A new neural linearizing control scheme using radial basis function network

Radial basis function 회로망을 이용한 새로운 신경망 선형화 제어구조

  • Kim, Seok-Jun ;
  • Lee, Min-Ho (Korea Maritime University) ;
  • Park, Seon-Won (Dept. of Chemistry Engineering, Korea Advanced Institute of Science and Technology) ;
  • Lee, Su-Yeong (Dept. of Electrical Electronic Engineering, Korea Advanced Institute of Science and Technology) ;
  • Park, Cheol-Hun (Dept. of Electrical Electronic Engineering, Korea )
  • 김석준 (주식회사 유공, 생산기술센터) ;
  • 이민호 (한국해양대학교 전기공학과) ;
  • 박선원 (한국과학기술원 화학공학과) ;
  • 이수영 (한국과학기술원 전기및전자공학과) ;
  • 박철훈 (한국과학기술원 전기및전자공학과)
  • Published : 1997.10.01

Abstract

To control nonlinear chemical processes, a new neural linearizing control scheme is proposed. This is a hybrid of a radial basis function(RBF) network and a linear controller, thus the control action applied to the process is the sum of both control actions. Firstly, to train the RBF newtork a linear reference model is determined by analyzing the past operating data of the process. Then, the training of the RBF newtork is iteratively performed to minimize the difference between outputs of the process and the linear reference model. As a result, the apparent dynamics of the process added by the RBF newtork becomes similar to that of the linear reference model. After training, the original nonlinear control problem changes to a linear one, and the closed-loop control performance is improved by using the optimum tuning parameters of the linear controller for the linear dynamics. The proposed control scheme performs control and training simultaneously, and shows a good control performance for nonlinear chemical processes.

References

  1. Computers Chem. Engng. v.14 Use of neural nets for dynamic modeling and control of chemical process systems N. V. Bhat;T. J. McAvoy
  2. IEEE Trans. on Neural Networks v.2 Orthogonal least square learning algorithm for redial basis frunction networks S. Chen;F. N. cowna;P. M. Grant
  3. AIChE Jl v.36 A critique of exact linearization strategies for process control M. A. Henson;D. E. Seborg
  4. Procs. of International Joint Conf. on Neural Network v.1 no.3 Optimal control with a recurrent neural networks and a priori knowledge of the system T. M. Hoshio, Kano;T. Endo
  5. Forward models: supervised learning with a digital teacher M. I. Jordan;D. E. Rumelhart
  6. Nenural Networks v.1 Feedback-error-learning neural network for trajectory control of a robotic manipulator M. Kawato
  7. World Congress on Neural Networks v.2 Neural linearizing control with radial basis function network for chemical process S. J. Kim;M. Lee;S. Park;S. Y. Lee;C. H. Park
  8. Neural networks for signal processing B. Kosko
  9. Neural Networks v.4 A gaussian potential function network with hierarchically self-organizing learning S. Lee;R. M. Kil
  10. ALChE Jl v.38 A new scheme combining neural feedforward control with model-predective control M. Lee;S. Park
  11. Electronics Letters v.28 no.3 Neural controller of nonlinear dynamic systems using higher order neural networks M. Lee;S.-Y. Lee;C. H. Park
  12. Procs. of International Joint Conf. on Neural Network v.3 Feedforward / feedback neural control stucture and its application to a robotic system M. Lee;S.-Y. Lee;C. H. Park
  13. Procs. of International Joint Conf. on Neural Network v.1 Learning rules for multilayer neural networks using a difference approximation Y. H. Maeda;Yamashita;Y. Kanada
  14. Ind. Engng. Chem Procedss Des. Dev. v.11 Dynamics of pH in CSTR T. J. McAvoy;E. Hsu;S. Lowenthal
  15. Computers Chem Engng. v.16 Nonlinear internal model control strategy for neural networks E. P. Nahas;D. E. Seborg
  16. IEEE Trans. on Neural Networks v.1 no.1 Identification and control of dynamic system using neural networks K. S. Narendra;K. Parthasarathy
  17. IEEE Control Systems Magazine v.10 no.3 Neural networks for self-learning control systems D. H. Nguyen;B. Widrow
  18. A. I. Memo no.1140 A theroy of networks for approximation and learning T. Poggio;F. Girosi