Assignment for AE4202 – CFD for Aerospace Engineers

Introduction

The assignment should be performed in groups of two students. You are free to choose your
partner. Both team members should equally contribute to the work and submit a joint
report. Though not recommended, you could also do the work alone and submit individual
reports.

The deliverable is a short report with max. 10 pages, excluding title page and list of
references. The report should be submitted through the Brightspace website before
November 13, 23:59 CET. If you submit the report late, one point per week of delay will be
subtracted from the assignment grade, which contributes 50% to your final AE4202 grade.
AE4202 - CFD for Aerospace Engineers
Problem description
Assignment to be submitted before November 14, 23:59 CET
Consider the compressible flow of air past a RAE2822 airfoil at 2.31 degrees angle of attack.
We want to reproduce a wind tunnel experiment and quantify effects of discretization and
modelling errors. The Reynolds number is approximately Re = 6 · 106 based on chord length
oblem description
and the Mach number is M = 0.729.
nsider the compressible flow of air past a NACA64A010 airfoil at an angle of attack ↵ = 4 . We
The numerical setup consists of an airfoil placed in a channel with frictionless walls, as
eproduce a wind
depictedtunnel
in Fig. experiment, where and
1. The airfoil geometry theexperimental
total temperature
values of is
theknown
pressureto be Ttotal = 293.15K
coefficient
merical setupare
consists
providedof as an airfoil placed in a and
"RAE2822_TO_ICEM.dat" channel with frictionless walls,
"RAE2822_cp_experiment.dat". Theas
airdepicted
is in Fig. 1
ynolds numberconsidered
equals toRe be =
an 12
ideal gas6with
· 10 andathe
dynamic
Machviscosity
number given
isbyMSutherland’s
= 0.789.law. TheTheairflow is
is considered to
in a steady state.
al gas with a dynamic viscosity given by Sutherland’s law. The flow is in a steady state.

Free-slip wall

MM=0.729
= 0.789

Free-slip wall

Figure 1: Sketch of the geometry.

sks
1. In ICEM, go to “File ! Geometry ! Open geometry” and select the file “naca64a010.tin” pro
with this assignment. Tick on/off Surfaces and Parts in the tree to familiarise yourself with th
main. Mesh the computational domain with a coarse structured grid with 15.000 to 20.000
(grid1).
2. Choose suitable boundary conditions. Impose the total pressure and temperature at the inlet an
static pressure at the outlet. Use the isentropic assumption to derive the appropriate values of
temperature and pressure.
Tasks

1. In ICEM, go to “File → Import Geometry → Formatted Point Data” and select the file
“RAE2822_TO_ICEM.dat” provided with this assignment. Add appropriate
boundaries to the geometry in order to define the computational domain. Verify
that the angle of attack is 2.31 degrees and adjust if necessary.

2. Mesh the computational domain with a coarse structured multi-block grid with
about 20.000 cells (grid1). Pay attention to orthogonality, smoothness and
resolution of the boundary layers; aim at y+ between 20 and 200. Visualize and
describe blocking and mesh in the report and discuss the mesh quality.

3. As boundary conditions, impose total temperature T0 = 291.15 K and total pressure

p0 = 101325 Pa at the inflow and a static pressure of p = 71150 Pa at the outflow.
This is easier if you set the CFX reference pressure to zero.
Use the ‘high-resolution’ scheme for the spatial discretisation and the k-omega SST
turbulence model. Choose a suitable fluid model and time-stepping scheme. Report
the simulation setup and discuss the reasons for your choices.

4. Perform a steady state simulation. Plot the solver residuals to verify convergence of
the simulation. Evaluate and discuss the flow conditions, i.e., Reynolds and Mach
number, with respect to the target values given above.

5. Visualize the flow field: produce contour plots for the Mach number and for a
second quantity of your choice. Plot the pressure coefficient along the airfoil
together with the experimental reference data. Extract the lift and drag forces and
summarize them in a table. Interpret your results.

6. Compute and plot the y+ value of grid1 along the airfoil. Refine the grid such that
y+ < 1 for the refined grid (grid2). Visualize and describe blocking and mesh, pay
attention to orthogonality and smoothness. Perform the simulation and plot the
solver residuals to verify convergence of the simulation. Plot y+ for grid2 to verify
that the resolution target was achieved. Critically discuss the mesh quality.

7. Perform the same post processing and visualizations as done for grid1. Include the
new Cp curve in the same figure and the lift and drag forces in the same table as the
results for grid 1 (same holds for point 8 and 9). Describe and evaluate the results.

8. In order to quantify the effect of numerical diffusion, repeat the simulations on both
grid1 and grid2 with a first-order upwind scheme. Plot the solver residuals and verify
convergence of the simulation. Repeat the same post processing and visualizations
as done before. Visualize and analyse the effect on the results.

9. In order to quantify the effect of the turbulence model, repeat the simulation on the
fine mesh (grid2) with a turbulence model that can better account for anisotropic
turbulence. Which model did you chose and why? Plot the solver residuals and verify
convergence of the simulation. Visualize and analyse the effect on the results.

10. Summarize your results, discuss them critically and suggest possible improvements.