ANN-based entropy prediction and statistical analysis of the hydromagnetic hybrid nanofluid flow over a rotating stretching disk

This article presents a machine-learning-based investigation of entropy generation in the magnetohydrodynamic flow of a hybrid nanofluid over a rotating disk embedded in a Darcy Forchheimer porous medium. The effects of a magnetic field on unsteady flow and heat transfer are examined for an incompressible hybrid nanofluid. The governing highly nonlinear partial differential equations are transformed into nonlinear ordinary differential equations using suitable similarity transformations and solved numerically via the spectral quasi-linearization technique. Local skin friction coefficients in the radial and tangential directions, along with the local heat transfer rate at the disk surface, are evaluated for various physical parameters. Results are presented through tables and dimensionless velocity and temperature profiles. Regression analysis is employed to estimate skin friction and heat transfer coefficients, while an artificial neural network model predicts average entropy generation with excellent accuracy (R=1). Increasing the magnetic parameter and Forchheimer number reduces fluid velocity, whereas higher radiation enhances temperature.