A new testing method for estimating the strength parameters c and φ of unsaturated soil is proposed. In this method, the undrained strength is evaluated using the ultimate uplift resistance force of a small buried plate. This testing method is called the plate uplift test. To investigate the applicability of the plate uplift test as a testing method, the results for both undisturbed natural ground and compacted artificial ground using remolded soil are compared with those measured by laboratory shear tests. For artificial ground, the range of cohesion by the plate uplift test was 20∼30% of cohesion measured by the direct shear test. The plate uplift test can also estimate the cohesion of undisturbed natural ground. The plate uplift test is an effective field testing method because the variation in measured cohesion is small and the testing equipment is portable and can be operated easily. As a case study, the plate uplift test was applied as the execution management test in an embankment with a height of 28 meters.

抄録全体を表示

The progressive failure and scale effect of trap-door problems were evaluated, by comparing experimental results with finite element analyses employing the elasto-plastic model including the shear band effect. The experiments were performed in passive and active mode by measuring load-displacement relationships and observing shear band developments using two sizes of apparatus (the trap-door width of the smaller one was 5 cm and of the larger one 10 cm.) The load was measured with soil stress measurement cells embedded in the face of the trap-door. The shear band developments were observed using thin horizontal colored sand layers placed in the sand mass. The distribution of earth pressure by finite element analysis achieved good agreement with experimental results. The localized narrow zone of shear strain by the analysis was approximately identical with the observed shear band in both passive and active modes. The scale effect was recognized by the analysis and the experiment only in passive mode.

抄録全体を表示

A new type of infinite element for groundwater analysis which is capable of including the infinitely distanced constant head boundary conditions from the area of disturbance (e.g. pumping) is proposed by employing a series function type of mapping functions derived based on the asymptotic far field behavior of an aquifer for typical boundary value problems. The proposed infinite element can be classified as a type of FESC (Finite Element with Singular Contraction) by Rajapakse and Karasudhi (1985). A figure to indicate the required size of the near field to obtain accurate drawdown at specified points is also proposed based on some calculations. The infinite element is also successfully applied to a real pumping test problem conducted to examine the performance of a recently developed continuous sheet cut-off wall, which shows excellent agreement with the observation.

抄録全体を表示

An effective method of carrying out a time-domain finite element analysis of footings subjected to dynamic axial load is presented. Infinite elements are used to model radiation effect in the horizontal direction. This allows the vertical boundary to be placed as close to the footing as desired. The restrictions are that the domain outside the radiation boundary must be horizontally layered and behave linearly during the loading. Shape functions for the infinite elements are obtained from frequency-domain analytical mode shapes, but in such a way that they are valid in time domain. The soil deposit is assumed to rest on a fixed horizontal plane. In the study reported herein, the entire domain is assumed to be isotropic and linear visco-elastic. The validity of the method is verified by comparing the results with those obtained by the standard finite element method with a vertical boundary placed sufficiently far from the footing.

抄録全体を表示

In this paper, eight joint geometry modelling schemes are suggested and applied to a set of Stripa mine joint data to build three-dimensional joint networks to a granitic rock mass. These modelling schemes include investigations for statistical homogeneity of the rock mass, corrections for sampling biases, and applications of stereological principles, to estimate 3D joint geometry parameters from 1D or 2D joint geometry parameter values. Results show the possibility of obtaining different estimates for both joint size and joint intensity parameters through these different schemes. This indicates the importance of performing validation studies for developed joint geometry modelling schemes. Validations performed indicated the need to try out different schemes in modelling joint geometry parameters in order to come up with realistic 3D joint geometry modelling schemes which provide good agreement with field data during verification. It is important to realize that different types of joint geometry modelling schemes are needed to model joint networks in different types of rock formations.

