BEGIN:VCALENDAR METHOD:PUBLISH PRODID:www-cbi-tf-fau-eu//Events//EN VERSION:2.0 BEGIN:VEVENT SUMMARY:CBI Colloquium "Non-Linear Super-Stencils for Turbulence Model Corrections" (Prof. Patrick Jenny\, ETH Zürich) UID:3a18-d63f-cc13-5703@www.cbi.tf.fau.eu DESCRIPTION:Accurate simulation of turbulent flows remains a challenge due to the high computational cost of direct numerical simulations (D NS) and the limitations of traditional turbulence models. This researc h explores a novel approach to augmenting standard models for Reynolds -Averaged Navier-Stokes (RANS) simulations using a Non-Linear Super-St encil (NLSS). The proposed method introduces a fully connected neural network that learns a mapping from the local mean flow field to a corr ective force term\, which is added to a standard RANS solver in order to align its solution with high-fidelity data\; an illustration of an NLSS is shown in Fig. 1. Figure 1: Non-Linear Super-Stencil (black dot s): The stencil is centered around the point at which the correction f orce shall be determined\, and it is aligned with the mean velocity at its center. A procedure is devised to extract training data from refe rence DNS and large eddy simulations (LES). To reduce the complexity o f the non-linear mapping\, the dimensionless local flow data is aligne d with the local mean velocity\, and the local support domain is scale d by the turbulent integral length scale. After being trained on a sin gle periodic hill case\, the NLSS-corrected RANS solver is shown to ge neralize to different periodic hill geometries and different Reynolds numbers\, producing significantly more accurate solutions than the unc orrected RANS simulations. For demonstration\, Fig. 2 shows RANS simul ation results with and without NLSS-correction along with high fidelit y references (mean velocity profiles in the top plots and wall shear s tresses on the bottom) of two test cases which were not used for train ing. For training a domain of lenth L = 9m\, a hill stretch factor of alpha = 1 and a Reynolds number of Re = 10595 were considered\, while L = 13:929m\, alpha = 1:5 and Re = 5600 were chosen for the left test case and L = 9m\, alpha = 1 and Re = 19000 for right test case. For bo th test cases it can be observed that the NLSS-corre DTSTART:20250724T161500Z DTEND:20250724T174500Z LOCATION:Hanns Hofmann Lecture Hall ( KS I)\, Cauerstr. 4\, 91058 Erlangen DTSTAMP:20260427T151956Z END:VEVENT END:VCALENDAR