10 April 2007 Real-time condition assessment of the Bill Emerson cable-stayed bridge using artificial neural networks
Author Affiliations +
Abstract
The Bill Emerson Cable-stayed Bridge is a newly built 1206 meter long structure crossing the Mississippi River. Due to its criticality and proximity to the New Madrid Seismic Zone, a seismic monitoring system consisting of 84 accelerometers was established for the bridge and its adjacent area. This paper is focused on a three-step artificial neural network strategy that was developed to identify the stiff of the bridge structure using the field measured dynamic response time histories without performing any eigenvalue analysis. The first step is to develop and train an emulator neural network for accurate prediction of the responses of the Bill Emerson Cable-stayed Bridge model, which represents the healthy state of the structure. A finite element model of the cable-stayed bridge was established, which represents the as-built bridge and was calibrated with the measured earthquake data from the seismic monitoring system. The second step is to establish and train a parameter evaluator neural network for relating the stiff reduction in the model bridge to the response prediction error by the emulator neural network. The third and last step is to identify the location and degree of stiff reduction in the Bill Emerson Cable-stayed Bridge.
© (2007) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Wenjian Wang, Genda Chen, Bryan A. Hartnagel, "Real-time condition assessment of the Bill Emerson cable-stayed bridge using artificial neural networks", Proc. SPIE 6529, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2007, 65291P (10 April 2007); doi: 10.1117/12.715244; https://doi.org/10.1117/12.715244
PROCEEDINGS
11 PAGES


SHARE
Back to Top