Numerical analysis of Cattaneo–Christov heat flux model over magnetic couple stress Casson nanofluid flow by Lavenberg–Marquard backpropagated neural networks

The purpose of study is to present a new nanofluidic model and its solution that describes steady couple stress magnetic Casson nanofluid flow in the presence of Cattaneo-Christov heat flux (CMCN-CCHF) system past a shrinking plat via aesthetics stochastic computing procedure based on supervised learning Lavenberg-Marquard technique as a backpropagated neural networks (LMT-BNNs). Physical quantities involve in the fluidic system i.e., magnetic parameter, Casson parameter, couple stress parameter, thermal relaxation coefficient and Prandtl number are assessed by suitable variations, which satisfy the dynamics of presented fluid model effectively. These are exploited for the construction of dataset for LMT-BNNs through a well-known deterministic homotopy analysis method. Testing, training and validation processes of LMT-BNNs are incorporated for the analysis of the CMCN-CCHF model. Accuracy of the proposed model through the statistical approaches ensures the reliability of newly developed LMT-BNNs computational intelligence solver. The reliability, stability and preciseness of the proposed LMT-BNNs for the solution of CMCN-CCHF via variety of statistical measures such as mean square error (MSE), error histogram, performance procedures, and regression/correlation curve fitting graphs are shown. Casson fluid parameter causes to enhance the skin friction profile but the presence of magnetic parameter retards the skin friction and local Nusselt number.