Recent advances in the prediction of thermophysical properties of nanofluids using artificial intelligence

Recently, artificial intelligence (AI) methods have been widely welcomed due to the weakness of traditional regression-based methods and their low accuracy in nonlinear problems related to the study of thermophysical properties of nanofluids. In recent years, various investigations have been devoted to applying AI in estimating the thermophysical properties of nanofluids and energy applications, most of which have focused on single nanofluids. The AI-based investigations on thermophysical properties of nanofluids demonstrated that most of the applications of machine learning and data-driven models are related to thermal conductivity and viscosity of mono fluids, and limited research has been conducted to model hybrid nanofluids. Given the increasing capabilities of AI methods and their integration with robust optimization algorithms, it can be hoped to solve nonlinear problems of hybrid nanofluids with many input variables to achieve promising results. In this direction, this chapter provides an overview of the recent advances in AI for predicting thermophysical properties of nanofluids.