A system theory approach to multi-input multi-output modal parameters identification methods
A system theory approach to multi-input multi-output modal parameters identification methods
复制标题
多输入多输出模态参数识别方法的系统论方法
DOI：
10.1006/mssp.1994.1014
复制
发表时间：
1994
影响因子：
8.4
通讯作者：
X. Wu
中科院分区：
文献类型：
--
作者：
Q. J. Yang;P. Q. Zhang;Cong;X. Wu
摘要 本文从系统论的角度提出了一种多输入多输出（MIMO）模态参数辨识技术的新理论。论文第一部分从可控性和可观性理论出发介绍了MIMO辨识的必要性。然后，通过系统状态方程的线性变换，给出了描述振动系统MIMO之间关系的数学模型；当通过测量振动系统的输入和输出来估计模型参数时，可以通过重建系统状态方程来识别模态参数。论文中还给出了识别公式。该理论可以作为一种统一的方法，许多著名的识别方法，如直接参数模型、多参考、Ibrahim时域方法和特征系统实现算法等都可以派生为特殊形式。这些在本文的最后部分中介绍，并显示了所有这些方法之间的关系。
Abstract In this paper, a new theory of multi-input multi-output (MIMO) modal parameters identification technique is developed from the view of system theory. In the first part of the paper, the necessity of the MIMO identification is introduced from controllability and observability theory. Then, by linear transformation of system state equations, a mathematical model which describes the relation between a vibration system's MIMO, is given; when the model parameters are estimated from the measurement of the vibration system's input and output, the modal parameters can be identified by means of reconstructing the system state equations. Formulae on the identification are also presented in the paper. This theory can serve as a unifying approach, as many well known identifying methods, such as the direct parameter model, polyreference, Ibrahim time domain method and Eigensystem realisation algorithm can be derived as special forms. These are presented in the last part of the paper, and the relationship among all these methods are shown.