24.12.2014 Views

Earthquake Engineering Research - HKU Libraries - The University ...

Earthquake Engineering Research - HKU Libraries - The University ...

Earthquake Engineering Research - HKU Libraries - The University ...

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

530<br />

niques for monitoring the health of structures have been borrowed from the aerospace industry<br />

to be refined and applied to civil infrastructure. <strong>The</strong>se techniques facilitate the<br />

monitoring of structural health in an automated mode, reducing the uncertainty in the integrity<br />

of the structure and providing valuable information to decision makers after a major<br />

event.<br />

One important issue in SHM is obtaining appropriate structural response data. For best results,<br />

one would typically excite the structure with an actuator or shaker, and measure the<br />

responses. However, for many structures, using forced vibration to induce response for<br />

SHM or system identification purposes may be infeasible or forbidden (due to concerns of<br />

the owners or danger that the structure may already be damaged). However, ambient excitation<br />

(e.g., microtremor, wind, traffic, etc.) can often provide sufficient response for system<br />

identification to be performed. Further, ambient input excitation is often unmeasured<br />

(or actually immeasurable) and methods that require knowledge of the input must be selected.<br />

Additional challenges arise from the nature of large, flexible structures. <strong>The</strong> characteristics<br />

of these structures include: they are continuous structures that are not wellrepresented<br />

with lumped masses; lateral and torsional motions can be highly coupled; using<br />

a limited number of sensors is feasible; and, they have closely-spaced modes.<br />

<strong>The</strong> focus of this paper is to examine the capabilities of the Natural Excitation Technique<br />

(NExT) combined with the Eigensystem Realization (ERA) algorithm to extract modal<br />

parameters from a structure with closely spaced modes. A numerical model of the Bill<br />

Emerson Memorial Bridge, developed for the benchmark problem on the control of cablestayed<br />

bridges, is used to perform simulations and obtain response records. For this initial<br />

study, the bridge is excited using a stationary broadband force as a rough simulation of<br />

ambient vibrations. Three sensor configurations are considered to investigate the dependence<br />

of the methodology on the sensor configurations.<br />

DESCRIPTION OF THE BRIDGE AND FINITE ELEMENT MODEL<br />

<strong>The</strong> cable-stayed bridge used for this benchmark study is the Bill Emerson Memorial<br />

Bridge spanning the Mississippi River (on Missouri 74—Illinois 146) near Cape Girardeau,<br />

Missouri, designed by the HNTB Corporation (Hague, 1997). <strong>The</strong> bridge is currently under<br />

construction and is to be completed in 2003. Instrumentation is being installed in the<br />

Emerson bridge and surrounding soil during the construction process to evaluate structural<br />

behavior and seismic risk (Qelebi, 1998).<br />

As shown in Fig. 1, the bridge is composed of two towers, 128 cables, and 12 additional<br />

piers in the approach bridge from the Illinois side. It has a total length of 1205.8 m (3956<br />

ft). <strong>The</strong> main span is 350.6 m (1150 ft) in length, the side spans are 142.7 m (468 ft) in<br />

length, and the approach on the Illinois side is 570 m (1870 ft). A cross section of the<br />

deck. <strong>The</strong> bridge has four lanes plus two narrower bicycle lanes, for a total width of 29.3<br />

m (96 ft). <strong>The</strong> deck is composed of steel beams and prestressed concrete slabs. <strong>The</strong> 128<br />

cables are made of high-strength, low-relaxation steel (ASTM A882 grade 270). <strong>The</strong> H-<br />

shaped towers have a height of 102.4 m (336 ft) at pier 2 and 108.5 m (356 ft) at pier 3.<br />

Each tower supports a total 64 cables. <strong>The</strong> cross section of each tower varies five times<br />

over the height of the tower. <strong>The</strong> approach bridge from the Illinois side is supported by 11<br />

piers and bent 15, <strong>The</strong> deck consists of a rigid steel diaphragm with a slab of concrete at<br />

the top. Sixteen 6.67 MN (1,500 kip) shock transmission devices are employed in the connection<br />

between the tower and the deck. <strong>The</strong>se devices are installed longitudinally to allow<br />

for expansion of the deck due to temperature changes. Under dynamic loads these<br />

devices are extremely stiff and behave as rigid links. Further details are provided in the paper<br />

describing the benchmark control problem statement Dyke et al, (2002),

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!