An RBF-based artificial neural network for prediction of dynamic viscosity of MgO/SAE 5W30 oil hybrid nano-lubricant to obtain the best performance of energy systems

Technological progress and complications in microfluidics usage have led researchers to use nanomaterials in different scientific fields. The properties and characteristics of hybrid Nanofluids are more enhanced compared to nanofluids based on single nanoparticles and conventional liquid. Recently, modeling methods have replaced most common statistical methods. Due to the high accuracy of the response and generalizability in various conditions, artificial neural networks (ANNs) to estimate nanofluids' viscosity and thermal conductivity have become common. Dynamic viscosity (mu) (estimation analyzes one of the key factors in determining the hydro-dynamic behavior of nanofluids. In this manuscript, an RBF-ANN is used to simulate the input-output relation of dynamic viscosity of the MgO-SAE 5W30 Oil hybrid nanofluid versus three important parameters, including volume fraction of nanoparticles, temperature, and shear rate. The results show that for this nanofluid, by increasing temperature and shear rate, the dynamic viscosity is decreased. In contrast, the volume fraction of nanoparticles directly affects the output, although this consequence can be neglected. By increasing the tem-perature from 5 degrees to 55 degrees C, the dynamic viscosity would decrease. Also, changing the shear rate from 50 to 1000 rpm decreases the dynamic viscosity from 400 cP to 25 cP. It is worth mentioning that the obtained trends and deviation of dynamic viscosity for MgO-SAE 5W30 Oil hybrid nanofluid versus temperature, the volume fraction of nanoparticles, and shear rate can be used by the academic community as well as an industrial section to obtain the best performance of energy systems based on this nanofluid.