Structural Damage Identification Of Bridge Based On The Virtual Distortion Method

Structural Health Monitoring is an important means to assure the safety for serve of bridge. It monitors the structural security situation in real-time by gathering environmental information and structural response. Structural damage identification is one of the core problems in structural health monitoring. This paper introduces the Virtual Distortion Method (VDM) which is improved to improve the computing efficiency of structural damage identification of bridge, and the damage location and damage degree are implemented accurately, and has carried on the preliminary analysis of bridge monitoring data. Main work contents and conclusions are as follows:(1) The structural damage identification method of bridge based on the static Virtual Distortion Method is researched. Firstly, based on the basic principle of VDM, the relationship between the damage factor and the virtual distortion was deduced; Secondly, Sequential Quadratic Programming (SQP) algorithm was used to optimize the objective function which is established by damage strain and actual damage strain, and the fast computation of structural damage identification is realized in numerical model of plane truss bridge; Finally, a finite element model of actual bridge is taken for instance, and the damage identification of suspenders are studied by the numerical simulation. Via two conditions of real time monitoring based on constant load and periodical detection based on vehicle static load, it's verified that the proposed method can greatly improve the computational efficiency, and identify the damage location and degree rapidly and accurately.(2) The structural damage identification method of bridge based on the dynamic Virtual Distortion Method is researched. In different with the static distortion method, the dynamic virtual distortion method is more complex, the damage factor, influence matrix, virtual distortion and so on need to be deduced and built anew. First of all, based on the theory of dynamic virtual distortion method, the simulated damage strain function whose variable is damage factor is built; Secondly, in order to implement the damage identification, it establishes objective function which is based on simulated damage strain and measured damage strain, and the SQP algorithm and genetic algorithm are used to optimize the objective function; Finally, a plane truss case are simulated, and the simulation is divided into single damage, multiple damage, noise-free, noisy conditions. The results show that noise resistance of Sequential Quadratic Programming algorithm is poor, because it can't identify the structural damage under noise conditions. However, the genetic algorithm has a good noise resistance, and it can accurately identify the damage location and degree even under the noise environment.(3) According to the design scheme and engineering practice, monitoring system of Boguan Bridge was established. And the data of structural health monitoring of bridge has a preliminary statistically analysis. Based on the contrastive analysis between the testing modal and the theoretical modal, it draws a conclusion that the stiffness of bridge is good, but the finite element model needs to be revised; The important control load-wind load of the bridge is statistically analyzed, and the wind speed data is turned into maximum average wind speed for 10minutes which is the standard of bridge design; The effect to the structure strain caused by the changing of the temperature strain is studied; The transformational relationship of Bridge GPS coordinates was deduced, and the change of displacement in GPS measuring points is analyzed. Through the statistical analysis, it shows that the whole structure of bridge is safe and stable, but because of its complicated structure form, the long-term monitoring is still needed.