Publication detail
THE CONTROL OF ACTIVE MAGNETIC BEARING USING TWO-PHASE Q-LEARNING
BŘEZINA, T. KREJSA, J.
Czech title
THE CONTROL OF ACTIVE MAGNETIC BEARING USING TWO-PHASE Q-LEARNING
English title
THE CONTROL OF ACTIVE MAGNETIC BEARING USING TWO-PHASE Q-LEARNING
Type
conference paper
Language
cs
Original abstract
Q-learning method proved to be usable in active magnetic bearing (AMB) control task, however the learning speed remains the main problem. Two-phase variant of the Q-learning speeds up the learning process by dividing the method into two stages – prelearning stage which uses mathematical model of AMB and is highly efficient and tutorage stage which uses conventional Q learning and is used for further fining of the Q-function. Comparison with PID controller is included to demonstrate algorithms control abilities. As table (step function) is used as Q-function implementation, the optimum configuration of the table (number and relative positions of node points) was also investigated.
Czech abstract
Q-learning method proved to be usable in active magnetic bearing (AMB) control task, however the learning speed remains the main problem. Two-phase variant of the Q-learning speeds up the learning process by dividing the method into two stages – prelearning stage which uses mathematical model of AMB and is highly efficient and tutorage stage which uses conventional Q learning and is used for further fining of the Q-function. Comparison with PID controller is included to demonstrate algorithms control abilities. As table (step function) is used as Q-function implementation, the optimum configuration of the table (number and relative positions of node points) was also investigated.
English abstract
Q-learning method proved to be usable in active magnetic bearing (AMB) control task, however the learning speed remains the main problem. Two-phase variant of the Q-learning speeds up the learning process by dividing the method into two stages – prelearning stage which uses mathematical model of AMB and is highly efficient and tutorage stage which uses conventional Q learning and is used for further fining of the Q-function. Comparison with PID controller is included to demonstrate algorithms control abilities. As table (step function) is used as Q-function implementation, the optimum configuration of the table (number and relative positions of node points) was also investigated.
Keywords in English
Q-learning, active magnetic bearing (AMB), control
RIV year
2003
Released
12.05.2003
Publisher
Institute of Theoretical and Applied Mechanics, Academy of Sciences of the Czech Republic, Prague
Location
Praha
ISBN
80-86246-18-3
Book
Engeneering Mechanics 2003
Edition number
1
Pages count
2
BIBTEX
@inproceedings{BUT7625,
author="Tomáš {Březina} and Jiří {Krejsa},
title="THE CONTROL OF ACTIVE MAGNETIC BEARING USING TWO-PHASE Q-LEARNING",
booktitle="Engeneering Mechanics 2003",
year="2003",
month="May",
publisher="Institute of Theoretical and Applied Mechanics, Academy of Sciences of the Czech Republic, Prague",
address="Praha",
isbn="80-86246-18-3"
}