全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...
化工学报  2015 

基于稳定Hammerstein模型的在线软测量建模方法及应用

DOI: 10.11949/j.issn.0438-1157.20141210, PP. 1378-1387

Keywords: Hammerstein模型,在线建模,软测量,预测,稳定学习,污水处理过程,稳定性

Full-Text   Cite this paper   Add to My Lib

Abstract:

针对复杂工业过程中由于存在未建模动态和不确定干扰,导致关键变量的软测量精度下降的问题,提出了一种基于稳定Hammerstein模型(H模型)的在线软测量建模方法。H模型的非线性增益采用带有时变稳定学习算法的小波神经网络模型,线性系统部分采用基于递推最小二乘的ARX模型,基于输入到状态稳定性理论证明了H模型辨识误差的有界性。其中小波神经网络具有表征强非线性的特性,稳定学习算法可抑制未建模动态和不确定干扰的影响,改善了模型的预测精度和自适应能力。以典型非线性系统和实际污水处理过程为例进行了仿真研究,结果表明,基于稳定H模型的软测量方法具有较高的在线软测量精度。

References

[1]  Chai Tianyou (柴天佑). Operational optimization and feedback control for complex industrial processes [J]. Acta Automatics Sinica (自动化学报), 2013, 39 (11): 1744-1757
[2]  Yu W. Nonlinear system identification using discrete-time recurrent neural networks with stable learning algorithms [J]. Information Sciences, 2004, 158 (1): 131-147
[3]  Sontag, E D, Wang Y. On characterizations of the input-to-state stability property [J]. Systems and Control Letters, 1995, 24 (5): 351-359
[4]  Feng Peidi (冯培悌). System Identification (系统辨识) [M]. Hangzhou: Zhejiang University Press, 1999
[5]  Wang Shuangjian (王双剑), Chu Jizheng (楚纪正). Hammerstein model identification method based on the new improved neural dynamics and its application to hybrid modeling [J]. Information and Control (信息与控制), 2012, 41 (3): 384-390
[6]  Kreinovich V S O, Cabmn S. Wavelet neural networks are asymptotically optimal approximators for function of one variable// Proceeding of IEEE ICNN [C]. Florida, USA, 1994: 299-304
[7]  Ye Lingjian (叶凌箭), Ma Xiushui (马修水), Song Zhihuan (宋执环). Real-time optimization for chemical processes based on on-line modeling of controlled variables [J]. CIESC Journal (化工学报), 2013, 64 (8): 2918-2923
[8]  Sontag E D. Input to state stability: basic concepts and results [J]. Lecture Notes in Mathematics, 2008, 1932: 163-200
[9]  Narendra K S, Li S M. Neural networks in control systems// Smolensky P, Mozer M C, Rumelhart D E, Mathematical Perspectives on Neural Networks [M]. New Jersey: Lawrence Erlbaum Associates, 1996: 347-394
[10]  Jia Li (贾立), Li Xunlong (李训龙). Identification of Hammerstein model: review and prospect [J]. Control Theory & Applications (控制理论与应用), 2014, 31 (1): 1-10
[11]  Jia L, Yang A H, Chiu M S. Special input signals based neurofuzzy Hammerstein-Wiener model and its application [J]. International Journal of System Control and Information Processing, 2012, 1 (2): 199-218
[12]  Hunt K J, Sbarbaro D, Zbikowski R, Gawthrop P J. Neural networks for control systems-a survey [J]. Automatica, 1992, 28 (6): 1083-1112
[13]  Chen Senfa (陈森发). Theory and Method of Complex System Modeling (复杂系统建模理论与方法) [M]. Nanjing: Southeast University Press, 2005
[14]  Cao Pengfei (曹鹏飞), Luo Xionglin (罗雄麟). Modeling of soft sensor for chemical process[J]. CIESC Journal(化工学报), 2013, 64 (3): 788-800
[15]  Yu W, Li X. Discrete-time neuro identification without robust modification [J]. IEE Proceedings -Control Theory and Applications, 2003, 150 (3): 311-316
[16]  Zhang N M. An online gradient method with momentum for two-layer feedforward neural networks [J]. Applied Mathematics and Computation, 2009, 212 (2): 488-498
[17]  Chen Y J, Huang T C, Hwang R C. An effective learning of neural network by using RFBP learning algorithm [J]. Information Sciences, 2004, 167 (1): 77-86
[18]  Narendra K S, Gallman P G. An iterative method for the identification of nonlinear systems using a Hammerstein model [J]. IEEE Transactions on Automatic Control, 1966, 11 (3): 546-550
[19]  Lakshminarayanan S, Shah S L, Nandakumar K. Modeling and control of multivariable processes: dynamic PLS approach [J]. AIChE Journal, 1997, 43 (9): 2307-2322
[20]  Wlodzimierz G. Stochastic approximation in nonparametric identification of Hammerstein systems [J]. IEEE Transactions on Automatic Control, 2002, 47 (11): 1800-1811
[21]  Zhao W X. Parametric identification of Hammerstein systems with consistency results using stochastic inputs [J]. IEEE Transactions on Automatic Control, 2010, 55 (2): 474-480

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133