Publications

Abstract : This article discussed the magneto-hydrodynamic Darcy–Forchheimer second-order velocity slip flow model (DFSO-VSFM) through employing the scheme of Levenberg–Marquardt with the procedure of backpropagation in neural networks (SLM-BNNs). Titanium dioxide and graphene oxide are the nanomaterials, where the base liquid is water. The governing PDEs are transformed into nonlinear ordinary DEs by utilizing appropriate transformations. Dataset used as a reference is computed using HAM for various scenarios. The absolute error graphs and solution plots for various variables are evaluated in MATLAB using this reference dataset. Performance of SLM-BNNs is validated through error histogram, regression analysis and MSE results. All through the computation, the target and output values are perceived to have R = 1 as a regression value of correlation. The solution of DFSO-VSFM is examined through the testing, validation and training processes. The influence of flow variables on velocity field, temperature distribution, concentration distribution, skin friction coefficients, Bejan number, Nusselt numbers and entropy generation are discussed graphically and in tabular form. Absolute error graphs are also depicted in the following figures, where the absolute error values for various scenarios are 10−4 to 10−8 , 10−4 to 10−7,10−1 to 10−5 , 10−3 to 10−7 , 10−2 to 10−10 , 10−4 to 10−8 , 10−4 to 10−7 , and 10−5 to 10−8.