China
Africa
Explore chapters and articles related to this topic
Seismic wave dispersion in high-rise historical building by interferometric analysis: The case history of Giotto's Bell Tower
Published in Renato Lancellotta, Carlo Viggiani, Alessandro Flora, Filomena de Silva, Lucia Mele, Geotechnical Engineering for the Preservation of Monuments and Historic Sites III, 2022
G. Lacanna, R. Lancellotta, M. Ripepe
The ambient vibrations were also used to investigate the wave propagation with seismic interferometry, a technique based on the correlation of waves recorded at different receivers. If we consider two signals u1(ω) and u2(ω) in the frequency domain, and we assume u2(ω) to be the input value, the Frequency Response Function (FRF) is defined as: FRFω=u1ω/u2ω
Passive seismic interferometry’s state-of-the-art – a literature review
Published in Jean-Pierre Tournier, Tony Bennett, Johanne Bibeau, Sustainable and Safe Dams Around the World, 2019
C.T. Rodrigues, A.Q. de Paula, T.R. Corrêa, C.S. Sebastião, O.V. Costa, G.G. Magalhães, L.D. Santana
As elucidated by Wapenaar et al. (2009), the term ‘seismic interferometry’ refers to the principle of generating new seismic responses of virtual sources by cross-correlating seismic observations at different receiver locations. The principle of this methodology can be distinguished between the use of controlled-source or passive seismic interferometry.
Identifying Dynamic Response of a Twenty-Story Instrumented Building to 2018 M7.1 Anchorage, Alaska Earthquake and Its Aftershocks
Published in Journal of Earthquake Engineering, 2022
One of the methods to perform the seismic interferometry of two signals is deconvolution. The deconvolution of the response at height by the response at height , is defined as
Characterizing Nonlinear Effects in Vertical Site Response of Dry Soils Using KiK-Net Data
Published in Journal of Earthquake Engineering, 2023
Beresnev, Nightengale, and Silva (2002) suggested the first constrained-modulus MRD model, based on an empirical study on ground motion records from the Japanese Kikban-Kyoshin Network (KiK-net). They analyze the P-wave window of their time-histories, assuming that is the only portion of the time-history to purely represent compressional waves. Their strong motion records represent five different stations and only one strong event. They compute the empirical transfer function (ETF) of the P-wave window (using surface to borehole spectral ratios – SBSR) and identify non-linearity as a shift in the predominant frequency, disregarding effects of groundwater levels or saturation on the soil response. Based on five data points, they conclude that the MRD of the constrained modulus is almost identical to that of the shear modulus. In a similar analysis, using a much more extensive dataset, Han et al. (2015) measured the shift in the predominant frequency, using ETFs from 29 KiK-net stations and looking at the entire VGM time-history, without cutting out the P-wave window. Although they include sites with very shallow GWD, they consider their results to represent the response of soils above the water table, because they only use VP of soils above the water table in their computation of strains. Tsai and Liu (2017) proposed an analytically based MRD for the constrained modulus, using elastic theory to compute M directly from G, while accounting for saturation conditions by introducing the bulk modulus of water into the computation of VP. Their analysis, also validated against recordings from five geotechnical downhole arrays, suggests that the MRD for M in dry soils is identical to that of G but that saturated soils show almost no non-linearity. Liu and Tsai (2018) continued with an empirical analysis of frequency shifts, using the same five downhole arrays, concluding again that saturated soils exhibit little to no non-linearity in the vertical direction. Finally, Shi et al. (2020) used deconvolution seismic interferometry to compute changes in the near-surface P-wave velocity before and after a strong shaking event, corelating the changes in the velocity to modulus degradation. They used recordings from 8 KiK-net stations, identifying a linear threshold on normal strains in the range of 1E-6 to 2E-05, which is similar to that of the shear strain linear threshold.
Nonlinear Response of Soil–Structure Systems using Dynamic Centrifuge Experiments
Published in Journal of Earthquake Engineering, 2019
Johanes Chandra, Philippe Guéguen
Figure 1a presents the plan view and cross section of the container used for the centrifuge test and the positions of the accelerometric sensors spread along the soil column and on the surface. Equivalent shear-beam (ESB) containers, 800 × 350 × 416 mm in dimension (model scale), were used to limit container edge effects on wave propagation [Zeng and Schofield, 1996]. Horizontal and vertical piezoelectric accelerometers (PCB type 200 A1 and Bruel & Kjaer type 4317) were used, attached to a thin plate to control orientation and position. Soil–structure models were excited by a shaker driven in displacement at the bottom of the soil column. Two different input signals, representing weak and strong excitation, were considered (Fig. 2). The first signal was a synthetic accelerogram representing strong ground motion with a realistic phase, corresponding to a Mw 5.5 earthquake at 15 km [Chazelas, 2010] and having a peak ground acceleration (PGA) equal to 0.43g (dominant frequency close to 1.86 Hz). The second signal was a moderate seismic ground motion recorded during the Mw 7.3 earthquake in 2007 in Martinique by one of the French Accelerometric Network’s stations [Péquegnat et al., 2008]. PGA was 0.07g with a dominant frequency close to 2.7 Hz. In the rest of the manuscript, strong and weak response will refer to the response of the soil and structure models under strong and weak excitation, respectively. For each setup, we prioritized experiment control and reproducibility of the trials [Chandra et al., 2016] rather than multiplication of the input signal. Moreover, during experiments, slight settlements usually occur that may slightly change the position of the sensor or the elastic properties of the sand. Without measurements of the sensor position and density during the experiments, the experimental sequence was therefore performed with two different containers and between three and six trials were done for each [building/container/input signal] setup. Chandra et al. [2015] showed the efficiency and stability of centrifuge tests, in terms of limitation of wave reflections on the container edges, equivalent response of the soil column between containers and reproducibility of the soil motion generated by the shaker. The soil was constituted of uniform, homogeneous, fine Fontainebleau sand (reference N234, emin = 0.55; emax = 0.86; γd = 15.42 kN/m3), poured into the container by dry pluviation to attain a relative density, Dr, equal to 57%. Chandra et al. [2015] computed the shear wave velocity (Vs) along the FF soil profile by applying the seismic interferometry by deconvolution method to the accelerogram recorded along the soil profile. According to Chandra et al. [2015], Vs values between the top and the bottom of the container equal to 242 m/s and 197 m/s were recorded under weak and strong excitation, respectively, the same as values provided by Li et al. [2013] for the same container configuration.