University of Cambridge > Talks.cam > Isaac Newton Institute Seminar Series > Wall turbulence drag reduction in the era of machine learning--From optimized distributed actuation to automated reduced-order modeling

Wall turbulence drag reduction in the era of machine learning--From optimized distributed actuation to automated reduced-order modeling

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact nobody.

TURW04 - Wall-bounded turbulence: beyond current boundaries

The control of wall turbulence is of paramount engineering importance. About 20% of the world energy is consumed by maritime, ground and airborne transport. Wall turbulence is a significant contribution to the parasitic drag of ships, trains and airplaines as well as the main resistance in pipe flows with oil, gas, water and other fluids. This talk focuses on machine-learned optimization and modeling of wall turbulence drag reduction with distributed actuation. First,  31% skin friction reduction of a turbulent boundary layer is achieved by traveling surface deformation in large eddy simulations. The actuation parameters are optimized with a machine learned self-similar response model and accurately predicted far outside the training data (Fernex et al. 2020 Phys. Rev. Fluids). Second, these simulations are modeled using a cluster-based network model (Fernex et al. 2021 Sci. Adv.). These data-driven reduced-order models have distinct advantages over POD models,like robustness, human interpretability, and automated development. Third,  first experimental results of a self-learning smart skin separation control over a smooth ramp are presented. The feedback laws of two-dimensional multi-modal actuator/sensor arrays are optimized with gradient-enriched machine learning control (Cornejo Maceda et al. 2021 J. Fluid Mech). The talk concludes with  perspectives of future developments. Co-authors: Songqi Li,  Jiayang Luo,  Guy Cornejo Maceda, Nan Gao (HIT, China);Daniel Fernex (EPFL) Richard Semaan (TU Braunschweig), Marian Albers, Wolfgang Schroeder (RWTH Aachen University)

This talk is part of the Isaac Newton Institute Seminar Series series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2022 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity