Component mode synthesis as a framework for uncertainty analysis
Component mode synthesis as a framework for uncertainty analysis
Component mode synthesis (CMS) is a well-established technique for the vibration analysis of built-up structures. It was originally developed as a method for reducing the size of a finite element model, hence reducing computational cost.
CMS also offers an appealing framework for the analysis of the dynamics of uncertain structures. The benefits concern the numerical costs, the way uncertainty is included, quantified and propagated. This paper reviews and discusses these issues. The fixed-interface (Craig–Bampton) method is described, while the number of interface degrees of freedom (dofs) can be further reduced using characteristic constraint modes.
Quantification and propagation of uncertainty is discussed. Uncertainties in properties can be naturally and straightforwardly introduced at the component level, either in terms of the component physical properties or the component modal properties, while the individual components are typically statistically independent, being made by different manufacturing processes.
CMS methods are also amenable to the inclusion of experimentally measured variability data, quantifying it in terms of component modal properties. An example is given. The application of perturbational techniques is considered. The CMS framework is particularly amenable to propagation of uncertainty through one or more of the analysis paths at component or at global level using perturbations.
Finally, qualitatively different uncertainty descriptions can be combined, with some components being described probabilistically, some possibilistically, the descriptions then being unified at the global level. Numerical examples are presented. Overall, CMS methods offer a strong physical insight into the analysis of structures with non-deterministic properties.
161-178
Hinke, Lars
486fa594-35f9-4098-b63e-6f8d4a78899f
Dohnal, F.
2593960f-9788-45d4-9699-06cb1f5f6a0f
Mace, Brian R.
cfb883c3-2211-4f3a-b7f3-d5beb9baaefe
Waters, Timothy P.
348d22f5-dba1-4384-87ac-04fe5d603c2f
Ferguson, Neil
8cb67e30-48e2-491c-9390-d444fa786ac8
10 July 2009
Hinke, Lars
486fa594-35f9-4098-b63e-6f8d4a78899f
Dohnal, F.
2593960f-9788-45d4-9699-06cb1f5f6a0f
Mace, Brian R.
cfb883c3-2211-4f3a-b7f3-d5beb9baaefe
Waters, Timothy P.
348d22f5-dba1-4384-87ac-04fe5d603c2f
Ferguson, Neil
8cb67e30-48e2-491c-9390-d444fa786ac8
Hinke, Lars, Dohnal, F., Mace, Brian R., Waters, Timothy P. and Ferguson, Neil
(2009)
Component mode synthesis as a framework for uncertainty analysis.
Journal of Sound and Vibration, 324 (1-2), .
(doi:10.1016/j.jsv.2009.01.056).
Abstract
Component mode synthesis (CMS) is a well-established technique for the vibration analysis of built-up structures. It was originally developed as a method for reducing the size of a finite element model, hence reducing computational cost.
CMS also offers an appealing framework for the analysis of the dynamics of uncertain structures. The benefits concern the numerical costs, the way uncertainty is included, quantified and propagated. This paper reviews and discusses these issues. The fixed-interface (Craig–Bampton) method is described, while the number of interface degrees of freedom (dofs) can be further reduced using characteristic constraint modes.
Quantification and propagation of uncertainty is discussed. Uncertainties in properties can be naturally and straightforwardly introduced at the component level, either in terms of the component physical properties or the component modal properties, while the individual components are typically statistically independent, being made by different manufacturing processes.
CMS methods are also amenable to the inclusion of experimentally measured variability data, quantifying it in terms of component modal properties. An example is given. The application of perturbational techniques is considered. The CMS framework is particularly amenable to propagation of uncertainty through one or more of the analysis paths at component or at global level using perturbations.
Finally, qualitatively different uncertainty descriptions can be combined, with some components being described probabilistically, some possibilistically, the descriptions then being unified at the global level. Numerical examples are presented. Overall, CMS methods offer a strong physical insight into the analysis of structures with non-deterministic properties.
This record has no associated files available for download.
More information
Published date: 10 July 2009
Venue - Dates:
Leuven Symposium on Applied Mechanics in Engineering (LSAME.08), Leuven, Belgium, 2008-03-31 - 2008-04-02
Organisations:
Dynamics Group
Identifiers
Local EPrints ID: 145233
URI: http://eprints.soton.ac.uk/id/eprint/145233
ISSN: 0022-460X
PURE UUID: bb2f197e-8742-4485-a29f-458a1669cbd2
Catalogue record
Date deposited: 16 Apr 2010 13:57
Last modified: 14 Mar 2024 02:32
Export record
Altmetrics
Contributors
Author:
Lars Hinke
Author:
F. Dohnal
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics