GARTEUR (AD) AG26: NAVIER-STOKES COMPUTATIONS OF 3D TRANSONIC FLOW FOR A WING/FUSELAGE CONFIGURATION

摘要

In this paper the work carried out within the GARTEUR Action Group (AD) AG26 on the validation of computational fluid dynamics (CFD) methods for transonic flow around transport type of aircraft is presented. The participants of AG26 are Aerospatiale (AS), BAE SYSTEMS (BAe), EADS (DASA), NLR, ONERA and Saab. They simulated the flow around the AS28G wing/body configuration of Aerospatiale for a transonic design and an transonic off-design condition with their CFD methods based on the Reynolds averaged Navier-Stokes equations (Refs 3, 4, 8, 10-13) using a common computational grid that made available by BAe. Various turbulence models have been employed (Refs 2, 7, 9 and 14). A code-to-code comparisons has been made on the basis of computed aerodynamic force and moment coefficients, pressure distributions, skin friction coefficients, and boundary layer profiles. The pressure distributions have been compared with the experimental data obtained in the ONERA S1-Modane wind tunnel (Ref 5). Agreement between the computational and experimental pressure distributions is generally good except for the shock position end the region close to the wing trailing edge. There is a significant variation in computed aerodynamic force coefficients (attributed to variation in predicted shock position and separation locations) which could not systematically be related to the numerical method and turbulence models employed for this study. Even though it was expected that the off-design would be a more difficult case to predict, the variation in the computational results is more or less the same for both test cases. It is concluded among other things that the common computational grid did not allow to distinguish between discretisation error and modelling errors and it is recommended for future CFD validation studies to adopt a grid convergence approach rather than a common grid strategy.