%0 Conference Paper
%A Bode, Mathis
%A Gauding, Michael
%A Göbbert, Jens Henrik
%A Liao, Baohao
%A Jitsev, Jenia
%A Pitsch, Heinz
%T Towards Prediction of Turbulent Flows at High Reynolds Numbers Using High Performance Computing Data and Deep Learning
%V 11203
%C Cham
%I Springer International Publishing
%M FZJ-2019-00776
%@ 978-3-030-02464-2 (print)
%B Lecture Notes in Computer Science
%P 614 - 623
%D 2018
%< High Performance Computing
%X In this paper, deep learning (DL) methods are evaluated in the context of turbulent flows. Various generative adversarial networks (GANs) are discussed with respect to their suitability for understanding and modeling turbulence. Wasserstein GANs (WGANs) are then chosen to generate small-scale turbulence. Highly resolved direct numerical simulation (DNS) turbulent data is used for training the WGANs and the effect of network parameters, such as learning rate and loss function, is studied. Qualitatively good agreement between DNS input data and generated turbulent structures is shown. A quantitative statistical assessment of the predicted turbulent fields is performed.
%B International Conference on High Performance Computing
%C 24 Jun 2018 - 28 Jun 2018, Frankfurt (Germany)
Y2 24 Jun 2018 - 28 Jun 2018
M2 Frankfurt, Germany
%F PUB:(DE-HGF)8 ; PUB:(DE-HGF)7
%9 Contribution to a conference proceedingsContribution to a book
%U <Go to ISI:>//WOS:000612998200051
%R 10.1007/978-3-030-02465-9_44
%U https://juser.fz-juelich.de/record/859970