EUVIMEDEUVIMED entdecken
Crossref

Predicting unsteady incompressible fluid dynamics with finite volume informed neural network

Tianyu Li, Shufan Zou, Xinghua Chang, Laiping Zhang, Xiaogang Deng

Physics of Fluids · 01.04.2024 · 36 (4)

DOI: 10.1063/5.0197425 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Physics of Fluids, online erschienen am 02.04.2024. Verfasst von Tianyu Li und 4 weiteren Autor:innen. Die Arbeit wurde bisher 39-mal zitiert (Crossref) und nennt 39 Literaturangaben. Gefördert von: Xiaogang Den, Liangzhong Che.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The rapid development of deep learning has significant implications for the advancement of computational fluid dynamics. Currently, most pixel-grid-based deep learning methods for flow field prediction exhibit significantly reduced accuracy in predicting boundary layer flows and poor adaptability to geometric shapes. Although graph neural network models for unstructured grid-based unsteady flow prediction have better geometric adaptability, these models suffer from error accumulation in long-term predictions of unsteady flows. More importantly, fully data-driven models often require extensive training time, greatly limiting the rapid update and iteration speed of deep learning models when facing more complex unsteady flows. Therefore, this paper aims to balance the demands for training overhead and prediction accuracy by integrating physical constraints based on the finite volume method into the loss function of the graph neural network. Additionally, it incorporates a twice-message aggregation mechanism inspired by the extended stencil method to enhance the unsteady flow prediction accuracy and geometric shape generalization ability of the graph neural network model on unstructured grids. We focus particularly on the model's predictive accuracy within the boundary layer. Compared to fully data-driven methods, our model achieves better predictive accuracy and geometric shape generalization ability in a shorter training time.

Förderung

  • Xiaogang Den
  • Liangzhong Che

Zitieren

APA

Li, T., Zou, S., Chang, X., Zhang, L., & Deng, X. (2024). Predicting unsteady incompressible fluid dynamics with finite volume informed neural network. Physics of Fluids, 36(4). https://doi.org/10.1063/5.0197425

Vancouver

Li T, Zou S, Chang X, Zhang L, Deng X. Predicting unsteady incompressible fluid dynamics with finite volume informed neural network. Physics of Fluids. 2024;36(4). doi:10.1063/5.0197425

BibTeX
@article{Li2024,
  author = {Li, Tianyu and Zou, Shufan and Chang, Xinghua and Zhang, Laiping and Deng, Xiaogang},
  title = {Predicting unsteady incompressible fluid dynamics with finite volume informed neural network},
  journal = {Physics of Fluids},
  year = {2024},
  volume = {36},
  number = {4},
  doi = {10.1063/5.0197425}
}
Kostenlos starten →