Study On The Health Diagnosis Of The Long-span Railway Steel-concrete Composite Trussed Beam Bridge

Under the action of environmental collisions and the loads, the aging phenomenon or thedamage of the bridge is unavoidable. If the damage to the bridge structure is continueddevelopment, it will make the bearing capacity of the bridge reduced. In serious cases it willmake the bridge structure destroyed. In recent years, many bridge collapse accidentoccurred in China, which will cause damages to people’s live and properties. Therefore, thebridge damage Identification is important.Taking80m steel-concrete composite truss bridge as engineering background,in this paper, we present this kind of bridge damage Characteristics based on the study ofthe structural characteristics of this bridge and the reason of which will cause damages to thebridge. Extracting damage index data of the bridge by establishing the model of the finiteelement damage of the bridge. RBF neural network structure is established by using thesoftware MATLAB and trained by the damage index data of training sample. The RBF neuralnetwork can be used for identifying the bridge’s damage locations and extentsThe main content consists of the following parts:1. Through researching national and international methods to identify bridge damagelocations and extents, their characteristics are summarized and analyzed.2. On the basis of neural networks theory, accompanying the features of steel-concretecomposite truss bridge, Design a RBF neural network which is used for identifying thebridge’s damage locations and extents and determine the steps for the bridge damagediagnosis. RBF neural network structure is established by using the software MATLAB(2013b version).3. In this chapter, we do the research of the structural characteristics of steel-concretecomposite truss bridge and the parts of the bridge which vulnerable damaged. analyzed thereason of which will cause damages to the bridge. Concluding correlate regularities thoughmake an analysis on the changes of force performance of the damaged bridge.4. According to the characteristic of steel-concrete composite truss bridge, find out theeffectively damage index. RBF neural network structure is established and trained by thedamage index data of training sample. The trained RBF neural network is used for identifyingthe bridge’s damage locations and extents. According to the diagnosis result validate theeffectiveness of this method.5. Taking the natural frequencies of the bridge structure as the damage index, The RBFNeural network method can be used for identify bridge Structure anomaly Phenomenon.Taking mode shapes as the damage index, The RBF Neural network method can be used for identify damage locations and extents of steel belly bar damage situations or node damagesituations or steel belly bar and node damage situations. The RBF Neural network method isan effective diagnosis method of bridge structure damage.