Damage Detection of Large-Scale Space Trusses Based on Changes in Static Flexibility Matrix

Detecting damages in large-scale space trusses poses a significant challenge due to numerous elements that can generate disturbances with relative ease. Based on changes in the static flexibility matrix, a two-stage method to localize and quantify damages of large-scale space trusses was thus proposed. The first-order sensitivity equation between damage severities and flexibility matrix changes is established in the damage localization stage. A singular value contribution index (SVCI) was constructed to truncate singular values of the flexibility sensitivity matrix properly. Then, an iterative process using SVCI was suggested to make damaged elements more outstanding, so that suspiciously damaged elements could be readily determined. In the damage quantification stage, the improved particle swarm optimization algorithm was employed to further estimate damage severities by merely setting suspiciously damaged elements as optimization variables. Take a space truss as the numerical example. Analysis results show that the proposed method has good antinoise ability and can accurately achieve damage localization and quantification by using an incomplete static flexibility matrix composed merely of vertical degrees of freedom.