NIED, Japan
NIED, Japan

Time filter

Source Type

Yamashita T.,NIED | Ohsaki M.,Hiroshima University | Kohiyama M.,Keio University | Miyamura T.,NIED | And 3 more authors.
Journal of Structural and Construction Engineering | Year: 2014

Modeling and numerical simulation techniques for performing a detailed finite element analysis of a composite beam are presented. The finite element analysis is performed using E-Simulator, which is under development at the Hyogo Earthquake Engineering Research Center (E-Defense) of the National Research Institute for Earth Science and Disaster Prevention (NIED). A piecewise linear isotropic-kinematic hardening law is used for the steel material, and heuristic and implicit rules are incorporated to simulate its complex cyclic elastoplastic behavior. The extended Drucker-Prager model is used for the concrete material. The steel beam, steel column and reinforced concrete slab are discretized into solid elements. The wire mesh consisting of steel bars in the slab is also modeled using solid elements and rigid beams are used for the stud bolts. Detailed analyses are carried out for both a steel beam and a composite beam subjected to static cyclic loading. Good agreement is found with experimental results such as strength degradation due to local buckling of the flange, and asymmetric behaviors resulting from contact between the slab and the column.

Minowa C.,Tomoe Research and Development | Mikoshiba T.,NIED | Hanazato T.,Mie University | Nitto K.,Tokyo University of the Arts
Journal of Structural and Construction Engineering | Year: 2014

Motion behaviors of the national heritage five story wooden pagoda at Hokekyouji temple are observed from 2007. Continuing part 1,modal and response analyses for the pagoda are conducted. The model is made by using the beam vibration modes.Response alyses are carried out for both 3/11 earthquake and strong wind. As results of linear numerical calculations to strong1 earthquake motion and stiong wind, the damping about 5%-10% would be agreed with observed response data. Modal analysis results showed the possibility of resonance between a pagoda frame and center column. However, in actual pagoda, it would be difficult to expect dynamic damper effects because of irregulariti's oe pago dastructure.

Takeuchi T.,Tokyo Institute of Technology | Yamamoto Y.,Tokyo Institute of Technology | Midorikawa M.,Hokkaido University | Kasai K.,Tokyo Institute of Technology | And 4 more authors.
Journal of Structural and Construction Engineering | Year: 2011

The authors have conducted a series of experimental studies on the response characteristics of controlled rocking systems. This system comprises components that include a rocking frame, post-tensioning anchorages, and energy dissipation fuses, eliminating the residual deformation after the earthquake. In this paper, the shaking table test of the controlled rocking system using a Buckling Restrained Brace as the energy dissipation fuse is discussed, including analytical simulations. A procedure to estimate maximum response is introduced based on the concept using maximum instantaneous input energy. The characteristics of the input energy and the accuracy of the proposed approach are demonstrated by comparing the prediction with numerical results as well as experimental results.

Mikoshiba T.,NIED | Suzuki T.,Brain Inc. | Sato T.,Brain Inc. | Terai M.,Fukuyama University | Chiba T.,Sanmai cho 161 20
American Society of Mechanical Engineers, Pressure Vessels and Piping Division (Publication) PVP | Year: 2011

By setting the isolation system between the foundation of the house and the foundation ground, it makes it possible to scale back the influence of the earthquake. A sliding isolator, which is compact and has high isolation characteristics, had been developed. This isolation system of the house have found effective and had high performance. This system was applied to the vending machine. Although the sliding isolator has been shown to be effective, it required the wide space around the vending machine. By using the base isolation combined with the air damper, it allowed to use in the narrow space area. It is useful when the vending machine is set in front of the wall. A sliding isolator combined with air damper has high isolation characteristics. Copyright © 2011 by ASME.

