CUED Finite Element Methods (3D7) Lab

This is the laboratory coursework activity for the Finite Element Course (3D7). The activity is contained in a Jupyter notebook and uses the open source FEniCSx libraries.

Running the coursework notebook in Colab is recommended. Google does very occasionally make changes to Colab that can break the required libraries. In this case the lab can be undertaken using Codespaces or run locally.

Colab (recommended)

At the start of each session it will take a few minutes for the required libraries to be installed.

Cambridge users should use their University credentials (CRSid) to log in.

Be sure to use "Copy to Drive" to keep a copy of your notebook.

Codespaces

You can return to your Codespace at https://github.com/codespaces. Codespaces core hours per month are limited, so it is helpful to stop running Codespaces at https://github.com/codespaces when you have finished working. You can restart the Codespace at any time.

Locally

VS Code and Docker

1. Open VS Code.
2. Clone the repository garth-wells/CUED-3D7-Lab from GitHUb using from VS Code.
3. Follow the VS Code prompts to install 'Dev Containers'.
4. Select the VS Code prompt 'Reopen in Container'.

You will be prompted to install Docker.

conda (macOS and Linux only)

The DOLFINx library used in this lab can be installed using conda.

conda create -n fenicsx-env
conda activate fenicsx-env
conda install -c conda-forge fenics-dolfinx=0.7.3 jupyterlab matplotlib mpich python-gmsh pyvista

Author

These notebooks are developed by Garth N. Wells (gnw20@cam.ac.uk).