ISSN: 0304-128X ISSN: 2233-9558
Copyright © 2024 KICHE. All rights reserved

Articles & Issues

Conflict of Interest
In relation to this article, we declare that there is no conflict of interest.
articles This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/bync/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright © KIChE. All rights reserved.

All issues

신경회로망을 이용한 비선형 화학공정의 실시간 학습제어

On-Line Learning Control of Nonlinear Chemical Processes Using Neural Networks

HWAHAK KONGHAK, April 1999, 37(2), 133-140(8), NONE
downloadDownload PDF

Abstract

본 논문은 신경회로망에 근거한 비선형 동적씨스템의 실시간 학습제어를 위한 직접 역제어의 개선된 구조와 학습방법 을 제시한다. 제안된 방법은 기존 신경망을 이용한 직접 역제어의 경우플랜트의 Jacobian을정확히 구하지 못해서 발생 하는 악영향을, 한 단계 앞 예측모델과 보다 강건하고 정확한 신경회로망의 연결가중치를 갱신하는 방법인 batch updating을 도입하여 해결하였다. 이 방법은 공정의 예측모델을 사용하므로 시간지연이 크고 품질변수를 실시간으로 측정할 수 없는 공정에도 직접 역제어를 적용할 수 있는 장점을 가진다. 제시된 개선된 구조의 학습빙법은 receding horizon control으로도 해석될 수 있다. 제시한 방법을 회분식 반응기 개시조업의 최적 되먹임 제어에 적용한 사례연구를 통하여 보였다. 사례연구 결과, 제안한 방법은 기존 방법에 비해 좋은 제어성능을 보였으며 측정잡음, 입력외란, 모델-플랜트 불일치 등에 대한 강건성도 뛰어남을 알 수 있었다.
This paper presents a modified neural network structure and a learning algorithm for direct inverse control of nonlinear and dynamic chemical processes. The major drawback of the typical direct inverse control, the difficuly of estimating the Jacobian for the plant, has been solved by introducing one-step ahead prediction model and batch updating rules. Using process prediction model offers the direct inverse control the ability to handle the processes with large time delays and quality variables that are difficult to measure on-line. The learning algorithm of the proposed architecture can be viewed as a receding horizon control. The proposed methods are illustrated by application case studies for optimal feedback control of batch reactor startup operations. The case studies have shown the proposed methods are better in terms of control performance and more robust to measurement noises, input disturbances, and model-plant mismatch than the previous ones.

References

Luss R, Int. J. Control, 52(1), 239 (1990)
Luus R, J. Proc. Cont., 4(4), 218 (1994) 
Lee MH, Han C, Chang KS, J. Control. Automation Systems Eng., 3(1), 46 (1997)
Goh CJ, Edwards NJ, Zoyama AY, Opt. Cont. App. Methods, 14, 1 (1993)
Morris AJ, Montague GA, Willis MJ, Trans. IChemE, 72(Part A), 3 (1994)
Hunt KJ, Sbarbaro D, Zbikowski R, Gawthrop PJ, Automatica, 28(6), 1083 (1992) 
Psaltis D, Sideris A, Yamamura AA, IEEE Cont. Syst. Mag., 4(17), 17 (1988)
Nguyen D, Widrow B, IEEE Cont. Syst. Mag., 10(3), 18 (1990) 
Yabuta T, Yamada T, IEEE Trans. Syst. Man Cybernet., 22(1), 170 (1992) 
Wang SD, Yeh HMS, Int. Joint Conf. On Neural Networks, June (1990)
Yen Y, Liu TZ, Liu DY, IEE Proc. Control Theory Appl., 42(5), 493 (1995) 
Kim SJ, Park S, Lee M, J. Chem. Eng. Jpn., 29(5), 805 (1996) 
Kim SJ, Lee M, Park S, Lee SY, Park CH, Comput. Chem. Eng., 21(2), 201 (1996)
Kim SJ, Park S, ICASE MAGAZINE, 2(1), 48 (1996)
Fausett L, "Fundamentals of Neural Networks: Architectures, Algorithms and Applications," Prentice Hall, New Jersey, NJ (1994)
Haykin S, "Neural Networks: A Comprehensive Foundation," Macmillan (1994)
Hagan MT, Demuth HB, Beale M, "Neural Network Design," PWS, Boston (1996)
Widrow B, Walach E, "Adaptive Inverse Control," Prentice Hall, New Jersey, NJ (1996)
Bode HW, Shannon CE, Proc. IRE, 38, 417 (1950)
Hornik K, Stinchcombe M, White H, Neural Networks, 2, 359 (1989) 
Jordan MI, Rumelhart DE, Cognitive Sci., 16, 307 (1992) 
Kawato M, Furukawa L, Suzuki R, Biological Cybernetics, 57, 169 (1987) 
Mejdell T, Skogestad S, Ind. Eng. Chem. Res., 30, 2543 (1991) 
Song JJ, Park S, J. Chem. Eng. Jpn., 26(4), 347 (1991) 
Bhat N, McAvoy TJ, Comput. Chem. Eng., 14(4-5), 573 (1990) 
Sontag ED, Technical Report, Rutgers Center for System and Control, Dept. of Mathematics, Rutgers University

The Korean Institute of Chemical Engineers. F5, 119, Anam-ro, Seongbuk-gu, 233 Spring Street Seoul 02856, South Korea.
Phone No. +82-2-458-3078FAX No. +82-507-804-0669E-mail : kiche@kiche.or.kr

Copyright (C) KICHE.all rights reserved.

- Korean Chemical Engineering Research 상단으로