Transactions on Control, Automation and Systems Engineering

Institute of Control, Robotics and Systems (제어로봇시스템학회)
권호 표지

Process Control Using n Neural Network Combined with the Conventional PID Controllers

  • Lee, Moonyong (School of Chemical Engineering and Technology, Yeungnam University) ;
  • Park, Sunwon (Dept. of Chemical Engineering, KAIST)
  • Published : 2000.09.01
Download PDF
⟨ Previous Next ⟩

Abstract

A neural controller for process control is proposed that combines a conventional multi-loop PID controller with a neural network. The concept of target signal based on feedback error is used fur on-line learning of the neural network. This controller is applied to distillation column control to illustrate its effectiveness. The result shows that the proposed neural controller can cope well with disturbance, strong interactions, time delays without any prior knowledge of the process.

Keywords

References

  1. IEEE Trans. Syst. Man, Cybern. v.SMC 13 no.5 Neuronlike adaptive elements that can difficult learning control problems A. G. Barto;R. S. Sutton;C. W. Anderson
  2. IEEE J. Robotics and Automation v.RA-3 no.2 Sensor-based control of robotic manipulators using a general learning algorithm W. Miller
  3. IEEE Int. Conf. on Neural Networks A neuromorphic controller with a human teacher A. Guez;J. Selinsky
  4. IEEE Control Systems Magazine A multi layered neural network controller D. Psalitis;A. Sideris;A. A. Yamamura
  5. IEEE Control Systems Magazine Hierarchial neural network model for voluntary movement with application to robotics M. Kawato;Y. Uno;M. Isobe;R. Suzuki
  6. AIChE J. v.38 A new scheme combining neural feedforward control with model-predictive control M. Lee;S. Park
  7. Proc. IEEE 1989 Intl. Conf. Neural Networks Theory of the backpropagation neural network R. Hecht-Nielsen
  8. Chem. Eng. Sci. v.28 Terminal composition control of a binary distillation column R. K. Wood;M. W. Berry