Using artificial neural network to predict thermal conductivity of ethylene glycol with alumina nanoparticle

Abstract

The correlations of thermal conductivity of alumina nanoparticle dispersed in pure ethylene glycol were proposed by neural network modeling using experimental data. The required input and target data have been taken from the experimental measurement to train artificial neural network (ANN). The temperatures were changed within 24–50 °C. Levenberg algorithm was used to train the ANN. Results showed that the thermal conductivity of nanofluid had a significant increase with increasing solid volume fraction of nanoparticles. The results also revealed that the ANN model can predict the thermal conductivity of Al2O3–EG nanofluid accurately with maximum deviation of 1.3 % and high correlation coefficient (R > 0.998).