Recording earthquakes

Seismie analysis is carried out for all important engineering structures such as dams, bridges and nuclear power plants. For regions where these are to be located the likely expectations of an earthquake as well as the extent of its magnitude must be assessed on the basis of the seismic history and the earthquake records of the region (Figures 14.12 to Figure 14.16). Based on these and other factors such as soil stratification, site dependent response spectra are determined. These are the RRS for equipment mounted... 

Fig. 5. Comparison of vertical component seismograms for southern Africa earthquakes recorded at two distances (a), (b) 1038 km (event 2 recorded at SLR, Fig. 1) and...

Beck, J. L. Determirung Models of Structures from Earthquake Records. Technical Report EERL 78-01, Cah-fomia Institute of Technology, Earthquake Engineering Research Laboratory, Pasadena, CA, 1978. 

Beck, J.L.(June 1978) Determining Models of Structures From Earthquake Records. EERL Report No.78-01, California Institute of Technology, Pasadena, Calif., pp.300. 

Another method of site response, which does not depend on the reference site, is based on horizontal to vertical spectral ratio (HVSR). In this method the horizontal component ofthe response spectra is normalized using the vertical component of the spectra for the site under consideration. This method can be applied for both the noise recordings (Nakamura, 1989 Field and Jacob, 1993) and the earthquake recordings. 

Chang, F.K., Krinitzsky, E.L. (1977). Duration, spectral content and predominant period of strong motion earthquake records from Western United States. Miscellaneous paper 5-73-1, U.S. Army Corps Engineers Waterways Experiment Station Vicksburg, Missipppi. 

Fig. 1.4. Seismograms of the 1906 San Francisco earthquake recorded in Gottingen, Germany, some 9100 miles away from the earthquake source (top) NS components, (bottom) EW component (from Wald et al. 1993).

To obtain the dynamic response of the stmcture, 9 different types of earthquakes have been used 6 synthetic earthquake records defined by seismic microzonation as well as records of three other earthquakes, namely, Petrovac 1979, Ulcinj 1979, and El Centro 1940. The response has been investigated for the maximum input grotmd acceleration of a ax = 0.16g and ax = 0.22g in accordance with the defined seismic hazard for return periods of 475 and 1,000 years. As a result of the dynamic analysis, displacements and ductility demanded by the earthquake =... 

The generated earthquake records in Table 18.2 have much larger velocity and displacement values due to the increased time step. The use of spline interpolation also introduced slightly larger maximum accelerations than the original data. This fact is expected from the numerical procedure and it can be neglected. 

The shaking table is one-dimensional, medium size (1.5 m x 2 m), light weight (made of 25.4 mm thick aluminum plate), and servo hydraulic controlled system. The table motion can be controlled from a control unit as well as from a computer. The control unit provides harmonic excitation options such as sinusoidal, step motion, triangular at various frequencies and amplitudes utilizing a plug-in function generator. On the other hand, earthquake records can be simulated from a computer... 

According to the earthquake record monitored by Chengdu station (CD2) of Chinese National Earthquake Bereau, the Wenchuan earthquake lasted for about 160 seconds and strong part are nearly 20 s. In this paper, earthquake wave of the front 20 seconds is used. Seismosignal software is used to make the filtration with range of 0.1-10 Hz and then do the linear baseline calibration. The velocity time history should be first transferred to velocity-stress time history and then set in the bottom viscous boundary. Stress time history is shown in Figure 7. 

Earthquake Record M Station Data Source Distance (km) Site Soil Condition PGA(g)... 

For nonlinear response history analyses (NRHA), three different earthquake records are employed with characteristics listed in Table 1. These records are appropriately scaled to match a smoothed elastic response spectrum as shown in Figure 3. In the analysis PERFORM-3D structural analysis program (CSI 2006) has been utilized. 

In general, dynamic response of any system depends on the seismic excitation characteristics (both in the time and in the frequency domain). In a recent preliminary investigation of DWSSI by Tsompanakis et al. (2006), both real earthquake records and pulses were used. In the present numerical study, in order to understand more clearly various aspects of the complex phenomena incorporated in the DWSSI, the excitations were limited to harmonic and simple pulses. Results provide a clear indication of the direct dynamic interaction between a retaining wall and its retained structures. That fact justifies the necessity for a more elaborate consideration of this interrelated phenomenon on the seismic design, not only of the retaining walls, but of the nearby structures as well. 

Vanmarcke, E. 1976. Structural response to earthquakes. In C. Lomnitz and E. Rosenblueth (Eds.), Seismic Risk and Engineering Decisions, pp. 287-337. Amsterdam Elsevier. Vanmarcke, E.H. Lai, S.P. 1980. Strong-motion duration and rms amplitude of earthquake records. Bulletin of the Seismological Society of America 70, 1293-1307. 

The treatment of uncertainties requires the use of probabilistic methods, estimating the probability of exceeding response targets for the different performance requirements, for example, on an annual basis. The dynamic structural responses are highly nonlinear, and their time history must be found by numerical (e.g., finite elements) analysis for the duration of the earthquake. In a thorough analysis, the nonlinearity of the response is further increased when the interactions between the structure and the foundations are included. It is not possible to establish an explicit relationship between the intervening variables and the dynamic responses, and results can only be obtained in a discrete manner, given specific values of the structural variables and a particular earthquake record. Reliability calculations depend on simulations... 

Vanmarcke, E.H., Lai, S.P. 1980. Strong-motion duration and rms amplitude of earthquake records. Bulletin of the Seismological Society of America 70, 1293-1307. 

Dynamic time history analyses are performed to demonstrate the validity of the proposed design methodology, using four earthquake records representative of near field and far field conditions. 

The base shear coefficient normalized with respect to the total weight ofthe building is shown in Figure 7, comparing the uncontrolled and the controlled structure. An average reduction of the base shear is observed for all the earthquake records of about 26%, while the higher reduction is obtained with Hachinobe earthquake (47%). 

For both indices instead different damper distributions can be obtained using different earthquake records that can be caused by the different frequency content of each earthquake (Figure 4). Hence it has been found that the optimal distributions obtained with J3 and J4 as objective functions strongly depend on the nature of the earthquake record at the site. Hence, the optimal damper distribution for seismic excited buildings can be obtained by minimizing J4 using the design earthquake of the particular site. 

This conclusion can be generalized, because the distribution is independent from the earthquake record selected, as shown in Figure 15, where there is almost no difference among SS, WOBI and ESPS when using the maximiun drift (J4) as index, while small differences can be found when index J3 is adopted as shown in Figure 16. 

The optimal device distribution is affected by the specific earthquake record employed in the analysis. 

---