Statistical damage detection based on strain mode shapes

摘要

Based on the strain mode shapes(SMS),a statistical damage detection approach is proposed and applied to the spatial frame structures.An element stiffness reduction parameter at a location is taken an index for damage at the corresponding structural location.To reduce the number of the parameters,a substructure approach is followed in which elements are divided into different groups.All elements are classified by similarity degrees between element perturbation vectors of SMS.A Bayesian statistical method employing SMS is followed to obtain the posterior probability density function(PDF)of the parameters.The most probable values and standard deviations of the parameters are determined through maximizing the PDF.By this means,damages are detected in two steps:the groups containing potential damaged elements are identified first,and then damage locations and sizes are detected.A numerical example is adopted to demonstrate performance of the proposed approach.