ABSTRACT
Identifying the number and location of damage is an important issue in the vibration health monitoring of bridges. This study aimed to improve the accuracy of damage detection using the natural frequencies utilized in conventional monitoring and the rotational angle of the bridge in the transverse direction. An analytical model was constructed based on a reduced arch model developed by the laboratory group. Damage in the hanger members, stiffening girders, and cross girders of the analytical model was investigated with reference to the corrosion reported for real arch bridges. Two selections were obtained from this damage, and the damage was modeled. Eigenvalue and static analyses were then carried out to investigate the accuracy of damage type and damage location detection based on the natural frequency and rotation displacement changes caused by each damage pattern. The results showed that the natural frequencies of damage at the location where the mode shape is predominant change significantly, and that the natural frequencies that change for each damaged member are different. It was also found that the damage that could not be identified by natural frequencies alone could be detected using the angle of rotation. These results indicate the possibility of classifying damage locations and types when there is more than one type of damage.
