On the treatment of uncertainty in experimentally measured frequency response functions

Meggitt, JWR ORCID: https://orcid.org/0000-0002-6665-2939 2018, 'On the treatment of uncertainty in experimentally measured frequency response functions' , Metrologia, 55 (6) , p. 806.

[img]
Preview
PDF - Accepted Version
Download (3MB) | Preview

Abstract

Frequency response functions (FRFs) are routinely used within experimental structural dynamics, among other fields. The uncertainty associated with their measurement is typically modelled on the simplified assumption that additive uncorrelated noise is present. This treatment of FRF uncertainty is questioned here by demonstrating that the notion of operator uncertainty (i.e. the uncertainty due to inconsistent force excitation) introduces a correlation between measured FRFs which is not accounted for under the assumption of additive noise. The origin of this correlation is illustrated and its experimental presence verified. The importance of this correlation on the propagation of uncertainty is subsequently investigated as part of a numerical and experimental case study, where its influence on the inversion of an uncertain FRF matrix is considered. It is shown that the neglect of such correlations can lead to large errors in uncertainty estimates, for example in dynamic sub-structuring and the solution of inverse problems.
Keywords: Frequency response function, structural dynamics, experimental, measurement, operator uncertainty

Item Type: Article
Schools: Schools > School of Computing, Science and Engineering
Journal or Publication Title: Metrologia
Publisher: IOP Science
ISSN: 0026-1394
Related URLs:
Funders: Engineering and Physical Sciences Research Council (EPSRC)
Depositing User: JWR Meggitt
Date Deposited: 08 Feb 2019 12:06
Last Modified: 19 Oct 2019 08:00
URI: http://usir.salford.ac.uk/id/eprint/49984

Actions (login required)

Edit record (repository staff only) Edit record (repository staff only)

Downloads

Downloads per month over past year