An intelligent approach to predicting the effect of nanoparticle mixture ratio, concentration and temperature on thermal conductivity of hybrid nanofluids

Abstract

Hybrid nanofluids are better heat transfer fluids than conventional nanofluids because of the combined properties of two or more nanoparticles. In this study, the thermal conductivity of Al2O3–ZnO nanoparticles suspended in a base fluid of distilled water is investigated. The experiments were conducted for three mixture ratios (1:2, 1:1 and 2:1) of Al2O3–ZnO nanofluid at five different volume concentrations of 0.33%, 0.67%, 1.0%, 1.33% and 1.67%. X-ray diffractometric analysis, X-ray fluorescence spectrometry and scanning electron microscopy were used to characterise the nanoparticles. The highest thermal conductivity enhancement achieved for Al2O3–ZnO hybrid nanofluids with 1:2, 1:1 and 2:1 (Al2O3:ZnO) mixture ratios was 36%, 35% and 40%, respectively, at volume concentration 1.67%. The study observed the highest thermal conductivity for Al2O3–ZnO nanofluid was achieved at a mixture ratio of 2:1. A “deeping” effect was observed at a mixture ratio of 1:1 representing the lowest value of thermal conductivity within the considered range. The study proposed and compared three models for obtaining the thermal conductivity of Al2O3–ZnO nanofluids based on temperature, volume concentration and nanoparticle mixture ratio. A polynomial correlation model, the adaptive neuro-fuzzy inference system model and an artificial neural network model optimised with three different learning algorithms. The adaptive neuro-fuzzy inference system model was most accurate in forecasting the thermal conductivity of the Al2O3–ZnO hybrid nanofluid with an R2 value of 0.9946.