Structural Damage Identification Method Of Considering Uncertainty Parameters Based On The Momentum Indicator

Civil engineering begin to deteriorate once they are built due to natural environment and human factors. Structural damage identification based on dynamic characteristics has thus attracted significant attention for assessment of current situation of civil structures. Nevertheless, the damage detection of civil structures still remains as a challenging task. The main obstacles are the complexity and particularity of civil structures, as well as measurement noise and a significant amount of uncertainty factors.This thesis is based on the dynamic indicator considered with the impact of parameter uncertainty for the structural damage identification. The basic process is that it utilizes the fourth-order statistical moments of displacement as a dynamic damage index, considering the impact of stochastic structure parameter uncertainty through probability density evolution theory and probability theory knowledge for the structure damage identification. Therefore, the location and extent of structural damage under the influences of parameter uncertainty can be determined.The main research content in this thesis is summarized as follows:(1) Firstly, it is based on preliminary research achievements with the application of statistical moment theory. The fourth-order statistical moments of displacement theory is applied to a single degree of freedom and multi-degree of freedom system. And it is indicated that the fourth-order statistical moments of displacement is an ideal dynamic damage index for framework through basic theoretical research and numerical simulation analysis. Meanwhile, considering the influences of environmental noise, damage identification under different noise levels are analyzed.(2) On the basis of the fourth-order of displacement statistical moments theory, it analysis and extends the generalized fourth-order of displacement statistical moments theory. Considering numerical simulation with any type of random excitation, incomplete test and measurement noise, the location and extent of structural damage for the plane frame model including frame beam and column members can be detected, which develop the theory.(3) Considering the parameters uncertainty for the structure, structural's optimal stiffness can be obtained by using the fourth-order of displacement statistical moments theory, integrating with the probability density evolution theory and probability theory for uncertainty of the damping ratio, the damage identification can be completed. A ten-story frame structure model is used to verify the correctness as mentioned above.(4) According to the analysis of the 12-story standard reinforced concrete shaking table test of Tongji University, it confirms that the entire damage detection system as all mentioned above has excellent feasibility and practicality. Moreover, in this thesis, it is the first time to use the above method for three-dimensional concrete frame structure, and the identified results are successful.