Analytical solutions are developed for calculating the harmonic steady-state axial stiffness, damping, and internal forces of pile groups embedded in multi-layered soil. Pile-soil interaction is represented through a dynamic Winkler model, with frequency-dependent spring and dashpot moduli. Pile-to-pile interaction is taken into account analytically by considering the wave field originating along an oscillating ("source") pile and the diffraction of this field by an adjacent ("receiver") pile. In the case of uniform and two-layer soil profiles, closed-form expressions are derived for both pile impedances and dynamic interaction factors between piles. A solution is finally presented for the additional axial forces developed in piles due to pile-to-pile interaction. The predictions of the model compare well with results of earlier studies, while its simplicity offers a versatile alternative to complicated numerical solutions.

View full abstract

Strong dependence of earth pressure coefficient on strain increment ratio was revealed based on triaxial loading tests along different constant strain paths. From this finding, a new methodology was developed for solving earth pressure problems under any boundary strain constraint. Using this method, new earth pressure equations were obtained by extending the formulas of Rankine and Coulomb theories. The equations proposed can be used to determine lateral earth pressures due to normally consolidated cohesionless soil for any lateral deformation between the active and passive states of stress, including the state of rest. Pragmatic charts corresponding to several simple cases were provided for actual design. Simplified methods were also suggested to determine the parameters in the proposed equations and to evaluate the earth pressures for different types of lateral deformation.

View full abstract

An analysis of settlements of cyclically loaded shallow foundations on air dry, non-cohesive compacting subsoil is presented. The starting point of the study was the results of experimental tests performed on a model foundations. It was assumed that the settlements are caused mainly by an oedometric compaction which takes place in a small zone beneath the foundation. The range of the compaction zone is estimated on the basis of some empirical observations. Settlements of the foundation are described within the framework of the simple compaction theory presented in this paper. The compaction law proposed suggests that there exists a common compaction curve which can be recognised as a material characteristic for non-cohesive soils. This idea is supported by the experimental results presented in this paper. This theory was applied to the analysis of some laboratory tests. A comparison of experimental and theoretical results shows that the method proposed may be applied to estimate settlements of cyclically loaded foundations with an accuracy sufficient for engineering practice.

View full abstract

A two-surface model is proposed to describe various volumetric behavior with change in the direction of a stress-controlled path. The model is formulated using the following concept of the isotropic-kinematic hardening. For overconsolidated soils, the nonassociated flow rule is used because of the occurrence of volumetric change, which is beyond the descriptive capabilities of the associated flow rule, during the loading with an abrupt change of stress path. The proposed model consists of nine material parameters that can be determined with a few triaxial tests. The model is verified by comparing model predictions with experimental results. In addition, the shear strength increases for the material that has the characteristic of stress path dependency are evaluated. Experimentally stabilized behavior which contradicts the theoretical condition for material stability by Drucker is discussed.

View full abstract

A time-dependent model for anisotropic cohesive soils is presented, where an existing invisting elasto-plastic bounding surface model is recast into an elastoplastic-viscoplastic format. In addition, in order to accurately simulate the complete creep response of cohesive soils, including creep rupture, a microstructure-based damage law is presented and incorporated into the time-dependent model. The model is numerically implemented and simulations of undrained creep response for both isotropically and anisotropically consolidated soils are provided. The model predictions compared favorably with experimental results, and soil anisotropy has been found to have a significant effect on the undrained creep behavior of anisotropic cohesive soils. The damage law has demonstrated excellent potential for simulating the entire undrained creep response including rupture.

View full abstract

The characterization of bridge-abutment-backfill-pile interaction is an important component in seismic bridge response evaluations. This paper presents a relatively simple and realistic methodology to evaluate the nonlinear translational (longitudinal, and transverse) spring stiffnesses of abutments founded on pile foundation. Many important factors such as the abutment dimensions, nonlinear pile-soil interaction, superstructure loads, and difference in soil behavior under active and passive conditions have been accounted for. Only routinely used soil properties are required in the model. The paper describes a calibration study undertaken to ascertain the predictive capability of the proposed approach using large-scale field tests. A very good agreement exists between the predictions and the large-scale field tests results. Both longitudinal and transverse field tests were utilized in the calibration. This paper also presents the results of a parametric study in which the influence of pile fixity conditions, relative contribution of piles to abutment stiffness, and deck loads were investigated.