Okazaki T.,Hokkaido University | Lignos D.G.,McGill University | Hikino T.,NIED | Kajiwara K.,Japan National Research Institute for Earth Science and Disaster Prevention
Structures Congress 2011 - Proceedings of the 2011 Structures Congress | Year: 2011

Large-scale shake table tests were conducted to examine the dynamic response of steel concentrically braced frames. The specimen was a single-bay, single-story frame with a pair of square-tube braces placed in a chevron arrangement. The gusset plates connecting the brace to the framing elements were provided with an elliptical clearance to permit free out-of-plane rotation. The specimen was subjected repeatedly to unidirectional ground motion with increasing magnitude until the braces fractured. The test results demonstrated excellent performance of the gusset plate connections. The specimen response was reproduced by a numerical model using fiber elements. This model was able to predict the occurrence of brace fracture and post-fracture behavior of the frame. Observations from the shake table tests and aspects of the numerical analysis are reported in this paper. © ASCE 2011.

Hikino T.,Steel Structures Engineering Division | Okazaki T.,NIED | Okazaki T.,Hokkaido University | Kajiwara K.,Japan National Research Institute for Earth Science and Disaster Prevention | Nakashima M.,Kyoto University
Structures Congress 2011 - Proceedings of the 2011 Structures Congress | Year: 2011

Large-scale shake table tests were performed to examine the out-of-plane stability of buckling-restrained braces (BRBs). Two specimens were subjected repeatedly to a near-fault ground motion with increasing magnitude. The test specimens comprised a single-bay, singlestory steel frame and a pair of BRBs placed in a chevron arrangement. The chevron braced frame specimens were not braced at the point of intersection in the beam, and was therefore susceptible to out-of-plane instability. Standard BRBs were used in the first specimen, while unique BRBs with a flexible segment at each end of the steel core were used in the second specimen. A simple stability model predicted the BRBs in the second specimen to fail due to out-of-plane buckling at a force smaller than the yielding strength of the steel core. The first specimen exhibited excellent ductility during the shake table tests, while the second specimen developed noticeable out-of-plane deformation. Observations from the tested are described along with design implications. © ASCE 2011.

