Artificial neural network and numerical analysis for performance enhancement of hybrid microchannel-pillar-jet impingement heat sink using Al2O3-water and CuO-water nanofluids

Abstract

This study investigates the effects of nanofluids and its particle volume fractions on performance enhancement of hybrid microchannel-pillar-jet impingement heat sink. The nanofluids used are Al2O3-water and CuO-water. A three-dimensional numerical model is applied to determine the fluid flow and heat transfer performance parameters such as pressure drop, temperature distribution, heat transfer coefficient, pumping power, and thermal resistance. Fluid flow through the jets and channel is assumed to be incompressible steady and laminar for a small range of low Reynolds number (Re ≤ 1000). The results obtained using nanofluid as the working fluid are compared with pure water, which shows that the performance is significantly increased using nanofluids as the heat transfer coefficient is increased and temperature-rise is reduced, however, the pumping power requirement is elevated. Moreover, CuO-water nanofluid provided better thermal management than Al2O3-water, and it is further improved with the increase in particle volume fractions in the base fluid. A trade-off between thermal resistance and pumping power has been obtained and discussed in view of energy consumption and capacity. Empirical correlations and artificial neural network model are developed to predict heat sink performance using water and Al2O3/CuO-water nanofluids as coolant. The model predictions are found within 15% deviation from the numerical data for nanofluids with particle volume fraction from 2-10% and Reynolds number from 100-1000.