Numerical and artificial neural network analysis of swirling Reiner–Rivlin fluid flow over thermally radiative stretchable disk

Source abstract

Abstract An incompressible time‐independent Reiner–Rivlin fluid (RRF) flow over a rotational stretching disk is considered. The laminar axi‐symmetric and thermal flow streams are influenced through thermal radiation and melting heat transfer. Self‐similar solution is assumed for the non‐Newtonian fluid flow problem by converting it into a dimensionalized system of ordinary differential equation through similarity transformations. The resultant system is assessed numerically using RKF‐45 (Runge–Kutta–Fehlberg) built‐in procedure with Maple software. The obtained physical factors are analyzed against flow and thermal fields. Furthermore, skin‐friction co‐efficient, moment torque, volumetric flow rate, and heat transfer rate are calculated in view of radial stretching and cross viscosity. A feed‐forward multi‐layer neural network with back propagation technique of Artificial Neural Networks (ANNs) is implemented on the numerical results obtained the physical quantities. The results indicate that radial velocity field is reduced by the Reiner–Rivlin parameter and is modified through stretching parameter at the disk surface. The magnitude of axial velocity is declined for larger values of melting parameter. It is also seen that the current ANN model is accurate, dependable, and convergent by measuring outputs.