New damage indices and algorithm based on square time-frequency distribution for damage detection in concrete piers of railroad bridges

What is it about?

This paper presents a new algorithm based on square time–frequency distribution to detect damages in concrete piers of railroad bridges. In addition, two new damage indices are proposed in this research. The algorithm and damage indices are described and corroborated using finite element model of Qotour Railroad Bridge. To create the finite element model according to the current condition of the Qotour Railroad Bridge, extensive in situ measurement results and laboratory tests were used. Some seismic damages were considered in the concrete piers of the model. The results indicate that the algorithm and damage indices can identify all damages from severe to small accurately. The proposed algorithm and damage indices are shown to be not sensitive to noise and even in case of noise-polluted response signal are working properly. The proposed algorithm only uses the response signals without the need for recording the excitation forces. In addition, according to the algorithm, there is no need to create an analytical model of the bridge for damage detection. This feature in modeling of bridge structures and especially complex bridge, which is associated with problems, is very useful.

Featured Image

Why is it important?

In this paper, a new algorithm based on square TFD is presented. In addition, two new damage indices are proposed. According to the algorithm, it is needed to vibrate the bridge before and after damage by exciting force and record responses at the piers. The suggested algorithm using the differences between energy distributions of the recorded signals on the time–frequency plane, identify damage and its location. In order to ensure that the differences are only due to damage in the piers, the excitation forces used to vibrate the intact and damaged bridge must be the same. However, there is no need to record the excitation force and the algorithm only uses the response signals to detect damage. To validate the algorithm and proposed damage indices, the analytical model of Qotour Railroad Bridge is used. To determine the damaged model, some earthquake records are applied to the model, and seismic damages in the bridge piers are evaluated. In addition, two different excitation forces are applied to the undamaged and damaged model, and the signals of pier responses are registered. Using the algorithm, the damages have been identified. Unlike many other methods, the proposed algorithm does not require any analytical model of the bridge. In this study, because there was no possibility of creating damage in the real bridge, the analytical model of the real bridge has been used. In addition, with a low number of sensors, the algorithm can detect damages.