Entropy Generation and Heat Transfer Prediction in Casson Nanofluid Flow Using OHAM and Artificial Neural Network (ANN) Approach

This paper presents a novel analytical—computational investigation of entropy generation and heat transfer characteristics in magnetohydrodynamic Casson nanofluid flow over an exponentially stretching sheet embedded in a porous medium with internal heat source/sink effects. The study is distinguished by the combined incorporation of Casson non-Newtonian rheology, exponential stretching dynamics, thermal radiation, Joule heating, and volumetric heat generation/absorption within a unified entropy-optimization framework. Using suitable similarity transformations, the governing nonlinear boundary-layer equations are reduced to coupled ordinary differential equations and solved semi-analytically via the Optimal Homotopy Analysis Method (OHAM), ensuring convergence-controlled solutions without relying on small or large perturbation parameters. Entropy generation and Bejan number analyses reveal that magnetic field strength, Brinkman number, and Eckert number significantly enhance thermodynamic irreversibility, whereas increased thermal radiation shifts entropy dominance toward heat-transfer irreversibility. The results further indicate that stronger magnetic fields and reduced permeability suppress velocity while elevating fluid temperature and entropy generation. Conversely, higher Prandtl numbers and radiation parameters improve surface heat transfer rates. To enhance predictive accuracy and computational efficiency, quadratic regression and Artificial Neural Network (ANN) surrogate models are developed using OHAM-generated data. The ANN predictions for skin friction and Nusselt number show excellent agreement with analytical results, demonstrating strong robustness and generalization capability. The proposed OHAM—ANN hybrid framework constitutes a novel and efficient predictive tool for analyzing entropy generation and thermal performance in Casson nanofluid systems, offering practical design insights for energy-efficient applications in polymer processing, biomedical transport, microfluidic devices, and advanced thermal management technologies.