Structural Damage Identification Research Based On Information Fusion

Owing to the influence of sudden factors, such as the hurricanes, the earthquake, the mudslides, the blizzard and so on, different degrees of damage will be happened.Even if it doesn’t occur, the structure will be aging with the increase of service life, which will lead to the damage of the structure. When the damage is accumulated to a certain extent, the safety of the structure, the applicability and the durability will be greatly reduced. What’s worse, the whole structure may be appear to be failure. Therefore, it’s important to take some scientific method to find the damage location and the extent of damage, and then take some actions to ensure the security of life and structure, minimize damage and avoid disaster. So the damage detection for the existing structure is a great urgency. Under this background, this paper based on summarizing and inheriting the former research results, takes advantage of the method of acceleration inner product vector and the theory of dissipation ratio of modal strain energy, combines with the improved D-S evidence theory, and then the research is done on the structural damage identification using information fusion.This paper, combined with the basic theory of structural dynamics, constructs the acceleration inner product vector with the acceleration response of the structure under the random excitation. It also comes from the cross correlation function value when T=0. on this foundation, the proposed damage detection method based on acceleration inner product vector is carried out. And this paper make use of the arch bridge model to simulate the damage of the structure under. various of damage conditions. And then, this paper discusses the influence of the different noise level to the recognition results. And finally the feasibility and validity of this method is proved. After that, method of acceleration inner product vector is got further e xcavation and innovation. With the help of D-S evidence theory, a more precise result is reached by fusing the reference of acceleration inner product vector. At the same time, another method called the dissipation rate of modal strain energy is used for the identification of the structural damage so that the two results can be fused in the paper. And also this paper make use of the same arch bridge model to simulate the damage of the structure under various of damage conditions. And finally the feasibility and validity of this method is proved.Combined with previous research results, the conflict coefficient K between evidences is modified and optimized for the problem that when D-S evidence theory process the highly conflict evidence.That is mean to weight the distance function and the classic conflict coefficient K which improved the D-S evidence theory. Finally, the paper fuses the dissipation rate of modal strain energy data and the acceleration vector inner product vector data with the help of the imp roved D-S evidence theory, which greatly improved the accuracy of damage identification, and at the same time it verified the feasibility and the effectiveness of this method which has a great application prospect in the field of damage identification.Finally, the relevant problems to research further were presented according to the content deficiencies in this paper.