A new certified hierarchical and adaptive RB-ML-ROM surrogate model for parametrized PDEs

This repository contains the supplementary material for the publication

A new certified hierarchical and adaptive RB-ML-ROM surrogate model for parametrized PDEs (2022)
B. Haasdonk, H. Kleikamp, M. Ohlberger, F. Schindler, T. Wenzel
SIAM Journal on Scientific Computing, Volume 45(3), A1039 - A1065, 2023
https://doi.org/10.1137/22M1493318

It was used to carry out the numerical experiments and generate the figures for the publication. The experiments were performed on Linux systems in 2022 and should work with Python versions of that time (e.g., python3.9). The steps below each indicate a setup to be executed in a terminal, resulting in a URL printed to be opened in a browser to view and run the indicated notebooks.

to reproduce the figures

The data collected during the experiments is contained within this repository (in the notebooks/KLAIM21__RBMLROM_* and notebooks/MC_adaptive_model folders), and generating the figures is straightforward. To setup an evironment for regenerating the figures, execute:

export BASEDIR=$PWD
virtualenv -p python3.9 venv-figures  # replace python3.9 with something suitable
. venv-figures/bin/activate
pip install -r requirements-figures.txt
cd $BASEDIR/pymor && pip install -e .
cd $BASEDIR && ./start_notebook_server.sh

After opening the printed URL (the one with 127....) in a browser, run

• for Figure 2: paper_section_4_1_reactive_flow_minimization_VKOGA_create_figures.md
• for Figure 3: paper_section_4_1_reactive_flow_adaptive_minimization_VKOGA_create_figures.md
• for Figure 5: paper_section_4_2_monte_carlo_DNN_create_figures.md

to reproduce the experiments

Reproducing the experiments is more involved and will not yield data coinciding with the original results (due to different hardware/software setup). To setup an evironment for reproducing the experiments, execute:

export BASEDIR=$PWD
virtualenv -p python3.9 venv-experiments  # replace python3.9 with something suitable
. venv-experiments/bin/activate
pip install -r requirements-experiments.txt
cd $BASEDIR/pymor && pip install -e .
cd $BASEDIR && ./start_notebook_server.sh

After opening the printed URL (the one with 127....) in a browser, run

• for Figure 2: paper_section_4_1_reactive_flow_minimization_VKOGA_run_experiment.md
• for Figure 3: paper_section_4_1_reactive_flow_adaptive_minimization_VKOGA_run_experiment.md
• for Figure 5: paper_section_4_2_monte_carlo_DNN_run_experiment.md