Heat transfer rate in Falkner–Skan fluid flow of ZnO-EG over a moving wedge: Intelligent backpropagated neural networks

In this study, the Falkner–Skan stream (FSS) of ZnO-EG across a moving wedge is investigated using artificial neural networks backpropagated using a Bayesian regularization strategy (ANN-BRS). The PDEs of the Falkner–Skan are converted into a set of ordinary differential equations (ODEs). The reference dataset is created using the mathematical solver in Mathematica and includes moving wedge boundaries, radiation boundaries, nanoparticle volume division boundaries and Falkner–Skan power-regulation boundaries for all proposed scenarios (ANN-BRS). Examined is the effect of practical cut off points on the stream field and temperature profiles, including the radiation limit, moving wedge limit, nanoparticle volume division and Falkner–Skan power. When nanoparticles are present, viscosity and heat conductivity rise, which can be physically explained. Increases in Falkner–Skan power have both positive and negative effects. When the settings are adjusted to their highest values, the maximum rate of heat transmission is realized. The effects of radiation boundary and so on are identical due to the linear augment. As a result, as the parameters are increased, the rate of heat transmission increases. Based on the intended data points that were obtained, the estimated answer is calculated for every scenario utilizing the testing, training and validation procedures. The mean square error data (MSE), error histogram and regression analysis are used to validate the performance of (ANN-BRS).