Are 500 Simulations Enough for Uncertainty Quantification on Turbulence and Transitional Model on low-Reynolds Airfoils?

This work presents an uncertainty quantification study on the DU89-134 airfoil at a Reynolds number of 5 × 10⁵ and an angle of attack of 5°. The uncertainty analysis is conducted using the k-ω SST model coupled with the γ - Re_θ transition model, focusing on the influence of selected inflow and empirical model parameters on the aerodynamic response. Monte Carlo sampling is applied independently to the distributions of the uncertain input parameters, while Polynomial Chaos Expansion is employed to propagate uncertainty and compute global sensitivity indices. Sobol indices are derived to quantify the relative importance of each variable. Results highlight the dominant role of α₁ and s₁ in determining the location of the separation and reattachment points on the suction side of the airfoil, whereas inflow-related variables such as turbulence intensity and velocity have a secondary influence. As expected, the lift coefficient C_L is primarily driven by the angle of attack.