Determination of the Heat Transfer Coefficient of Metal Oxide Based Water Nanofluids in a Laminar Flow Regime Using an Adaptive Neuro-Fuzzy Inference System

In order to enhance the thermal properties of turbine oil (TO), three different nanoparticles (CuO, Al₂O₃, and TiO₂) are loaded into the TO. To measure the thermal performance of nanoparticle-based TO nanofluids at laminar flow and under constant heat flux boundary conditions, an experimental setup was applied. The obtained data clearly demonstrate the positive effect of all nanoparticles on the heat transfer rate of TO. As the most important factor, the heat transfer coefficient of the abovementioned two-phase systems is increased upon increasing both the volume concentration and the flow rate. An adaptive neuro-fuzzy inference system (ANFIS) is applied for modeling the effect of critical parameters on the heat transfer coefficient of nanoparticle-TO based nanofluids numerically. The results are compared with experimental ones for training and test data. The results suggest that the developed model is valid enough and promising for predicting the extant of the heat transfer coefficient. R² and MSE values for all data were 0.990208751 and 108.1150734, respectively. Based on the results, it is obvious that our proposed modeling by ANFIS is efficient and valid, which can be expanded for more general states.