detect traces of transitional effects. Complementary, the following set of codes were applied to cases 1.1 and 1.2 (axial inflow with 10 m/s and 15 m/s respectively) – elsA, CFX, OpenFOAM (with 2 different turbulence/transitional models), Ellipsys, (with 2 different turbulence models and eN transition prediction tool), FLOWer and TAU – to search for detection of laminar parts by means of simulation. Obviously, the flow around a rotating blade is much more complicated than around a simple 2D section. Therefore, results for even integrated quantities like thrust and torque are varying strongly. Nevertheless, visible differences between fully turbulent and transitional set-ups are present. We discuss our findings, especially with respect to turbulence and transition models used.