Research On Bridge Damage Detection Method Based On RBF Neural Networks

On the basis of collection and analysis of the data about structural damage detection, artificial neural networks, three global damage identification methods are summarized, including: signature analysis approaches, model updating approaches, neural networks approaches. The theories, formulations and usages are discussed systematically; the predominance and shortcomings of each method are compared and analyzed. Combined with the prospect of bridge damage detection and artificial neural networks, a bridge damage detection method based on RBF neural networks is presented in this paper.Through analyzing the research findings of structural damage identification based on artificial neural networks and comparing the capabilities of BP, RBF and probabilistic neural networks, this paper presents that bridge damage detection can be researched based on RBF neural networks and the process is given.According to the design and construction data of Dan Shan reservoir cable-stayed bridge, damage-detection-oriented finite element model of the bridge is established and the free vibration analysis is then carried out. The focus of the research is placed on three instances, including: one component of the bridge is damaged; two components and three components are damaged. Modal frequencies, mode shapes, curvature mode are used as RBF neural networks import vector respectively; sample data of each damaged state are collected; 9 RBF neural networks models are established for researching bridge damage detection. The result indicates RBF neural networks can detect not only the damage position but the damage degree.