V. INTERNATIONAL TURKIC WORLD CONGRESS ON SCIENCE AND ENGINEERING (TURK-COSE 2023) PROGRAMME, Bishkek, Kırgızistan, 15 - 17 Eylül 2023, ss.235, (Özet Bildiri)
The most fundamental property expected from nanofluids is to improve the
thermal conductivity and therefore the heat transfer properties of the base fluid. Since
its discovery in 2004, studies on graphene have progressed very rapidly in every field.
Graphene's high thermal conductivity has made it a high potential particle for
nanofluids. Due to the high cost and time consumption of experimental measurements,
researchers are trying to find a theoretical relationship to estimate the thermal
conductivity of nanofluids. Nowadays, artificial neural networks (ANN) inspired by the
behavior of the human brain are one of the suitable methods to predict and model the
thermal conductivity of different materials such as gases, liquids, and solids. In this
study, experimental data of graphene-water nanofluid belonging to 4 different mass
ratios (0.001, 0.005, 0.015 and 0.045) found in the literature were trained in artificial
neural networks with MSE and R values of 9.3916E-04 and 0.9998, respectively, and
a new correlation based on temperature and mass ratio was proposed using the
obtained outputs.