Computational & Technology Resources
an online resource for computational,
engineering & technology publications
PROCEEDINGS OF THE ELEVENTH INTERNATIONAL CONFERENCE ON COMPUTATIONAL STRUCTURES TECHNOLOGY
Edited by: B.H.V. Topping
Dealing with Uncertainty in Advanced Frequency-Domain Operational Modal Analysis
B. Peeters1, M. El-Kafafy2 and P. Guillaume2
1LMS International, Leuven, Belgium
B. Peeters, M. El-Kafafy, P. Guillaume, "Dealing with Uncertainty in Advanced Frequency-Domain Operational Modal Analysis", in B.H.V. Topping, (Editor), "Proceedings of the Eleventh International Conference on Computational Structures Technology", Civil-Comp Press, Stirlingshire, UK, Paper 105, 2012. doi:10.4203/ccp.99.105
Keywords: operational modal analysis, noise, uncertainty, confidence bounds, PolyMAX Plus.
This paper presents a new advanced frequency-domain operational modal analysis method that is able to deal with uncertainties. The concept of the method is that uncertainties in the measurements (noise) should be taken into account when estimating the modal parameters of the tested structure, and that also confidence bounds on the modal parameters are estimated. The starting point for the new developments is the PolyMAX method that became quite popular in the field of experimental and operational modal analysis. Operational modal analysis, which is the domain in which this paper is situated, refers to the fact that the structure is tested in its operational conditions and that only response data are available (e.g. bridge acceleration response data arising from unmeasurable wind or traffic excitation). The main advantages of PolyMAX are its computational efficiency and the very clear stabilization diagrams it yields even in the case of highly-damped systems and noisy measurements. Moreover, the numerical stability of the algorithm allows for a large-bandwidth and high-model order analysis and makes it suitable both for lowly- and highly-damped structures. In this paper, an iterative maximum likelihood estimation method is introduced as an extension to Polymax of which the main benefits become clear in case of very noisy data and in cases where the estimation of confidence bounds on the modal parameters is of interest. This "PolyMAX Plus" approach will be illustrated using railway bridge vibration data that was acquired during the execution of the FADLESS research project.
purchase the full-text of this paper (price £20)