Using Machine Learning for formulating new wall functions for Detached Eddy Simulation

Paper i proceeding, 2023

Machine Learning (ML) is used for developing wall functions for Detached Eddy Simulations.I use Improved Delayed Detached Eddy Simulations (IDDES) in fully-developed channel flow at a frictional Reynolds number of 5200 to create the database. This database is used as a training set for the ML method (support vector regression, SVR, in Python is used). The input (i.e. the influence parameters) is y^+. The ML method is trained to predict U^+.

The trained ML model is saved to disk and it is subsequently uploaded into the Python CFD code pyCALC-LES. IDDES is carried out on coarse wall-function meshes. The wall-shear stress (using the local y^+ and u) is predicted using the developed ML model. The test cases are channel flow at Re_tau=16 000 and flat-plate boundary layer at Re_theta=2550.