Modal identification using the frequency-scale domain decomposition technique of ambient vibration responses

被引：14

作者：

Thien-Phu Le ^{[1
,2
,3
]}

Argoul, Pierre ^{[3
,4
]}

机构：

[1] Univ Clermont Auvergne, Inst Pascal, Sigma Clermont, BP 10448, F-63000 Clermont Ferrand, France

[2] CNRS, UMR 6602, IP, F-63178 Aubiere, France

[3] Univ Paris Est, Lab Navier ENPC IFSTTAR CNRS, Ecole Ponts ParisTech, 6 & 8 Ave Blaise Pascal, F-77455 Champs Sur Marne, Marne La Vallee, France

[4] Univ Paris Est, IFSTTAR MAST SDOA, 14-20 Blvd Newton, F-77447 Champs Sur Marne, Marne La Vallee, France

来源：

JOURNAL OF SOUND AND VIBRATION | 2016年 / 384卷

关键词：

CONTINUOUS WAVELET TRANSFORM; NATURAL FREQUENCIES;

D O I：

10.1016/j.jsv.2016.08.019

中图分类号：

O42 [声学];

学科分类号：

070206 ; 082403 ;

摘要：

This paper proposes a new modal identification method of ambient vibration responses. The application of the singular value decomposition to continuous wavelet transform of power spectral density matrix gives singular values and singular vectors in frequency-scale domain. Analytical development shows a direct relation between local maxima in frequency-scale representation of singular values and modal parameters. This relation is then carried on for the identification of modal parameters via a complete practical procedure. The main novelties of this work involve the new formulation in frequency-scale domain and the capacity for the identification of modal parameters without the step of ridges extraction in comparison with previous wavelet-based modal identification methods. (C) 2016 Elsevier Ltd. All rights reserved.

引用

页码：325 / 338

页数：14

共 26 条

[1]

[Anonymous], 1988, ADV ENG MATH

[2]

[Anonymous], 1999, WAVELET TOUR SIGNAL

[3] Linear dynamical identification: An integral transform seen as a complex wavelet transform [J].