The present study develops an intelligent computational framework to enhance heat and mass transfer in solar thermal systems using a penta-hybrid nanofluid (PHNF) comprising Al2O3, Cu, TiO2, SiO2 and MWCNT nanoparticles dispersed in water. The electromagnetic hydrodynamic flow of a Maxwell PHNF over a permeable stretching surface is investigated under the effects of thermal radiation, viscous dissipation, Joule heating and porous media. The novelty lies in combining multi-component penta-hybrid nanoparticles with an artificial neural network (ANN)-based predictive model for simultaneous estimation of the skin friction coefficient and Nusselt number. The transformed nonlinear governing equations are solved numerically using MATLAB's bvp4c solver. Results reveal that increasing the magnetic parameter reduces velocity due to enhanced Lorentz force resistance, while higher thermal radiation and a higher Biot number significantly improve the temperature distribution and heat transfer. The PHNF model enhances the heat transfer rate by 26.62% compared with the base fluid. The ANN model trained using Bayesian regularisation achieves high predictive accuracy, with correlation coefficients above 0.95 and low RMSE values. Sensitivity analysis identifies magnetic field strength and thermal radiation as dominant factors influencing momentum and thermal transport. These findings support applications in solar collectors, photovoltaic/thermal systems, thermal energy storage and renewable-energy cooling technologies.
Mohanty et al. (Sun,) studied this question.