抄録
This paper describes a design method of an I-PD (proportional-derivative preceded integral) controller. This method, which consists of two stages, is simple compared with conventional ones. Specifically, first an original high-order plant is reduced to a second-order system in terms of the ν-gap metric as a criterion of the modeling error, and next the control gains are obtained as state feedback gains of the integral-type optimal servomechanism (IOS) for an augmented system. The resultant controller is exactly an I-PD one. Although perfect closed-loop stability and control performance are not guaranteed due to the modeling error, if the ν-gap is small enough, it is probable for these properties to be adequately preserved. Moreover, it can be transformed into a one-degree-of-freedom PID (proportional-derivative-integral) controller and a PI-D (derivative preceded proportional-integral) one in order to use a suitable type of the controllers depending on the plant or control objective. Design examples and simulation results are shown to demonstrate the effectiveness and usefulness of the proposed method.
