Semi-Empirical Modeling of Turbulent Anisotropy for Airfoil Self-Noise Predictions

摘要

In the present paper an approach to derive anisotropic turbulence properties from standard Reynolds-averaged Navier-Stokes simulations is proposed. The theory includes models for anisotropic velocity spectra and scaling functions that can be used to derive other turbulence parameters required for the turbulent boundary-layer trailing-edge interaction noise prediction. This type of noise is supposed to be affected by the inhomogeneous and anisotropic flow character near the trailing edge of an airfoil or wing. For this purpose, two-point turbulent velocity correlation measurements in the near wake of two airfoils (NACA 0012 and NACA 64_3-418) at high Reynolds numbers (Re = 1.5 × 10~6 and 2.5 × 10~6) were conducted in the Laminar Wind Tunnel of the University of Stuttgart A detailed postprocessing along with assessment of the experimental data has been performed for the modeling of different isotropic and anisotropic velocity spectra and related length scales. For this modeling, an anisotropy amplitude scaling method and Kolmogorov local-isotropy hypothesis have been applied. Furthermore, two different semi-empirical functions have been correlated to the resultant anisotropy scaling factors to model the effects of turbulence anisotropy in numerically predicted Reynolds-averaged Navier-Stokes data. The proposed spectrum modeling allows an accurate evaluation of the turbulence mean dissipation rate. The new models were validated and applied to a turbulent boundary-layer trailing-edge interaction noise prediction method (Rnoise) to analyze the impact of anisotropy on the total predicted noise spectra.