A novel damage identification method based on short time Fourier transform and a new efficient index

Structural failure can be corrected and restored with the preliminary review of the status of the structures, and thus prevent the spread of corruption and collapse of structures. Among the non-destructive detection methods that are vibration-based, time-frequency distributions are taken into consideration about their ability in the signal processing. In this research, a new method was proposed for damage detection in concrete piers of bridges. According to the proposed method, a small impact load is applied to the piers, and the responses were recorded in acceleration at mid-height. The recorded signals are processed by the short-time Fourier transform. The calculated time-frequency matrices are examined using a new damage index. The index was used from the Levenberg-Marquardt algorithm and Chebyshev distance. The proposed index has been evaluated using the finite element model of the Vadi-e-Rahmat bridge. The results show that the proposed method has been identified the damages and diagnosed their location accurately. The proposed method, while easy to use, significantly reduces the cost of field surveys by reducing the data needed for calculations, as the baseline response is recorded at only one point. Also, the method is output-only and does not require the creation of the analytical model of the bridges.

Automated summary

The research proposes a new method for damage detection in concrete piers of bridges, which accurately identifies damages and diagnoses their location. The method is based on short-time Fourier transform and a new damage index, using small impact loads and recorded responses. This approach is not only easy to use, but also significantly reduces the cost of field surveys.