Additional teaching videos for PhD students and other interested parties
The Horizon 2020 project Conflex provides a number of interesting videos on various topics also of interest to ModConFlex.
Teaching - First Network Meeting (2023):
Electric Grid Influences on Floating Wind Turbines - Part 1
Electric Grid Influences on Floating Wind Turbines - Part 2
Modelling, and Model Predictive Control, for Flexible Structures and Fluids - Part 1
Modelling, and Model Predictive Control, for Flexible Structures and Fluids - Part 2
Modelling, and Model Predictive Control, for Flexible Structures and Fluids - Part 3
Virtual Synchronous Machines - Part 1
Virtual Synchronous Machines - Part 2
Virtual Synchronous Machines - Part 3
Virtual Synchronous Machines - Part 4
Slides
PDE modelling of large floating structures - Part 1
Teaching - Second Network Meeting (2024):
PDE modelling of large floating structures - Part 2
PDE modelling of large floating structures - Part 3
PDE modelling of large floating structures - final part
Slides
Enhancing machine learning through control and numerics - Part 1
Enhancing machine learning through control and numerics - Part 2
Enhancing machine learning through control and numerics - Part 3
Enhancing machine learning through control and numerics - Part 4
Enhancing machine learning through control and numerics - Part 5
Enhancing machine learning through control and numerics - Part 6
Teaching - Third Network Meeting (2025):
Port-Hamiltonian Systems from Analysis to Numerics
Course taught in October 2025 by Hans Zwart (Universiteit Twente) and Denis Matignon (Institut Supérieur de l'Aéronautique et de l'Espace, Toulouse):
In modelling of (physical) systems we often encounter conserved quantities, like mass, charge, energy, etc. Since these quantities are independent of the models chosen for the subparts, it indicates that this could be an underlying structure in which this conservation is encoded.
In this series of lectures we discuss this structure, i.e., the Dirac structure. We show that this structure reveals already many properties that the model will have, independently of whether is discrete or continuous time model, independently of if it is an ordinary or partial differential equation, and whether it is a discretised model. However, as is shown by examples, the structure does not automatically give you well-posedness, i.e., existence of solutions.
The course aims at anyone with a basiv mathematics, physics, and or engineering background, who wants to learn more about the joint underlying properties of models.
The knowledge of the underlying structure has shown to be very helpful in designed numerical discretisation which preserves quantities like energy, and on the other hand has simplified well-posedness problem of partial differential equations. Results on these two topics are still a very active field of research.
Slides and Board Photos
Port-Hamiltonian Systems from Analysis to Numerics 1
Port-Hamiltonian Systems from Analysis to Numerics 2
Port-Hamiltonian Systems from Analysis to Numerics 3
Port-Hamiltonian Systems from Analysis to Numerics
Boundary Controlled Port-Hamiltonian Systems - from reversible to irreversible thermodynamic formulations
Course taught in October by Yann Le Gorrec (Université Marie et Louis Pasteur, Besançon) and Hector Ramirez (Universidad Técnica Federico Santa Maria)
