Experimental Investigation and Specific Heat Capacity Prediction of Graphene Nanoplatelet-Infused SAE10W Oil Using Artificial Neural Networks

Abstract

This research aims to develop an artificial neural network (ANN) model to predict the specific heat capacity (SHC) of graphene/SAE10W oil nanofluid. An experimental investigation of the SHC of graphene nanoplatelets infused in SAE10W oil was carried out using a thermal constant analyzer. In experimental testing the volume percentage and fluid temperature range vary from 0.05 to 0.15 and 293 K to 353 K, respectively. The experimental data shows the graphene/SAE10W oil nanofluid exhibited a reduction in SHC relative to the base fluid. The ANN model was developed using experimental data to predict the specific heat capacity of graphene/SAE10W oil nanofluid. During ANN model training, the correlation coefficient and mean square error of 0.999 and 6.592 × 10 ^–6 were achieved, respectively. Compared to experimental values, the ANN model predicts SHC with a 0.45 error percentage. Additionally, a mathematical model has been developed to predict the SHC of graphene/SAE10W oil nanofluids using curve fitting. The data obtained from the developed mathematical model showed excellent correlation with all experimental values, with an error percentage of ±0.42. Hence, it is concluded that both models provide an optimal approach for estimating their SHC.