Prediction the dynamic viscosity of MWCNT-Al2O3 (30:70)/ Oil 5W50 hybrid nano-lubricant using Principal Component Analysis (PCA) with Artificial Neural Network (ANN)

Abstract

In this study, the prediction of dynamic viscosity (µnf) of MWCNT-Al2O3 (30:70)/ Oil 5W50 hybrid nano-lubricant using Artificial Neural Network (ANN) is performed. The objective of the present research is to investigate the effect of temperature and solid volume fraction (SVF) to predict the shear rates (SR) and µnf using ANN. The feed-forward ANN consists of a multilayer perceptron network (MLP), which is capable of predicting µnf in connection with experimental data of temperature, SR and SVF. Sensitivity analysis is used to evaluate the importance and role of temperature, SR, and SVF in experimental µnf variations. ANN is generated and tested with experimental data sets and the results show that there was a good agreement between the actual and predicted ANN values. Moreover, the results of ANN simulation are compared with other data processing methods such as Support Vector Machine (SVM), Partial Least Squares (PLS), Principal Component Regression. In addition, the results of the residual value of ANN with seven neurons for µnf can be very small and close to the expected normal value. From this, it can be concluded that the given model can expect exact values.

Keywords

• Dynamic viscosity
• Hybrid nano-lubricant
• ANN
• Support Vector Machine (SVM)
• Partial Least Squares (PLS)
• Principal Component Regression