Prediction of thermal conductivity of alumina water-based nanofluids by artificial neural networks

Abstract

The aims of the present study are to develop and validate an artificial neural network (ANN) approach to estimate the thermal conductivity ratio (TCR) of alumina water-based nanofluids as a function of temperature, volume fraction and diameter of the nanoparticle. The ANN parameters are adjusted by back propagation learning algorithm using 285 collected experimental data sets from various literatures. Statistical accuracy analysis confirms that a two-layer feed forward ANN model with fourteen hidden neurons is the best architecture for modeling the considered task. The developed ANN approach has predicted the experimental data with the absolute average relative deviation (AARD%) of 1.27%, mean square error (MSE) of 4.73×10−4 and regression coefficient (R2) of 0.971875. Comparison of predictive capability of the proposed technique with some recommended correlations in the literatures confirmed that the ANN model is more superior to other published works and therefore can be considered as a practical tool for estimation of the thermal conductivity ratio of alumina water-based nanofluids.