EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Transfer learning from two-dimensional supercritical airfoils to three-dimensional transonic swept wings

Runze LI, Yufei ZHANG, Haixin CHEN

Chinese Journal of Aeronautics · 2023-09 · 36 (9) · 96-110

DOI: 10.1016/j.cja.2023.04.008 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Chinese Journal of Aeronautics, 2023-09. Verfasst von Runze LI und 2 weiteren Autor:innen. Die Arbeit wurde bisher 69-mal zitiert (Crossref) und nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Machine learning has been widely utilized in fluid mechanics studies and aerodynamic optimizations. However, most applications, especially flow field modeling and inverse design, involve two-dimensional flows and geometries. The dimensionality of three-dimensional problems is so high that it is too difficult and expensive to prepare sufficient samples. Therefore, transfer learning has become a promising approach to reuse well-trained two-dimensional models and greatly reduce the need for samples for three-dimensional problems. This paper proposes to reuse the baseline models trained on supercritical airfoils to predict finite-span swept supercritical wings, where the simple swept theory is embedded to improve the prediction accuracy. Two baseline models for transfer learning are investigated: one is commonly referred to as the forward problem of predicting the pressure coefficient distribution based on the geometry, and the other is the inverse problem that predicts the geometry based on the pressure coefficient distribution. Two transfer learning strategies are compared for both baseline models. The transferred models are then tested on the prediction of complete wings. The results show that transfer learning requires only approximately 500 wing samples to achieve good prediction accuracy on different wing planforms and different free stream conditions. Compared to the two baseline models, the transferred models reduce the prediction error by 60% and 80%, respectively.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

LI, R., ZHANG, Y., & CHEN, H. (2023). Transfer learning from two-dimensional supercritical airfoils to three-dimensional transonic swept wings. Chinese Journal of Aeronautics, 36(9), 96-110. https://doi.org/10.1016/j.cja.2023.04.008

Vancouver

LI R, ZHANG Y, CHEN H. Transfer learning from two-dimensional supercritical airfoils to three-dimensional transonic swept wings. Chinese Journal of Aeronautics. 2023;36(9):96-110. doi:10.1016/j.cja.2023.04.008

BibTeX
@article{LI2023,
  author = {LI, Runze and ZHANG, Yufei and CHEN, Haixin},
  title = {Transfer learning from two-dimensional supercritical airfoils to three-dimensional transonic swept wings},
  journal = {Chinese Journal of Aeronautics},
  year = {2023},
  volume = {36},
  number = {9},
  pages = {96--110},
  doi = {10.1016/j.cja.2023.04.008}
}
Kostenlos starten →