Consider the turbulent flow around a truck. Say that the truck is 10m long, 2m wide and 2m
high. You have to determine the drug force acting on this truck as it moves with a speed of
100 km/h
a. You will make an experimental measurement with a 1/10 model of the truck.
i. What should be the minimum size of the wind tunnel you are using?
ii. What is the wind speed at the inlet of the wind tunnel?
iii. What measurements would you take (pressure, velocity, etc.? )
iv. How do you post-process the recorded measurements to calculate the drug
force on the actual truck.
b. You will make a simulation to calculate the drug force.
i. Would you use the actual 3D model of the truck, or would you simplify the
geometry? Why?
ii. What should be the minimum size of the flow domain?
iii. What are the boundary conditions set?
iv. Would you prefer RANS, unsteady RANS, LES or DNS? Why?
v. Assume you have to use RANS, which turbulence model would you use, and
why?
vi. What should the first wall distance (∆????????1) be in your mesh? What is the
corresponding ???????? +? Do you use a wall model?
vii. Which data do you extract from the results of the simulation to calculate the
total drag force?
c. Estimate the time and money you would need to conduct
i. Experiment
ii. Simulation
iii. A back-of-the-envelope type estimation based on analytical methods.
d. How would these methods compare in their accuracy?