Modeling and prediction of rheological behavior of Al2O3-MWCNT/5W50 hybrid nano-lubricant by artificial neural network using experimental data

In this paper, the artificial neural network model and new correlation based on experimental data are proposed to predict Rheological behavior of Al2O3-MWCNT/SW50. The ANN model has three inputs including temperature, volume fraction and share rate. Predictions of suggested models were evaluated by using statistical and graphical validations approaches. The results revealed that the maximum values of margin of deviation are 0.07% and 7.3% for ANN and correlation outputs, respectively. The findings showed that an artificial neural network can predict the relative viscosity of the nanofluid more accurately than empirical correlation. (C) 2018 Published by Elsevier B.V.