Repository for the novel CFD approach called AI4CFD and no traniing of the network is needed. More detailed implementation and possible applications can be found in as follow:
- "Using AI libraries for incompressible Computational Fluid Dynamics" https://www.imperial.ac.uk/people/boyang.chen16/document/10322/AI_for_CFD/?AI_for_CFD.pdf
- "Using Graph Neural Networks for Incompressible CFD on Unstructured Meshes" \
- "Solving the Dsicretised Diffusion Equation in Neutron Transport using Neural Networks" https://www.imperial.ac.uk/people/boyang.chen16/document/10422/A_neural_network_solver_for_Neutron_Transport/?A_neural_network_solver_for_Neutron_Transport.pdf-
- "Solving the Dsicrete Ordinate Equations in Neutron Transport using Neural Networks"
https://drive.google.com/file/d/1dm4cXBBX-fcFS33Zi7IIFjTU7qywDKWK/view?usp=sharing
To install requirements:
$ conda env create -f environment.yml
$ conda activate ten
$ python -m ipykernel install --user --name=python3 (optional)
Finally, start Jupyter with a local machine:
$ jupyter notebook
Or, start Jupyter with a remote server:
$ jupyter notebook --no-browser --port=<number>