Research on damage identification of truss structures based on BP neural network
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.
BP neural networktruss structuredamage identificationstatic loading test
万方数据
维普
