Research on Multi-stage Scheme for Identifying Damage in Steel Truss Arch-bridge Based on Neural Network
|Keywords||Steel truss arch bridge Damage Identification GRNN neural network Substructure|
Steel truss arch bridge is an important structure in the form of bridge structures , steel truss arch bridge majestic appearance , spanning capacity , higher carrying capacity somewhat in recent years has been the rapid development , such as of Tianjin Cathay bridge and Chongqing Chaotianmen Yangtze River Bridge and so on. Chaotianmen Yangtze River Bridge opened to traffic in 2009 , the main bridge arrangement 190 552 190m, is the world's largest span steel truss arch bridge . Many bridge structure in the course of different degrees of damage may occur due to various reasons , cracking and aging, which lay hidden dangers , causing a major accident , endangering the lives and safety of the people and lead to loss of property . Therefore, the bridge structure damage identification and early warning , to grasp the health of important significance under the bridge operation . Identify damage to the bridge structure (especially steel truss arch bridge structure) has become an important research topic in the field of civil engineering . The study pointed out that the damage to the bridge structure will cause a corresponding change in the structure of the overall power characteristics , therefore , if we can establish a mapping relationship between the dynamic characteristics of the structure changes and structural damage can realize the damage identification . The neural network has the advantages of fault tolerance , nonlinear mapping ability , so it is very suitable for structural damage identification . On the basis of previous studies , as the research object , the application of neural network technology to large -span steel truss arch bridge proposed large-span steel truss arch bridge structure damage step-by-step identification method using APDL language in ANSYS and MATLAB language programming make mutual convergence , efficient neural network method is suitable for the steel truss arch bridge structure damage identification . First warning of the damage to the bridge , and to determine the appearance of damage ; then initial positioning of the components of the bridge damage , discriminant the damage bar where sub structure ; then further damage identification on the basis of preliminary positioning , which determine the damaged members specific location ; Finally, to identify the degree of injury to injury bar .