A Two-Stage Algorithm for Combined Iterative Learning Control with Real-Time Feedback; A State Space Formulation
| Authors: | Lee Kwang Soon, Sogang University, Korea, Republic of
Chin Insik, Samsung Co., Korea, Republic of
Cho Moonki, Conwell Ltd. Co., Korea, Republic of
Qin S. Joe, University of Texas at Austin, United States |
|---|
| Topic: | 6.1 Chemical Process Control |
|---|
| Session: | Advances in Process Control |
|---|
| Keywords: | Iterative Learning Control, Run-to-run Control, Batch Process Control, LQG Control, Disturbance Rejection |
|---|
Abstract
A new combined iterative learning control (ILC) and real-time feedback control (RFC) algorithm has been proposed on the basis of the two-stage implementation and LQG control. The proposed methodassumes Gaussian disturbances and deal with the batch-wise recurrent and nonrecurrent disturbances by separate LQG controls for ILC and RFC, respectively. In this way, the problem with the existing combined ILC-RFC methods that nonrecurrent real-time disturbances cause ILC to digress from its convergence track along the run index could be overcome. The performance of the proposed technique has been demonstrated using numerical simulation.
