Free convection effect on oscillatory flow using artificial neural networks and statistical techniques

In the present paper, we present the estimated impact of the free convection on oscillatory flow using data to predict the value of the attributes to promote concentration for mass transfer. Various statistical methods can predict concentration of free convection oscillatory flow. We provide exact analytic expressions for the underlying concentration profile using the regular perturbation technique and decision tree method. Moreover, we show how these expressions can help decrease the governing equations to some uncoupled and coupled linear ordinary differential equations. Results were obtained concerning the concentrations constructed and plotted graphically, taking into consideration three values of the time. We then focused on the statistical technique used to analyze the neural network problem of predicting the concentration of free convection oscillatory flow. Both Decision Tree and MLP were used as the best models according to the value of sMAPE, MASE, and MAPE. The obtained results were compared with the corresponding estimation in Abo-Dahab and Hatem (2020), which matched and satisfied the boundary conditions of the free convection impact on the oscillatory flow. Finally, the exact analytical results that have several applications in chemical industries, medicine, and engineering were concluded. (C) 2020 The Authors. Published by Elsevier B.V. on behalf of Faculty of Engineering, Alexandria University.