Engineering Mechanics

Proceedings Vol. 11 (2005)

ENGINEERING MECHANICS 2005

Copyright © 2005 Institute of Mechanics of Solids, Faculty of Mechanical Engineering, Bruno University of Technology, Brno

 

Březina T., Pulchart J., Turek M.
pages 59 - +7p., full text

Algorithm of locally weighted regression is presented in this contribution. Local approximator repeatedly uses the locally linear model based on least square method. Simulation study describing both behavior and learning of small active magnetic bearing control through nonlinear PD controller which uses parallel nonlinear compensation realized by given local approximator is further presented. The goal of on-line learning process is to minimize the mean square of control error. Obtained results are compared with results reached with nonlinear analytically designed PD controller. Here described controller is robust to high level noise and also able to work with relative high delay.

Text and facts may be copied and used freely, but credit should be given to these Proceedings.

All papers were reviewed by members of the scientific committee.