Research On Damage Detection Of The Girder For Cable-stayed Bridge Based On The Most Sensitive Cable Force Index

As a combination system of cable and beam, a large number of cable-stayed bridges has been built all over the world because of its aesthetic appeal and its spanning capacity. But along with the increase of use age, damage is arosen inevitably in bridge structure. The problem of damage identification of large span bridge has become the hot and focus problem to ensure the safety of bridge operations. Damage identification for large span cable-stayed bridge is a very complication process and a interdisciplinary problem, which involves the structure dynamic response simulation,health monitoring, state identification,local damage analysis and so on.The main girder’s damage identification of cable-stayed bridge is to be further study in this paper. The research focuses on the dynamic analysis of inclined cable, the design of the test model, model testing, finite element updating and damage identification basen on the cable tension index of the most sensitivity cable.(1) A precise analysis of dynamic character has been done according to the dynamic equation which considers the influence of the tension variety along the inclined cable. Conclusion is made that the influence of the tension variety along the inclined cable must be taken into account for super-long stayed cables. Then the concept of relative sensitivity has been introduced to the dynamic model of inclined cable. The sensitive analysis of the design parameters for the frequency of long cable and short cable has been done. The most sensitive parameter has been determined from these design parameters which include the tension of cable, tensile stiffness, bending stiffness and linear density. Finally, combined with optimization theory, the dynamic updating method based on the most sensitive parameter for stayed cable has been proposed.(2) A new method based on pressure ring sensor is proposed to test the tension of inclined cable. The operation of this method is simple and the performance is stable and measuring results are reliable. Cable tension can be tested quickly and in real-time by using this method, not changing the stiffness of stay cables.(3) A test model of the single pylon cable-stayed bridge oriented damage identification has been designed and manufactured. The main girder of the test model is made from aluminum alloy with a box-shaped cross section. The girder is devided into several segments and each segment has a deck, web and stiffening ribs, and all members are bolted. The simulation of the bridge’s damage was achieved by changing the thickness of the girder’s deck. All types of girder damage of the test model could be easily simulated by changing the length of the damaged segments or changing the location of the damaged segments.On the basis, the bridge model is built in the laboratary and the test platform is constructed.(4) A new model updating method based on the measured tension of stay cables is put forward. In this method, cable forces are taken as the main updating object and displacements of the main girder are selected as the secondary updating object, the main and secondary updating parameters are obtained according to the sensitivity analysis and the initial finite element model of the model bridge has been updated by using ANSYS optimization module. All the updated parameters carry explicit physical meanings and the results of the finite element analysis after updating are consistent with the measured responses from the static tests. The baseline finite element model is obtained for bridge damage detection research.(5) This paper presents a new damage identification method for cable-stayed bridge’s girder based on the combination of the most sensitive cable tension index and neural network technique. Taking the updated finite element model of the model bridge as example, the most sensitive cable is obtained by sensitivity analysis.Taking cable tension indices as inputs of neural network for both training and testing, damage locations of the main girder are indicated by the outputs of the network. The results show that this method can identificate and locate the damage of the main girder effectively and the efficiency and precision are high. The problem of the blindness and uncertainty for stay cable’s choice in former methods has been solved and measurement numbers of cable tension is the least, this method has great practical value.The results will further enrich engineering application theories and experimental basis of the cable-stayed bridge’s damage location. And it is of great theoretical significance and practical engineering value. In addition, it will generate significant social and economic benefits.