抄録全体を表示

In order to examine the capability of the effective stress model proposed by the authors, a finite element analysis was conducted on the field performance of two quay walls during an earthquake. The quay walls are of anchored steel sheet piles. Though the cross sections were similar to each other and the locations were adjacent to each other, one quay wall suffered serious damage while the other did not. The effective stress model used in the analysis consists of a multiple shear mechanism defined in strain space. The model has the capability to represent essential features in the cyclic behavior of sand such as the effects of rotation of principal stress axis directions. For estimating the model parameters, soils were taken from the site for laboratory tests. The record of the earthquake motion was recovered from the site and digitized for the analysis. Results of the finite element analysis are basically consistent with the observed performance of the quay walls; the model demonstrates the potential ability to differentiate between serious large deformations (i.e. damage) from negligibly small deformations (i.e. no damage) in similar types of sheet pile quay walls.

抄録全体を表示

High-quality undisturbed gravel samples, 30 cm in diameter and 60 cm high, are obtained by the in-situ freezing method at four sites. Both the undrained cyclic shear strength and drained shear strength of these samples are measured by using a large scale triaxial test apparatus. The standard penetration test (SPT) and large penetration test (LPT) are conducted at each site to obtain possible correlations between their penetration resistances and the in-situ shear strengths. The data of gravels obtained by other investigators are also reviewed and discussed. The laboratory test results and the review of previous studies lead to the following conclusions. (1) The liquefaction resistance (the shear stress ratio required to cause 2 or 2.5% double amplitude axial strain in 15 load cycles) of the high-quality undisturbed gravel samples shows a good correlation with the value of N_L1(normalized for the effective vertical stress of 1 kgf/cm²) over a range of N_L1 between 20 and 50 in such a way that the liquefaction resistance increases with an increase in N_L1. The correlation appears to be independent of effective vertical stress. (2) The liquefaction resistance of the high-quality undisturbed gravel samples is overestimated particularly for N₁-values larger than 30, if correlations between the liquefaction strength and the blowcount from the conventional SPT obtained for sands are directly applied.

抄録全体を表示

By theoretically studying the behavior of sand through the mechanics of assembly of spheres, a micromechanical background is offered for a constitutive model of a generalized plasticity-multiple mechanism type defined in strain space (Iai et al., 1992a). The study confirms the existence of the hierarchical structure in the stress, concept of which played a fundamental role in deriving the aforementioned model. The basic level of partial stress contribution is associated with the contact forces of a specific direction. This level of partial stress contribution represents a fundamental "element" of stress. The second level of partial stress contribution is the one associated with the virtual plane strain mechanism and obtained by combining the basic level of partial stress contributions. The stress in the three dimensional space, which is nothing but a stress of usual definition, is the highest level and obtained by combining the second level of partial stress contributions. This type of structure is believed to play a crucial role on the cyclic behavior of sand involving the effects of rotation of principal stress axis directions.

抄録全体を表示

A series of plane-strain biaxial compression tests on dry coarse sand is described in which the state of shear-band formation, shear-band orientation, and shear-band growth are investigated. The results show that both the shear strain at shear banding and the shear-band orientation depend on the confining pressure. The experimental results are compared with the theoretical predictions of equilibrium bifurcation theory for a number of local and non-local incremental constitutive equations. Significant deviation between the experimental results and theoretical predictions for local constitutive equations involving the Mohr-Coulomb yield condition is observed at higher confining pressures. Much better agreement is obtained if a yield function possessing a corner is introduced. The analysis of the shear-band growth beyond the onset of localization reveals that simple purely kinematic models cannot describe adequately the observed evolution in the shear-band thickness.

抄録全体を表示

Geotechnical aspects of the ground damage caused by widespread occurrence of liquefaction at the time of the 1900 Luzon, Philippines earthquake are highlighted by examining the geomorphological regime and boring data obtained in the city area of Dagupan. The comparison of old and present locations of the river courses indicates that the liquefaction in the city of Dagupan had taken place in two geomorphologically different areas : namely, the terrain of naturally buried old river channels and the area of point bar deposits. Although impossible to distinguish soil profile characteristics between these two types of terrains, the common feature is that there exists a surface layer composed of unliquefiable silty soils, underlain by sand deposits which are loose enough to provoke liquefaction. As a result of simple analysis of liquefaction, it was shown that the upper part of this sand deposits developed liquefaction, leading to the disastrous consequences in the city of Dagupan. By compiling a number of SPT data and the data of Swedish cone penetration tests, an attempt was made to clarify the relationship between the thickness of unliquefiable surface layer and that of the underlying liquefied layer which can or can not bring about surface manifestation of liquefaction, accompanied by disastrous damage on the ground surface. Lastly, the lateral spreads of the ground near the riverside were analyzed by considering the soil profiles established by the Swedish cone tests. The residual strength determined by back-calculation at three typical cross sections indicated a general coincidence with those estimated previously for other data sets.

