Statistical subspace based damage localization on Saint-Nazaire Bridge mock-up

The subject of damage localization is an important issue for Structural Health Monitoring (SHM) particularly in mechanical or civil structures under ambient excitation. In this paper, the statistical subspace-based damage localization method has been applied on a benchmark application, namely a 1/200 scale model of the Saint-Nazaire Bridge, which is a cable-stayed bridge located on the Loire River near the river's mouth. The employed damage localization method combines data-driven features with physical parameter information from a finite element model in statistical tests, avoiding typical ill-conditioning problems of FE model updating. Damage is introduced in the mockup for cable failures on some of the 72 cables. The purpose of the experiment is to assess the capability of damage assessment methods to find a cable failure.