Laboratory for Flow Instabilities and Dynamics
© Fernex, Daniel, Bernd R. Noack, and Richard Semaan. "Cluster-based network modeling—From snapshots to complex dynamical systems." Science Advances 7.25 (2021): eabf5006

Novel Techniques in Model Building for Gas Turbine Combustion Systems

In order to fully exploit the potential of modern gas turbines, further design steps are necessary to optimize the operational performance. In this context, the possibility of alternative fuels such as hydrogen also places new demands on gas turbines. In our research, we use machine learning methods to create mathematical models of various components of the combustion system that can be used for analysis and optimization. With the participating students, we will pursue the new approach of Cluster-Based network modeling to analyze and evaluate experimental flow data from a combustion chamber. For this purpose, we will give an introduction to the method as well as to combustion technology in general. In the subsequent project-oriented part of the course, participants will then apply the method to more complex data in small groups.

Ideally, participants should have a solid basic knowledge of mathematics. In addition, basic knowledge of programming (ideally Python) is highly recommended. Knowledge of fluid dynamics, combustion, or machine learning may be beneficial, but is not a mandatory requirement. The module can be credited with 6 ECTS in the free elective area. You can register for the module by sending us an email (xstudent(at)fd.tu-berlin.de). If you have any questions, we will be happy to answer them by mail.

We are looking forward to meet you !

This research project will be offered in English and only during the summer semester 2022.

 

 

Content

In order to fully exploit the potential of modern gas turbines, further design steps are necessary to optimize the operational performance. In this context, the possibility of alternative fuels such as hydrogen also places new demands on gas turbines. In our research, we use machine learning methods to create mathematical models of various components of the combustion system that can be used for analysis and optimization. With the participating students, we will pursue the new approach of Cluster-Based network modeling to analyze and evaluate experimental flow data from a combustion chamber.

Time and Location

The X-Student Research Groups are implemented in the form of research seminars and run for one semester.

The lecture, exercise or question time take place online:

Wednesdays: 10:00 - 12:00.

The first date is the 20.04.22.

Credits

6 ECTS

The course is credited with 6 ECTS.

More information on crediting can be found on the website of the  Berlin-University-Alliance.

Requirements

Contact

Notes

The X-Student Research Group is funded by the Berlin University Alliance.
More information about the research groups can be found here.

There will be a mix of lecture, tutorial and independent group work.
The group work will be concluded with a presentation.

The module can (as far as we know) be credited in the free elective area.

If you want to participate please send us an email to:
xstudent(at)fd.tu-berlin.de