The Three-Dimensional Grid Adaptive Generation And Its Application In The Bridge Crack Injury Detection

With the increasing of unexpected accidents of bridge, People increasingly recognize that it is important to improve the reliability and security of bridge structure. But monitoring online and fault diagnosis for bridge are helpful to detect the state of the structure and to find and locate faults position and injury extent. So they play an important role in bridge maintenance timely and prolonging service life and preventing calamitous accident. We can execute bridge crack damage detection by grid division and modal analysis. The quality of the mesh will determine the results of finite element analysis; as the important tool of the bridge damage identification, modal analysis requires many damage index data which directly or indirectly come from the finite element analysis results, thus the high-quality of mesh division is the basis for the successful modal analysis, while the modal analysis services for the bridge health monitoring.Currently in multitudinous mesh generation algorithm, the most widely used is the Delaunay algorithm. Because it has perfect mathematics rationale and its judgment criterion, good flexibility, easy to control and its application can be extended from two-dimensional to three-dimensional. So compared with other algorithms, it has incomparable advantages.Modal analysis as an important means for the bridge damage identification has become a hot research at home and abroad. Modal is the inherent vibration characteristics, which has a specific natural frequency, damping ratio and modal shape. When the bridge suffers injury, the material stiffness will drop that causes the bridge modal parameters change. Therefore, we can accomplish damage detection by the modal analysis method.Some aspects have been researched in the paper as follows:1. Three-dimensional Delaunay grid optimization algorithm for bridge application is studied in this paper. This section describes the principle, limitation of Delaunay algorithm firstly, and then puts forward a simple force-based mesh quality smoothing algorithm for the bridge structure model in order to make the data better reflect the true state of the bridge model. This method finds point position using force equilibrium, distance function and changes the topology by the Delaunay triangulation. It is not only simple, but also improving the quality of the grid that ensures the convergence of the results and solution accuracy. So it establishes the foundation for modal analysis of bridge.2. In this paper a new approach based curvature mode ratio (CMR) and natural frequency with BP neutral network for crack injury detection is discussed. Taking a simple cantilever beam with multiple damage and perfect structure as study object, injury indices are gained by way of analysis the merit and drawback of natural frequency and the curvature mode, then use BP neutral network training tests that regards the natural frequency, the third-order natural frequency relative decline and the first-order maximum CMR as the input parameter, crack location and injury degree as the output parameter. The simulated results show that the proposed method can be very well applied to the crack damage identification.