Improving the performance of LPG with graphene-nanolubricant in a domestic refrigerator: an artificial intelligence approach

Abstract

This work investigated the possibility of improving liquefied petroleum gas (LPG) of propane and butane (60:40) mixture with graphene nanolubricant (GN) in a domestic refrigerator system. An artificial intelligence model with Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference (ANFIS) was also developed to predict the performance of the LPG with graphene nanolubricant in the system and validate it with experimental results. The graphene nanolubricant was investigated in 50–70 g charge of LPG refrigerant. Four Type K thermocouples were attached to the refrigerator components to track the temperature of the system. Two pressure gauges were also attached to the compressor to determine the pressure at suction and discharge of the domestic refrigerator. A digital watt-meter was used to measure the refrigerator’s compressor power consumption. The obtained result showed a higher COP with graphene nanolubricant concentrations. The power consumption and cooling capacity improved within the range of 16.3%–20.1% and 6.4%–19.6% respectively with graphene nanolubricant concentrations. Also, the ANN and ANFIS model predicts the COP of the refrigerator with a Root Mean Square Error (RMSE) value of 0.620 and 0.303 and Mean Absolute Percentage Error (MAPE) of 29.4% and 14.9% respectively.