In multivariable systems, controlled variables interact each other in general. Therefore it is difficult for the classical control theory such as PID control to control multivariable systems. The Nyquist Array method as the practical design technique has been used for the long time. However the method becomes fairly complicated when we design a precompensator and PID controller.
Therefore we propose the PID controller design method for multivariable systems which can realize PID control relatively easily in view of utility. We design a compensator which meets the diagonal dominance using Real Code Genetic Algorithms. Next, we design the PID controller using Coefficient Diagram Method which takes simultaneously good balance of stability, response, and robustness by using coefficient diagram. Finally, the effectiveness of the proposed method is confirmed with the simulation results.
J-STAGEがリニューアルされました! https://www.jstage.jst.go.jp/browse/-char/ja/