Journal of Chemical Engineering of Japan, Vol.43, No.2, 196-208, 2010
Analytical Design of Multi-Loop PI Controllers for Interactive Multivariable Processes
In this paper, a simple and efficient design method of multi-loop PI controllers is proposed by extending the Maclaurin series approach to a multi-loop control system. The controller parameter of a multi-loop system is related to that of a SISO non-interacting system with an interaction factor in the simple multiplication form. Based on this relation, analytical tuning rules for a multi-loop PI controller are derived for several representative process models. In order to improve both performance and robustness of the multi-loop control system, the multi-loop Ms criterion is utilized as a performance cost function. The simulation studies confirm the effectiveness of the proposed method.
Keywords:Interactive Multivariable Process;Multi-Loop PI Controller;Generalized IMC-PID Tuning;Relative Gain Array;Closed-Loop Dynamic Interaction Matrix;Multi-Loop Maximum Sensitivity Criterion