News Article | December 9, 2015

Large fault slips, which can produce strong ground motion and a huge tsunami, are associated with a remarkable decrease in rock friction at seismic slip rates. Numerous laboratory experiments1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13 have investigated the underlying mechanisms of reduction of friction (weakening) at high slip rate. Although various weakening mechanisms—such as gouge or powder lubrication8, 9, 11, silica gel lubrication2, 4 and melting1, 6, 10, 13—have been suggested, all of them seem to be triggered by a rapid temperature rise caused by frictional work on the fault5. Therefore, the mechanical work rate per unit area (that is, the power density) seems likely to be a key parameter. Another question then arises: are the frictional properties independent of length scale? It is known that rupture propagation on a fault is affected by the fault surface topography14, and thus should be dependent on scale. Presumably, steady-sliding rock friction is also affected by scale. However, it has not been possible to investigate such a dependence because it has been difficult to conduct a larger-scale friction experiment. Although some friction experiments using metre-sized rock specimens have been conducted15, 16, 17, 18, the apparatuses used could not produce large slip distances with high slip rates. We therefore conducted experiments with metre-sized rock specimens by using a second-generation large-scale biaxial friction apparatus at the National Research Institute for Earth Science and Disaster Prevention (NIED) in Japan (Fig. 1a)19. We used a pair of quadrangular prismatic rock specimens made of Indian metagabbro, whose nominal contact area was 1.5 m long and 0.1 m wide. The contact surfaces were flattened to an undulation of within 10 μm before the first experiment. We repeated several experiments using the same rock specimens under the conditions of constant loading rates up to 3 × 10−2 m s−1 and with constant normal stresses up to 6.7 MPa (see Supplementary Information section 1 and Supplementary Table 2 for detailed methods and conditions, respectively). After each experiment, we found indications of localized damage on the fault surface, including gouge and grooves (Fig. 1b–d). The grooves began with very sharp edges (Fig. 1c) and widened with continued slip. The grooves were filled with the gouge, which looked heavily comminuted and partly consolidated (Fig. 1d). Consolidated gouge within the grooves often swelled up and spilled into the surrounding fault surface. The fault surface without grooves became smoother and smoother with experiments, which was also confirmed by topography measurements (see sections 2 and 3 in Supplementary Information for detailed observation of the fault surface). After each experiment, we removed and inspected all gouge, including from inside the grooves. Therefore, the groove areas were not in contact at the beginning of the next experiment. To investigate the distribution of damage, we took pictures of the fault surface just after each experiment as well as after the gouge collection. Detailed methods of the damage analysis are described in Supplementary Information section 4. Figure 2 shows a clear dependence of rock friction in the metre-sized rock specimens on the work rate per unit area (or power density), which we term ‘work rate’ for simplicity hereafter, obtained from the present experiments. At a low work rate, the rock friction obeys Byerlee’s law20, but it falls dramatically at high work rates. Such work-rate dependence has already been reported, based on the data of rotary shear experiments5, 12. To directly investigate whether there is any size effect on the frictional properties, we also conducted rotary shear experiments with specimens of metagabbro and monzodiorite whose inner and outer diameters were 1.7 cm and 4.0 cm, respectively. The two rock types yield a similar frictional dependence on work rate, as shown in Fig. 2. The experimental data are listed in Supplementary Table 3. As shown in Fig. 2, the findings clearly demonstrate a scale dependence of rock friction, as follows. The friction in centimetre-sized rock specimens temporarily decreases at the intermediate work rate (~5 × 10−3 MJ m−2 s−1), increases with the increase in the work rate from 10−2 MJ m−2 s−1 to 10−1 MJ m−2 s−1 and suddenly falls at a work rate of 10−1 MJ m−2 s−1, whereas the friction in metre-sized rock specimens starts to decrease at a work rate that is one order of magnitude smaller (10−2 MJ m−2 s−1) than that of the centimetre-sized rock specimens and without showing the reduction in the intermediate work rate. Such differences are not likely to be explained by the differences in experimental conditions (see Supplementary Information section 3). Figure 3a and b shows the macroscopic friction data of experiments LB05-004 and LB04-007, respectively. The work rate of LB05-004 is larger than that of LB04-007 since the loading rate of LB05-004 is higher (Supplementary Table 2). Figure 3c–f shows the evolution of the spatial distribution and its standard deviation of local shear stresses, which were measured by a local shear strain gauge array (see Fig. 1a and section 1 in Supplementary Information). We observed local shear stress concentrations and a related increase in spatial heterogeneity (standard deviation of local shear stress) with the fault slip during LB05-004, as illustrated in Fig. 3c and e. In contrast, we did not observe such an increase in local stress variation during LB04-007, in which the friction coefficient did not drop with slip (Fig. 3b, d, f). These observations suggest that the decrease in the macroscopic friction of metre-sized rocks is associated with the increase in spatial stress heterogeneity as well as the work rate. Similar results can be seen in other experiments (Supplementary Fig. 7). We also evaluated the degree of spatial heterogeneity on the fault by mapping newly generated grooves and gouge using the photo image data. A negative correlation was observed between the final friction coefficient and the spatial heterogeneity; the friction coefficients tended to be low when the grooves and gouge were heterogeneously generated during the experiments (see Supplementary Information section 5 and Supplementary Fig. 8). We found clear evidence of melting in the gouge collected in the large-scale biaxial experiments (see Supplementary Information section 5 and Supplementary Fig. 9), though frictional melting did not occur in centimetre-scale rotary shear experiments under similar mechanical conditions12. This material observation suggests intense stress concentrations with locally high work rate. From these observations, we propose a stress localization