Python Reservoir Simulation Toolbox

The Python Reservoir Simulation Toolbox is a Python version of the MATLAB Reservoir Simulation Toolbox. The goal is to clone the functionality found in MRST, and add tests.

MRST-2015b features available in PRST

plotting

• plotGrid - partial support by exporting to ParaView
• plotCellData - partial support by exporting to ParaView

gridprocessing

• grid_structure (now a class Grid)
• tensorGrid
• cartGrid
• computeGeometry
• makeLayeredGrid

incomp

• fluid_structure (now a class F

utils

• rlencode
• rldecode

params

• rock_struture (now a class Rock)
• wells_and_bc boundary conditions (now a class BoundaryCondition)

Installation

Currently there are no Python or Conda packages for this project. Download the whole repository and copy the "PRST/pyrst" folder into your project. Any scientific Python3 distribution such as Anaconda will satisfy the requirements.

An experimental package is available on PyPI. It is recommended to install this within the full Anaconda2 distribution.

pip install prst

The dependencies are not fully defined, and may be missing. Fix this using conda or pip.

pip install numpy_groupies

To run the tests, additional dependencies are needed:

conda install numpy scipy six pytest pip
conda install -c https://conda.anaconda.org/bokeh pytest-cov
py.test

Linux installation (using Miniconda/Anaconda)

The installation procedure used for automatic deployments (Travis CS) is the following:

# Download PRST
git clone https://github.com/roessland/PRST.git

# Download and update Miniconda for Python 2
wget http://repo.continuum.io/miniconda/Miniconda-latest-Linux-x86_64.sh -O miniconda.sh
chmod +x miniconda.sh
./miniconda.sh -b -f $HOME/miniconda2
export PATH=$HOME/miniconda2/bin:$PATH
conda update conda

# Install Conda packages necessary
conda install python=2.7 numpy scipy six pytest pip
conda install -c https://conda.anaconda.org/bokeh pytest-cov

# Install pip packages necessary (using conda's pip!)
pip install numpy_groupies

# Run tests to see that everything is working correctly
cd PRST
py.test

Linux installation (using Pip only)

To setup the development environment, and run tests, do the following:

git clone https://github.com/roessland/PRST.git
cd PRST
virtualenv --python=python3 venv3
source venv3/bin/activate
pip install -r requirements.txt
py.test

To resume working after the initial steps have been completed:

cd PRST
source venv3/bin/activate
py.test

SciPy needs to be compiled when installed using pip, and this can take up to 30 minutes. Use conda to avoid this.

Windows installation

Use Anaconda2 or Miniconda2.

Troubleshooting installation

• py.test executable does not exist: Remember to add the Miniconda environment to your $PATH, as done in the steps above.

• py.test: error: unrecognized arguments: --cov-report=term-missing ...: This means that pytest-cov has not been installed, or that you are running the wrong py.test executable. Run type -a py.test to see where py.test is located. It should be in the miniconda2-folder if you followed the instructions above, or in a virtual environment folder if using the method below.

Code style

To remain similar to MRST, functions and variables should be camelCased. All code should be tested.

License

MRST is GPLv3-licensed, and since PRST is a derivative work, it is also GPLv3-licensed.

Publish to PyPI

These steps are only needed by the maintainer of the PyPI package.

pip install twine wheel
git tag 0.0.1 # Version to be released
python setup.py sdist bdist_wheel
twine upload dist/*