MIMO SYSTEM CONTROL

Abstract

The processes encountered in the real world are usually MIMO. And designing a control system implies identification of control objectives; selection of appropriate measurements and manipulations, as well as the determination of loops connecting these; and identification of proper control laws. Present study is concerned with many control aspects of a binary distillation column. In present study load rejection performance is evaluated for Wood and Berry binary distillation column for different controller tuning models i.e. Ziegler Nichols, Empirical and Biggest log modulus-tuning model. Then closed loop performance is with two PI controllers with above mentioned controller settings. And stability analysis of the system is also carried out. Then decoupler is designed. For analysis purpose MATLAB programs are written for open loop response, closed loop response, for load rejection, for simulation of distillation column, for nyquist multivariable stability criteria and decoupler. The results obtained show that load rejection performance is best with BLT tuned controllers. But system is more stable with empirical controller settings and gives best closed loop response. The system is highly coupled, but in presence of decoupler it gives better response in comparison to without decoupler control.