Damage Identification For Structure Node Crack Propagation

As the state's large infrastructure, Civil engineering is the basic guarantee of national economic activity and economic lifeline. But structure of civil engineering tends to be big and heavy, to detect the damage is not as convenient and feasible as mechanical structure, so its development has been slow. However, after nearly decades efforts of academia, the safety monitoring of civil engineering structure had also got very big development. The safety monitoring of some long-span bridges, high-rise buildings, large gymnasium had made great progress.Railway steel truss bridge is a commonly used form in big span railway bridge. Railway steel truss bridge has been long operating under dynamic load cause by passing train, so the cumulative fatigue damage of the weld in the node of steel truss bridge is inevitable during the operation and maintenance of the bridge, even sometimes will endanger the railway bridge and cause serious safety problems. So far, there is already some bridge structure has been ruined by the accumulation of fatigue damage. This paper studies the identification of weld crack propagation in the node of the railway steel truss bridge. In the process of research, this paper adopts the entity modeling method considering the node area welds of Poyang steel truss bridge.Identifying the damage area early and monitor the damage regional during operation is an important method to guarantee the safety work of civil engineering structure. In the process of modeling, give full consideration to the later the computation efficiency and reliability of post-treatment data processing, adopted to build isolation body. Using isolation body as the transition zone between the crack zone and the no crack zone is a good method. It not only ensures fine elements classification requirements in crack zone, but also effective control the total number of the element which ensures the analysis efficiency.This paper introduced straight through crack length in the welding area of nodes as the identification parameters. First established a fuzzy mode library, then by definition membership functions and use the membership degree assess the amount to identify the model in fuzzy mode library, so as to achieve the purpose of damage identification.It is particularly important to choose damage identification variation for damage identification. This article chooses the nodes nearby the crack zone, defines the sensitivity of these nodes with the change of crack length, through comparing the sensitivity of various points on these node could selected the most sensitivity point. The time-history stresses on these sensitivity points during the variation of the crack length was defined as fuzzy mode samples, through contrast fuzzy sample amount of each sample and calculate the membership, thus identify the crack parameters. Finally, this paper selected two groups of crack as identification samples on each condition, and the result is satisfactory. Explain the use of this method of crack propagation recognition is very effective.