基于BP神经网络的桁架结构损伤识别研究

Research on damage identification of truss structures based on BP neural network

为分析验证BP神经网络在结构损伤识别定位上应用的有效性,设计了一个桁架结构模型的静力加载试验,通过在桁架结构中更换不同直径的杆件来模拟杆件损伤,并记录其所引起的静力响应;采用ABAQUS分析软件建立了对应的桁架有限元模型,并分析了该桁架在不同荷载工况下的受力性能与损伤情况;在此基础上建立了用于预测桁架损伤状态的BP神经网络,该网络以有限元分析中桁架结构的荷载和节点位移作为输入参数,桁架杆件的损伤状态作为输出参数,建立了桁架结构杆件损伤状态与荷载、节点位移之间的关系;采用桁架结构模型静力加载试验所得数据对BP神经网络算法的正确性进行验证。结果表明:BP神经网络用于桁架结构杆件的损伤识别定位能够达到较高的精度,可以实现对结构进行实时准确的安全状态评价。

In order to analyze and verify the effectiveness of BP neural network application in structural damage identification and location,a static loading test of a truss structure model was designed.The damage of the member was simulated by replacing the member with different diameters in the truss structure,and the static response caused by it was recorded.ABAQUS analysis software was used to establish the corresponding finite element model of the truss,and the mechanical performance and damage of the truss under different load cases were analyzed.On the basis,BP neural network was established to predict the truss damage state.The network took the load and node displacement of the truss structure in the finite element analysis as the input parameters,and the damage state of the truss member as the output parameters.The relationship between the damage state of the truss member and the load and node displacement was established.The correctness of BP neural network algorithm was verified by the data obtained from the static load test of the truss structure model.The results show that the BP neural network can achieve high accuracy in damage identification and location of truss members,and can realize real-time safety state evaluation of structures.