EUVIMEDEUVIMED entdecken
Crossref

Deep learning-driven physics-informed neural networks for rotating fluid flow over a vertical cylinder with Hall currents and thermal effects

Vijay Kumar Sukariya, Anurag Shukla, Anand Kumar

Physics of Fluids · 01.06.2025 · 37 (6)

DOI: 10.1063/5.0275514 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Physics of Fluids, online erschienen am 25.06.2025. Verfasst von Vijay Kumar Sukariya und 2 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 36 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper presents a physics-informed neural network approach to solve the coupled system of partial differential equations governing the three dimensional transient flow of a rotating fluid over an infinitely long vertical cylinder in the radial direction. The flow distribution is influenced by thermal buoyancy forces, external thermal effects, and Hall currents resulting from an externally applied radial magnetic field. The study begins by transforming the governing equations of the mathematical model into their dimensionless forms using appropriate nondimensional parameters. These equations are then solved using the advanced deep learning framework known as physics-informed neural networks and validated through numerical methods. This study revels a noteworthy behavior regarding Hall current parameter, where the primary velocity initially diminishes. However, after reaching a critical value (m=0.6) at time t=3, it suddenly rises within the boundary layer region. Furthermore, we observe contrasting effects on both velocities as the rotation parameter increases, leading to a reduction in primary velocity while simultaneously enhancing secondary velocity. The skin friction coefficient in the axial direction increases by 5.80% as the Hall parameter is raised from 0.4 to 1.0, while in the θ− direction, this coefficient rises by 91.61%.

Zitieren

APA

Sukariya, V. K., Shukla, A., & Kumar, A. (2025). Deep learning-driven physics-informed neural networks for rotating fluid flow over a vertical cylinder with Hall currents and thermal effects. Physics of Fluids, 37(6). https://doi.org/10.1063/5.0275514

Vancouver

Sukariya VK, Shukla A, Kumar A. Deep learning-driven physics-informed neural networks for rotating fluid flow over a vertical cylinder with Hall currents and thermal effects. Physics of Fluids. 2025;37(6). doi:10.1063/5.0275514

BibTeX
@article{Sukariya2025,
  author = {Sukariya, Vijay Kumar and Shukla, Anurag and Kumar, Anand},
  title = {Deep learning-driven physics-informed neural networks for rotating fluid flow over a vertical cylinder with Hall currents and thermal effects},
  journal = {Physics of Fluids},
  year = {2025},
  volume = {37},
  number = {6},
  doi = {10.1063/5.0275514}
}
Kostenlos starten →