Development of a feedback controller tuner using virtual fuzzy sets

This paper presents the development of a fuzzy tuner for tuning PI and PID feedback controllers. The implementation is based on the concept of virtual fuzzy sets and an adaptive fuzzy inference scheme. The tuner uses the standard Ziegler-Nichols tuning rules for initial gain setting according to the open-loop step response of the process. Then the fuzzy system fine tunes the controller iteratively based on the performance of the closed-loop controlled system response. The tuning knowledge is extracted automatically from the tuning of a representative process. Expert rules of thumb can be incorporated into the rule base, and explanatory capability is also available through the use of the IF-THEN rule base. Finally, examples covering a wide range of process dynamics are tested to demonstrate the excellent performance of the tuner.