Seismic deconvolution interferometry applied to the monitoring of dams structural integrity

The aim of the thesis is to test the possibility of using a set of preliminary parameters to decide when and how to organize further detailed investigations to assess the structural integrity of dams. The parameters researched should be the outcome of a method that must be cost effective and therefore simplified. This work proposes to use the shear wave velocity and the structural damping obtained by interferometric analysis based on deconvolution of recorded seismic events or ambient vibrations. In this work, I discuss whether this method could be profitably used as a first explorative approach to dam dynamical characterization. To verify the applicability of the deconvolution interferometry approach on dams, a concrete arch-gravity dam located in Central Italy is assessed as a case study, using different techniques. On the dam several survey campaigns were performed in different time periods: two dynamic forced vibrations tests were performed in the past (July 1988 and May 1993), on the basis of which a finite element model was developed. In 2015 an ambient vibration survey was performed; one year later the Central Italy earthquake of the 24th August 2016 hit the structure, causing non-structural damage. After the earthquake the Civil Defence Department repeated the ambient vibration tests on the dam and installed a permanent dynamic monitoring system, part of the Seismic Observatory for Structures (OSS – Osservatorio Sismico delle Strutture). The monitoring system recorded five seismic events in 2017, which are analysed in this thesis using deconvolution interferometry. The large amount of data recorded on this specific dam was used in order to compare the results obtained with the different techniques, working as a consistency check on the outcomes of the interferometric approach based on deconvolution. The results obtained by the seismic deconvolution interferometry are promising. The shear wave velocity inside the structure, in the central section of the dam, was estimated to be around 900 m/s. This value of velocity could be expected, since shear wave velocity inside regular buildings is usually around 300 m/s, but the investigated dam is much stiffer and does not present any empty spaces except the tunnels, which are negligible in comparison to the geometric dimensions of the dam. The consistency of this velocity value was also checked through the formula, valid for the shear beam model, which relates the fundamental frequency to the measured velocity, obtaining a value of frequency very close to the one determined by forced and ambient vibration tests, although the excessively simplification assumed to apply this formula. The damping factor was calculated too, using the attenuation of the recorded wavefield, resulted to be 3%.