A new correlation for predicting the thermal conductivity of liquid refrigerants

• Published in: Journal of thermal analysis and calorimetry 2021, Vol.143 (1), p.795-800
• Main Authors: Chen, Zhixiong, Akbari, Mohammadreza, Forouharmanesh, Forouzan, Keshani, Mojtaba, Akbari, Mohammad, Afrand, Masoud, Karimipour, Arash
• Format: Article
• Language: English
• Subjects: Analysis; Analytical Chemistry; Chemistry; Chemistry and Materials Science; Compressors; Correlation; Design optimization; Deviation; Empirical analysis; Fourier's equation; Heat conductivity; Heat exchangers; Heat transfer; Inorganic Chemistry; Measurement Science and Instrumentation; Parameters; Physical Chemistry; Polymer Sciences; Refrigerants; Regression analysis; Regression models; Thermal conductivity; Variance analysis
• ISSN: 1388-6150; 1588-2926
• DOI: 10.1007/s10973-019-09238-w

The material ability to conduct the heat transfer is called thermal conductivity which is defined by Fourier's equation. Thermodynamic data on environmentally acceptable refrigerants have maximum interest for industries to optimize and design equipment of refrigeration such as exchangers and heat compressors. Because source empirical findings are not applicable for all temperature ranges in industries, correlation approaches are usually preferred. In this research, a novel simple correlation has been developed to predict the thermal conductivity of liquid refrigerants using regression approaches. The variance analysis was applied to study the rationality of regression model. Around 15,874 experimental data of 27 refrigerants were examined to obtain the main effects between the independent parameters. Independent parameters are temperature, boiling and reduced temperatures. The calculations show that the accuracy of the proposed correlation using the average absolute relative deviation (AARD) and root mean square deviation has priority over the previous relations. The results indicated that the AARD of the proposed model is 1.1% which is 68% lower than of the most accurate previous model (Latini–Sotte).