Research On Damage Identification Method Of Bridge Structure Based On Bayesian Dynamic Linear Model

Bridge health monitoring system collects a large number of data in the process of bridge operation.Reasonable processing and analysis of these data can effectively grasp the operation status of bridge structure.The performance degradation of bridge structure and the influence of various uncertain factors will be reflected in the structural health monitoring data.How to consider this influence and then effectively judge the damage of the structure is one of the key problems to be solved in the field of bridge structural health monitoring.In order to improve people’s management and maintenance level of bridge,based on Bayesian prediction theory,this paper systematically studies the damage identification method of bridge structure system by using simulation test data and actual monitoring data of a bridge.The main work is as follows several aspects:(1)The basic theory of Bayesian dynamic linear model is introduced,and the specific modeling process of the model is deduced in detail,including the recursion process of the algorithm and the determination method of the main parameters of the model,which provides the modeling basis for the research of damage identification methods.(2)The bridge damage assessment index describing the performance degradation of bridge structure is established.Based on the Bayesian dynamic linear model,combined with different damage simulation conditions,the difference distribution function of the structure in different states is fitted.Then,the maximum cumulative Bayesian factor is established to judge whether there is damage in the structure,and the start and end time of damage is estimated;furthermore,the damage degree of the structure can be judged by the abnormal data fitting process.(3)The effectiveness of damage evaluation index is verified by simulation test.By introducing a model of prestressed simply supported beam,the performance degradation of the structure over time is simulated by using the prestressing reduction formula.At the same time,the random load and the damage of the structure are introduced to obtain the deformation value under different conditions.The Bayes dynamic linear model is established to verify the effectiveness of the damage evaluation index.(4)The above damage assessment indexes are applied to the condition assessment of the actual bridge to verify the applicability of the damage assessment indexes in the actual project.The Bayesian dynamic linear model is established by using the monitoring data.At the same time,the different damage data of the bridge are simulated,and the calculation results of the damage evaluation index are obtained to determine the actual operation state of the bridge.The applicability of the damage evaluation index proposed in this paper is verified.The results of this paper improve the damage identification method of bridge structure using Bayesian theory,and provide technical support for the operation and maintenance of bridge structure.