The industry has started to introduce time-dependent
numerical simulation such as large eddy simulation (LES),
unsteady RANS or hybrid approaches such as DES in their
development process. The advantage of these techniques is
much higher accuracy of the prediction compared with steady
RANS simulations but also possibility to replace costly
physical testing and simulate flow situations where
experiments are difficult or impossible. Although most
commercial (and open source) CFD software have option to run
unsteady methods, their usage requires knowledge in running
unsteady simulations which puts different requirements than
steady RANS on all steps of the process. Being resolving
techniques, these techniques put higher requirements on all
steps in the project, from the quality of the computational
grids, numerical discretization, performing the CFD
simulation to analysis of large amount of data obtained from
a time-dependent simulation.

This aim of this course is to equip an engineer or
researcher with the knowledge to run unsteady CFD
simulations for the application of his interest. After the
course, the participant will be able to:

- choose the adequate unsteady technique

- prepare computational grid for that is required for the
chosen technique

- chose appropriate spatial and temporal discretization and
other parameters in the simulation

- run the simulation and do appropriate monitoring of the
results

- postprocessor the results and evaluate the quality of the
achieved prediction

We will start with introduction in how time-dependent
techniques such as Large-Eddy Simulations (LES),
Detached-Eddy Simulations (DES), Partially-Averaged
Navies-Stokes (PANS) and unsteady RANS (URANS) can be used
in vehicle aerodynamics. We will discuss differences between
these techniques and give best practical guidelines for
usage of each technique. We will show when each of the
techniques could/should be used and how to use it.

We believe that the course is useful for engineers and
researcher in all industrial sectors where unsteady flows
are of importance. The practical example that we will study
during the course come from vehicle aerodynamics and bluff
body aerodynamics but the method is by no means limited to
these two fields. The same methodology can be used in many
other industrial fields where separated and unsteady flows
occur.

The course will cover three important applications of CFD
techniques in industry: prediction using unsteady CFD
simulations (LES, DES, PANS, URANS), CFD for flow control
and aerodynamic shape optimization.

PRELIMINARY PROGRAM

DAY 1 (9.00 -- 17.00)

09.00-12.00  Introductions to time-dependent numerical
techniques for vehicle aerodynamics LES, URANS, DES, PANS,
numerical aspects, modeling, grids etc.
12.00-13.00  Lunch
13:15-17:00  Workshop: Implementation of LES for external
vehicle aerodynamics

DAY 2 (9.00 -- 17.00)
09.00-10.00 CFD for flow control.
10.00-12.00 Workshop: Application of LES and PANS for active
flow control.
12.00-13.00 Lunch 13.15-15.00 Active flow control, passive
flow control.
15.15-17.00 Workshop: Application of LES and PANS for active
flow control. 18:30 Course dinner

DAY 3 (9.00 -- 17.00)

09.00-10.00 Aerodynamic shape optimization of vehicles using
CFD.
10.00-12.00 Workshop: Optimization of vehicle shapes using
RANS and URANS, response surfaces, neural networks and
genetic algorithms.
12.00-13.00 Lunch
13.15-15.00 Optimization using surrogate-based modeling,
automatic shape optimization.
15.15-17.00 Workshop: Optimization of vehicle shapes using
RANS and URANS, response surfaces, neural networks and
genetic algorithms.

Computer Workshops

The course consists of one theoretical and one practical
part. The practical part contains three computer exercises.
The flow cases used in our exercises are taken directly from
our research which makes the exercises realistic and thus
more relevant for participants than pure academic simplified
cases.

For the description of the computer exercises please see:

http://sinisa-aerodynamics.se/courses-training-2/computer-exercises/