Designing an Artificial Neural Network (ANN) to predict the viscosity of Silver/Ethylene glycol nanofluid at different temperatures and volume fraction of nanoparticles

By Davood Toghraie, Nima Sina, Niyusha Adavoodi Jolfaei, Mehdi Hajian, Masoud Afrand. Physica A: Statistical Mechanics and its Applications, 2019

150 citations recorded by the source corpus.

In the current work, we investigate the dynamic viscosity of Ag/Ethylene glycol nanofluid within the temperature range of 25–55 ° C with volume fraction of nanoparticles range of 0.2%–2%. The experimental data includes 42 samples. At first, an Artificial Neural Network (ANN) is designed to predict the dynamic viscosity of this nanofluid and finally the results of ANN and correlation has been compared. The algorithm of generating the best architecture of ANN has been proposed and the best ANN has been used to predict the dynamic viscosity of Silver/Ethylene glycol nanofluid. It is found that the ANN can predict the viscosity of Ag/Ethylene glycol nanofluid with good precision compared to the correlation method. Also, in the correlation method, MSE is 0.0012, SSE is 0.0512 and the maximum value of error is 0.0858.

Subfields: Analytical Chemistry, Biomedical Engineering, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/a5af000e36f7077b486ee335e352bdd3c197cf88