View full abstract

An experimental investigation was carried out to study the traces of slip surfaces in geotextile reinforced cohesive soil retaining structures when the slope angle changes from vertical to the slope of 63.4°(1H : 2V) using centrifuge modeling technique. The results indicate that as the length of the reinforcement increases from unreinforced to a length equal to 0.75 of the height, the slip surface moves toward the face of the reinforced structure, regardless of the slope angle. At all times, however, the slip surfaces were behind the locations of failure surfaces of unreinforced models. In addition, as the slope angle decreases from vertical to the slope of 63.4°, the slip surfaces have a tendency to move toward the face of the slope, regardless of the length of reinforcement. The locations and traces of the slip surfaces obtained from physical tests were consistent with those from limit equilibrium analyses.

View full abstract

The mechanical characteristics of an EPS block light weight fill, which consists of assembled EPS (Expanded Poly-Styrol) blocks, was investigated. The mechanism of load propagation and deformation of the fill was examined in two series of laboratory tests on EPS block model fills ; Load Propagation and Deformation Tests were conducted on the EPS block model fills with various internal structures. The test results obtained, revealed that the characteristic mechanical behavior of the fill is a type of discrete media ; the mechanical properties of EPS block fill is very much influenced by its internal structure. The influence of fasteners was also investigated experimentally. The behaviors of the fills observed in the laboratory tests were analyzed by two numerical calculation methods : DEM (Distinct Element Method) and FEM (Finite Element Method). As a result of the comparison of the calculation results with the observed behaviors, the superior applicability of DEM to EPS block fills was shown. It was found that DEM can explain the load propagation properties with regard to the effect of fastener, and simulate the deformation behavior of the fill.

View full abstract

A series of 69 cyclic tests was conducted to study the damping properties of two reconstituted sands and three laboratory-made clays at small cyclic shear strain amplitudes γ_c≈0.001-0.04%. A recently developed constantvolume equivalent-undrained direct simple shear device for small-strain testing was employed. The specific effect of the shape of cyclic strain-time history on the equivalent viscous damping ratio, λ, was investigated. The results show that λ can be significantly affected by the shape of the cyclic strain-time history because of the viscous nature of soil material response and the associated effects of creep and relaxation. As the shape of the cyclic strain-time history is changed from triangular to sinusoidal, and further from trapezoidal to square, λ at small strains may increase by a factor of two or more. The test results show that this effect was largest for clays, smaller for silty sand and negligible for clean sand.

View full abstract

Theoretical predictions for passive earth pressure have been developed by some researchers based on the rigid plasticity theory. However, those predictions have yet to be adequately confirmed by experiments particularly for the case of a large wall oblique angle α and large wall friction angle δ. In this paper, the results of laboratory passive earth pressure model tests are compared with predictions based on the rigid plasticity theory. The purpose of this paper is to examine how far the rigid plasticity theory can be applied to passive earth pressure problems. The test results demonstrated that the observed failure zone was similar to that predicted by the characteristics method based on the rigid plasticity theory. Except for the lowest part of the passive wall, the earth pressure increased linearly with depth. For a small wall friction angle δ, passive earth pressure coefficients K_p were almost equal to the theoretical predictions, whereas in the case of a large angle δ those were smaller than the theoretical ones. Because of this difference, progressive failure is suggested. It was demonstrated in the figures of strain distribution and displacement contours of the sand before the peak load.

View full abstract

Seismic records obtained from a seismometer array located in downtown Tokyo Japan for about ten years were inversely analyzed to estimate the dynamic soil parameters. Due to the illposed nature of the problem, the simple and often used "least square method" does not properly estimate the parameters. The Extended Bayesian Method combined with the Akaike Bayesian Information Criterion was introduced to overcome this difficulty. The results obtained were compared with dynamic triaxial test results obtained at the time of the seismometer installation. The shear moduli agree quite well with the estimated results, however the damping ratios estimated are slightly higher than the ones obtained in the laboratory.

View full abstract

To reduce long-term maintenance costs arising from the deterioration of concrete bridges due to leaky expansion joints, integral bridges have been becoming increasingly popular in many countries. Abandoning expansion joints leads to complex soil-structure interaction problems including those caused by cyclic temperature changes in the deck. These have adverse effects on the performance of integral bridges in terms of deformation mechanisms, lateral earth pressures acting on the abutment wall, bending moment and axial stress in the deck. All these are of great concern to engineers. Centrifuge model tests were conducted on spread-base integral bridge abutments to simulate these temperature effects on the soil-structure interaction. Thermal expansion and contraction of the deck were modelled by imposing controlled cyclic displacements at the top of the abutment wall. Substantial horizontal sliding and the rocking of the abutment due to soil densification and "strain" ratchetting were observed. The measured lateral earth pressure increased with the amplitude of the displacements into the fill (in the passive sense) and the number of cycles, but at a decreasing rate. For the ultimate limit state and the 1 in 120 years return event, the measured lateral earth pressure coefficient was 3.7 and 4.2 for the abutment backfilled with dense and loose sand respectively. The measured bending moments varied fairly linearly with depth, rather than varying as a cubic function with depth as would be expected for a relatively flexible wall subjected to a triangular distribution of lateral pressure. The measured value appeared to reach the design bending capacity of the reinforced concrete wall and base for the 1 in 120 years return event.

