Studies on the online intelligent diagnosis method of undercharging sub-health air source heat pump water heater

Due to its significant energy savings, the use of air source heat pump water heater (ASHPWH) has increased rapidly in recent years. Pipeline seal issues, improper installation, and other reasons can cause refrigerant leakage. Minimal refrigerant leakage causes slight changes in system characteristics and is difficult to notice, thus people always determine the system to be normal. Refrigerant leakage can cause the system to deviate from the reasonable working conditions for a long time, resulting in the decline of system performance, efficiency drop, and increase in energy consumption. In this paper, the concept of sub-health operation of heat pump systems is proposed to distinguish the transition state between normal and faulty. Based on this concept, the research on the changes of sub-health is proposed with an online intelligent diagnosis method. This diagnosis method utilizes the data of normal system operations to train a diagnostic model, and it does not need fault marking data, which reduces the difficulty of data acquisition. The proposed method can achieve accurate fitting of unsteady systems with resistance to heat transfer environment fluctuations and is better suited for online diagnosis. It has been verified by experiments that this method can achieve online diagnosis of refrigerant leakage of ASHPWH, and it is a feasible and efficient sub-health diagnosis method.