Computational Investigation of the Airfoil Self-Noise in a Low Reynolds Number Flow

Abstract

The present study explores the numerical simulation of airfoil self-noise using Computational Fluid Dynamics. Direct Noise Computations (DNC) are conducted for a NACA0012 airfoil set at near-zero incidences (𝜶 = 0˚, 𝜶 = 0.5˚) within a low Reynolds number flow (Re = 6.6×10 4 ). The DNC rely on 2D, compressible, wall-resolved Large Eddy Simulations, which allow capturing both the unsteady aerodynamic flow around the airfoil and the subsequent acoustic emission at its trailing edge. The numerical results are favorably compared against experimental data involving a similar configuration (airfoil type, Reynolds number, flow incidence). Results reveal that the primary self-noise tone and its upper harmonics are accurately predicted, for both angles of attack. Some harmonic tones are found to emerge only when the incidence is not nil, being "silenced" when the angle of attack is zero. Special focus is put on the tonal noise generation mechanisms to occur in each flow condition. It is inferred that two distinct mechanisms co-exist, namely, a vortex-shedding and a convective boundary layer instability. The latter is believed to relate to a classical feedback loop phenomenon, whose approximate model is found to match the current results.