Abstract
In Japan, many bridges have passed long time after construction and are progressing aging by corrosion and other factors. In order to maintain the soundness of bridges, efficient and reliable inspection technique for screening inspection point is required. In this study, we discussed damage identification method focused on deterioration degree of bridge members and position of damaged members by using Self-Organizing feature Map (SOM) which is a kind of Neural Networks. We considered damages by changing the thickness of stiffening girder as deteriorated member. We were using the Power spectrum as learning and recognition data to SOM. In order to calculate the Power spectrum, FEM analysis model was created based on the small arch bridge experimental model. In conclusion, it is found that there is a possibility of performing damage identified by the proposed method.
