Bryne is a python framework to build coupled multiphysics models. It also facilitates the creation of reproducible simulation setups by storing the full human-readable simulation setups with the results.
The current implementation heavily leverages the DUNE-FEM python API to set up models as FEM problems and solve them using JIT generated C++ code.
Installation
There are several ways to run bryne, see the documentation pages for ways to get started.
- for reproduction, we recommend the bryne docker images
- for developers, we provide a VSCode development container
License
The bryne code is free open-source software, licensed under version 3 or later of the GNU General Public License. See the file LICENSE for full copying permissions.
Reference
Please cite Bryne as (see BibTex entry at end of Readme):
Terschanski, B., Klöfkorn, R., Dedner, A., & Kowalski, J. (2025). Bryne: sustainable prototyping of finite element models - Software release (1.0.0). Zenodo. https://doi.org/10.5281/zenodo.15789249
Publications
Models that are part of bryne have been described in the following publications:
-
Terschanski et al. (2025), "Stable across regimes: A mixed DG method for Darcy–Brinkman–Stokes type flows" is implemented in
models.stokes_brinkman_dg_model
Acknowledgements
Funding
Development of bryne was funded by the German Research Foundation (DFG) - grant number 460819306 - from December 2021 through April 2025.
Contributor credits
Robert Klöfkorn: continuous assistance with the dune-fem backend. Also contributed a Newton-solver template for petsc4py.
Andreas Dedner: continuous assistance with the dune-fem backend.
Mohammad Ghanem: supported development as a student research assistant from 2023 to 2024. He helped in testing the initial framework and developing a first testing infrastructure, including test automation with Gitlab CI/CD.
BibTex Citation
@software{terschanski_2025_15789249,
author = {Terschanski, Benjamin and
Klöfkorn, Robert and
Dedner, Andreas and
Kowalski, Julia},
title = {Bryne: sustainable prototyping of finite element
models - Software release
},
month = jul,
year = 2025,
publisher = {Zenodo},
version = {1.0.0},
doi = {10.5281/zenodo.15789249},
url = {https://doi.org/10.5281/zenodo.15789249},
}