Affiliations: Department of Civil Engineering, The University of Tokyo, Tokyo 113-8656, Japan | Bridge Maintenance Consultant Corp., WBG Marive West 25th Floor, 2–6 Nakase, Mihama-ku, Chiba, 261-7125, Japan | Bridge Structure Team, Public Works Research Institute, 1–6, Minamihara, Tsukuba City, Ibaraki, 305-8516, Japan
Abstract: Processing and interpreting a large amount of data represents a great challenge for infrastructure health monitoring. This study demonstrates how to apply some available algorithms to the data measured by a practical structural health monitoring system and how to evaluate the results. Short-term and long-term field tests were conducted on a steel highway bridge to monitor the structural behavior, including strain, displacement and temperature. The short-term test was carried out with a special vehicle crossing the bridge at different speeds. Measured dynamic strain responses are used to extract the dynamic characteristics of the structure by Eigensystem Realization Algorithm and to analyze the vehicle – bridge interaction properties by Wavelet Transform. In the long-term monitoring, strain time history and displacement versus temperature were recorded at several points. Cumulative fatigue damage model, extreme value distribution and linear regression model are, respectively, applied to the data and several indices are presented to evaluate the structural condition. Results indicate that the structural condition is deficient and further special inspection is required at the monitored bridge. The algorithms used in this paper can be applied as general structural health monitoring tools for processing of data measured on bridges.
Keywords: Health monitoring, Dynamic response, Eigensystem Realization Algorithm, Vehicle – bridge interaction, Wavelet Transform, Fatigue damage, Extreme value distribution, Linear regression model
.jpg)