View full abstract

Inherent and stress state-induced anisotropy in the elastic deformation properties of granular materials were investigated experimentally. Both axial and lateral principal strains of large square-prismatic and small solid cylindrical specimens were measured locally. Very small strain-amplitude cyclic normal stresses were applied in the vertical and horizontal directions at various isotropic and anisotropic stress states. Elastic deformation properties are inherently anisotropic with the Young's modulus being larger in the vertical direction than in the horizontal direction at isotropic stress states. The Young's modulus is stress state-dependent, becoming more anisotropic as the stress state becomes more anisotropic. The elasticity is of hypo-elastic type. Deformations do not become totally elastic even after the application of a large number of stress cycles of relatively large amplitude along a fixed stress path.

View full abstract

A dense sand layer is usually assumed to be a bearing stratum in foundation design. It is, however, difficult for engineers to judge if a sand layer with limited thickness can be used as a bearing stratum when it is underlain by a thick soft clay deposit. Many important parameters must be considered in this problem, for example, the thickness of the sand, strength of clay and width, shape and embedment of footing. Several methods have been proposed to calculate the bearing capacities of the sand overlying clay, in which limit equilibrium of forces acting on an imaginary sand block between the base of the footing and the sand/clay interface was considered. The bearing capacities calculated based on existing methods are not the same for different assumptions adopted in each method. The shape of the sand block and the forces acting on the surface of the block are the determining factors in the bearing capacity calculation. In this paper, these factors as well as calculated bearing capacities are compared with the results of well-conditioned centrifuge tests (Okamura et al., 1997) to verify the validity of the methods. It has been confirmed that reasonable assumptions in which the variation of the shape of the sand block and the forces related to the parameters are taken into consideration are important to obtain a reasonable prediction. An alternative method is proposed and the bearing capacity charts are presented for easy reference.

View full abstract

This study proposes two methods of stability for earth structures resisting earthquakes based on the shakedown theory and discusses their applicability in practice. One is a stability method for repetitive static loads based on the seismic coefficient concept (Method A) and the other is a stability method which considers the acceleration time history of the earthquake accurately (Method B). The applicability of Method B was evaluated for both sine and irregular acceleration waves. The applicability of the seismic coefficient method (Method A) is discussed in comparison with Method B. The conclusions from this study are as follows : 1) The effects of periods of enforced accelerations on embankment stability are suitably evaluated with Method B. The stability depends not only on the enforced acceleration but also the dynamic properties of the subsoil profile. The resonance of embankment to the enforced acceleration was simulated. 2) The result of Method A was coincident with that of Method B for the long period case.

View full abstract

The paper describes how research into the effective stresses developed on displacement piles has led to a new design procedure for driven piles in clay. The new approach calculates local shaft friction capacity from a failure criterion derived from (i) drained interface shear tests, combined with (ii) expressions which link the radial effective stresses at failure to the initial vertical effective stress (σ'_vo), the apparent overconsolidation ratio (or YSR), the clay's sensitivity (S_t) and the relative depth to the pile tip (h/R^*). Base resistance may be assessed by conventional methods, or estimated from a new relationship involving the local average CPT resistance, q_c. The new method is demonstrated and verified using data from two Japanese case histories involving tests on large instrumented driven pipe-piles.

View full abstract

This paper reports deformation and strength characteristics for a geogrid-reinforced granular soil on a large-scale plane strain compression apparatus. The influence of intensity and types of geogrid reinforcement is specifically discussed. It was found that as more reinforcement layers are used, the strength of the reinforced soil increased. In cases where reinforcement is most intensely used, the reinforcement layers in the samples were eventually subject to breakage due to plane strain compression and the reinforced soil failed. In cases where reinforcement is less intensely used, the reinforced soil reached failure without any sign of reinforcement breakage. It was also found that types of reinforcement have a significant influence on the lateral deformation of the reinforced soil, where stiffer reinforcement provides more constraints on lateral deformation of the reinforced soil. It can be superceded, however, by the influence of reinforcement intensity. The small stain measurement by means of LDT observed that the stiffness of non-reinforced soil begins to decrease more rapidly at a strain level of 0.05∼0.15%, compared to the reinforced soil. The experimental results also indicate that the relationship between the apparent angle of internal friction and the geometry of the structure of the reinforced granular soil seems to exist for the geogrid reinforcement used in the tests reported in this paper.

