Soft computing paradigm for heat and mass transfer characteristics of nanofluid in magnetohydrodynamic (MHD) boundary layer over a vertical cone under the convective boundary condition

In this article, the Levenberg-Marquardt back propagation technique under the framework of neural networks (LMBT-NN) is incorporated. The heat and mass transfer characteristics of nanofluid (HMT-CNF) in a magneto-hydrodynamic (MHD) boundary layer over a vertical cone under convective boundary conditions have been investigated. The similarity transformation has been employed with the goal of transforming nonlinear partial differential equations (PDEs) into the system of ordinary differential equations (ODEs). A set of suggested data (LMBT-NN) is produced for some scenarios by modifying the radiation parameter (R), magnetic field parameter (M), buoyancy ratio parameter (Nr), chemical reaction parameter (Cr), thermophoresis factor (Nt), Lewis number (Le), and Brownian motion factor (Nb) within the applicability of the state-of-the-art Adams numerical technique. Using the (LMBT-NN) training, testing, and validation technique, the approximate solution of distinct instances has been validated, and for excellence, the proposed model has equated. To justify the proposed methodology (LMBT-NN), different error plots and numerical illustrations based on mean square errors, histogram plots, and regression analysis representations are prepared. With a correctness level ranging from E-9 to E-10, the recommended method has been observed based on the closeness of the suggested and reference outcomes.