Support vector regression for structural identification via component-mode synthesis

被引:0
|
作者
Zhang, J.
Sato, T.
Iai, S.
机构
[1] Univ Calif San Diego, Dept Struct Engn, La Jolla, CA 92093 USA
[2] Kobegakuin Univ, Sect Disaster Management & Social Serv, Kobe, Hyogo 6512180, Japan
[3] Kyoto Univ, Dept Civil & Earth Resources Engn, Kyoto 6110011, Japan
关键词
D O I
10.12989/sem.2007.25.5.631
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
A support vector regression (SVR)-based structural identification method is developed through a component-mode synthesis (CMS) technique. The CMS method transforms the structural identification equation for the whole structural system in the original coordinate to several uncoupled sub-structural formulas in the normal co-ordinate. The CMS methods are classified to free interface, fixed interface, and hybrid methods according to whether the degrees of freedom at the interfaces are free, constrained or partially constrained. The measurement of the input force is not necessary in this approach, because the effect of the input force can be expressed in terms of the interface node responses. The CMS reduces the number of unknown parameters comprised in the identification equation, guaranteeing the SVR work efficiently in a low dimension.
引用
收藏
页码:631 / 636
页数:6
相关论文
共 50 条
  • [1] Support vector regression for structural identification
    Zhang, J
    Sato, T
    [J]. Smart Structures and Materials 2005: Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace, Pts 1 and 2, 2005, 5765 : 720 - 730
  • [2] Inverse component-mode synthesis method for damped large structural systems
    Takewaki, I
    Uetani, K
    [J]. COMPUTERS & STRUCTURES, 2000, 78 (1-3) : 415 - 423
  • [3] Inverse component-mode synthesis method for redesign of large structural systems
    Takewaki, I
    [J]. COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING, 1998, 166 (3-4) : 201 - 209
  • [4] IMPROVED COMPONENT-MODE REPRESENTATION FOR STRUCTURAL DYNAMIC ANALYSIS
    RUBIN, S
    [J]. AIAA JOURNAL, 1975, 13 (08) : 995 - 1006
  • [5] Improved component-mode synthesis for nonclassically damped systems
    Liu, M. H.
    Zheng, G. T.
    [J]. AIAA JOURNAL, 2008, 46 (05) : 1160 - 1168
  • [6] Novel support vector regression for structural system identification
    Zhang, Jian
    Sato, Tadanobu
    Iai, Susumu
    [J]. STRUCTURAL CONTROL & HEALTH MONITORING, 2007, 14 (04): : 609 - 626
  • [7] Component-mode synthesis method variants in the dynamics of coupled structures
    Biondi, B
    Muscolino, G
    [J]. MECCANICA, 2000, 35 (01) : 17 - 38
  • [8] A PROCEDURE FOR IMPROVING COMPONENT-MODE REPRESENTATION IN STRUCTURAL DYNAMIC ANALYSIS
    JEZEQUEL, L
    TCHERE, ST
    [J]. JOURNAL OF SOUND AND VIBRATION, 1991, 144 (03) : 409 - 419
  • [9] Efficient Mode of Action Identification by Support Vector Machine Regression
    Bevilacqua, Vitoantonio
    Pannarale, Paolo
    [J]. EMERGING INTELLIGENT COMPUTING TECHNOLOGY AND APPLICATIONS, 2012, 304 : 191 - 196
  • [10] Dynamic analysis of cooling tower by using component-mode synthesis
    Huang, ZL
    Wu, JK
    Wang, DJ
    Ren, ZM
    [J]. NATURAL DRAUGHT COOLING TOWERS, 1996, : 383 - 387