Fuzzy model-based fault-tolerant control of air-conditioning systems

The paper describes the development of a su pervisory control scheme that adapts to the presence of degradation faults and  minimises any resulting increase in energy consumption or deterioration  in  occupant  comfort   Since  there is a high degree of uncertainty associated with the  results  of  any   fault  identi cation  scheme in information poor systems of this type the supervisory control scheme uses fuzzy models to predict fuzzy measures of the overall performance and a fuzzy decision maker to determine the most appropriate set points This fault tolerant control scheme is  being  developed  and  evaluated using a detailed computer  simulation of a multi zone variable air volume air conditioning system The fuzzy models relate the performance of the air handling unit and the chiller to the supply air and chilled water temperature set points and to fuzzy measures of the amount of air side and water side fouling Results are presented that demonstrate the ability of the fuzzy models to predict the performance and show the e ect of both water side and air side fouling on the performance Problems associated with training the fuzzy models are also discussed