Predicting drag on arbitrary rough surfaces
Time: Fri 2022-10-14 10.30 - 11.30
Location: Faxén, Teknikringen 8
Participating: Pourya Fooroghi (Aarhus University)
Abstract: The aim is to predict roughness-induced drag on an arbitrary rough surface based on the topography of roughness. We employ direct numerical simulations of turbulent flow over mathematically generated roughness as numerical experiments. An ensemble neural network is trained through an active learning framework to minimize the number of simulations required for a universal model. The trained model successfully predicts equivalent sand roughness height for a wide range of unseen realistic rough surfaces.
Bio: P. Forooghi received his PhD in Mechanical Engineering from The University of Queensland, Australia. He then joined Institute of Fluid Mechanics at Karlsruhe Institute of Technology as a post-doc, where his main focus was on prediction of drag and heat transfer over rough walls. Currently he is an Assistant Professor in Mechanical Engineering at Aarhus University, Denmark, where he leads the Heat and Fluid Flow research group with focus on understanding and control of momentum, heat and mass transfer in wall-bounded fluid flows using numerical methods.