Abstract
Recently, the suspension of hybrid nanoparticles in conventional fluids has been investigated as a technique for improving the thermophysical properties of nanofluids. The dearth of documentation on the trio influence of volume concentration, base fluid, and temperature on the electrical conductivity and viscosity of hybrid alumina-ferrofluids [Al2O3-Fe2O3 (25:75 mass%)] has led to this study. The effective viscosity and electrical conductivity of the deionized water (DW)-based and ethylene glycol (EG)-DW-based (50:50 vol%) hybrid alumina-ferrofluids were measured at temperatures of 20-50 degrees C and volume concentrations of 0.05-0.75%. Based on the importance of soft computing methods to engineers, adaptive neuro-fuzzy inference system (ANFIS) and artificial neural network (ANN) were used for predicting the relative viscosity and electrical conductivity of the two types of hybrid ferrofluids. The measured data for viscosity and electrical conductivity were used in the modeling. Model performances were evaluated using the root mean squared error index. Viscosity was enhanced by 3.23-43.64% and 2.79-49.38%, while electrical conductivity was increased by 163.37-1692.16% and 717.14-7618.89% for the DW- and EG-DIW-based hybrid ferrofluids, respectively, compared with the respective base fluids. Increasing volume concentration augmented the viscosity and electrical conductivity of all the hybrid alumina-ferrofluids, whereas a rise in temperature enhanced their electrical conductivity and detracted the viscosity. DW-based hybrid alumina-ferrofluid was observed to have a lower viscosity and higher electrical conductivity than the EG-DW-based counterpart. The results showed that the optimum ANN and ANFIS models have a maximum error of less than 4.5% and 3.9% for relative viscosity and electrical conductivity, respectively, which were lower than those proposed using regression analysis. With the hybrid alumina-ferrofluids possessing a lower viscosity relative to single-particle ferrofluids, they are recommended for engineering application.