Machine learning implication for thermal scrutiny of ternary inclined magnetized nanofluid with convergent divergent geometries: Two hidden neural layer procedure
International Journal of Modern Physics B · 20.08.2025 · 39 (21)
DOI: 10.1142/s0217979225501991 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Modern Physics B, online erschienen am 05.06.2025. Verfasst von Saleh Chebaane und 6 weiteren Autor:innen. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 16 Literaturangaben. Gefördert von: University of Hail.
Abstract (Originalsprache)
The root purpose of the study is to investigate the thermal transport properties of ternary non-Newtonian nanofluids in inclined magnetized environments. This study involves convergent divergent geometries which have interesting engineering applications. The research aims to evaluate the efficacy of combining numerical methods like bvp4c with neural networks to address the challenges posed by nonlinear governing equations associated with such fluid flows. This study employs a combined approach utilizing bvp4c and a three-hidden-layer neural network to model and analyze the fluid dynamics. Governing equations are derived using non-Newtonian fluid dynamics principles and are solved with the bvp4c technique to establish a baseline. The neural network, with three hidden layers, further refines the solution and trains the network and further predicts the numerical solution. Larger opening angles cause fluid to spread over a broader area, reducing velocity gradient and overall flow speed. The velocity increases with higher magnetic parameter values, as the magnetic field induces stabilizing forces that enhance flow uniformity and reduce turbulence. Higher thermophoresis values enhance nanoparticle movement along temperature gradients, driving particles away from hotter regions and promoting dispersion.
Förderung
- University of Hail
Zitieren
APA
Chebaane, S., Haider, A., Shah, S. Z. H., Ayub, A., Dafhalla, A., Affan, H., & Manai, L. (2025). Machine learning implication for thermal scrutiny of ternary inclined magnetized nanofluid with convergent divergent geometries: Two hidden neural layer procedure. International Journal of Modern Physics B, 39(21). https://doi.org/10.1142/s0217979225501991
Vancouver
Chebaane S, Haider A, Shah SZH, Ayub A, Dafhalla A, Affan H, et al. Machine learning implication for thermal scrutiny of ternary inclined magnetized nanofluid with convergent divergent geometries: Two hidden neural layer procedure. International Journal of Modern Physics B. 2025;39(21). doi:10.1142/s0217979225501991
BibTeX
@article{Chebaane2025,
author = {Chebaane, Saleh and Haider, Ali and Shah, Syed Zahir Hussain and Ayub, Assad and Dafhalla, Alaa and Affan, Hira and Manai, Leila},
title = {Machine learning implication for thermal scrutiny of ternary inclined magnetized nanofluid with convergent divergent geometries: Two hidden neural layer procedure},
journal = {International Journal of Modern Physics B},
year = {2025},
volume = {39},
number = {21},
doi = {10.1142/s0217979225501991}
}
EUVIMED