Yapay zekâ destekli nanoakışkan modellemesi: Termal iletkenlik ve viskozite için stabiliteye bağlı korelasyon geliştirilmesi.

In recent years, artificial intelligence-based modeling studies have been attempted to overcome the reliance on experimental results in nanofluid research. These modeling studies have mainly focused on the fundamental thermal properties of nanofluids, namely thermal conductivity and viscosity, while stability, which is one of the most important properties for nanofluids, has often been neglected. In this study, experimental data for TiO2 nanofluid has been utilized to model thermal conductivity and viscosity values using Artificial Neural Networks (ANNs). The performance metrics of the ANN model, MSE (Mean Squared Error), and R (Correlation Coefficient), are 4,2484E-06 and 0,99982, respectively. Using the model results, correlations have been established temperature, mass ratio, and pH variables for the thermal conductivity and viscosity properties of nanofluids. The deviation rates of the proposed correlations from the experimental values are within the range of ± 3,5% for thermal conductivity and ± 9% for viscosity. [ABSTRACT FROM AUTHOR]