Neuro-computing networks for entropy generation under the influence of MHD and thermal radiation

Abstract In this research article, artificial neural networks back-propagated with Levenberg Marquardt scheme (ANN-BLMS) is presented to analyze the entropy generation of carbon nanotubes (CNTs) between two rotating stretching discs under the influence of thermal radiation and magneto-hydrodynamic nano-fluid flow model. (MHD-NFM). The fluid flow is initially represented by system of PDEs is then transformed into system of ODEs. A set of data for proposed ANN-BLMS is generated for various scenarios by variation of stretching parameters of lower and upper disks A1 and A2 respectively, suction injection parameter (Ws), rotational parameter (Ω), the radiation parameter (R), the Eckert number (Ec), the Hartmann number (M) by using Adams Numerical method. The approximate solution of different cases is determined by testing, training and validation process of ANN-BLMS and comparison for verification of correctness of proposed model. Later on, regression analysis, mean square error and histogram studies endorse the performance of proposed ANN-BLMS.