Cross-Validation of Numerical and Experimental Studies of Transitional Airfoil Performance

The aerodynamic performance characteristic of airfoils are the main input for estimating wind turbine blade loading as well as annual energy production of wind farms. For transitional flow regimes these data are difficult to obtain, both experimentally as well as numerically , due to the very high sensitivity of the flow to perturbations, large scale separation and performance hysteresis. The objective of this work is to improve the understanding of the transitional airfoil flow performance by studying the S826 NREL airfoil at low Reynolds numbers (Re = 4.10 4 and 1.10 5) with two inherently different CFD methodologies, in combination with wind tunnel experiments. Large-Eddy Simulations (LES) performed with a novel high order code based on the Discontinuous Galerkin Method are compared to LES from the well established wind turbine CFD code EllipSys3D. Both codes are considering natural transition. The similarity of the results obtained by these two very different simulation methods seems to demonstrate the validity of the computations. Differences are however observed with the experimental results. To understand these discrepancies, further analyses have been performed on both the numerical and the experimental sides. On the numerical side, the span sensitivity study showed that span lengths of 10 and 40% of the chord were leading to similar results. On the experimental side, the flow visualizations using oil streaks indicated strong 3D effects under the form of stall cells on a significant part of the span, as well as walls effects for Re = 1.10 5. Considering the sensitivity of the measurements to the tunnel environment, the strong similarity of the LES results inspires confidence in the validity of the computations. .