Thermal performance, entropy generation, and machine learning insights of Al₂O₃-TiO₂ hybrid nanofluids in turbulent flow.

Abstract

This study investigates the heat transfer performance of water-based Al₂O₃-TiO₂ (50:50) hybrid nanofluids under turbulent flow conditions. Al₂O₃ and TiO₂ nanoparticles (13 and 21 nm, respectively) were dispersed in water to prepare the nanofluids in the concentrations range of 0 to 1 vol%. Thermal conductivity and viscosity are measured in the temperature range of 30 to 60^oC for the prepared nanofluids. Experimental and numerical analyses explored the effect of concentration and Reynolds number on Nusselt number, entropy generation, and friction factor. The results demonstrate that maximum viscosity enhancement of 15.77 and 14.76% is observed for 1 vol% of hybrid nanofluid and Al₂O₃ nanofluid compared to base fluid at 30 ^oC, respectively. The maximum Nusselt number enchantment is 70.4% for 1 vol% of hybrid nanofluid compared to the water. Similarly, hybrid nanofluids achieved a remarkable reduction in total entropy generation of 46% in contrast to the base fluid. New correlations are proposed to predict both the Nusselt number and friction factor for hybrid nanofluids. Furthermore, employing machine learning techniques, highly accurate models are developed. These findings highlight the promising role of hybrid nanofluids in achieving efficient thermal management in various applications.