PD-type vs. PID-type fuzzy controllers

A fuzzy controller considering the error and the charge rate of error of the output variable can be regarded as a nonlinear PD controller. The authors show that the PD-type fuzzy controller is deficient in structure and can be improved by incorporating the summation of the output errors. The proposed PID-type fuzzy controller is capable of achieving zero steady-state error with very little overshoot for difference reference points, which is a very difficult task for a fixed PD-type fuzzy controller. Several examples are shown to illustrate the weakness of the PD-type fuzzy controllers and the improvements obtained by the addition of summation-of-errors in the premise of the rule base.<>