拉伤
结构工程
计算机科学
材料科学
工程类
生物
解剖
作者
Ru-shou Liu,Saeed Babanajad,Todd Taylor,Farhad Ansari
标识
DOI:10.1088/0964-1726/24/11/115038
摘要
A defect detection method of civil structures is studied. In order to complete the task, the proposed detection method is based on the analysis of distributed dynamic strains using Brillouin scattering based fiber optic sensors along large span structures. The current challenges in the detection of localized damage fundamentally include monitoring the dynamic strain as well as eliminating the system noise and the distortion of the changing distributed strain. Due to the capability of Brillouin scattering based methods in distributed monitoring of large structures, Brillouin optical time-domain analysis approach is implemented for assessing damage. In order to highlight the singularity at the damage location, Fourier as well as dual tree complex wavelet transform approaches were conducted. During the processing, the dynamic distributed strain in the time domain was transformed into the frequency domain for extraction of natural and forced frequencies. Then, the data was decomposed, filtered for extraction of crack features and reconstructed. The feasibility of the proposed method is evaluated through an experimental program involving the use of pulse-pre-pump Brillouin optical time domain analysis for the distributed measurement of dynamic strain with 13 Hz sampling speed and detection of simulated cracks in a 15 m long steel beam. The beam mimics a bridge girder with two artificial cracks along its length subjected to free and forced vibrations. The results indicate that the method based on the discontinuities in the strain distribution is applicable in the detection of very small damage as small as 40 micro strains. A crack gauge independently monitored the crack opening displacements during the experiments, and the limit of detected crack openings based on the first appearance of strain singularities was 30 μm.
科研通智能强力驱动
Strongly Powered by AbleSci AI