Standard Unstructured Grid Solutions for Cranked Arrow Wing Aerodynamics Project International F-16XL

Abstract

Steady and unsteady viscous flow simulations of a full-scale, semispan, and full-span model of the F-16XL-1 aircraft are performed with three different computational fluid dynamics codes using a common unstructured grid. Six different flight conditions are considered. They represent Reynolds and Mach number combinations at subsonic speeds, with and without sideslip. The steady computations of the flow at these flight conditions are made with several Reynolds-averaged Navier-Stokes turbulence models of different complexity. Detached-eddy simulation, delayed detached-eddy simulation, and an algebraic hybrid Reynolds-averaged Navier-Stokes/large-eddy simulation model are used to quantify unsteady effects at the same flight conditions. The computed results are compared with flight-test data in the form of surface pressures, skin friction, and boundary-layer velocity profiles. The focus of the comparison is on turbulence modeling effects and effects of unsteadiness. The overall agreement with flight data is good, with no clear trend as to which physical modeling approach is superior for this class of flow. The Reynolds-averaged Navier-Stokes turbulence models perform well in predicting the flow in an average sense. However, some of the flow conditions involve locally unsteady flow over the aircraft, which are held responsible for the scatter between the different turbulence modeling approaches. The detached-eddy simulations are able to quantify the unsteady effects, although they are not consistently better than the Reynolds-averaged Navier-Stokes turbulence models in predicting the flow in an average sense in these flow regions. Detached-eddy simulation fails to predict boundary-layer profiles consistently over a range of flow regimes, with delayed detached-eddy simulation and hybrid Reynolds-averaged Navier-Stokes/large-eddy simulation models offering a remedy to recover some of the predictive capabilities of the underlying Reynolds-averaged Navier-Stokes turbulence model. Nonetheless, the confidence in the predictive capabilities of the computational fluid dynamics codes with regard to complex vortical flowfields around high-performance aircraft of this planform increased significantly during this study.