ESTIMATION OF DYNAMIC SYSTEM'S EXCITATION FORCES BY THE INDEPENDENT COMPONENT ANALYSIS

被引:8
|
作者
Akrout, Ali [1 ]
Tounsi, Dhouha [1 ]
Taktak, Mohamed [1 ]
Abbes, Mohamed Slim [1 ]
Haddar, Mohamed [1 ]
机构
[1] Natl Sch Engineers Sfax, Mech Modeling & Prod Res Unit, Dept Mech Engn, Sfax 3038, Tunisia
关键词
Blind source separation; independent component analysis; excitation forces; beams; mass-spring system; vibratory responses; BLIND SOURCE SEPARATION; BEHAVIOR;
D O I
10.1142/S1758825112500329
中图分类号
O3 [力学];
学科分类号
08 ; 0801 ;
摘要
This paper deals with a numerical investigation for the estimation of dynamic system's excitation sources using the independent component analysis (ICA). In fact, the ICA concept is an important technique of the blind source separation (BSS) method. In this case, only the dynamic responses of a given mechanical system are supposed to be known. Thus, the main difficulty of such problem resides in the existence of any information about the excitation forces. For this purpose, the ICA concept, which consists on optimizing a fourth-order statistical criterion, can be highlighted. Hence, a numerical procedure based on the signal sources independency in the ICA concept is developed. In this work, the analytical or the finite element (FE) dynamic responses are calculated and exploited in order to identify the excitation forces applied on discrete (mass-spring) and continuous (beam) systems. Then, estimated results obtained by the ICA concept are presented and compared to those achieved analytically or by the FE and the modal recombination methods. Since a good agreement is obtained, this approach can be used when the vibratory responses of a dynamic system are obtained through sensor's measurements.
引用
收藏
页数:26
相关论文
共 50 条
  • [41] Model order estimation for independent component analysis of epoched EEG signals
    Rasmussen, Peter Mondrup
    Morup, Morten
    Hansen, Lars Kai
    Arnfred, Sidse M.
    BIOSIGNALS 2008: PROCEEDINGS OF THE FIRST INTERNATIONAL CONFERENCE ON BIO-INSPIRED SYSTEMS AND SIGNAL PROCESSING, VOL II, 2008, : 3 - 10
  • [42] Noisy independent component analysis, maximum likelihood estimation, and competitive learning
    Hyvarinen, A
    IEEE WORLD CONGRESS ON COMPUTATIONAL INTELLIGENCE, 1998, : 2282 - 2287
  • [43] Multicore Power Estimation using Independent Component Analysis based Modeling
    Sagi, Mark
    Nguyen Anh Vu Doan
    Wild, Thomas
    Herkersdorf, Andreas
    2019 IEEE 13TH INTERNATIONAL SYMPOSIUM ON EMBEDDED MULTICORE/MANY-CORE SYSTEMS-ON-CHIP (MCSOC 2019), 2019, : 38 - 45
  • [44] Large MIMO Channel Estimation Study Based on Independent Component Analysis
    Takahashi, Yuya
    Ikegami, Tetsushi
    2023 25TH INTERNATIONAL CONFERENCE ON ADVANCED COMMUNICATION TECHNOLOGY, ICACT, 2023, : 85 - 88
  • [45] Estimation for conditional higher moments risk based on independent component analysis
    Xu, Qi-Fa
    Jiang, Cui-Xia
    PROCEEDINGS OF 2006 INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND CYBERNETICS, VOLS 1-7, 2006, : 2358 - +
  • [46] Improving Channel Estimation In OFDM Systems via Independent Component Analysis
    Homayounzadeh, A.
    Shirazi, M. A. Masnadi
    PROCEEDINGS OF THE 2009 INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING SYSTEMS, 2009, : 196 - 200
  • [47] Dynamic heart rate estimation using principal component analysis
    Yu, Yong-Poh
    Raveendran, P.
    Lim, Chern-Loon
    Kwan, Ban-Hoe
    BIOMEDICAL OPTICS EXPRESS, 2015, 6 (11): : 4610 - 4618
  • [48] Estimation of dynamic models on the factors of marginal principal component analysis
    Mayo, Nathanael
    STATISTICS, 2011, 45 (01) : 101 - 120
  • [49] Dynamic kernel independent component analysis approach for fault detection and diagnosis
    Feng, Lin
    Sun, Rongrong
    2017 CHINESE AUTOMATION CONGRESS (CAC), 2017, : 2193 - 2197
  • [50] An algorithm for real-time independent component analysis in dynamic environments
    Ding, SX
    Wei, DM
    Omata, S
    2004 47TH MIDWEST SYMPOSIUM ON CIRCUITS AND SYSTEMS, VOL III, CONFERENCE PROCEEDINGS, 2004, : 105 - 108