Determining the optimal structure for accurate estimation of the dynamic viscosity of oil-based hybrid nanofluid containing MgO and MWCNTs nanoparticles using multilayer perceptron neural networks with Levenberg-Marquardt Algorithm

In this paper, prediction of viscosity (mu nf) of MWCNT-MgO/SAE40 engine oil nanofluid (NF) using artificial neural network (ANN) in different conditions (temperature, solid volume fraction (SVF or phi) and shear rate) was investigated. This research was used to evaluate and predict the viscosity of NF by ANN from a multilayer perceptron (MLP) ANN with the Levenberg-Marquardt (ML) learning algorithm. Among 400 different ANN structures, the optimal was selected from a set. It includes two hidden layers with an optimal structure of 10 neurons in the first layer and 4 neurons in the second layer. Concentration, shear rate and temperature are considered input parameters and predicted mu nf is considered as an output parameter in ANN modeling. The re-sults show that the optimal ANN with 8 neurons per layer has the least mean square error (MSE) and the maximum regression coefficient R close to 1 for predicting mu nf. The range of MODs is-2% < MOD<2%. data predicted by ANN is much more accurate than the new correlation.

Automated summary

This study investigated the prediction of viscosity of MWCNT-MgO/SAE40 engine oil nanofluid using artificial neural network (ANN) under different conditions. The optimal ANN structure consisted of two hidden layers with 10 neurons in the first layer and 4 neurons in the second layer. Concentration, shear rate, and temperature were considered as input parameters for the ANN modeling. The results showed that the optimal ANN with 8 neurons per layer had the least mean square error (MSE) and the maximum regression coefficient R close to 1 for predicting viscosity.