Artificial intelligence model and correlation for characterization and viscosity measurements of mono & hybrid nanofluids concerned graphene oxide/silica

Abstract

Graphene oxide/silica composite’s rheological behavior was studied in this investigation. This composite was made to reduce the cost of industrial usages. The volume fractions investigated from 0.1% to 1.0% (GO 30%–SiO2 70%), the shear rates investigated from 12.23 to 122.3 s−1, and the temperatures investigated from 25 to 50 °C. To study the characterization of each solid and the composite, the XRD and the FESEM tests were done. The results of the viscosity investigation revealed the non-Newtonian behavior. After that, a numerical study was done to present a correlation and train an artificial neural network model. These numerical studies were done for both 12.23 and 122.3 s−1 shear rates. The novel equation tolerances were 1.932% and 1.338% for 12.23 and 122.3 s−1 shear rates, while for the artificial neural network model, the tolerances were 1.46196% and 1.25386% for 12.23 and 122.3 s−1 shear rates. This means, after the model was trained, the deviation decreased around −0.46999% and −0.08467% for 12.23 and 122.3 s−1 shear rates. This nanofluid can be employed in industrial systems.