대한전기학회:학술대회논문집 (Proceedings of the KIEE Conference)

대한전기학회 (The Korean Institute of Electrical Engineers)
권호 표지

신경망을 이용한 PID제어기의 적응동조에 관한 연구

A Study on Adaptive-Tuning of PID Controller Using a Neural Network

  • 김상원 (한국기술교육대학교 전기전자공학과) ;
  • 이홍규 (한국기술교육대학교 전기전자공학과)
  • Kim, Sang-Won (Dept. of Electric and Electronic Eng., Korea Univ. of technology and Education) ;
  • Lee, Hong-Kyu (Dept. of Electric and Electronic Eng., Korea Univ. of technology and Education)
  • 발행 : 1999.07.19
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

In this thesis, We implement the controller system only using the neural network to identify the plant characteristics with keeping the PID controller structure. The neural network has learned by the adaptive learning rates that has suggested by Chao-Chee Ku and the DBP algorithm. We proposed the on-line tuning algorithm about the unknown plant using the adaptive tuning technique. As a result of executing the parameters has tuned from the initial value to more suitable ones and the output of the Plant has improved and also it is appeared that the convergence is guaranteed.

키워드