About

This repository contains the software supporting the paper

Lestang, T., Bouchet, F., & Lévêque, E. (2020). Numerical study of extreme mechanical force exerted by a turbulent flow on a bluff body by direct and rare-event sampling techniques. Journal of Fluid Mechanics, 895, A19. doi:10.1017/jfm.2020.293 arxiv:2005.09323

The repository is organised as follow

figures/
	# Contains scripts for derived data and figures
	fig1/
		fig1.py
	fig2/
		fig2.py
	...
jfm_paper/
	# python package for generating derived data and figures
texfiles/
	# LaTeX sources

Many scripts currently rely on raw data for which the generating code is not yet made available, but will added to this repository soon. If you have any issues or queries, please create a new issue on the this repository's issue tracker.

Installing the python package

Requirements

• Python 3.6 and above
• A C++ compiler (ex: g++)
• The Python header files

On Debian-based GNU/Linux distribution (e.g. Ubuntu, Linux Mint), the above can be installed with

$ apt install python3 python3-dev gcc

On MacOS, using homebrew,

$ brew install python3 gcc

Installation

To install the python package, first clone this repository

$ git clone https://github.com/tlestang/paper_extreme_drag_fluctuations.git

We highly recommend installing the package inside a python virtual environment, so as to keeps it and its dependencies separate from your system's python.

# In directory paper_extreme_drag_fluctuations
$ python3 -m virtualenv venv
$ source venv/bin/activate

the move into the directory and install with pip

# In directory paper_extreme_drag_fluctuations
(venv) $ pip install .

Depending on you platform, the python installation may or may not come with the virtualenv and pip. In most cases, these can be installed via

apt install python3-pip python3-venv

Fluid flow data generation

All flow simulation rely on a in-house C++ library implementing the Lattice Boltzmann Method. It is available here.

GKTL algorithm implementation

AMS algorithm implementation