SHM method for locating damage with incomplete observations based on substructure's connectivity analysis

In this study, a structural health monitoring (SHM) method is proposed for damage localization with incomplete observations, which combines a substructuring approach and connectivity analysis of a substructure. The substructuring approach aims to isolate and monitor several substructures individually. The connectivity among the substructural degree-of-freedom (DOFs) was calculated based on the multivariate and conditional Granger causality analysis, and the health of the substructure was determined by analyzing connectivity variations. This method determines whether the substructure is damaged, and it also allows the location of damage within the substructure. It is not necessary to have a dynamic model of the structure; it only requires the use of accelerometers. The proposed method was evaluated numerically and experimentally using a three-dimensional truss structure.

Automated summary

In this study, a structural health monitoring method is proposed for damage localization. It combines a substructuring approach and connectivity analysis of a substructure, aiming to isolate and monitor multiple substructures individually. The health of the substructure is determined by analyzing connectivity variations using multivariate and conditional Granger causality analysis. The proposed method can determine whether the substructure is damaged and locate the damage within the substructure, without the need for a dynamic model of the structure.