A method of utilizing the principle of magnetostriction to perform non-destructive diagnosis to determine whether or not there is residual stress produced by an earthquake in a bent pipe has been developed. The method is used to measure total stress of the principal stress difference. Because large stress produced during the manufacture of a bent pipe remains in the pipe, an effort was made to separate the residual stress from the measured value by regressing the measured value to a theoretical stress calculation equation for bent pipe by Rodabaugh and George. As a result, only the stress produced by external force was precisely obtained divided into longitudinal and circumferential stress. Because the upper limit stress value that can be diagnosed by the method is approximately 55% of the yield stress, it has been proposed as a method to be used to make a primary judgment of whether or not stress produced by an earthquake remains.

View full abstract

In seismic areas of the 2004 Mid Niigata earthquake of Japan, the authors carried out questionnaire surveys on inhabitants' impression about performance of seismically isolated buildings. The questionnaire mainly deals with floor responses. And its items are concerned about the ability of evacuation action, feeling and aspects of floor vibration, valuation of seismically isolated buildings and so on. In the 6⁺ seismic intensity area, some inhabitants felt discomfort like seasickness during the earthquake and a few light or unstable interior things fell down or overturned on the floor. But on the whole, seismically isolated building effectively reduced floor responses and most of the inhabitants highly appreciated its performance.

View full abstract

Grouped-drilled shaft foundations are one of the most common foundation types in highway bridge foundations, and it is expected to develop a nonlinear numerical model to predict their deformation capacities during earthquakes. In this paper, numerical simulations of model single and grouped drilled-shafts are conducted using a fiber model that considers the effects of material nonlinearity including lateral con- finement from surrounding soil to pile concrete and reinforcement swelling. Numerical results agree well with experimental results in single drilled shaft cases. However, the capability of the numerical simulation for grouped-drilled shafts is worse than that for single drilled shafts, while the calculation captures overall tendencies as observed in the experiment. The numerical results indicate that the compressive stress-strain model for concrete can strongly affect calculated foundation behavior, because, in groupedpile foundations, the axial load in each pile varies with an increase in the horizontal displacement of the foundation, and the piles become subjected to large axial loads.

View full abstract

Accuracy of earthquake response analyses of ground is examined focusing on damping. In order to compare frequency domain analysis and time domain analysis under the same condition, functions to consider various frequency dependent damping are installed in the improved equivalent linear analysis, and modal damping, which has more frequency in determining damping with respect to frequency, is installed in the dime domain analysis. Moreover, a method to build modal damping in the elastic base problem is proposed. Comparisons between frequency domain and time domain analyses and between constant damping and damping proportional to frequency are made by an example ground under both elastic and nonlinear conditions. It is found that both frequency domain and time domain analysis give nearly identical result under elastic response, which indicates that modal damping can be used for frequency dependent problem in the time domain analysis. In the comparison of nonlinear analysis, behaviors are also almost identical in frequency lower than several Hz, and slight difference appears in the high frequency regions. In addition, proportional damping has a tendency to evaluate maximum strain or maximum displacement smaller than constant damping.

View full abstract

We have constructed a three-dimensional (3D) model of geological structure of a seismic gap region south of the 2004 Mid Niigata Earthquake with a recently-proposed 3D modeling method using constraints of geotectonic history revealed by geological studies. The large-scale, say 64 km x 33 km, 3D geological structure was successfully modeled on the basis of surface geological maps, a relatively small amount of exploration data and the information of the important geotectonic processes. We have poin ed out an advantage and a disadvantage of the method: This method is considered to be a powerful tool to construct and revise a three-dimensional model of a region without sufficient exploration data, and to be appropriate for such a large-scale region in view of the simplification incorporated to the model. We consider that the resultant 3D model can be used for a strong ground motion prediction in the future earthquake region.

View full abstract

frequency range are examined and high-cut filters with a cut-off frequency, f_maxare proposed for small and large events. The obtained high-cut filter of large event is compared with the filters proposed in previous studies. Moreover, a filter to correct difference of spectral attenuation between small event and large events in high frequency range is proposed. Finally, effects of the proposed filters on strong ground motion prediction are examined through simulations by stochastic Green's function method. The results obtained in this paper may contribute to strong ground motion prediction in high frequency range.

View full abstract

We conduct a systematic and quantitative evaluation of effects of a deep sedimentary structure model between the engineering and seismic basements on the theoretical broadband site-amplification spectra, taking sites in the Osaka basin, Japan, as a test site. First, from a parametric study on theoretical evaluation of site amplification, adopting a one-dimensional S-wave propagation theory, we show that spectral peaks in lower frequency rang are controlled by velocity structure of the deep sediments, while a spectral amplification trend in higher frequency range is controlled by attenuation structure of the deep sediments. On the contrary, a shallower sedimentary structure model above the engineering basement mainly controls a shape of site-amplification spectra in higher frequency range. Next, we validate a macroscopic modeling for the deep sedimentary structure, based on an inversion of empirical site-amplification spectra. Finally, we show that the spectral reproduction in low frequency ranges is improved by a change in a macroscopic-model representation of deep sedimentary structure from a stack of three homogeneous velocity layers to a gradual velocity increasing model.

View full abstract