Application of level-set type topology optimization for defect topology identification analysis using hammering response data

Abstract

In this study, we presented the defect identification analysis based on the level-set type topology optimization using hammering testing data. To simulate the oscillation behavior of concrete structure, the equation of motion in three dimensions is introduced as the governing equation. The performance function is defied by the time integral of the squared residual between the computed and observed displacements. The problem is to find the defect topologies so as to minimize the performance function. In the model with a small number of observation points, it was found that nonexistent defects disappeared by using the weighted sensitivity method. In addition, it was confirmed that the model with a large number of observation points did not show the same tendency as the model with a small number of observation points because the accuracy of defect identification increased as the number of observation points increased.