Neural network thermal analysis of electromagnetic water and blood based nanofluids flow in a channel with suction

In this paper, a Bayesian Regularization Artificial Neural Network (ANN-BR) assisted numerical investigation is conducted for two nanofluids flow through a suction channel including a magnetic field effect are developed. Three different nanoparticles, namely copper (Cu), aluminum oxide (Al₂O₃), and titanium oxide (TiO₂), dispersed in water and blood as base fluids, are considered. In this framework, we analyze the interplay of base fluid with nanoparticles, and represent the rate at which momentum and heat transfer between the two phases precisely. Next, the associated governing nonlinear equations are converted into a dimensionless form with the application of similarity variables, which then the coupled equation set is solved numerically using the Bvp4c so that we can evaluate the parameters. The study reveals that when the magnetic parameter is elevated, the velocity field is altered: for one thing, the flow velocity within the channel core is enhanced, and the other thing is, the flow velocity at the wall is reduced due to Lorentz force. The separate ANN-BR models have been formed to determine the skin-friction coefficient and Nusselt number from the obtained simulation data. Blood based nanofluid showed greater performance regarding heat transfer than water-based nanofluid in the frame work utilized here. These calculated results also agree very closely with those previously published. This provides a large measure of credibility and validation to the results presented here. This model results that are crucial for the development and enhancement of biomedical transport applications, smart delivery of therapeutics, and cancer hyperthermia treatments.