Predict the thermal conductivity of SiO2/water–ethylene glycol (50:50) hybrid nanofluid using artificial neural network

Abstract

In the current work, after generating experimental data points for different volume fraction of nanoparticles ( $$\phi$$ ) and different temperatures, an algorithm to find the best neuron number in the hidden layer of artificial neural network (ANN) method is proposed to find the best architecture and then to predict the thermal conductivity ( $$k_{\text{nf}}$$ ) of SiO2/water–ethylene glycol (50:50) nanofluid. This ANN is a feed-forward network with Levenberg–Marquardt for the learning algorithm. Regarding the experimental data points, a third-order function is obtained. In the fitting method, the mean square error is 2.7547e−05, and the maximum value of error is 0.0125. The correlation coefficient of the fitting method is 0.9919. This surface also shows the behavior of nanofluid based on the $$\phi$$ and temperatures, and finally, the results of these methods have been compared. It can be seen that for 8 neuron numbers, the correlation coefficient for all outputs of ANN is 0.993861.