Structural health monitoring of a masonry arch bridge: modal identification and model updating

Historic masonry arch bridges are widespread across the European countries and the assessment of their current safety under operative conditions is a hard task. The dynamic loads due to the vehicular traffic and the degradation of the materials often speeded up by harsh environmental conditions and scarce maintenance, which may lead to critical situations. As a consequence, the scientific community is working in order to develop and propose reliable methodologies of analysis and non-invasive technologies of investigation to assess their current state of conservation. In this respect, this paper discusses an experimental activity performed in a two-span historic masonry arch bridge. Output-only modal identification was carried out both in time and frequency domains using the stochastic subspace iteration (SSI) and the frequency domain decomposition (FDD), respectively. The results provided by the two procedures are first compared, and subsequently employed for FE modal updating. To calibrate the FE model an optimisation approach based on genetic algorithms (GAs) was employed.