Modal Identification of Flexible Structures using Improved Multivariate Empirical Mode Decomposition
Author(s): |
Ayan Sadhu
Arndt Goldack Sriram Narasimhan |
---|---|
Medium: | conference paper |
Language(s): | English |
Conference: | Footbridge 2014 - Past, Present & Future, London, 16-18 July 2014 |
Published in: | Footbridge 2014 - Past, Present & Future |
Year: | 2014 |
Abstract: | In this paper, an improved multivariate empirical mode decomposition (MEMD) method is proposed to perform ambient modal identification of highly fl exible structures like footbridges. Unlike univariate empirical mode decomposition (EMD), multivariate EMD (MEMD) is pursued to estimate joint info rmation of intrinsic modal functions (IMFs) utilizing multichannel vibration measurements of structures. Resulting mode mixing in IMFs are then circumvented using ensemble EMD ( EEMD). The proposed hybrid MEMD-EEMD method is validated using a suite of numerical studies containing low energy modes and closely spaced frequencies that typically appear in flexible pedestrian bridges. |
Keywords: |
modal identification empirical mode decomposition (EMD) EEMD MEMD mode mixing
|