Viscosity of Alumina Water-Based Nanofluids Modeling by Artificial Neural Network

Objectives: To develop efficient models for the prediction of viscosity of nanofluids. Methods/Statistical Analysis: Artificial Neural Network (ANN) toolbox for Matlab and an experimental data set of effective viscosity of alumina waterbased nanofluids are used. ANN is mathematical model of artificial intelligence product, inspired by the structure and functioning of the human nervous system. Experimental data set are divided into two groups: train and test. The train instructed ANN and the results were compared with the test. Findings: ANN viscosity results were compared with the experimental data points. The expected values were in excellent agreement with the measured ones, viewing that the developed model is accurate and has the great ability for predicting the viscosity. 0.9994 and 0.9998 are the values of R2 linear regressions for training and testing data set, respectively and 2.7187*10-4 and 1.2461*10-4 are respective values of mean square errors. Applications: Artificial Neural Networks to model thermal characterization of nanofluids.