About

Hi there! My name is Andre Weiner. I am a research scientist working on computational fluid dynamics (CFD) and machine learning (ML). Besides ML and CFD, I am also interested in making research reproducible by employing containerization and version control (Docker, Singularity, Git, Github). Reproducibility also implies full transparency regarding source code and workflows, which is why my work as well as the work of students I supervise is available on Github. I am also an advocate for self-determined, lifelong learning.

That’s me in 2017.

Profile pages

If you are interested in my academic curriculum, have a look at the following profile pages:

• Google Scholar (journal articles)
• Technical University of Dresden (since Sep 2023)
• Technical University of Braunschweig (May 2020 - Jul 2023)
• Technical University of Darmstadt (Oct 2014 - Apr 2020)

Scientific software packages

• flowTorch - a Python library for analysis and reduced-order modeling of fluid flows, Github
• drlFoam - Deep Reinforcement Learning with OpenFOAM, Github

University courses

• Machine learning in computational fluid dynamics, TU Braunschweig and TU Dresden, since winter term 2021/2022; I designed this course from scratch; lecture notes, slides, exercises, and datasets are freely available
• Modeling and simulation of turbulent flows, TU Braunschweig, summer term 2023; I designed this course mostly from scratch (syllabus, lecture slides, exercises)

Presentation and training slides

• A brief introduction to Bayesian optimization, PSM seminar, Oct 2024, Dresden
• Best practice guidelines for the modal analysis of numerical and experimental data, DLRK 2024, Sep 2023, Hamburg
• An optimized dynamic mode decomposition for flow analysis and forecasting, 19th OpenFOAM workshop, June 2024, Beijing
• Machine learning in computational fluid dynamics - an overview, invited talk, automotive industry, July 2024
• Combining machine learning with computational fluid dynamics using OpenFOAM and SmartSim, 19th OpenFOAM workshop, June 2024, Beijing
• Analyse turbulenter Strömungen - wie man Ordnung im Chaos findet, Lange Nacht der Wissenschaft, June 2024, Dresden
• Modal analysis of fluid flows - best practices, ISM seminar - invited talk, May 2024, Braunschweig
• Model-based reinforcement learning for active flow control, 94th GAMM, Mar 2024, Magdeburg
• Extracting coherent structures and reduced-order models from flows, Bosch - invited talk, Jan 2024, Stuttgart
• DMD analysis of the XRF-1 wing undergoing transonic shock buffet, DLRK 2023, Sep 2023, Stuttgart
• Advances in the application of DRL to flow control, 18th OpenFOAM workshop, Jul 2023, Genoa
• Setting up a new flow control problem in drlFoam, 18th OpenFOAM workshop, Jul 2023, Genoa
• Model-based deep reinforcement learning for flow control, 22nd Computer Fluids Conference, Apr 2023, Cannes
• Data-driven modeling and validation of reactive mass transfer at rising bubbles, OpenFoam multiphase short course - invited talk, Mar 2023, Darmstadt
• ProDiGI - Machine learning in computational fluid dynamics lecture, ProDiGI event, Feb 2023, Braunschweig
• Analysis of shock buffet PSP measurements by means of dynamic mode decomposition, Airbus XRF-1 workshop, Dec 2022, virtual
• Activities of the data-driven modeling special interest group, OpenFOAM-v2206 release webinar, Jul 2022, virtual
• Modal analysis of transonic shock buffets on a NACA-0012 airfoil, 17th OpenFOAM workshop, Jul 2022, Cambridge
• Analyzing coherent structures by means of dynamic mode decomposition, 17th OpenFOAM workshop, Jul 2022, Cambridge
• Active flow control via deep reinforcement learning implemented in OpenFOAM, OpenFOAM combustion seminar - invited talk, May 2022, Singapore, virtual
• Simulation and modal analysis of transonic shock buffets on a NACA-0012 airfoil, Euromech Colloquium 612, Mar 2022, Aachen, virtual
• Machine learning in computational fluid dynamics - an overview, keynote lecture chemical industry, Feb 2022
• Simulation and modal analysis of transonic shock buffets on a NACA-0012 airfoil, AIAA SciTech Forum, Jan 2022, San Diego, virtual
• Active control of the flow past a cylinder using deep reinforcement learning, OpenFOAM conference, Oct 2021, virtual
• flowTorch - A platform for analysis and reduced-order modeling of high-speed stall flow phenomena, DLRK 2021, Aug 2021, Bremen, virtual
• Machine learning-aided CFD with OpenFOAM and PyTorch, training at 16th OpenFOAM Workshop, Jun 2021, Dublin, virtual
• Machine learning-aided CFD with OpenFOAM and PyTorch, SSD Seminar - invited talk, Jun 2021, RWTH Aachen, virtual
• Sparse Spatial Samling - S³, AIAA SciTech Forum, Jan 2021, Nashville, virtual
• Creating data-driven workflows with OpenFOAM and PyTorch, 8th ESI OpenFOAM conference, Oct 2020, Berlin, virtual
• An introduction to supervised learning by example: path regime classification, internal training, Aug 2020, Braunschweig, virtual
• A hybrid approach to compute convection-dominated mass transfer at rising bubbles, 4th GOFUN, Apr 2020, Braunschweig, virtual
• Modeling and simulation of convection-dominated species transfer at rising bubbles, Ph.D. defense, Jan 2020, Darmstadt
• A brief introduction to machine learning and its potential application to CFD, 14th OpenFOAM workshop, Jul 2019, Duisburg
• Data-driven subgrid-scale modeling for convection-dominated concentration boundary layers, 14th OpenFOAM workshop, Jul 2019, Duisburg

