Self-detection of thermal ambient parameters for air-conditioned space by learning operation data and self-prediction of indoor temperature and power consumption
For efficient energy use of air conditioning unit (AC), there is a need to find an optimal control method for operation. To deduce an optimal control algorithm, indoor temperature and power consumption need to be predicted according to given operating conditions. Among the key factors, the thermal ambient parameters such as thermal resistances of the operation sites are most difficult to identify. This study proposed a method for an AC to deduce these parameters for itself by learning operation data obtained from the embedded sensors and autonomously predict indoor temperature and power consumption by solving the transient energy conservation equation incorporating refrigeration cycle simulation. A novel aspect of this study is that the AC can autonomously perceive thermal ambient parameters of its surroundings and utilize these parameters to predict indoor temperature variations and power consumption. These data could be fed back to the control software of the AC in real time. To validate the proposed method, experiments were conducted in model house test facility using a real air-conditioner, and the results were compared with the simulation results. It was confirmed that the time to reach the air temperature setpoint was within 3 min and the power consumption could be predicted within 9 %.
