Neural Network–Based Heat Transfer Optimization of Casson Nanofluid Flow for Sustainable Energy Applications

The thermal interaction between a moving material and the surrounding environment can be analyzed through the fluid flow over a vertically placed plate that begins motion impulsively. The study is carried out to explore the heat transfer from the moving plate by using Casson hybrid nanofluid. The mathematical model is derived and solved using machine learning techniques. The machine learning model is carried out using physics-informed machine learning. The constructed results such as velocity and temperature profiles are plotted against the normal to the plate. Also, the heat transfer and fluid friction against the boundary are interpreted using Nusselt number and skin friction coefficient, respectively. The results are examined under the variations of dimensionless parameters such as Eckert number, Grashof number, volume fraction of the hybrid nanoparticles, suction, magnetic, and radiation parameters. The study shows that the use of Casson hybrid nanofluid with a higher volume fraction of nanoparticles is good in heat transfer applications.