Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
267812 | Engineering Structures | 2011 | 8 Pages |
Signal analysis techniques (supervised-type learning filter in combination with a Chebyshev filter) constrained and tested by independent accelerometer data were used to process noisy GPS measurements of oscillations of 40 m long steel footbridge excited by coordinated jumps of a group of people. This approach permitted to de-noise the geodetic displacement record and reconstruct a minimum bias waveform for the dynamic displacement of this stiff bridge (4.3 Hz modal frequency, ∼6 mm oscillation amplitude). This result indicates that properly processed high-frequency satellite geodetic data may be used to measure dynamic displacements not only of high-rise buildings, cable-stayed bridges and other flexible structures, but of stiff civil engineering structures as well and may be useful for the Structural Health Monitoring, analysis and design of a large range of engineering structures. It was also found that although currently used 10 Hz sampling rate GPS receivers may underestimate certain high-frequency peak displacements, this will not be a problem for the recently introduced 50–100 Hz receivers.