2001
DOI: 10.1115/1.1410919
|View full text |Cite
|
Sign up to set email alerts
|

Output-Only Subspace-Based Structural Identification: From Theory to Industrial Testing Practice1

Abstract: We address the problem of structural model identification during normal operating conditions and thus with uncontrolled, unmeasured, and nonstationary excitation. We advocate the use of output-only and covariance-driven subspace-based stochastic identification methods. We explain how to handle nonsimultaneously measured data from multiple sensor setups, and how robustness with respect to nonstationary excitation can be achieved. Experimental results obtained for three real application examples are shown.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

1
73
0
4

Year Published

2003
2003
2016
2016

Publication Types

Select...
9
1

Relationship

1
9

Authors

Journals

citations
Cited by 99 publications
(78 citation statements)
references
References 19 publications
1
73
0
4
Order By: Relevance
“…The methods, based on output measurements only, assume that the input can be well represented by a vector white noise process. Recent developments are reported in De Roeck (1999, 2001) and Basseville at al. (2001) using time domain stochastic subspace identification methods, in Beck et al (1994) using time domain least-squares methods based on correlation functions of the output time histories, in Verboten et al (2002), Gauberghe (2004) and Brincker et al (2001) using frequency domain least-squares methods based on full cross-power spectral densities (CPSD), and in Peeters and Van der Auweraer (2005) based on half spectra.…”
Section: Introductionmentioning
confidence: 99%
“…The methods, based on output measurements only, assume that the input can be well represented by a vector white noise process. Recent developments are reported in De Roeck (1999, 2001) and Basseville at al. (2001) using time domain stochastic subspace identification methods, in Beck et al (1994) using time domain least-squares methods based on correlation functions of the output time histories, in Verboten et al (2002), Gauberghe (2004) and Brincker et al (2001) using frequency domain least-squares methods based on full cross-power spectral densities (CPSD), and in Peeters and Van der Auweraer (2005) based on half spectra.…”
Section: Introductionmentioning
confidence: 99%
“…Because of the non-stationarity of the unknown excitation, we advocate for the use of an output-only and covariance-driven subspace-based stochastic identification method [15]. The difference between the covariance-driven form of subspace algorithms which we use and the usual data-driven form is minor, at least for eigenstructure identification.…”
Section: Stochastic Subspace-based Structural Identificationmentioning
confidence: 99%
“…These considerations are extensively treated in [32][33][34]. An estimate of the Hankel matrix can be written aŝ…”
Section: A Classical Output-only Subspace Identification Algorithmmentioning
confidence: 99%