A theoretical refrigerant charge prediction equation for air source heat pump system based on sensor information. (August 2019)

• Record Type: Journal Article
• Authors: Hong, Sung Bin; Yoo, Jin Woo; Kim, Min Soo
• Is Part Of: International journal of refrigeration. Volume 104(2019)
• Page Start: 335
• Page End: 343
• Publication Date: 2019-08
• Subjects: Refrigerant charge amount -- Charge amount prediction -- Air source heat pump -- Theoretical analysis -- Sensor information; Quantité de charge en frigorigène -- Prédiction de la quantité de charge -- Pompe à chaleur aérothermique -- Analyse théorique -- Information du capteur; Refrigeration and refrigerating machinery -- Periodicals; 621.56
• DOI: 10.1016/j.ijrefrig.2019.05.031
• Languages: English
• ISSNs: 0140-7007

Abstract

Highlights: A new prediction method of refrigerant charge amount in the air source heat pump system is proposed. Theoretical analysis of the refrigerant charge in the system is performed. Charge prediction method requires only few experimental data and installed sensors. The accuracy of the method is verified by both cooling and heating mode experiments. Abstract: The performance of the heat pump system varies greatly depending on the refrigerant charge amount. However, the refrigerant in the system slowly leaks during operation. Therefore, in order to guarantee system performance and life span, it is important to predict the charge amount in real time and recharge the shortage. Nevertheless, since the configuration and control of the heat pump system becomes more complicated, it is very difficult to predict the refrigerant charge amount. Various previous studies have been carried out to predict refrigerant charge amount, but most of them are empirical methods or theoretical, but require much experimental data for high accuracy. In this paper, a refrigerant charge amount prediction method which requires only a few experimental data for high prediction accuracy is proposed through theoretical analysis of the refrigerant charge. The proposed method was verified by experiments, and the root mean square error was 3.7% in cooling mode and 8.2% in heating mode.