Experimental evaluation and ANN modeling of thermal conductivity of graphene oxide nanoplatelets/deionized water nanofluid

Abstract

In this research study, the thermal conductivity of graphene oxide nanoplatelets/deionized water nanofluid is studied in different temperatures and weight fractions using artificial neural network (ANN) and experimental data. For the purpose of training the ANN, the thermal conductivity of nanofluid is measured in temperatures between 25 and 50°C and weight fractions equal to 0.001, 0.005, 0.015 and 0.045. For the purpose of evaluating the accuracy of the proposed model by ANN, root mean square error (RMSE), R2 and also mean absolute percentage error (MAPE) are utilized. The best ANN model has two hidden layers and one output layer and also utilizes tansig, logsig and pureline functions and the number of neurons is 4–8–1 in the mentioned layers respectively. The inputs of the ANN model are weight fraction and nanofluid temperature and the output of the network is the thermal conductivity of the nanofluid. The results indicate that the proposed model by ANN can precisely predict the thermal conductivity of the nanofluid.