A Hilbert Transform Method for Parameter Identification of Time-Varying Structures with Observer Techniques

Abstract

This paper presents a recursive Hilbert transform method for the time-varying property identification of large-scale shear-type buildings with limited sensor deployments. An observer technique is introduced to estimate the building responses from limited available measurements. For an n-story shear-type building with l measurements (l≤n), the responses of other stories without measurements can be estimated based on the first r mode shapes (r≤l) as-built conditions and l measurements. Both the measured responses and evaluated responses and their Hilbert transforms are then used to track any variation of structural parameters of a multi-story building over time. Given floor masses, both the stiffness and damping coefficients of the building are identified one-by-one from the top to the bottom story. When variations of parameters are detected, a new developed branch-and-bound technique can be used to update the first r mode shapes with the identified parameters. A 60-story shear building with abruptly varying stiffness at different floors is simulated as an example. The numerical results indicate that the proposed method can detect variations of the parameters of large-scale shear-type buildings with limited sensor deployments at appropriate locations.

Department(s)

Civil, Architectural and Environmental Engineering

Keywords and Phrases

Branch And Bounds; Damping Coefficients; Hilbert Transform; Identified Parameter; Mode Shapes; Multistory Building; Numerical Results; Property Identification; Sensor Deployment; Shear Buildings; Structural Parameter; Time-Varying Structure; Varying Stiffness; Floors; Hilbert Spaces; Numerical Methods; Buildings

International Standard Serial Number (ISSN)

0964-1726

Document Type

Article - Journal

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2012 IOP Publishing Ltd., All rights reserved.

Publication Date

01 Oct 2012

Link to Full Text

Share

 
COinS