Modified internal model controller design based on OKID for multivariable system with time delays

Most industrial processes have multiple time delays and due to the existence of these time delays and couplings within these systems, the control performances are usually not satisfied. In this paper, a modified internal model controller (MIMC) design procedure is employed to deal with this obstacle based on a model discretization and simplification method, together with a decoupling controller. First, the multivariable system with time delays is decoupled into subsystems. Then, a model reduction method of observer/Kalman filter identification (OKID) is used to simplify the complicated subsystems, without involving the common time delay factor. Based upon the discrete-time simplified models obtained by OKID method, the modified internal model controllers are designed to control the decoupled subsystems to track the set-point values. Finally, simulation results of a chemical process are given to illustrate the effectiveness of the proposed method.