High-Order Internal Model Based Indirect-Type Iterative Learning Control Design for Batch Processes with Batch-Varying Factors

Abstract

This paper proposes a high-order internal model (HOIM) based indirect-type iterative learning control (ILC) scheme for batch processes subject to batch-varying initial condition and reference along with external disturbance. A widely used proportional-integral (PI) control structure in practical applications is taken as the inner loop, while the set-point related indirect-type ILC updating law is designed independent of the inner loop to robustly track the desired output trajectory. In comparison with the existing indirect-type ILC methods, the set-point commands and output tracking errors over more than one previous batches are used for the ILC design in terms of an augmented HOIM associated with the initial process state, reference, and external disturbance. By using an equivalent 2D Roesser system description of the closed-loop ILC system, a sufficient condition in terms of linear matrix inequality is established to ensure asymptotic stability of the resulting 2D system together with a 2D $\mathcal{H}_{\infty}$ performance under non-zero boundary conditions. Finally, the obtained results are validated by an illustrative example of injection molding.