About the Workshop

This one-day workshop aims to bridge mathematical theory, numerical analysis, and practical computation by bringing together foundational concepts and hands-on tools. We focus on the derivation of gradient flow formulations and the incompressible Navier-Stokes equations and subsequent time discretisation techniques, providing participants with a unified perspective on modern PDE modeling and simulation. Complementing the theoretical sessions, practical tutorials on Git and NGSolve equip participants with essential skills for reproducible, collaborative, and efficient scientific computing. The workshop is designed to foster understanding, cross-disciplinary dialogue, and immediate applicability in research and teaching.

Date & Location

  • Date: 21-22 September 2026
  • Location: Universität Regensburg, Germany

Schedule

Monday

Time Topic Speaker
13:15–14:45 Lecture: Gradient flows Brunk
15:15–16:45 Lecture: Navier–Stokes Brunk

Tuesday

Time Topic Speaker
08:15–10:00 Lecture: Zeitdiskretisierung Brunk
10:30–12:00 HandsOn: Git + Tutorial Höhn
13:00–15:00 HandsOn: NGSolve + Tutorial Höhn

Prerequisites

Participants are asked to complete the installations below before the workshop.

1. Git (Version Control)

Please install either Git (command line) or GitHub Desktop (GUI).

Windows

  • Install Git for Windows https://git-scm.com/download/win
  • OR GitHub Desktop https://desktop.github.com/

macOS

  • Install Git (if not already present): brew install git
  • OR GitHub Desktop https://desktop.github.com/

Ubuntu / Linux

  • sudo apt update
  • sudo apt install gi
  • git –version

2. NGSolve (Numerical PDE Library)

The hands-on session uses NGSolve installed via Python.

Requirements

  • Python 3.9 or newer
  • Internet connection

Virtual Environment

  • sudo apt install python3 python3-pip python3-venv

Win

  • python -m venv ngsolve-env
  • ngsolve-env\Scripts\activate

macOS/Linux

  • python3 -m venv ngsolve-env
  • source ngsolve-env/bin/activate

Installation

  • python -m pip install –upgrade pip
  • python -m pip install numpy scipy matplotlib jupyter
  • python -m pip install –upgrade ngsolve webgui_jupyter_widgets

Contact

Email: abrunk@uni-mainz.de