摘要：AbstractHeating, ventilation, and air conditioning (HVAC) system is an important system in the building. If the HVAC system can run in a fault-free condition, energy consumption will be reduced directly. In HVAC system, the air handling units systems (AHUs) are used to deal with the heat change between outdoor environment and the building, which is a key subsystem of the HVAC system. If the AHUs run under a fault condition, energy consumption will be increased. In fault detection, the state of AHUs will be affected by outdoor environment. The influence from outdoor may mislead the alarm of fault detection and reduce the precision of detecting. By dynamically monitoring the fluctuation of AHUs parameters, we can get the state of AHUs in time, and there will be an alarm song if the fault occurs. This paper introduces a fault detection method to provide a new perspective for AHUs fault detection. Through this method, we obtain the states of the AHUs through unscented Kalman filter, and set the dynamic-control limit by Statistical Process Control (SPC) method to get alarm of fault condition. The method can get system’s states accurately, and the false alarm will be well controlled.
