.png){kind=icon}
.png){kind=icon}
Return
Machine learning based modeling of triple-diffusive MHD hyperbolic nanofluid flow with heat transfer optimization in Darcy-Forchheimer porous medium
.png){kind=icon}
OA
AI
N
W
M
F
DOI：10.1016/j.jppr.2026.05.001.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This study investigates magnetohydrodynamic (MHD) flow of a tangent hyperbolic nanofluid (HNF) over a stretching surface, incorporating thermal radiation, chemical reactions, and Soret-Dufour effects within a Darcy-Forchheimer porous medium. The governing partial differential equations (PDEs) are first converted into ordinary differential equations (ODEs) using appropriate similarity transformations. The MATLAB bvp4c algorithm is used to solve and generate a dataset for the Levenberg-Marquardt backpropagation artificial neural network (LMBP-ANN) numerically. The ANN model is trained, tested, and validated using comprehensive datasets generated for various fluid parameters, with accuracy assessed through regression analysis, error histograms, and curve fitting. The dataset was divided into three subsets, training, validation, and testing, 80%, 10%, and 10%. Key findings reveal that the velocity profile declines with increasing the value of the Hartmann number, Weissenberg number, and power-law index but rises with buoyancy parameters and the Deborah number. The temperature profile is enhanced by the Hartmann number, Brownian motion, power-law index, thermophoresis, Dufour number, and thermal radiation, while being reduced by the Prandtl number. The solute concentration profile decreases with the Schmidt number but increases with the Soret number. The numerical results for skin friction, Nusselt number, and Sherwood number are presented in tabular form.
Keywords:
MHD tangent hyperbolic
Stretching sheet
Soret-Dufour effects
Nanofluid
Thermal radiation
Darcy-Forchheimer
ANN
Journal
IF:
6.3
Papers:
336
Citations:
1.7K
Organization
Cited Papers