Cross-coupling design of generalized predictive control with reference models

In this paper, a cross-coupled generalized predictive control algorithm is proposed with reference models on inputs and outputs of the process. It is shown that such an algorithm provides a combined feedback-feedforward control resulting in pole-zero cancellation of poles which do not correspond to the reference models. Hence, the system behaves asymptotically as the reference model provided that the cancelled poles are stable. In addition, the cross-coupling control action introduced in an optimal way will effectively reduce the synchronization deviations of processes with multiple motion axes. The proposed algorithm is applied to the control of a servo-system with two motion axes, and simulation results show that the control algorithm gives improved performance.