Uncertainty Quantification and Sensitivity Analysis of Spalart Allmaras and Menter Shear Stress Model Coefficients for NACA 0012 Airfoil

This study quantifies epistemic uncertainty associated with the coefficients of two closure models used in Reynolds-Averaged Navier–Stokes simulations. The incompressible flow over a NACA0012 airfoil at angles of attack ranging from 2° to 17° is simulated using the Spalart–Allmaras and Menter’s Shear Stress Transport (SST) turbulence models available in OpenFOAM. For each model, the three most influential coefficients are treated as uncertain parameters. The quantities of interest are the lift and drag coefficients. Uncertainty propagation is performed using non-intrusive polynomial chaos expansion (NIPCE), and global sensitivity analysis is conducted using Sobol indices. The results show that uncertainty increases with the angle of attack, exhibiting a sharp growth near and after stall. Overall, the Menter’s SST model displays higher uncertainty than the Spalart–Allmaras model, and the drag coefficient is more sensitive to model-based uncertainty than the lift coefficient. In the Spalart–Allmaras model, the von Kármán coefficient and �� are the most influential parameter before and post stall, respectively. In contrast, for the Menter’s SST model, the ranking of the most influential parameters on the resulting uncertainty varies with angle of attack.