In this study a comparison between ILC and feedback control is made in an experimental manner. The experiments are done with a Linear Motor Motion System (LiMMS), which suffers from non-linear disturbances . the cogging effect. | Universiteit Twente Iterative Learning Control vs. Feedback Control. - An experimental study M. Eglence Pre-Masters assignment (Pre-Doctoraalopdracht) Author: M. Eglence Supervisors: Prof. dr. ir. J. van Amerongen Dr. ir. . de Vries Ir. . Verwoerd Date: October 14, 2002 Control Laboratory Faculty of EEMCS University of Twente . Box 217 7500 AE Enschede The Netherlands Report Nr: 025CE2002 II III Preface This project is done as a pre-assignment for the master thesis. I would like to thank my primary supervisor Ir. Mark Verwoerd for his good support and suggestions during this project. I also want to thank my fellow students for making it a pleasant time. Especially I want to thank Ali Mohamoud, a fellow student and good friend, for his support during the “hard times” and for reading and correcting this report. Enschede, October 14, 2002 M. Eglence IV V Summary Iterative Learning Control (ILC) is now known for about 2 decades and has proven to improve the performance of classical feedback control. In this study a comparison between ILC and feedback control is made in an experimental manner. The experiments are done with a Linear Motor Motion System (LiMMS), which suffers from non-linear disturbances . the cogging effect. When the ILC converges, it is shown that the feedback control and the ILC together can be seen as a new feedback controller, which is totally equivalent and should therefore give the same results. First a feedback controller is designed for the LiMMS to obtain a stable closed-loop system. Then an ILC part is added to the controller, which has improved the performance considerably. When the equivalent controller was derived it gave numerical problems because of the great difference between the numerator and the denominator. Therefore an alternative equivalent controller is derived that is mathematically the same. Applying the alternative equivalent controller gives improvement in comparison with the feedback .
