Designing a neural network for predicting the heat transfer and pressure drop characteristics of Ag/water nanofluids in a heat exchanger

Nanofluids are advanced liquids with a relatively high production cost. Experiments that are performed to detect nanofluid characteristics in various thermal systems could be replaced by modeling tools to reduce the expenses. The present paper deals with the post-processing of experimental data on the flow and heat transfer in a nanofluid-based double tube heat exchanger using an artificial neural network. Relative values of Nusselt number and pressure drop in the heat exchanger are modeled where Ag/water nanofluids with volume fractions up to 1% have been exploited as the working fluid. The results of present work unveil the ability of neural network to predict the data with excessive noise. The data regression coefficients for the relative Nusselt number and relative pressure drop are 99.76% and 99.54%, respectively, which show the high accuracy of the applied method.