Scientific computing of radiative heat transfer with thermal slip effects near stagnation point by artificial neural network

This article employs an artificial neural network technique to approximate the solution for the stagnation point flow with velocity and thermal slip effects, as well as radiative heat transfer. The PDE governing system is transformed into a set of coupled ordinary differential systems by incorporating similarity variables. The shooting method is used to obtain a dataset in Mathematica. To test the precision of the suggested model, the operations of training, testing, and validation are performed, and the results are compared to a reference dataset. The Levenberg-Marquardt backpropagation neural network model is utilized to solve the system of equations under different scenarios, and its output is evaluated using mean square error, state transition dynamics, error histograms analysis, and regression illustrations. The findings indicate that the neural network approach achieves a high level of accuracy when predicting thermal analysis. Moreover, the current Artificial Neural Network model has an advantage over other numerical techniques in that it can handle more intricate mathematical models while minimizing the resources required for problem-solving.