Publication detail
Selflearning controller of active magnetic bearing
TUREK, M. BŘEZINA, T. PULCHART, J.
Czech title
Učící se kontroler aktivního magnetického ložiska
English title
Selflearning controller of active magnetic bearing
Type
conference paper
Language
en
Original abstract
The active magnetic bearing control through self learning controller is described in this contribution. controller's coefficient (parameter) values come from actions of Continuous Action Reinforcement Learning Automatas (CARLAs) whic continuously update the controller's coefficients according to behavior of the active magnetic bearing. The goal of this on-line training is formulated as achievement of minimum mean square of control error. It is shown that CARLA method is capable of learning better parameters than standard method of optimal control design called LQ (linear quadratic) design. Described concept of control is proved by control of the active magnetic bearing.
Czech abstract
V tomto příspěvku je popsáno řízení aktivního magnetického ložiska pomocí učícího se kontroleru. Parametry kontroleru jsou učeny pomocí metody CARLA, která je aktualizuje na základě chování aktivního magnetického ložiska. Cílem on-line učení je minimalizace kvadrátu chyby řízení. Je ukázáno, že pomocí metody CARLA lze nalézt lepší parametry než jsou parametry navržené metodou optimálního návrhu řízení zvanou LQ (linear quadratic) design. Popsaný koncept řízení je poté ověřen na řízení reálného aktivního magnetického ložiska.
English abstract
The active magnetic bearing control through self learning controller is described in this contribution. controller's coefficient (parameter) values come from actions of Continuous Action Reinforcement Learning Automatas (CARLAs) whic continuously update the controller's coefficients according to behavior of the active magnetic bearing. The goal of this on-line training is formulated as achievement of minimum mean square of control error. It is shown that CARLA method is capable of learning better parameters than standard method of optimal control design called LQ (linear quadratic) design. Described concept of control is proved by control of the active magnetic bearing.
Keywords in English
Active Magnetic Bearing, Continuous Action Reinforcement Learning Automata
RIV year
2006
Released
15.05.2006
Book
Book of extended abstracts
Pages from–to
392–393
Pages count
2
BIBTEX
@inproceedings{BUT19037,
author="Milan {Turek} and Tomáš {Březina} and Jaroslav {Pulchart},
title="Selflearning controller of active magnetic bearing",
booktitle="Book of extended abstracts",
year="2006",
month="May",
pages="392--393"
}