This Jupyter notebook demonstrates how machine-learning metamodels can be used to solve fluid mechanical design optimisation problems. Here, design optimisation refers to the process of finding the best solution to an engineering problem within constraints using mathematical techniques. Metamodels allow us to explore parameter space to find optimal solutions to Partial Differential Equations (PDEs) while only running a relatively small number of costly Computational Fluid Dynamics (CFD) simulations. This notebook contains an example application to coronary bypass anastomosis, a surgical procedure used to treat coronary artery disease (Lassila et al., 2013).
Binder and Colab buttons
Will launch this tutorial in binder or Google Colab.
Running locally
If you're already familiar with Git, Anaconda and virtual environments, the environment you need to create is found in DO.yml and the code below will install, activate and launch the notebook. The .yml file has been tested on the latest version of Ubuntu Linux. Windows and macOS are not officially supported, although Windows users can alternatively use the Windows Subsystem for Linux.
git clone [email protected]:cemac/LIFD_DesignOptimisation.git
cd LIFD_DesignOptimisation
conda env create -f DO.yml
conda activate DO
jupyter-notebookThis notebook is designed to run on a laptop with no special hardware required. Even on modest hardware, the CFD simulations should only take a few minutes to run. It is recommended to do a local installation as outlined in the repository howtorun and jupyter_notebooks sections. Otherwise, online compute platforms (e.g. Google Colab) are also supported.
LIFD_ENV_ML_NOTEBOOKS by CEMAC are licenced under a Creative Commons Attribution 4.0 International License.
Thanks to Toni Lassila for providing code and material for this notebook. This tutorial is part of the LIFD_ENV_ML_NOTEBOOKS series. Please refer to the parent repository for full acknowledgements.