Research On Forecasting Function Model For Damage Identification Of Cable-Stayed Bridge Based On The Deflection Monitoring

In recent years, with the booming development of transportation, we put a lot of effort in the construction of bridges, and the bridge plays an important role in traffic hub, following the bridge collapse accidents have continued to increase, the safety of the bridge has aroused people’s concern. In order to ensure the operation of bridge safely and healthily, we need to put forward higher demands of the bridge damage identification, the method and the technology of the bridge damage identification has become the focus in the present research.This thesis is based on the(180+312+180) m cable-stayed bridge of Wuhu Yangtze River Bridge as the research background, it mainly carried on the following works:At first, utilizing the ANSYS engineering structrue simulation software to establish the finite element model of the cable-stayed bridge and revise the model, get the bridge model which is relatively reliable and consistent with the actual bridge structure; secondly, the related calculation is according to this model, extract the deflection and the tension of the stay cable in the force of gravity, the different temperature, the load of the train, and analysis the regulation; then the model sample data are analyzed by making use of the multiple regression analysis method which is based on the principle of least square method inmathematical statistics. Utilizing the MATLAB software to establish the prediction function model between the dependent variable(the deflection、the tension of the stay cable) and the the independent variable(the temperature、the load of the train), and evaluate this model; at last, according to the statistical comparison diagnostic method, utilizing the prediction function model to identify whether the bridge is damaged or not, and two diagnostic standards are determined according to the bridge examination specifications.After the research and analysis we know that, the prediction function model which is established with the multiple regression analysis method has throught the model test, it indicates that the model has the availability, it can be used in the damage identification of bridge structures. This evaluation method may take comprehensive multi-factor influence on the structure repose variables into consideration, it can effectively identify whether the structure is damaged or not, the recognition effect and applicability is good. Identifing the damage of Wuhu Yangtze River Bridge in different load cases, the result shows that this method can identify whether the cable is damaged or not. If it identifies that the damage has occurred, it can also identifies the damage location preliminary, but it can not accurately identify the damage degree yet,which needs further research.