Estimate the shear rate & apparent viscosity of multi-phased non-Newtonian hybrid nanofluids via new developed Support Vector Machine method coupled with sensitivity analysis

The Support Vector Machine method is employed to predict the thermo-physical properties of hybrid nanofluid composed of TiO2 and ZnO nanoparticles and ethylene glycol as the base fluid at different temperatures, shear rates, and nanoparticle volume fractions. The present work novelty is to use a new sensitivity analysis based on this method which has been widely utilized in the regression and function approximation fields. In addition, the SVM method advantages are its unique solutions, high accurate outcomes at the training data points, and appropriate generalization. Regarding the obtained results, effects of different mentioned working conditions on apparent viscosity and shear stress have examined besides the highest values of sensitivities and pertinent independent parameters are reported. The new statistical/ math proposed model can predict the apparent viscosity, shear stress and shear rate of TiO2/ZnO/EG non-Newtonian hybrid nanofluid which implies its suitable performance for multi-phase non-Newtonian fluids. (C) 2019 Elsevier B.V. All rights reserved.