A dynamic relative gain array based on model predictive control

Loop pairing is a major concern in the stage of decentralized control system design of complex industrial processes. Without regard to the effects of the dynamic interaction, RGA was transformed to other impoved dynamic relative gain array. A new approach to defining a new dynamic RGA (s-DRGA) is presented based on multivariable state feedback predictive control(SFPC). The approach assumes the availability both of steady state and a dynamic process model. The new DRGA is defined based on the dynamic variable relation matrix and steady state array. Several cases are studied and the comparisons with existing methods indicate that the proposed method is a useful tool to give the best pairing scheme.