Advanced ANN computational procedure for thermal transport prediction in polymer-based ternary radiative Carreau nanofluid with extreme shear rates over bullet surface

Bullet surface has a significant role in many engineering and industrial sectors, due to its wide fluid-based thermal management systems. The current approach emphasizes heat transfer mechanism in flow of ternary hybrid nanofluid over a bullet shape geometry. The integration of infinite shear rate viscosity-based model of Carreau explored the predictive capabilities of enhanced heat transport in ternary hybrid nanofluid. The purpose of the study is to seek an advanced predictive model that accurately captures the thermal prediction in ternary hybrid nanofluid under varying conditions of shear rate. By utilizing artificial neural networks (ANNs), the aim of this study is to simulate and analyze how these fluids respond to the combined effects of viscous dissipation, non-uniform heat sink source, thermal radiation, and infinite shear rate viscosity when interacting with bullet-shaped geometry. The physical model initially generated a set of partial differential equations, based on assumption in this study, and then this system is converted into ordinary differential equations (ODEs) using similarity transformations. This conversion simplifies the system into a more manageable form. The resulting ODEs are then numerically solved using the bvp4c method. The solutions obtained from this process are compiled into a dataset, which is then used to train through ANN. This neural network is designed to predict advanced solutions. The increase in velocity magnitude increases for stretching ratio and infinite shear rate parameter while it decreases for location parameter and velocity slip parameter. On the other hand, temperature profile decreased with augmentation in the numeric values of radiation parameter and Eckert numbers while it demonstrates the opposite trend for heat generation number and magnetic parameter. The rate of temperature increment is highest in ternary hybrid nanofluids compared to nanofluids and hybrid nanofluids.