open-flash 1.0.21

A Python package for semi-analytical hydrodynamics modeling including matched eigenfunction expansion method

OpenFLASH ⚡️

Open-source Flexible Library for Analytical and Semi-analytical Hydrodynamics

About The OpenFLASH Project

The OpenFLASH project is a Python package designed for solving hydrodynamic boundary value problems using eigenfunction expansion methods. It provides a modular framework for defining complex geometries, setting up multi-domain problems, performing numerical computations, and analyzing results, particularly for linear potential flow hydrodynamics. It can be significantly faster than Boundary Element Method calculations although is restricted to certain geometries (currently axisymmetric compound cylinders).

When referencing this work, please reference our citation.cff.

This project is licensed under the MIT License. See the LICENSE file for details.

Installation

Three common installation options are shown below. For more details, see the installation section of the docs.

Option 1: via pypi (recommended for users who manage environments with venv or similar)

pip install open-flash

Option 2: via conda (recommended for users who manage environments with conda)

conda create -n openflash-env sea-lab::open-flash
conda activate openflash-env

Option 3: via git (recommended for developers)

Note - if you are a developer outside of the SEA Lab, please create a fork and clone your fork.

1. Clone the repository:

git clone https://github.com/symbiotic-engineering/OpenFLASH.git
cd OpenFLASH

2. Install the package:

pip install -e .

3. Install dependencies:

pip install -e .[dev]

Usage

Please see our documentation and tutorial notebook. The documentation provides detailed instructions and API reference for the different modules and classes within the openflash package.

If you prefer not to utilize the package programatically, the model can also be run with an interactive web app (the site may take several seconds to load).

Theory

Please see our equations documentation and references for mathematical background and derivations as well as validation information.

MATLAB

We also have a MATLAB code version, although the Python package is intended as primary for future development. MATLAB only supports bodies consisting of 2 concentric cylinders, rather than the arbitrary N concentric cylinders in the Python package.

See matlab/src/run_MEEM.m for the symbolic and numeric code, see matlab/test/ for some scripts to get results, and matlab/dev for various matlab experiments.