Prediction of thermophysical properties of mixed refrigerants using artificial neural network

• Published in: Energy conversion and management Vol. 52; no. 2; pp. 958 - 974
• Main Authors: SENCAN, Arzu, NOSE, Ismail Ilke, SELBAS, Reşat
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01.02.2011; Elsevier
• Subjects: ANN; Applied sciences; Artificial neural networks; Compressing; Density; Dynamic viscosity; Dynamics; Energy; Energy. Thermal use of fuels; Exact sciences and technology; Heat conduction coefficient; Kinematic viscosity; Learning theory; Neural networks; Refrigerant; Refrigerants; Refrigerating engineering; Refrigerating engineering. Cryogenics. Food conservation; Refrigeration; Specific heat capacity; Thermal diffusivity; Thermophysical properties; Viscosity; ANN; Refrigerant; Heat conduction coefficient; Thermal diffusivity; Dynamic viscosity; Kinematic viscosity; Density; Specific heat capacity
• ISSN: 0196-8904; 1879-2227
• DOI: 10.1016/j.enconman.2010.08.024

The determination of thermophysical properties of the refrigerants is very important for thermodynamic analysis of vapor compression refrigeration systems. In this paper, an artificial neural network (ANN) is proposed to determine properties as heat conduction coefficient, dynamic viscosity, kinematic viscosity, thermal diffusivity, density, specific heat capacity of refrigerants. Five alternative refrigerants are considered: R413A, R417A, R422A, R422D and R423A. The training and validation were performed with good accuracy. The thermophysical properties of the refrigerants are formulated using artificial neural network (ANN) methodology. Liquid and vapor thermophysical properties of refrigerants with new formulation obtained from ANN can be easily estimated. The method proposed offers more flexibility and therefore thermodynamic analysis of vapor compression refrigeration systems is fairly simplified.