In this paper, numerical study and thermal prediction for a nanofluid jet impingement cooling of an isothermal hot surface with an adiabatic rotating cylinder were performed. Finite volume method was used for the solution of resulting governing equations along with the boundary conditions. Influence of various pertinent parameters such as Reynolds number (between 100 and 400), angular rotational velocity of the cylinder (between -0.1 and 0.1), horizontal location of the cylinder (between 0 and 3.75w) and solid particle volume fraction (between 0 and 0.04) on the fluid flow thermal characteristics were examined. It was observed that cylinder rotation and its location affect the cooling performance of the hot surface. It can be used as control element for heat and fluid flow. At the highest angular rotational speed as compared to motionless cylinder case, average Nusselt number reduces by about 20.16% for clockwise rotation. Solid particle addition to the base fluid affects the variation of first and secondary peaks in the Nusselt number along the hot wall. At the highest solid when the cylinder is away from the inlet slot and average Nusselt number enhancement is by about 8.08% at the highest volume fraction. An efficient modeling strategy was developed based on proper orthogonal decomposition and radial basis neural networks for thermal predictions. Accurate and fast results were achieved as compared to high fidelity computational fluid dynamics simulation results. (C) 2018 Elsevier Ltd. All rights reserved.