GNGTS 2014 - Atti del 33° Convegno Nazionale

GNGTS 2014 S essione 2.3 317 formulation recently proposed by Lagomarsino (2014): (1) (2) where T k , ψ k and γ k are the period , modal shape and coefficient of participation of mode k , respectively; η is a damping correction factor [e.g., assumed as proposed in Eurocode 8 (CEN, 2005)]; ζ b is the damping of the building and ζ that of the mechanism. The coefficient of participation may be estimated by the approximate formula 3 N /(2 N +1), being N the number of stories. By assuming ψ k equal to 0.7 and a damping correction factor equal to 1 (since the response is analyzed in elastic phase), expressions (1) and (2) provide values of )0( a S and ) ( 1 TS a equal respectively to 0.0009 and 0.00418 m/s 2 . The complete acceleration spectrum, which accounts for the contribution of the first mode only, is plotted in Fig. 3. This assumption justifies the lower value of amplification estimated by the analytical floor spectrum with respect to the experimental ones. Conclusions. In this study spectral ratio techniques have been applied to study the response of three buildings in the historic town of Genoa using noise and blast recordings. Results for one building are presented in this study [the results for the remaining structures can be found in the article by Barani et al. (2014)]. Rather than remarking what described above, we want here to make general considerations deriving from this experiment: 1. H/V spectral ratios from noise recordings are consistent with those calculated using blast signals, thus indicating that the eigenfrequency of vibration of a structure is insensitive (or almost insensitive) to source signal (i.e., input excitation). 2. Although ambient noise H/V measurements represent an economical and flexible alternative to SSRs, because of the less equipment involved and because no earthquake recordings are needed, they could lead to misleading results in the sense that it may not be possible to distinguish the eigenfrequencies of the building from the resonance frequencies of foundation soil (unless they are known from free field measurements). Therefore, the application of the SSR technique may be preferable in densely urbanized environments where free field H/V measurements may be affected by the presence of buildings. 3. The amplitude of the peaks obtained fromH/V spectral ratios differ from those determined by using a reference station. As a consequence, H/V curves can not be assumed as representative of the absolute amplification level. 4. H/V spectral ratios and SSRs may be very useful to support the seismic assessment of buildings, in order to calibrate both numerical models and the seismic demand to be adopted in the evaluation of local mechanisms that involve the upper parts of the structures. References Barani S., De Ferrari R., Eva C., Lagomarsino S., Cattari S. and Pieracci A.; 2014: Structural response of Historical and residential buildings during dredging activities in the genoa harbour . Proceedings of the 2 nd European Conference on Earthquake Engineering and Seismology, Istanbul, 25-29 August, Paper No. 459. Bard P.Y.; 1999: Microtremor measurements: a tool for site effect estimation? In: Irikura, Kudo, Okada and Sasatani, (eds), Second International Symposium on the Effects of Surface Geology on seismic motion, Yokohama, December 1-3, 1998, Balkema 1999, 3 , 1251-1279. Beijing Jiaotong University; 2007: Report on cultural relic protection along the subway line . Bindi D., Parolai S., Spallarossa D. and Cattaneo M.; 2000: Site effects by H/V ratio: comparison of two different procedures . Journal of Earthquake Engineering, 4 , 97-113.