View full abstract

The author has proposed a reduction coefficient of differential settlements in a soil-structure interaction and has shown that it is very useful for predicting differential settlements. In such a prediction, the soil model has been assumed to be elastic and linear and until now consolidation settlements were evaluated as equivalent elastic settlements. It is very significant, however, to extend the elastic model to a consolidation model, because practical cases are frequently related to consolidation settlements in reclaimed land. This paper, therefore, attempts to formulate the characteristics of consolidation settlements in soil-structure interaction, in order to incorporate the consolidation settlements into the prediction method of differential settlements suggested by the author. Terzaghi's one-dimensional consolidation was interpreted mechanically to be expressed as an infinite linear summation of Kelvin type viscoelastic models. The soil consisting of an infinite linear summation of Viscoelastic Pasternak Models was then proposed. Using this model, the degree of differential consolidation settlements was introduced and the reduction coefficient of differential consolidation settlements was formulated. Some characteristics of differential settlements are discussed. The results indicate that differential consolidation settlements occur faster than the average settlements and the final differential consolidation settlements correspond to equivalent elastic differential settlements.

View full abstract

A method is proposed to utilize vertical walls that surround a buried structure for mitigating damage due to complete liquefaction of the soil both inside and outside the walls. A method of static analysis is presented for the design of the structure and the walls based on two simplified models that are expected to provide upper limits of the forces on and displacements of the structural elements. The first model in which the walls are assumed to be fixed to the structure and are perfectly flexible and inextensible permits the determination of the upper limits of the uplift pressure on the bottom of the structure and the tensile force in the walls. The second model for which a frictionless boundary is assumed between the structure and the walls permits the determination of the upper limits of flexural stresses in the walls and the upward displacement of the structure. The second model is expected to offer a less costly alternative than the first model where the width of the structure and/or the depth to the bottom of the structure are large.

View full abstract

The finite element method has been used to compute the bearing response of shallow foundations penetrating deeply into non-homogeneous soil, for comparison with classical plasticity solutions. The numerical approach uses remeshing and interpolation techniques, together with conventional small strain analysis. The computed bearing capacity has been compared with lower and upper bound plasticity calculations, based on idealised modelling of the soil conditions around the foundation, for both smooth and rough conditions at the foundation-soil interface. It is shown that classical bearing capacity solutions that ignore the strength contribution of the adjacent soil above the plane the plane of the foundation provide good estimates of the bearing capacity of a foundation penetrated from the surface. This is due to the compensating effect of weak soil trapped beneath the foundation. By comparison, foundations that are pre-embedded show appreciably higher bearing capacity. The effect of local soil heave is also shown to be significant in increasing the bearing capacity of foundations penetrated from the soil surface.

View full abstract

Earth structures constructed of clayey soils, which have a tendency to shrink and swell, develop desiccation cracks when subjected to periods of drying and wetting. Over time, the seasonal shrinking and swelling may result in sloughing and shallow slides. One possible solution to this problem is the inclusion of randomly distributed short fibers in the clay. The purpose of the research presented herein was to assess the feasibility of using commercially available short fibers to reduce the development of desiccation cracks in clay. The test results showed that the fibers were effective in reducing the amount of desiccation cracking that occurred in clays subjected to drying. However, when subjected to wet/dry cycles attempting to simulate environmental conditions, the effectiveness of the fibers was not as evident. The inclusion of fibers increased the tensile strength of the clay and provided a ductile behavior that was not present in the samples without fibers.

View full abstract

Several charts have been proposed so far for soil classification using either cone penetration test (CPT) or CPT with pore pressure measurement (CPTU) data. In the present paper, a soil classification method has been proposed based on the fuzzy theoretical method using CPTU data. The proposed method has been applied at different field sites where a soil boring was carried out at an adjacent site in order to obtain data on the soil profile. Reasonable agreement was observed between the boring soil profile and the soil profile estimated from the proposed method. It has also been shown that the proposed method is applicable to reasonably estimate a very thin sand layer overlying a soft clay stratum.

View full abstract