Study Of Damage Detection For Bridges Based On The Static Measured Data

With the rapid development of modern science and technology, more and more high-rise buildings and long-span bridges have been built and developed. At the same time, many structures become aging with the growth in service time. In order to protect the people's life and property, it's necessary to detect the possible damage area and damage extent efficiently and hold the healthy state of the structure. Recently, the research of the structural health evaluation and monitor is one of the foreland problems of the civil engineering area. The health evaluation theory based on the static measured data has important theoretical and applicant values because the static measured data can be gotten easily with a relatively high degree of accuracy. Supported jointly by the National Natural Science Foundation of China(No.50378041)and the Specialized Research Fund for the Doctoral Program of Higher Education (No. 20030487016) the thesis studies the static measured data based damage detection methods and the application in complicated structures. On the base of systematic summary of the recent research in the field, the damage localization methods based on the incomplete measured data are discussed deeply and a new damage localization method based on static measured data is put forward. The detection of the damage magnitude is solved as the optimization problem of the parameters identification, so the damage magnitude can be detected rapidly and accurately by the genetic algorithms although the measured information is very little. On the basis of the traditional genetic algorithms, the genetic algorithms and the fuzzy theory are combined, and the traditional genetic algorithms are improved, furthermore, the detection of structural damage magnitude based on the improved genetic algorithm is developed. The main research contents in the thesis are organized as follows: Firstly, the static displacement curvature assurance coefficient (SDCAC) based on the grey relation theory is put forward for the damage localization. The damages of the beam-like and the plane-like structures are correctly located by using the SDCAC. After the simulation research on the damage localization of the beam-like and plane-like structures is studied, the conclusion can be drawn that the SDCAC based on the static measured data is sensitive to local damages. Through analyzing the change of the SDCAC along one direction, the damage area along this direction can be detected clearly, so the static displacement curvature assurance coefficient can be used as the criteria of the damage localization. The structural curvature is in inverse proportion to the structural stiffness, so the structural curvature can be considered as the basis of the structural stiffness identification. The structural damage can be represented by the stiffness change. That is, the SDCAC can be as the basis of the damage localization. Secondly, the improved genetic algorithm by combining the traditional genetic algorithm and the fuzzy optimization theory is developed and is used to detect the damage magnitude. Through applying the parallel selection technique in the algorithm, the effective way to identify the damage magnitude is formed so that the computation speed and stability of the algorithm are improved. Thirdly, the static displacement curvature assurance coefficient and the improved genetic algorithm are applied to the damage detection of several numerical examples in order to verify the effectiveness of the proposed methods above. The single and multiple damages of one cantilever beam and one bridge with two fixed ends are detected respectively by using the proposed methods. When results of the damage localization of the beam by using the static displacement curvature assurance coefficient is compared to the result using the other damage localization methods, it can be found that the damage localization method proposed in this thesis is simple and the localization precision is high. After the damages are located, the improved genetic algorithm proposed in this thesis is used to identify the damage magnitude of the beam-like and plane-like structures. The numerical results show that the improved genetic algorithm is suitable for the damage detection of large-scale structures. Finally, the proposed method in the thesis is used for numerical and experimental studies of the damage identification of Xinyang Bridge, Henan Province. The numerical result shows that the proposed damage localization method and the improved genetic algorithm for damage content identification are suitable for the large-scale structures. Moreover, the damage of Xinyang Bridge is located with a relatively high degree of accuracy by the measured data in site, and the accurate damage detection offers a basis for the structural health evaluation.