オフライン学習のリカレントニューラルネットワークを用いたパフォーマンス駆動型制御系の提案

抄録

To maintain high performance of controlling and minimize the total cost for readjustments of the controller parameters, determination the appropriate timing for readjustment the controller parameters is important. This paper proposes a design of performance-driven control system to maintain the system performance by detecting the system change using the recurrent neural networks and readjusting the control parameters based on Fictitious Reference Iterative Tuning (FRIT). This paper conducts some numerical experiments to verify the availability of the proposed method to some systems.