Designing an artificial neural network using radial basis RBF-ANN) to model thermal conductivity of ethylene glycol-water-based TiO2 nanofluids

作者：Esfe Mohammad Hemmat^*

来源：Journal of Thermal Analysis and Calorimetry, 2017, 127(3): 2125-2131.

DOI：10.1007/s10973-016-5725-y

• Thermal conductivity
• Artificial neural network
• Temperature
• Modeling

In this article, thermal conductivity of ethylene glycol-water-based TiO2 nanofluids has been modeled by artificial neural network. For this purpose, thermal conductivity of nanofluids with volume fractions of 0.2-1 % has been collected in temperatures of 30-70 A degrees C. These data were modeled by artificial neural networks. So two common types of neural networks were used, and the results were compared with each other. One of these networks was multilayer perceptron, and the other one was radial basis approximation. Finally, an experimental relationship was suggested for calculating thermal conductivity of this nanofluid and the results were compared with the results of radial basis neural network. This comparison shows that neural networks are very powerful in modeling the nanofluids experimental data and are able to follow the patterns of these data with a high precision.

• 出版日期2017-3