Master/Project Thesis: Wall-modeled LES and Machine Learning
Background:
The Institute of Fluid Mechanics is involved in a European project called CEEC (https://ceec-coe.eu/). CEEC’s ambition is to enable the use of exascale computers for key computational fluid dynamics (CFD) applications and demonstrate their capabilities through key light-house cases. The light-house case we are working on is the flow around a merchant ship hull, i.e., the Japanese Bulk Carrier. For this well-known benchmark case, the Reynolds numbers are expected in a range that even with the help of exascale computing wall-modeling is required for the LES simulations to somehow “bridge” the wall closest layers. The objective of this task is to work on wall-modeling, either algebraic wall models or wall-models enriched by machine learning, implement these approaches into our flow solver Neko and validate the developments basing on particular test cases.
Specific tasks:
- Literature study on the topics of wall-models and machine-learning enriched wall-models
- Implement algebraic and / or machine-learning enriched models into Neko
- Setting up of validation test cases and perform simulations
- Data analysis and reporting of the results
Requirements:
- Interest in turbulent flows, LES, wall-models and coding
- Successfully taken part in NMTFD1+2 or CFD1+2 respectively
- Independence working style
Starting date: immediately
Advisor:
