Structural Parameter Identification Of Concrete Bridges And Research On Safety Assessment Method

In our daily life and travel,bridges are an indispensable part of highway traffic.To ensure the safety of bridges in their service life,it is necessary to identify the parameters of bridge structures by using the long-term data collected by the bridge health monitoring system,and to evaluate the safety of bridges through the identified structural parameters,so as to find the problems of bridges in time and prevent the occurrence of major safety accidents.The main contents of this thesis are as follows:1.The algorithm of stochastic subspace principle is improved,and a more efficient optimization algorithm is deduced,avoiding solving projection matrix,and using observable matrix to construct sub-block matrix to complete modal identification.The efficiency and accuracy of the algorithm are verified by an example,and the modal parameters of Haierwa Bridge are successfully identified,which provides basic data for on-line model updating and damage identification.2.Modal parameters reflect the characteristics of the structure itself.Modifying the original finite element model according to the identification results of modal parameters so that its frequency matrix is within the limit difference can make the modified finite element model simulate the actual state of the bridge.Through the modified finite element model,the structural parameters of the current state of the bridge can be obtained,so as to know whether the current state of the bridge is normal or not.3.The corrected finite element model is used to load the measured six-axle vehicle,and the dynamic stress data waveform of the key section is extracted.The accuracy of the finite element model is verified by comparing the dynamic stress waveform measured by the dynamic weighing system and calculating the correlation coefficient.The relationship between maximum dynamic stress,vehicle speed and vehicle load is summarized to evaluate the health status of bridges.4.A long-term monitoring and evaluation model for concrete bridges is established,which is mainly used to isolate the significant influencing factors of monitoring effects.By reducing the data acquisition interval of sensors,the vehicle loads with strong randomness can be separated.Then,according to the respective characteristics of temperature and creep,the appropriate expression is selected to establish a multivariate regression mathematical model.The temperature component and time component are separated from the monitoring effect component,and then the qualitative analysis and safety evaluation of the bridge can be carried out.5.Short-term prediction model of concrete bridge is established.The application of grey GM(1,1)model in bridge health monitoring is studied,and its applicable conditions for predicting static strain are obtained.