ANALYSIS OF VARIOUS TURBULENCE MODELS FOR NACA 23012 AIRFOIL

Abstract

This paper compares Reynolds Averaged Navier-Stokes (RANS) turbulence models and evaluates the NACA 23012 airfoil based on these test situations. ANSYS-Fluent version 22.1 was used to compare turbulence models for high-fidelity CFD simulations. A turbulence model should be used to anticipate turbulence's effect on flow. The turbulence model is a set of constitutive equations used to close the flow-governing Navier-Stokes equations. Most engineering turbulence models are based on Boussinesq hypothesis (Spalart – Allmaras, k- Standard, k- SST, k- Standard, k- Realizable and k- RNG) because it gives a low-cost calculation for solving turbulence viscosity. In this work, three representative turbulence models are used: Spalart-Allmaras, k- SST, and k- Standard. The flow is investigated at various angles of attack (AoA), from -2° to 18°. These AoA correlate to Re = 3×106 and Mach number M∞ = 0.13. Simulations are steady-state and incompressible. Flow velocity, pressure, and density are not time dependent. Comparing these CFD scenarios to Abbott's [2] experimental data reveals that these turbulence models offer close results with a 5% error margin for low and medium AoA.