model in which the contacting fault surfaces are composed of patch and off-patch areas with rather high and low normal stresses, respectively. The high normal stress on a patch causes high shear stress, high mechanical work and a high work rate upon loading, which lead to a high production rate of wear material (that is, gouge). The gouge generated is dragged by the dislocating fault surface, which will cause further extension of the grooves, and thus will produce further gouge and stress heterogeneities. This is because the gouge layer on the groove (patch) will become thicker than the gouge layer on the surrounding area (off-patch). Therefore, these processes form a positive-feedback loop. As a result of high work rate in the patch areas, the local frictional strength decreases rapidly with subsequent slip. Since most of the macroscopic shear load is sustained by the patch areas, the macroscopic frictional strength will be dominated by the rapid reduction of frictional strength in the patch areas. We note that the decrease in the frictional strength decelerates the gouge production, which causes a negative feedback. Therefore, these two opposite feedback effects compete during fault slips and so the macroscopic frictional behaviour depends on the competition. We also note that the positive-feedback effect enhances spatial stress heterogeneity as the fault slips even if the initial heterogeneity is small before the slip. In fact, deviation of the stress on the fault before the experiment was small in the present study (see section 3 of the Supplementary Information). To quantitatively verify this idea, we numerically simulated a slip-evolved friction for both the patch and the off-patch areas, as well as the macroscopic friction. Calculations were done according to the flow chart shown in Supplementary Fig. 10. In this computation, we assumed that the local friction coefficients on both the patch and off-patch areas follow the work-rate-dependent friction law estimated from the centimetre-scale rotary shear experiments (Fig. 2). We also assume that the local friction does not immediately attain the target friction coefficient determined by the work rate, but gradually approaches it until the breakdown work is equal to the fracture energy (E ), which is estimated from the centimetre-scale rotary shear experiments. Therefore, the friction on the patch and off-patch areas (μ ) is written as μ  = F (μ , σ , V, s, E ), where σ is the local normal stress, V is the slip rate, and s is the slip distance. The asterisk denotes either ‘p’ or ‘op’, for the patch or off-patch area, respectively. The expression for F can be found in section 6 of the Supplementary Information (as for the below-mentioned F , F and F ). The normal stresses on the patch and off-patch areas (σ ) depend on the thickness of the gouge layer as well as the area ratio: σ  = F (σ, R , ΔT), where σ is the macroscopic normal stress, R is the fraction of the patch areas over the entire fault area, and ΔT is the difference between the thicknesses of the gouge layer in the patch and off-patch areas. We estimated R using the areas occupied with the gouge (which were obtained from the image analyses), and assumed that R was constant during the experiments. The thickness of the gouge layer (T ) is a function of the input frictional work and the wear rate (R ) as follows: T  = F (μ , σ , s, R ). R was evaluated from the mass of the collected gouge materials per unit input work. The variables were updated at each step and the macroscopic friction (μ ) was calculated from the equation μ  = F (μ , μ , σ , σ , σ, R ). The detailed procedures of the simulation, including how the related parameters were determined, are described in section 6 of the Supplementary Information. The simulations successfully reproduced a concentration of normal stress and work rate on the patch, and also a weakening in the macroscopic friction with the slip that fits well with the selected experimental data (Fig. 4). By conducting simulations under various conditions, which are the same as those for the present biaxial experiments, we confirmed that the work-rate dependence of the simulated friction was fairly consistent with that of the biaxial experiments (Fig. 2). From these results, we also confirmed that the slip-evolved spatial heterogeneity caused the scale-dependent rock friction. Rock samples collected from fault zones are frequently investigated in the laboratory in an attempt to understand the frictional behaviour of natural faults21, 22. However, the present study suggests that the estimated frictional properties might not properly represent the behaviour of larger, natural faults because spatial heterogeneity, which is common or even dominant in nature, can greatly affect the macroscopic friction. The amount of gouge production on the patch area was a key process in the present study, but frictional melting may be dominant under natural conditions. Even in such a melt-dominant situation, a similar mechanism to that presented here will work, as shown by an earlier investigation13. We also note that the normal stress concentration can cause not only the decrease in dynamic friction but also an unstable fault slip by reducing the critical nucleation length L . If we assume a friction law in which frictional force is roughly proportional to normal force—such as the rate- and state-dependent friction law—then L will be inversely proportional to the normal stress23, 24. If the area of stress concentration is large enough relative to the diminished L and the frictional properties are uniform there, unstable slip will start to occur in that area. Small foreshocks during slow slip before a main shock have been observed in laboratory experiments and on natural faults18, 25, 26, 27, 28. These foreshocks should be related to the spatial heterogeneity, as observed in the present study. We conclude that spatial stress heterogeneity can facilitate the occurrence of an earthquake. In the present large-scale experiments, we found that the rock friction is scale-dependent, and we believe that spatial stress heterogeneity and the resulting localized high work rate cause this dependence. Although some earlier work13, 29, 30 qualitatively suggested the idea that local frictional phenomena can control the macroscopic friction, here we have quantitatively confirmed it using metre-scale experiments. If large rocks lose their frictional strength at a small work rate, a natural fault may rapidly weaken at an unexpectedly early stage during slip. More studies on large-scale friction experiments, increasing normal stress for example, are needed to explore scale dependence further and to understand better the frictional properties of natural faults.

