Neuro-computing intelligent networks for entropy optimized MHD fully developed nanofluid flow with activation energy and slip effects

In this paper, the main focus of this research is to represent an intelligent computing model through an artificial backpropagated Levenberg-Marquardt neural network (ABP-LMNN) for entropy optimized magnetohydrodynamic fully developed nanofluid flow with slip and activation energy effects. In mathematical modeling, dimensionless non-linear ODEs represent the magnetohydrodynamic nanofluid flow model (MHD-NFM). A reference dataset of ABP-LMNN is constructed for diverse situations of MHD-NFM by discrepancy of parameters. The attained reference dataset (RD) is randomly utilized for validation, testing and training processes for ABP-LMNN are employed to examine the approximate solution of MHD-NFM is demonstrated by comparison of outcomes. The authentic performance of the ABP-LMNN is validated through accuracy in the phrase of error histogram, mean square error and regression learning. The thermal and solutal parameters upsurge both the thermal and the concentration gradients. Moreover, the velocity profiles are declined owing to an increase in the second-order slip parameter in the tangential direction of the flow. [Display omitted] •Here entropy optimized magnetohydrodynamic fully developed nanofluid flow is examined.•Slip and activation energy effects are considered.•Artificial backpropagated Levenberg-Marquardt neural network (ABP-LMNN) method is used for the numerical results.