15
Identification of Faults in Air-Conditioning Systems
Early detection of the faults in air-conditioning systems can prevent energy wastage and avoid occupant discomfort (Qin and Wang, 2005). When faced with high levels of uncertainty however, there is a real risk of incorrect diagnosis and the cost of failing to diagnose a fault must be weighed against the cost of having to respond to a false alarm (Dodier et al., 1998). The plant operator may even turn off the fault diagnosis system if there are too many false alarms (Visier et al., 1997). One of the main requirements of any fault diagnosis scheme is therefore that it should generate very few false alarms.
15.1 Main Sources of Uncertainty and Practical Considerations
The problems associated with isolating and identifying faults in air-conditioning systems are more severe than those that occur in most process control applications. The behaviour of heating, ventilating and air-conditioning (HVAC) plants and buildings is more difficult to predict. Accurate mathematical models cannot be produced because most HVAC designs are unique and financial considerations restrict the amount of time and effort that can be put into deriving a model (Lo et al., 2007). Detailed design information is seldom available (Haves et al., 1996a) and measured data from the actual plant are often a poor indicator of the overall behaviour, since test signals (Haves et al., 1996b) cannot usually be injected during normal operation and buildings are subject ...
