A New Two-Step Method for Damage Identification in Beam Type Structure

摘要

A two-step method based on the difference of normalized strain mode combined with BP neural network for damage identification of cantilever structure is presented in this paper. The number and location of the crack is identified from the difference between the normalized strain mode of the intact and cracked beams in the first step. Then, the structural damage degree is identified using BP neural network. To verify the effectiveness and efficiency of the two-step method in this paper, the simulation model of the cantilever beam structure is established using MATLAB software. Both single crack and double cracks recognition are simulated, the damage identification result is compared with the transfer matrix method. Compared with transfer matrix equation method, the two-step method presented in this paper has higher recognition accuracy and lower computational cost. After that, the damage identification experiments of single crack and double crack aluminum alloy cantilever structure based on PVDF sensors array were conducted. Simulation and experimental results showed that the proposed two-step method in this paper is versatile for both single crack and multiple cracks identification.