Session: 05-07-02: CFD Development and V&V

Paper Number: 132960

132960 - On the Challenges of Validation Exercises of Two-Dimensional Models 

Modeling and Simulation (MS) has become an engineering tool used on a daily basis. Therefore, the assessment of the modeling accuracy of MS, i.e. validation, is essential to establish its credibility. In Computational Fluid Dynamics (CFD), assuming two-dimensional flow, as for example in airfoils, enables the reduction of numerical uncertainties to a negligible level using modest hardware resources. Therefore, two-dimensional validation studies have been widely performed in CFD. However, the evaluation of modeling errors requires measurements of the physical reality that cannot correspond exactly to a two-dimensional situation. Furthermore, boundary conditions required by the two-dimensional simulations are not always available from the experiment that provides the comparison data. In many validation exercises, the difficulties with the specification of the boundary conditions lead to changes in the domain geometry and simplifed assumptions in the specification of the boundary conditions.

In this paper, we present a validation study for boundary-layers developing in non-equilibrium boundary-layers simulated with the RANS equations. Experimental data has been measured in the Virginia Tech wind-tunnel that has a rectangular geometry for the center plane section. In the simulations, the flow is assumed to be two-dimensional and so a modified domain geometry with a tilted top wall to take into account the effect of the boundary-layers of the lateral walls of the wind-tunnel has been proposed to the validation exercise. Furthermore, no slip conditions are suggested for the tilted top wall. In this study, we compare the results obtained in the proposed domain and boundary conditions with simulations performed using the geometry of the wind-tunnel center plane and the no-slip condition applied at the two wind-tunnel walls. Furthermore, we investigate the influence of the post-processing required to obtain the quantities of interest in the comparison of experimental and simulations data.

Presenting Author: Luis Eca IST