Hybrid low-fidelity and scale-resolving simulations on the CRM high-lift aircraft configuration

What is it about?

Recent studies on the high-lift configuration of NASA’s Common Research Model (CRM-HL) showed a wide spread of results in the attempt to predict low-speed stall. The scatter is attributed to several choices regarding numerical setup and methodology, which lead to significant sensitivity to turbulence modeling, initialization strategies and wind-tunnel wall confinement effects. In the present study we perform calculations using a hybrid Reynolds-Averaged Navier-Stokes (RANS) and Large Eddy Simulation (LES) technique to determine the effects of the numerical approaches selected on the prediction of low-speed stall. While the aerodynamic forces near stall are well predicted with a 5% error with respect to the experiments, a better characterization of the experimental inflow is required to improve the prediction of separation on the main wing of the half-span model when including the wind-tunnel walls in the simulation. The initialization strategy has a significant impact on the evolution of the flow dynamics that can lead to very long transients before statistics can be collected. Difficulties on controlling the Mach number in the test section have also been encountered and some possible remedies proposed. While further investigations are needed, this study contributes to the creation of best-practices for the certification of aircraft using numerical simulations.

Featured Image

Why is it important?

The prediction of low-speed stall via numerical simulations is a difficult challenge towards the realization of the aircraft life-cycle digital transformation and certification by analysis. In this paper, we looked at different aspects of the simulation method and setup that can affect the results. Knowing these sensitivities is of fundamental importance to reliably predict low-speed stall and quantify uncertainties.

Perspectives