论文部分内容阅读
首先根据桥梁结构的动力特性分析,构造了用于结构损伤识别的损伤标示量,并从理论上分析了该参数用于结构损伤识别的可行性.然后,从径向基函数(RBF)神经网络结构、网络设计和网络训练算法等方面论述了RBF神经网络理论,着重说明RBF网络的调用及径向基函数中心和宽度的确定步骤.最后,以一座装配式预应力钢筋混凝土系杆拱桥为工程实例,通过改变构件的弹性模量降低单元刚度来模拟结构损伤程度,并以任意三组向量对网络进行测试,说明了基于频率参数和RBF网络方法的结构损伤识别的可行性和准确性.
Firstly, according to the analysis of the dynamic characteristics of the bridge structure, the amount of damage marking for the structural damage identification is constructed and the feasibility of the parameter identification for the structural damage is analyzed theoretically. Then, the radial basis function (RBF) neural network Structure, network design and network training algorithm, the RBF neural network theory is elaborated, and the procedure of RBF network call and radial basis function center and width determination are described.Finally, with a prefabricated prestressed concrete tied arch bridge as a project An example is given to simulate the degree of structural damage by reducing the element stiffness by changing the elastic modulus of the component. The network is tested with any three sets of vectors, and the feasibility and accuracy of structural damage identification based on frequency parameters and RBF network method are illustrated.