Illustration of low oscillating magnetic field on sodium alginate-based hybrid nanofluid flow between two revolving disks: An artificial neural network-based study

The application of hybrid nanofluids is pivotal in improving heat transfer, and aiding in the thermal regulation of electronic devices, heat exchangers, and cutting-edge manufacturing processes. The integration of hybrid nanofluids and artificial neural networks enhances the development of smart cooling systems, optimizing thermal regulation. This study examines the flow of magnetite $\left(F{e}_3{O}_4\right)$ and graphene ($Gr$) nanoparticles with a base fluid sodium alginate over a porous coaxial disk in a low oscillating magnetic field in conjugation with thermal radiation. The Runge-Kutta 4th order scheme is operated to analyze the characteristics of the flow fields followed by the suitable transformation used for the conversion of dimensional governing equations to their corresponding non-dimensional form. In a novel way, this research utilizes artificial neural networks (ANN) to design an effective predictive model for thermal transfer rate, comparing its performance with response surface methodology (RSM) and validating the results through analysis of variance (ANOVA). The outcomes are the velocity distribution decreases with a higher magnetization parameter and a large Reynolds number results in a higher velocity. Nusselt number achieves it most efficient thermal transfer rate at 88 epoch with gradient of $9.9805\times {10}^{-8}$.