A design of neural networks to study MHD and heat transfer in two phase model of nano-fluid flow in the presence of thermal radiation

The main objective of research is to design a numerical computational solver based two-layers structure of backpropagation Levenberg-Marquardt scheme with artificial neural networks i.e. BLMS-ANN for the analyses of the MHD effects on thermal radiation in a two phase model (MHD-TRTM) of nano-fluid flow with heat transfer between two horizontal rotating plates through varying involved parameters including the Reynolds number, radiation parameter, magnetic parameter, rotation parameter, thermophoretic parameter, and the Schmidt number for various scenarios. The MHDTRTM model is mathematically formulated as system of PDEs that are converted in desire system of ODEs by means of suitable transformation. Software tools are used to simulate numerical behavior. The data-sets are constructed by Homotopy analysis method (HAM) technique that are exploited as a target dataset for the learning of BLMS-ANN based on the process of validation, training and testing to determines the solution of MHD-TRTM model for various physical scenarios. Validation, convergence, stability and verification of BLMS-ANN for solution predictive strength of the MHD-TRTM problem are certified in terms of achieved accuracy, regression index measurements, and analysis of error histogram illustrations. With a level of accuracy ranging from 10−10 to 10−12, the recommended approach is distinguishable from the proposed and reference outcomes.