Modeling Subscale Rotor Wake in Ground Effect with Accurate Turbulent Length Scales

Abstract

An accurate prediction of the tip vortex flowfield is required to understand the viscous interactions of the rotor wake and ground prevalent during helicopter brownout. To achieve this goal, a compressible, structured, overset Reynolds-averaged Navier–Stokes-based solver is used to simulate a subscale rotor in ground effect. Novel mesh-gridding techniques are used to avoid prohibitive computational costs. The Reynolds-averaged Navier–Stokes equations solved with the Spalart–Allmaras turbulence model show that a smeared-out flowfield at an early wake ages due to excessive turbulence levels. To further investigate that the source of a smeared rotor wake is the use of turbulence modeling, the computations are performed with laminar flow assumption that show a preserved rotor wake field for longer wake ages. However, an accurate understanding of the cause of brownout requires an accurate prediction of turbulence levels. Therefore, a hybrid Reynolds-averaged Navier–Stokes/large-eddy simulation methodology such as delayed detached-eddy simulation is explored. The flowfield obtained from the delayed detached-eddy simulation methodology still does not preserve the rotor wake. Because the wake capturing grids are highly anisotropic, a modified length scaling is used. The delayed detached-eddy simulation-based length scale with correction for anisotropic grids helps in preserving the rotor wake and shows a close agreement with experimental results.