Improved PI based indirect-type ILC for batch processes with time-varying uncertainties: A new perspective

This paper proposes an improved PI based indirect-type iterative learning control for batch processes with time-varying uncertainties from a new perspective. The proposed ILC law can be designed independent of the initial PI tuning which primarily aims at holding the robust stability of the closed-loop system due to that only the setpoint command is adjusted by ILC. The feedforward control used in the recently developed indirect-type ILC methods is replaced by retuning the closed-loop PI controller, while the redundant learning information from the current and previous cycles could be reduced compared with the existing literature. Correspondingly, the control structure is significantly simplified with less design parameters, while achieving the similar convergence performance along the batch direction. A robust ILC updating law along with retuned PI parameters are derived by solving a linear matrix inequality and results in the robust stability of the closed-loop system. An illustrative example from the literature is adopted to demonstrate the effectiveness of the proposed method.