Stochastic analysis through Levenberg Marquardt backpropagation neural networks for radiative Carreau nanofluid flow subject to chemical reaction

The aim of this research work is to estimate and analyze the solution of rheological chemical reactive Carreau nanofluid (CNRFM) induced by exponentially extended surface (EES) subject to variable physical attributes by using stochastic analysis on Levenberg Marquardt backpropagation neural networks (SALMBNNs). The non-linear Partial Differential Equations (PDEs) are transformed by using the similarity transformation variables into their corresponding ODEs. The reference values are created with ARK (adaptive Runge-Kutta) scheme. The ensuing results are explained for the variable viscosity, Weissenberg number (material number), Brownian movement factor, LRF (local rotation factor), LN (Lewis number) and activation energy with chemical reaction in addition. Numerical calculations of different physical quantities are approximated with artificial intelligence based SALMBNNs from dataset created with ARK method. The convergence, accuracy, and efficiency of the proposed stochastic analysis on Levenberg Marquardt backpropagation neural network (SALMBNNs) are established and endorsed through iterative learning curves at each incremental step in epoch, statistical instance distribution studies of error-histograms, analysis of adaptive controlling parameters of SALMBNNs, and evaluation of regression metric.