Model structure selection for nonlinear system identification using feedforward neural networks

被引:0
|
作者
Petrovic, I [1 ]
Baotic, M [1 ]
Peric, N [1 ]
机构
[1] Univ Zagreb, Fac Elect & Comp Engn, Dept Control & Comp Engn Automat, HR-10000 Zagreb, Croatia
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
A nonlinear black-box structure for a dynamic system is a model structure that is prepared to describe virtually any nonlinear dynamics. The majority of nonlinear models based on neural networks are of the black-box structure. A nonlinear system can be nonlinear in many different ways, thus the nonlinear black-box model structure must be very flexible. This means that it must have many parameters. A model offering many parameters usually creates problems, and the variance contribution to the error might be high. For a particular identification problem. only a subset of the parameters might be necessary and the main topic in nonlinear system identification is how to select a model structure that describes the system dynamics with the minimum number of parameters. This paper discusses nonlinear input-output models that are suitable for implementation of feedforward neural networks. The proposed model structures were tested and compared using the identification procedure of a pH process. The results indicated that it would be worthwhile using the simplest model structure that can satisfactorily represent the investigated process.
引用
收藏
页码:53 / 57
页数:5
相关论文
共 50 条
  • [1] Nonlinear system identification and model reduction using artificial neural networks
    Prasad, V
    Bequette, BW
    [J]. COMPUTERS & CHEMICAL ENGINEERING, 2003, 27 (12) : 1741 - 1754
  • [2] Nonlinear System Identification Using Neural Networks
    Purwar, S.
    Kar, I. N.
    Jha, A. N.
    [J]. IETE JOURNAL OF RESEARCH, 2007, 53 (01) : 35 - 42
  • [3] Method for the selection of inputs and structure of feedforward neural networks
    Saxen, H.
    Pettersson, F.
    [J]. COMPUTERS & CHEMICAL ENGINEERING, 2006, 30 (6-7) : 1038 - 1045
  • [4] A priori nonlinear model structure selection for system identification
    Petrick, MH
    Wigdorowitz, B
    [J]. CONTROL ENGINEERING PRACTICE, 1997, 5 (08) : 1053 - 1062
  • [5] Variable selection using feedforward and recurrent neural networks
    Munoz, A
    Czernichow, T
    [J]. ENGINEERING INTELLIGENT SYSTEMS FOR ELECTRICAL ENGINEERING AND COMMUNICATIONS, 1998, 6 (02): : 91 - 102
  • [6] NONLINEAR-SYSTEM IDENTIFICATION USING NEURAL NETWORKS
    CHEN, S
    BILLINGS, SA
    GRANT, PM
    [J]. INTERNATIONAL JOURNAL OF CONTROL, 1990, 51 (06) : 1191 - 1214
  • [7] Nonlinear model identification using soft neural networks
    Zhang, YQ
    [J]. WORLD MULTICONFERENCE ON SYSTEMICS, CYBERNETICS AND INFORMATICS, VOL 1, PROCEEDINGS: ISAS '98, 1998, : 672 - 678
  • [8] Hammerstein model identification by multilayer feedforward neural networks
    AlDuwaish, H
    Karim, MN
    Chandrasekar, V
    [J]. INTERNATIONAL JOURNAL OF SYSTEMS SCIENCE, 1997, 28 (01) : 49 - 54
  • [9] Feedforward neural networks using RPROP algorithm and its application in system identification
    Zhou, LR
    Han, P
    Jiao, SM
    Lin, BH
    [J]. 2002 INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND CYBERNETICS, VOLS 1-4, PROCEEDINGS, 2002, : 2041 - 2044
  • [10] A simple method for selection of inputs and structure of feedforward neural networks
    Saxén, H
    Pettersson, F
    [J]. ADAPTIVE AND NATURAL COMPUTING ALGORITHMS, 2005, : 9 - 12