Application of artificial neural networking to scrutinize the three-dimensional stagnation-point flow with variable physical properties

This work examines the steady three-dimensional stagnation point of an electrically conducting Newtonian fluid under Oberbeck–Boussinesq approximation. This article is concerned with the boundary-layer formation over the vertical sheet. An analysis has been carried out to investigate the influence o...

Full description

Saved in:
Bibliographic Details
Published in:Physics of fluids (1994) 2024-10, Vol.36 (10)
Main Authors: Saleem, Sana, Ul Haq, Rizwan, Mustafa, M., Soomro, F. A.
Format: Article
Language:English
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This work examines the steady three-dimensional stagnation point of an electrically conducting Newtonian fluid under Oberbeck–Boussinesq approximation. This article is concerned with the boundary-layer formation over the vertical sheet. An analysis has been carried out to investigate the influence of variable fluid properties in mixed convection three-dimensional flow of viscous fluid by a vertical surface with heat transfer. The mathematical model incorporates by considering temperature-dependent variations in viscosity and thermal conductivity. The governing equations are transformed into nonlinear partial differential equations by appropriate transformation and admit local-similar solutions. The flow has to satisfy ordinary differential equations whose solution depends upon different parameters such as mixed convection parameter λ and variable viscosity θr, and c1 shows 3-dimensional motion of flow. Quantitative analysis of the flow field and heat transfer characteristics is conducted using graphs and numerical values obtained through Bvp4c MATLAB. A novel approach using Artificial Neural Networks (ANNs) has been used to accurately predict heat transfer parameters. The ANN model is trained using a comprehensive dataset obtained from numerical simulations and experimental measurements. The inputs to the ANN include relevant flow parameters such as Reynolds number, Prandtl number, and geometrical characteristics, while the outputs are the corresponding skin friction and the Nusselt number. The results indicate that the ANN model exhibits excellent predictive accuracy compared to traditional empirical correlations and computational fluid dynamics simulations. The graphical representation of emergent parameters has been explored, along with a corresponding discussion. The comparison is being made between the effects of constant and variable fluid properties.
ISSN:1070-6631
1089-7666
DOI:10.1063/5.0227095