Machine learning based analysis of heat transfer in tangent hyperbolic fluid at heat generating magnetized surface

An artificial intelligence-based study is conducted to examine the heat transfer in the tangent hyperbolic fluid using a stretchable magnetized surface. To be more specific, heat and mass transfer are considered in the flow regime of tangent hyperbolic fluid. The magnetic field is applied externally. The heat generation, velocity, and thermal slip effects are considered. The flow is formulated in terms of coupled non-linear PDEs. Lie groups of transformations are constructed to reduce the order of PDEs. The reduced equations are solved by using the shooting method. The impact of the Weissenberg number, power law index, Prandtl number, and heat generation parameter is evaluated on the heat transfer coefficient by using artificial intelligence. 88 samples are divided at random into training 62 (70 %), validation 13 (15 %), and testing 13 (15 %). The hidden layer contains 10 neurons. The Levenberg-Marquadt backpropagation algorithm is used to train the model. The developed model is evaluated by mean square error and regression analysis. According to ANN anticipated values, the heat transfer coefficient shows decreasing trends towards higher values of Weissenberg number, power law index, and heat generation parameter. The tangent hyperbolic fluid temperature admits an increase for heat generation and magnetic field parameter while the opposite is the case for thermal slip parameter