Three-Dimensional Navier-Stokes Simulations for Transport Aircraft High-Lift Configurations

Computations of lift and drag polars for a transport aircraft wing/fuselage high-lift cone guration using the MEGAFLOWcodesystemarecarriedoutandcomparedtowind-tunnelexperiments.Themainemphasisislaidon acomparisonoftheblock-structuredandtheunstructuredcodemodulesforsuchtypeofapplication.FortheblockstructuredFLOWercodeincombinationwith a k‐! turbulencemodel,thenumericalresultsarein good agreement with the available experimental data in the linear CL range. Beyond 15-deg incidence, a strong separation near the e ap cut-out is simulated, leading to an underprediction of total lift near CL; max compared to the experimental data. In contrast to this, the results of the unstructured TAU code utilizing the Spalart ‐Allmaras turbulence model are characterized by a nearly constant lift overestimation up to maximum lift without the aforementioned separation tendency at moderate incidences. The lift overprediction in the unstructured results is attributed to the main wing and the slat upperside suction peaks, which are higher resolved by the unstructured grid. Neither code reproduces the lift breakdown beyond CL; max according to the experiments. The use of preconditioning in conjunction with theFLOWercodeshowsonly minorimprovement of theaccuracy,but considerabledeterioration of the convergence properties, requiring improvements for routine use. Further studies will focus on the ine uence of geometry simplie cations at the wing root in the theoretical models and its impact on the experimental evidence.