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
  • [1] Estimation of system harmonic impedance using independent component analysis
    Lian, S. (suo08@eis.osakafu-u.ac.jp), 1600, Institute of Electrical Engineers of Japan (133):
  • [2] CO COMPONENT ESTIMATION BASED ON THE INDEPENDENT COMPONENT ANALYSIS
    Ichiki, Kiyotomo
    Kaji, Ryohei
    Yamamoto, Hiroaki
    Takeuchi, Tsutomu T.
    Fukui, Yasuo
    ASTROPHYSICAL JOURNAL, 2014, 780 (01):
  • [3] Amplitude and Frequency Estimation of Power System Signals Using Independent Component Analysis
    Uzunoglu, Cengiz Polat
    Ugur, Mukden
    Turan, Faruk
    Cekli, Serap
    2013 21ST SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU), 2013,
  • [4] Generalized independent component analysis as density estimation
    Palmieri, F
    Budillon, A
    NEURAL NETS, 2002, 2486 : 83 - 89
  • [5] Sequential Independent Component Analysis Density Estimation
    Aladjem, Mayer
    Israeli-Ran, Itamar
    Bortman, Maria
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2018, 29 (10) : 5084 - 5097
  • [6] Harmonic State Estimation for a Simple Power System Model using Independent Component Analysis
    Supriya, P.
    Nambiar, T. N. P.
    PROCEEDINGS OF 2015 IEEE 9TH INTERNATIONAL CONFERENCE ON INTELLIGENT SYSTEMS AND CONTROL (ISCO), 2015,
  • [7] System harmonic impedance estimation method based on sub-band component decomposition and independent component analysis
    Lin S.
    Yan X.
    Zhong L.
    Li D.
    Fu Y.
    Dianli Zidonghua Shebei/Electric Power Automation Equipment, 2021, 41 (01): : 179 - 185
  • [8] R-Estimation for Asymmetric Independent Component Analysis
    Hallin, Marc
    Mehta, Chintan
    JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 2015, 110 (509) : 218 - 232
  • [9] Independent component analysis based on nonparametric density estimation
    Boscolo, R
    Pan, H
    Roychowdhury, VP
    IEEE TRANSACTIONS ON NEURAL NETWORKS, 2004, 15 (01): : 55 - 65
  • [10] Design of estimation/deflation approaches to independent component analysis
    Douglas, SC
    Kung, SY
    CONFERENCE RECORD OF THE THIRTY-SECOND ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS & COMPUTERS, VOLS 1 AND 2, 1998, : 707 - 711