抄録全体を表示

Limit loads on finite layers of weak soil over a rigid base are analyzed using the upper bound approach. The soil is considered cohesive, and the Tresca yield condition is used to describe its strength. Strength increasing with depth is considered, and different strengths of the base interface are accounted for. The analysis is performed having the design of embankments in mind, and it makes it possible to account for the arbitrary horizontal component of the load generated by the horizontal thrust in the embankment. Results are compared to the lower bound solutions of particular cases available in the literature. Even though the upper bound approach was used, the solutions presented fall very close to the lower bound solutions. The advantage of the solutions presented in this paper is in easier calculations of bearing capacity, especially when the soil strength increases with depth, and in accounting for the arbitrary horizontal load transmitted from the embankment to the foundation soil. A proposal for arriving at critical heights of both unreinforced and reinforced embankments is given.

抄録全体を表示

A series of model tests has been conducted to study the effect of footing embedment and load eccentricity on the load-settlement and moment-rotation relationships as well as on the bearing capacity. A method was developed for predicting the nonlinear response of an embedded footing, which uses elasticity theory to model soil behavior at low load levels and then incorporates soil nonlinearity in compliance with the experimental behavior. In addition, the effect of embedment on the interaction diagrams between vertical loads and moments that cause failure of the footing is analyzed.

抄録全体を表示

The determination of a curved failure surface of a granular soil can be a time and cost consuming exercise. The fitting of a failure curve to failure points obtained by testing different, but supposedly replicate, samples can also be an error-prone process. A new, but simple, method based on post-failure strain path testing is presented. Micro-processor control techniques are used to control the strain path followed by a sample in a post-failure domain. This new method enables the determination of the failure curve from a one test. The pre-requisite of the success of this method is elimination of platen restraint to a minimal level.

抄録全体を表示

Available experimental data on dynamic shear moduli and damping ratios of various soils including non-plastic sands to highly plastic clays are collected. Those are reanalyzed and brought into simple unified formulas. The unified formulas express the dynamic shear moduli and the damping ratios in terms of maximum dynamic shear modulus, cyclic shear strain amplitude, mean effective confining pressure and soil's plasticity index. Although the availability of experimental data on clay is still limited at this time, the formulas fit those data reasonably well and could be conveniently utilized in dynamic analyses such as seismic ground response and soil-structure interaction problems.

抄録全体を表示

A theoretical backup is offered for the strain space multiple mechanism model of cyclic behavior of sand presented by the author and his colleagues (Iai et al., 1992). The model presented previously was of a plane strain mechanism. By superposing the plane strain mechanisms around axes of arbitrary orientations, the present study demonstrates that the model can be extended into the three dimensional space. The study also shows that the model satisfies the objectivity, which is essential to a realistic constitutive model.

抄録全体を表示

The degree of grain orientation given by the magnitude of vector strength by visual measurements is compared with that of the birefringence ratio by an indirect one. To link both expressions, a concentration degree of distribution is introduced on the assumption that the grain arrangement obeys the circular normal distribution. The longitudinal section illustrating the shear-induced fabrics is put under a polarizing microscope, and quartz grains in it are measured by the trigonometric treatment. The grain orientation brought forth during shearing is evaluated as a random orientation, because grains rotate as the horizontal displacement increases. However, it is also shown that the parallel grain arrangement exists in a certain direction such as the slip line and its conjugate at the top of crack occurring during shearing.

抄録全体を表示