Master Thesis

In this study, Generalised Richardson Extrapolation (GREP) is examined in terms of effectivity and efficiency in a Oscillating Wall Flow Control Analysis. The analysis also used standard channel flow as a comparison for the study. To analyse the effects of the oscillating wall flow control as well as the effect of turbulence, DNS is used as the simulation method. Incompact3d is a fast, yet accurate software to run DNS flow simulations.

Incompact3d Compiling

Before compiling, cmake and mpich modules must be installed into Ubuntu, using sudo apt install mpich and sudo apt install cmake. If they are unable to be downloaded, make sure Ubuntu is already up to date using sudo apt update and then sudo apt upgrade. If you are using HPC environment, loading cmake and mpich module (or openMPI in some cases) is important.

To compile and run the incompact3d (substitute ncores with the number of desired cores for compiling):

git clone https://github.com/wwdhd/master_thesis.git
cd Incompact3d_osc180
export FC=mpif90
cmake -S . -B build
cd build
cmake --build . -j {ncores}
cd ../examples/Channel/

To run the channel flow code in the local computer:

mpirun -np {ncores} ../../build/bin/xcompact3d > x3d2.log &
tail -f x3d2.log

In HPC environment, the same command is used, just ensure the job scheduler format is followed.

Py4Incompact3d Download and Compiling

Before compiling, it is important to have the Python module, downloaded using sudo apt install python3-pip. To compile and install Py4Incompact3d in Ubuntu:

git clone https://github.com/xcompact3d/Py4Incompact3D
cd Py4Incompact3D
pip install .
pip install scipy
pip install matplotlib

To run the code in the examples, or the one given in this repository, just type:

python3 {name-of-the-code}

Graphics display software also needs to be downloaded. Personally, I use "feh", installed using sudo apt install feh.