Thermal entropy generation and exergy efficiency analysis of rGO/water nanofluid in a tube under turbulent regime using experimental and fully connected neural network

Analyzing entropy generation is one of the most effective methods for examining how well thermal systems operate. To determine the ideal operating conditions, numerous researchers have examined the entropy generation of thermal systems. In this work, the effects of concentrations of reduced graphene oxide (rGO) nanoparticles on the formation of entropy in water-rGO nanofluid flow through a circular pipe with a constant wall heat flux boundary condition in turbulent regimes are investigated experimentally. rGOs were prepared from Hummer's technique. Experiments are conducted at particle volumes from 0.5 % to 2.0 %, mass flow rates from 0.0333 kg/s to 0.25 kg/s, and Reynolds numbers from 1800 to 21,000, respectively. Later, estimated data points were anticipated with a convolutional neural network (CNN). The model was analyzed through several performances like root-mean-square-error (RMSE), mean-square-error (MSE), mean-absolute-percentage-error (MAE), and the coefficient of determination (R2). Results indicated, with used by 2.0 vol% of nanofluid, entropy generation-thermal is reduced to 39.24 %, entropy generation-frictional is higher by 16.92 %, entropy generation-total is decreased by 37.82 %, and second law efficiency is augmented to 34.17 % at Reynolds number of 13,705, against base fluid data. The correlation coefficient R2 evaluated from CNN for thermal entropy generation, frictional entropy generation, total entropy generation, and exergy efficiency are 0.99297, 0.99924, 0.9989, and 0.99273, respectively.