Abstract
The goal of this study is to analyze the heat transfer characteristics of a Casson hybrid nanofluid flow system by incorporating EMHD (electromagnetohydrodynamic) and entropy production using the stochastic solver and Levenberg-Marquardt backpropagation neural networks. Self-similarity transformations are used to turn the mathematical model of a system of partial into a series of ordinary differential equations. The datasets are built through Runge–Kutta fourth order with a shooting approach, which aids in the generation of a continuous neural network mapping. To grasp the neural network mapping, the validation, training and testing processes are used to estimate the solutions of various physical constraints. For this hybrid model, the determination, convergence, verification and stability of Levenberg-Marquardt backpropagation neural network mappings are evaluated using regression-based statistical analysis, mean squared error and error histograms. To acquire high possibilities in drug delivery systems, hybrid nanofluids are regarded as blood-based nanofluids with dispersion of hybrid ( and
) nanoparticles. Due to this reason, this model might be beneficial for medication delivery systems, particularly if pharmaceuticals are injected into the bloodstream.
Disclosure statement
No potential conflict of interest was reported by the author(s).