Identification of stiffness degradation of steel highway bridges using wavelet analysis.

摘要

Fatigue, which is perhaps the most important failure mode of steel highway bridges, results from cyclic loading conditions. In recent years, structural health monitoring (SHM) technology is developed to assess the performance and to measure the response of bridges. Most approaches, however, can only provide data based on which change of global parameters may be obtained. Very few techniques are available to identify structural damages, particularly those with gradual degradation over a long period of time, such as fatigue. The Wavelet Transform (WT) is a technique that has been applied successfully to detect sudden structural stiffness loss. However, it has not been shown that it can be used to assess long-term stiffness degradation. This dissertation presents a new approach by modifying the WT technique to identify stiffness degradation from response data over a long period of time. Numerical simulations for both single degree of freedom (SDOF) systems and multi-degree of freedom (MDOF) systems are presented. This new approach is validated through field experimental data obtained from 2004 to 2006 on an I-990 steel girder highway bridge in Western New York, under ambient traffic conditions.