Aarhus Universitets segl

English

Du er her: AU » Om AU » Organisation » Predicting drag on rough surfaces by transfer learning of...

Om AU

Predicting drag on rough surfaces by transfer learning of empirical correlations

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avis › Tidsskriftartikel › Forskning › peer review

Links

DOI

  • Sangseung Lee, KTH Stockholm
    • ,
  • Jiasheng Yang, Karlsruhe Institute of Technology
    • ,
  • Pourya Forooghi
  • Alexander Stroh, Karlsruhe Institute of Technology
    • ,
  • Shervin Bagheri, KTH Stockholm

Recent developments in neural networks have shown the potential of estimating drag on irregular rough surfaces. Nevertheless, the difficulty of obtaining a large high-fidelity dataset to train neural networks is deterring their use in practical applications. In this study, we propose a transfer learning framework to model the drag on irregular rough surfaces even with a limited amount of direct numerical simulations. We show that transfer learning of empirical correlations, reported in the literature, can significantly improve the performance of neural networks for drag prediction. This is because empirical correlations include 'approximate knowledge' of the drag dependency in high-fidelity physics. The 'approximate knowledge' allows neural networks to learn the surface statistics known to affect drag more efficiently. The developed framework can be applied to applications where acquiring a large dataset is difficult but empirical correlations have been reported.

OriginalsprogEngelsk
ArtikelnummerA18
TidsskriftJournal of Fluid Mechanics
Vol/bind933
Antal sider21
ISSN0022-1120
DOI
StatusUdgivet - 25 feb. 2022

Se relationer på Aarhus Universitet Citationsformater

ID: 229259181