Kajiwara K.,NIED | Enokida R.,Kyoto University | Nakashima M.,Kyoto University
Journal of Structural and Construction Engineering | Year: 2012

Many control techniques are used for shaking table tests to reproduce the target wave on the table. These controllers are basically designed in frequency domain by referencing a transfer function of a system to be controlled and the controllers are greatly influenced by the linearity of the system. If the system has a severe nonlinearity, these controllers have difficulty following the nonlinearity and the accuracy of the control decrease largely. To control the systems having severe nonlinearities accurately, control methods in the time domain may be more adequate than conventional methods in the frequency domain. This paper proposes a nonlinear control method in the time domain improving Minimal Control Synthesis (MCS) and it does not need any transfer functions or dynamic properties of plant to construct the controller, although original MCS is based on the reference model of the system to be controlled. In this paper, the efficiency of the proposed method is verified numerically and experimentally comparing with the conventional method such as H∞ method.

Mikoshiba T.,NIED | Minowa C.,NIED | Chiba T.,Sanmai cho 161 20
American Society of Mechanical Engineers, Pressure Vessels and Piping Division (Publication) PVP | Year: 2010

Five storied pagoda is built of wood and is known as a high earthquake proof structure. The pagoda is composed of the frame and the center column. The center column is structurally independent of the frame structure. Many researchers have mainly been interested in its dynamic characteristics under the earthquake. They have focused on the acceleration response of the pagoda. This time, we observed not only its dynamic characteristics under seismic load but also under strong wind load. We focused on the displacement response of the pagoda. By integrating the acceleration data twice, the displacement of the structure was obtained. Under low level earthquake, the pagoda swayed about the central axis of the structure. On the other hand, the pagoda was statically deformed by the wind pressure and swayed in the deformed central axis of the structure. The dynamic behavior of the pagoda in the wind is distinctly different from that of the earthquake. Copyright © 2009 by ASME.

Konishi C.,OYO Corporation | Ishizawa T.,NIED | Danjo T.,NIED | Sakai N.,NIED
Near Surface Geoscience 2015 - 21st European Meeting of Environmental and Engineering Geophysics | Year: 2015

We conduct the rainfall experiment using a large scale rainfall simulator and an artificial embankment consists of sand and silt in order to evaluate S-wave velocity monitoring for a prediction of a slope failure. The intensity of the rainfall is controlled to maintain 15 to 200 mm/hour for a certain time frame. S-wave velocity cross sections are acquired by MASW survey before, during, and after the controlled rainfall. The obtained cross sections show little change in the S-wave velocity and that is considered to reflect water filtration process in subsurface. The difference is not so significant, but it is confirmed by both waveforms and the dispersion curves calculated from common shot gathers at a fixed shot point. The dispersion curves are included various higher modes and that makes it difficult to evaluate the result; however, the complex higher mode would be useful to derive more information in the future. Compared to the resistivity monitoring, the S-wave velocity monitoring is not so sensitive to the water content; instead, it is relatively easy to be obtained by surface wave method. Therefore, it will be an effective monitoring tool to assess vulnerability of slopes and mitigate damage by natural disasters. © (2015) by the European Association of Geoscientists & Engineers (EAGE).

Loading NIED collaborators
Loading NIED collaborators