The Design Of Guaranteed PID Performance Fuzzy System

At present, fuzzy control has been successfully applied in many fields, but the study of fuzzy control theory is expecting the steady, foundation, in order to resolve problems of fuzzy system, such as control performance, systematization design method, accordingly strict and systematic fuzzy control theory will be built. In addition, it is full of disputation which control performance is better between fuzzy control and traditional PID control. Although, large numbers of researches have demonstrated "fuzzy controller's performance is better than traditional linear controller", so how to design fuzzy controller and make its performance better than traditional linear controller is not only a serious subject in research of fuzzy controller, but also a difficult problem needed to be resolved urgently in the application fields. Aimed at the problem, Hu advise that a conservative design strategy should be used in fuzzy controller, which indicates "any fuzzy PID controller can realize the linear input and output relation through adjusting parameters. The system according with the strategy is called "Guaranteed-PID-Performance Fuzzy System", or GPP system. The strategy is very meaningful in theory and application, it would lead that fuzzy theory can quote or reference the theorem and strategy in traditional control theory as lower limit performance index for analysis, and make fuzzy control technology based on traditional control theory steadily. This paper deeply analyzed and studied the mechanism of linear PID controller and linear fuzzy controller, completed the following questions, meanwhile an design method of GPP fuzzy system was presented: [1] Linear control algorithm is realized with fuzzy controller. [2]An design method of linear fuzzy controller and simulation realization was presented. [3]The design process of GPP fuzzy controller was completed, and made the performance comparison between GPP fuzzy controller and traditional PID controller, gave the simulation and software realization. The result demonstrated that fuzzy controller can be designed according to GPP, through adjusting fuzzy rule, control performance can be better than linear optimum controller.