LAUSR

Abstract The objective of this study is to offer a suitable nano-lubricant (engine oils containing nanoparticles) to use in light-duty automotive industries in order to reach a higher ability and efficient oil in comparison to ordinary engine oils, in order to reduce cold start engine damages. Therefore, in present study a feasibility study of using a new nano-engine oil containing a combination of MWCNT (multi wall carbon nanotubes)-ZnO nanoparticles with the ratio of 30–70% has been arranged. Results of experimental study show a considerable decrease in viscosity of nano-engine oil (in comparison to viscosity of pure 5W50 oil) after adding 0.05% and 0.1% nanoparticles to 5W50. This viscosity reduction, reduces the damage caused by starting up the engine in cold start condition. In order to predict the viscosity of this applied nano-engine oil (obtained from experimental studies), the efficiency of using a mathematical correlation using response surface methods (RSM) was investigated for viscosity prediction. For the proposed correlation, R2 is equal to 0.9715 that shows its acceptable accuracy. An artificial neural network has also been used as the second method to predict viscosity in the range of 5 °C–55 °C and in solid volume fractions of 0.05%–1%. Selected structure of Artificial Neural Network with R = 9.999e−01 is the most optimal and precise structure among 100 studied structures.