Efficient Q-learning modification aplied on active magnetic bearing control

Efficient Q-learning modification aplied on active magnetic bearing control

journal article - other

en

The paper is focused on use of Q-learning for active magnetic bearing (AMB) control. Q-learning belongs to the reinforcement learning methods which are the part of real time machine learning approaches. The essence of proposed method is in separating the Q-learning into two phases – prelearning phase, which use mathematical model of real system and tutorage phase, which works with the real system and is used for further improvement of Q-values found during the prelearning phase. Proposed learning method, used in prelearning phase, proved to be highly efficient. Controller based on Q-learning show better results (regarding the number of successful trials) than referential PID controller after only 1000 table passes. The control quality criterion results are comparable. The policy found by learning also shows high robustness against errors of system variables observations, even only very simple reinforcement function in shape of simple reduced penalty is used.

The paper is focused on use of Q-learning for active magnetic bearing (AMB) control. Q-learning belongs to the reinforcement learning methods which are the part of real time machine learning approaches. The essence of proposed method is in separating the Q-learning into two phases – prelearning phase, which use mathematical model of real system and tutorage phase, which works with the real system and is used for further improvement of Q-values found during the prelearning phase. Proposed learning method, used in prelearning phase, proved to be highly efficient. Controller based on Q-learning show better results (regarding the number of successful trials) than referential PID controller after only 1000 table passes. The control quality criterion results are comparable. The policy found by learning also shows high robustness against errors of system variables observations, even only very simple reinforcement function in shape of simple reduced penalty is used.

The paper is focused on use of Q-learning for active magnetic bearing (AMB) control. Q-learning belongs to the reinforcement learning methods which are the part of real time machine learning approaches. The essence of proposed method is in separating the Q-learning into two phases – prelearning phase, which use mathematical model of real system and tutorage phase, which works with the real system and is used for further improvement of Q-values found during the prelearning phase. Proposed learning method, used in prelearning phase, proved to be highly efficient. Controller based on Q-learning show better results (regarding the number of successful trials) than referential PID controller after only 1000 table passes. The control quality criterion results are comparable. The policy found by learning also shows high robustness against errors of system variables observations, even only very simple reinforcement function in shape of simple reduced penalty is used.

Reinforcement learning, Q-learning, Active magnetic bearing

2004

01.11.2004

Association for engineering mechanics, Czech Republic

1210-2717

Inženýrská mechanika - Engineering Mechanics

11/2004

2

14