|
Unsteady Simulations: LES, DES, hybrid LES/RANS and Machine Learning
|
|
LES is suitable for bluff-body flows or flows at low Reynolds numbers.To extend LES to cover industrial flows at high Reynolds numbers, new approaches (hybrid LES-RANS, URANS, DDES, SAS, PANS, must be used. They are all based on a mix of LES and RANS. The course will give an introduction to LES and these new methods. Day 3, an introduction will be given on how to use Machine Learning for improving underlying RANS turbulence models and wall-functions.
|
| Date: |
December 1, 2025 - December 5, 2025
|
| Location: |
Zoom, Sweden
|
| Web Page: |
https://www.cfd-sweden.se/
|
| Contact Email: |
lada@flowsim.se
|
| Organizer: |
fliowsim AB
|
| Application Areas: |
Turbomachinery, Electronics Cooling, Automotive, Process Industry, Aerospace, Power Generation, Nuclear Reactor Thermal Hydraulics, Wind Turbines
|
| Special Fields: |
Turbulence Modeling, Turbulence - LES Methods, Heat Transfer, Aerodynamics, Fluid Mechancis, Turbulence - Hybrid RANS-LES Methods
|
| Software: |
Python
|
| Deadlines: |
November 7, 2025 (registration)
|
| Type of Event: |
Course, International
|
| |
| Description: |
A three-day ONLINE course on Unsteady Simulations: LES, DES, hybrid LES/RANS and Machine Learning
LES is suitable for bluff-body flows or flows at low Reynolds numbers. To extend LES to cover industrial flows at high Reynolds numbers, new approaches (hybrid LES-RANS, URANS, DDES, SAS, PANS, must be used. They are all based on a mix of LES and RANS.
The course will give an introduction to LES and these new methods. Day 3, an introduction will be given on how to use Machine Learning for improving underlying RANS turbulence models and wall-functions.\n\nThe lectures will be given on-line (Live) using Zoom. During the workshops, the articipants will get supervision.
In the workshops, the participants will use Python/Matlab/Octave for analyzing SGS models, SAS, PANS, DES and DDES using DES databasis. These will be the topics Day 1 and
2.
Day 3 is devoted to Machine Learning. Neural network and KDTRee in Python's Pytorch will be used to improve RANS turbulence models and wall functions.
The number of participants is limited to 16.
The course fee is 14 700 SEK (approx 1300 Euro).
|
| |
|
Event record first posted on August 29, 2025, last modified on September 22, 2025
|
