A Study on the Auto-Tuning of a PID Controller using Artificial Neural Network

인공신경망에 의한 PID 제어기 자동동조에 관한 연구

  • 정종대 (한국기술교육대학교 전기공학과)
  • Published : 1996.06.01

Abstract

In this paper, we proposed a PID controller, which could control unknown plants using Artificial Neural Network(ANN) for auto-tuning of the PID parameters. In the proposed algorithm, the parameters of the controller were adjusted to reduce the error of the controlled plant. In this process, the sensitivity between input and output of the unknown plant was needed. So, in order to obtain this sensitivity, the ANN's learnig ability was used. Computer simualtions were performed for the regulation problems, and the results were compared with those of Ziegler-Nichols PID controller. As a result, it was shown that the proposed algorithm outperformed Ziegler-Nichols controller in rise time, overshoot, undershoot, and setting time.

Keywords

References

  1. Proc. ACC Simulation studies of an analytical fuzzy tuner for a PID servo C.W.de Silva
  2. Proc. ACC System identification and control using diagonal redurrent neural networks Chao-Chee Ku;Kwang,Y.Lee
  3. IEEE Tr. Neural Networks v.1 no.1 Identification and control of dynamic systems using neural networks Kumpati,S.Narendra; Kannan Parthasarathy
  4. Proc. IEEE Realization of PID controls by fuzzy control methods M.Mizumoto
  5. Proc. ACC Optimal PID controller setting using integral performance criteria M.Zhuang;D.D.Atherton
  6. Proc. ACC Novel neural network based self-tuning PID controller which uses pattern recognition technique Roman,W.Swiniarski
  7. IEEE Tr. Neural Networks v.3 no.1 Comparison of four neural net learning methods for dynamic system identification Si-Zhao Qin;Hong-Te Su;Thomas,J.McAvoy
  8. Proc. ACC Closed-loop tuning of PID controllers Tor Steinar Schei
  9. IEEE Tr. on Sys., man and Cyber. v.23 no.5 Fuzzy gain scheduling of PID controllers Zhen-Yu Zhao(et al.)
  10. 전자공학회논문지 v.27 no.2 직류 서어보 전공기제어를 위한 직업 극배치 PID 자기 동조 제어기의 설계 남문현;이규영