Fast Training of Neural Networks for Nonlinear Dynamic System Identification

被引：1
|
作者：
Deflorian, Michael [1 ]
Kloepper, Florian [2 ]
机构：
[1] BMW Grp, Forsch & Innovat Zentrum, D-80788 Munich, Germany
[2] BMW AG, Forsch & Innovat Zentrum, D-80788 Munich, Germany
来源：
AT-AUTOMATISIERUNGSTECHNIK | 2011年 / 59卷 / 02期
关键词：
Recurrent neural networks; autocorrelated error; NARX; NOE; LEAST-SQUARES ALGORITHM; REGRESSION;
D O I：
10.1524/auto.2011.0898
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
The time consuming training of dynamic neural networks often prohobits the industrial application. This paper presents a fast method to train neural networks with external dynamics for nonlinear dynamic system identification. The proposed method selects first a suitable set of neurons and trains the resulting network in a second step. This proposed procedure leads to faster convergence compared to standard training algorithms and comparable accuracy.
引用
收藏
页码：75 / 83
页数：9
相关论文
共 50 条
  • [1] Nonlinear dynamic system identification with Dynamic Recurrent Neural Networks
    Calderon, G
    Draye, JP
    Pavisic, D
    Teran, R
    Libert, G
    [J]. INTERNATIONAL WORKSHOP ON NEURAL NETWORKS FOR IDENTIFICATION, CONTROL, ROBOTICS, AND SIGNAL/IMAGE PROCESSING - PROCEEDINGS, 1996, : 49 - 54
  • [2] NEURAL NETWORKS FOR NONLINEAR DYNAMIC SYSTEM MODELING AND IDENTIFICATION
    CHEN, S
    BILLINGS, SA
    [J]. INTERNATIONAL JOURNAL OF CONTROL, 1992, 56 (02) : 319 - 346
  • [3] Nonlinear filtering design using dynamic neural networks with fast training
    Becerikli, Y
    [J]. COMPUTER AND INFORMATION SCIENCES - ISCIS 2003, 2003, 2869 : 601 - 610
  • [4] An efficient parameterization of dynamic neural networks for nonlinear system identification
    Becerra, VM
    Garces, FR
    Nasuto, SJ
    Holderbaum, W
    [J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 2005, 16 (04): : 983 - 988
  • [5] Dynamic neural networks with hybrid structures for nonlinear system identification
    Deng, Jiamei
    [J]. ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2013, 26 (01) : 281 - 292
  • [6] Nonlinear system identification using embedded dynamic neural networks
    Yazdizadeh, A
    Khorasani, K
    [J]. IEEE WORLD CONGRESS ON COMPUTATIONAL INTELLIGENCE, 1998, : 378 - 383
  • [7] Robust training algorithm of multilayered neural networks for identification of nonlinear dynamic systems
    Song, Q
    [J]. IEE PROCEEDINGS-CONTROL THEORY AND APPLICATIONS, 1998, 145 (01): : 41 - 46
  • [8] Nonlinear system identification and trajectory tracking using dynamic neural networks
    Poznyak, AS
    Sanchez, EN
    [J]. PROCEEDINGS OF THE 35TH IEEE CONFERENCE ON DECISION AND CONTROL, VOLS 1-4, 1996, : 955 - 960
  • [9] Error analysis for nonlinear system identification using dynamic neural networks
    Poznyak, AS
    Sanchez, EN
    Acosta, G
    [J]. PROCEEDINGS ISAI/IFIS 1996 - MEXICO - USA COLLABORATION IN INTELLIGENT SYSTEMS TECHNOLOGIES, 1996, : 403 - 407
  • [10] Dynamic multilayer neural networks for nonlinear system on-line identification
    Yu, W
    Poznyak, AS
    Sanchez, EN
    [J]. PROCEEDINGS OF THE 2000 IEEE INTERNATIONAL SYMPOSIUM ON INTELLIGENT CONTROL, 2000, : 25 - 30
12345