BibTex format
@inproceedings{Mahfoze:2019,
author = {Mahfoze, OA and Wynn, A and Whalley, RD and Laizet, S},
title = {Bayesian optimisation of intermittent wall blowing in a turbulent boundary layer for net power saving},
year = {2019}
}
Publications
Citation
RIS format (EndNote, RefMan)
TY - CPAPER
AB - A Bayesian optimisation framework is used to optimise low-amplitude wall-normal blowing control of a turbulent boundary-layer (TBL) flow in order to achieve skin-friction drag reduction and net-power saving. The study is carried out using Direct Numerical Simulations (DNS) and Implicit Large Eddy Simulations (ILES). Control performance is assessed by using the power consumption from two different sets of experimental data from two different types of blowing device. The simulations demonstrate that wall-normal blowing control can generate a local skin-friction drag reduction of up to 75%, which persists far downstream of the control. This slow spatial recovery of the skin-friction coefficient back to its canonical counterpart can generate net-power savings up to 5% in the present study. When combined with DNS or ILES, Bayesian optimisation, with its fast convergence (within a dozen iterations with three parameters to optimise) is an ideal tool to find the optimal set of parameters to maximise net-power saving. The evolution of the skin-friction coefficient is decomposed using the Fukagata-Iwamoto-Kasagi (FIK) identity, which shows that the generation of the net-power savings is due to changes in contributions to both the convection and streamwise development terms of the turbulent boundary-layer flow.
AU - Mahfoze,OA
AU - Wynn,A
AU - Whalley,RD
AU - Laizet,S
PY - 2019///
TI - Bayesian optimisation of intermittent wall blowing in a turbulent boundary layer for net power saving
ER -