Usually, fuzzy controller design is based on expert knowledge of plant input/output behavior.Very often the control knowledge of the expert is not complete or, even worse, it consists of contradictory rules. Thus, the performance of the closed-loop system often does not meet requirements. We give a theorem about obtaining the closed-loop system fuzzy relation using the fuzzy relation of the controller and the fuzzy relation of the plant. Based on this theorem and using a fuzzy rule model of the plant and a desired fuzzy rule model of the closed-loop system, we present a new technique to synthesize the fuzzy controller. Using the peak pattern technique, we develop a complete design procedure to obtain the fuzzy controller, and then apply it to an example showing that the given technique is easy to apply.
展开▼
