Employing response surface methodology and neural network to accurately model thermal conductivity of TiO2-water nanofluid using experimental data

In this research, the thermal conductivity of the H2O-titania nanofluid is modeled versus the particle concentration and temperature via the Artificial Neural Network (ANN) and Response Surface Methodology (RSM). The experimental data include six particle concentrations and five temperatures from 30 to 70 degrees C. The thermal conductivity augments by the increment in nanoparticle concentration and temperature, such that the maximum thermal conductivity increment happens at the highest temperature and nanoparticle concentration (i.e., T = 70 degrees C and phi = 1%). It is observed that the impact of temperature on the thermal conductivity is more noticeable than the influence of particle concentration, however, the thermal conductivity demonstrates a more non-linear trend versus nanoparticle volume fraction compared with the temperature. The best structure of the neural network has 2 hidden layers with 2 and 4 neurons, respectively in the 1st and 2nd hidden layers. The results show that the prediction precision of the ANN correlation is better than that of the RSM correlation.

Automated summary

In this study, the thermal conductivity of H2O-titania nanofluid was modeled against particle concentration and temperature using Artificial Neural Network (ANN) and Response Surface Methodology (RSM). The results indicate that temperature has a more significant impact on thermal conductivity than particle concentration, and that thermal conductivity shows a non-linear trend with nanoparticle volume fraction compared to temperature.