Estimation of the Mixed Convection Heat Transfer of a Rotating Cylinder in a Vented Cavity Subjected to Nanofluid by Using Generalized Neural Networks

In this study, numerical investigation of mixed convection in a square cavity with ventilation ports filled with nanofluids in the presence of an adiabatic rotating cylinder is conducted. The governing equations are solved with a commercial finite element code (COMSOL). The effects of Grashof number (Gr = 103 to Gr = 105), Reynolds number (Re = 50 to Re = 300), nanoparticle volume fraction (φ = 0 to φ = 0.05), and cylinder rotation angle (Ω = −5 to Ω = 5) on the flow and thermal fields are numerically studied for a range of different parameter sets. The generalized neural network (GRNN) is used to predict the thermal performance of the system. It is observed that the heat transfer increases almost linearly with increasing the nanoparticle volume fraction. The increasing rotation angle in the clockwise direction generally enhances the heat transfer. Moreover, the validation results with artificial neural networks show that generalized neural nets show better performance compared to radial basis and feed-forward networks.