Machine Learning in Fluid Mechanics

Lecture notes and exercises from the Institute of Fluid Mechanics at TU Dresden.

The course was established in 2020 at TU Braunschweig by Dr. Andre Weiner, drawing on his research in machine learning for fluid mechanics. It has been taught at TU Dresden since 2023. Its development previously received support through TU Braunschweig’s ProDiGI funding scheme, funded by the Foundation for Innovation in Higher Education.

The course continues to grow as new topics are added and existing materials are revised and kept up to date. Its content is closely connected to current research at the Institute of Fluid Mechanics.

Lectures and exercises

#	Topic	Lecture	Exercise	Slides
0	How to learn with generative AI	Read	—	—
1	Course overview and motivation	Read	Exercises	Slides
2	Introduction to machine learning	Read	Exercises	Slides
3	Surrogate models for discrete predictions
4	Surrogate models for continuous predictions
5	Analysis of coherent structures
6	Reduced-order modeling
7	Bayesian optimization and exploration
8	Data assimilation
9	Active flow control

Optional material

Use these readings to refresh background knowledge as needed. They are separate from the numbered lectures and exercises.

Topic	Material
Getting started with Python	Read
Training neural networks	Read