Publication detail
The active magnetic bearing control with automatic parallel compensation
BŘEZINA, T. TUREK, M. PULCHART, J.
Czech title
Řízení aktivního magnetického ložiska s automatickou paralelní kompenzací
English title
The active magnetic bearing control with automatic parallel compensation
Type
journal article - other
Language
en
Original abstract
The active magnetic bearing control through analytically designed linear PD regulator, with parallel nonlinear compensation represented by automatic approximator is described in this contribution. Coefficient (parameter) values come from actions of Continuous Action Reinforcement Learning Automata (CARLAs). Modified algorithm for automata implementation is used which continuously updates learning parameters according to former learning process. The goal of this on-line training is formulated as achievement of minimum mean square of control error. Described concept of control is proved by simulation study. It is shown that the significant improvement of whole system behavior can be achieved.
Czech abstract
V tomto příspěvku je popsáno řízení aktivního magnetického ložiska pomocí analyticky navrženého lineárního PD regulátoru s nelineární kompenzací vytvořenou pomocí automatické aproximace. Koeficienty (parametry) jsou získávany pomocí metody Continuous Action Reinforcement Learnint Automata (CARLA). Pro implementaci je použit modifikovaný algoritmus, který aktualizuje parametry učení na základě předchozího průběhu učení. Cíl učení je formulován jako dosažení minimální hodnoty kvadrátu chyby řízení. Popsaný koncept řízení je ověřen simulační studií. Je ukázáno, že je možné dosáhnout významného zlepšení chování systému.
English abstract
The active magnetic bearing control through analytically designed linear PD regulator, with parallel nonlinear compensation represented by automatic approximator is described in this contribution. Coefficient (parameter) values come from actions of Continuous Action Reinforcement Learning Automata (CARLAs). Modified algorithm for automata implementation is used which continuously updates learning parameters according to former learning process. The goal of this on-line training is formulated as achievement of minimum mean square of control error. Described concept of control is proved by simulation study. It is shown that the significant improvement of whole system behavior can be achieved.
Keywords in English
Key words: continuous action reinforcement learning automata, active magnetic bearing
RIV year
2005
Released
20.09.2005
Publisher
Enginering Academy of the Czech Republic
Location
Praha, Czech Republic
ISSN
1210-2717
Journal
Inženýrská mechanika - Engineering Mechanics
Volume
2005
Number
12-3
Pages count
7
BIBTEX
@article{BUT42742,
author="Tomáš {Březina} and Milan {Turek} and Jaroslav {Pulchart},
title="The active magnetic bearing control with automatic parallel compensation",
journal="Inženýrská mechanika - Engineering Mechanics",
year="2005",
volume="2005",
number="12-3",
month="September",
publisher="Enginering Academy of the Czech Republic",
address="Praha, Czech Republic",
issn="1210-2717"
}