Fuzzy air handling system controller

Modem air conditioning systems for commercial buildings commonly employ the concept of a ‘central all-air system’ and the variable air volume (VAV) system in particular is widely used in Hong Kong and elsewhere other places around the world for energy conservation. In the lengthy wet summer season of Hong Kong, centralised air handling units (AHUs) dehumidify and cool down the appropriate mixture of return air and outdoor fresh air to feed a ducting network to various VAV boxes. A good controller for the AHUs is extremely desirable from both human comfort and energy saving points of view. In this paper, a simulation model for a practical air handling system is presented. Its behaviour under a conventional system of PID controllers is studied. A new controller based on fuzzy logic has been developed where all the parameters including input variables and actuating variables are represented as fuzzy subsets with carefully assigned membership functions. Such a controller is expected to directly replace the conventional PID controllers with identical input and output points. Based on an expert system developed from the recommendations of experienced human operators on air conditioning control, decision rules for proper control actions are laid down. Computer simulation as detailed in this paper has proved that such a fuzzy logic-based controller is superior to a conventional, cvell-tuned PID controller in at least three aspects. Fuzzy logic-based control is more robust since slight changes in the values of parameters do not greatly affect the performance compared with detuned PID control. The response rate of fuzzy logic-based control is faster when there is a sudden change in the environment. Energy saving is another merit of using fuzzy logic-based control.