Development of knowledge management for viscosity of nanolubricant in hot and cold lubrication conditions

Abstract

The optimization and prediction of SAE40 based nanofluid (NF) viscosity with a knowledge management approach have been discussed using Response surface methodology (RSM). The used data are independent variables (SR,T,SVF) and viscosity as the response or the dependent variable. For this purpose, in the first step, six RSM models were analyzed compared to select a model accurately in ANOVA terms and statistical plots and Fifth model with values of R2 = 0.9832, R2-adjusted= 0.9768 and Margin of Deviation −17 <MOD< +15 has the most accurate among the selected models. The findings have shown that, according to the correlation relationship presented based on the selected model, behavior of NF is non-Newtonian. Also, in interaction between the independent variables, NF viscosity gradually decreases with increase in T and gradually increases with the increase in SVF. Thus, the highest viscosity (in SVF= 0.0625 % and T = 25 °C) is 497 mPa.sec and lowest value (in T = 50 °C and SVF=0.125 %) is 69.7 mPa.sec. In optimization section, to find the most optimal viscosity in hot and cold temperature conditions with the lowest amount of ZnO nanoparticles (NPs) use, the viscosity value for two cases was obtained as 202.07 mPa.sec and 190.36 mPa.sec, respectively. Also, the implementation of knowledge management steps in NFs, in addition to collecting information, storing and accessing various resources, brings a wide range of benefits to researchers and industrialists.