Supervised student projects

• Learning of optimized multigrid solver settings for CFD applications, Master thesis, Janis Geise, 2023, Github
• A comparison of reduced-order models for wing buffet predictions, course project, Anton Schreiber, 2023, Github
• Prediction of ice formation on heat exchanger surfaces and defrost control using machine learning, master project, together with Coldsense, Tom Krogmann, 2023
• Robust model-based deep reinforcement learning for flow control, course project, Janis Geise, 2023, Github
• Optimal sensor placement for active flow control with deep reinforcement learning, course project, Tom Krogmann, 2022, Github
• Design and implementation of a data-driven wall function for the velocity in RANS simulations, master project, Jihoo Kang, 2022, Github
• POD-based parametric reduced-order models for time-dependent fluid flows, course project, Jan Schlüter, 2022
• Model-based reinforcement learning for accelerated learning from CFD simulations, course project, Jan Erik Schulze, 2022, Github
• Reduced-order modeling based on cluster-based network modeling applied to the latent variables of an autoencoder, course project, Niels Formella, 2021, Github
• Active control of the flow past a cylinder under Reynolds number variation using deep reinforcement learning, Bachelor thesis, Fabian Gabriel, 2021, Github
• Numerical investigation of 2D transonic shock-buffet around a NACA 0012-34 airfoil using OpenFOAM and flowTorch, course project, Tushar Anil Gholap, 2021, Github
• Active flow control in simulations of fluid flows based on deep reinforcement learning, course project, Darshan Thummar, 2021, Github
• Simulation of Fluid Flows based on the Data-driven Evolution of Vortex Particles, Master thesis, Vemburaj Chockalingam Yadav, 2021, Github
• Datenbasierte Subgridskalen-Modellierung reaktiver Konzentrationsgrenzschichten an freiaufsteigenden Einzelblasen, Master thesis, Alexander Kiefer, 2020
• A comparative study of different mesh types for transport processes near gas bubbles regarding accuracy, stability, and run time, Bachelor thesis, Jan-Alexander Kleikemper, 2018
• Numerical simulation of single rising bubbles influenced by soluble surfactant in the spherical and ellipsoidal regime, Master thesis, Matthias Steinhausen, 2018
• Numerical simulation of reactive species transfer at a spherical gas bubble, Bachelor thesis, Tim Jeremy Patrick Karpowski, 2017