Application of Artificial Neural Networks (ANNs) for Predicting the Viscosity of Tungsten Oxide (WO3)-MWCNTs/Engine Oil Hybrid Nanofluid

Abstract

In the current work, the viscosity of hybrid WO3-MWCNTs/Engine Oil nanofluids for the different volume fraction of nanoparticles (φ), temperatures, and shear rates has been measured, and then using Artificial Neural Network (ANN) has been predicted. The statistical reports of ANN have been presented in figures and tables, and it could be found that the designed ANN has an acceptable accuracy in predicting the viscosity of hybrid nanofluids. It can be seen that the optimum neuron number for this data set is 39 and the ANN outputs are very close to the experimental data points. Finally, it is found that in the fitting method, the correlation between experimental data points and the output values is 0.999 424, but in the ANN method, the correlation was 0.999 837 454.