Investigation of heat and mass transfer for an MHD hybrid nanofluid flow over a rotating disk: a numerical investigation with supervised neural network

The key intention behind this research work is to investigate the heat and mass transfer phenomenon in the flow of hybrid nanofluid over a revolving disk with magnetohydrodynamic effects under consideration (MHD-HNRD). The flow model of the problem has been formulated in the form of PDEs which are further transmuted into ODEs with proper transforming equations. The influence of Magnetic parameter, Brinkman number, Prandtl number, Rotation parameter, and Concentration of the nanoparticles on velocities and thermal profile has been studied and reference data sets for the MHD-HNRD model are computed with the help of the renowned Lobatto-IIIA solver by varying these parameters. This reference solution is then subjected under the proposed solution methodology based on Levenberg-Marquardt Supervised Neural Networks (LMSNNs). An increase in heat transfer rate has been observed with the larger magnetic force and rotation rate. Graphical and numerical results based on mean-square errors (MSEs), time series response, error distribution plots, and regression plots endorse the precision, validity, and consistency of the proposed solution methodology. MSEs up to the level of 10-12 and performance learning curves predict the smoothness of the curves with the number of epochs confirming the accuracy, convergence, and validity of the designed solver.