MHD williamson nanofluid flow in the rheology of thermal radiation, joule heating, and chemical reaction using the Levenberg–Marquardt neural network algorithm

Various studies have been conducted on the topic of predicting thermal conductivity nanofluids. Here, nanofluids is determined using artificial neural networks since this approach rapid and accurate, as well cost-effective. To forecast magnetohydrodynamic Williamson flow through a vertical sheet, feed-forward network with various numbers neurons has evaluated, best based performance selected. The fluid model incorporates effects Joule heating, heat generation absorption, radiation, chemical reaction (MHD-WNF-HGA). A combination radiation reactive species improves energy solute profiles. magnetic Reynolds number assumed to be so small; therefore, generated field no effect. postulate similarity variables used convert physical in form nonlinear partial differential equations an ordinary equation system. supervised Levenberg–Marquardt backpropagation algorithm possesses multilayer perceptron that for training network, which one top algorithms machine learning. bvp4c numerical technique adopted build datasets construction continuous mapping. Flow, energy, concentration profiles fluidic are constructed by adjusting several quantities such parameter, Eckert number, Darcy Brownian motion, thermophoresis parameter. Analytical techniques error histogram graphs regression-based statistical examine accuracy suggested method. It found mappings’ derivation, convergence, authentication, consistency proven. Furthermore, assists distribution increase smoothly. Fluid velocity drops whereas impact enhances strength nanofluid density.