International Journal of Geomechanics

Various testing methods, such as the static triaxial test, cyclic resilient modulus test, repeated-load static triaxial test, repeated-load California Bearing Ratio (CBR) test, and repeated-load oedometer test, were explored to characterize pavement granular materials. Using such methods, an experimental data set of the stress–strain response of a typical granular material was collected. The modulus values as obtained from the cyclic triaxial tests and the various repeated-load tests were compared. The feasibility of predicting the behavior of the granular material using the hardening soil model was analyzed. Hardening soil model parameters were calibrated using simple static triaxial test data. The effectiveness of the model in predicting the material behavior was then verified by predicting the other experimental results using the hardening soil model with a finite-element implementation. The stress–strain behavior as predicted by the model matched well with the experimentally measured stress–strain values; hence the validity of the hardening soil model was ascertained.

This paper presents a simple method to predict the settlement of the composite foundation with sparse prestressed tubular concrete (PTC) capped-piles under embankment. In this study, it is assumed that the distribution of skin frictions on PTC capped-piles is simplified as two force triangles, and a calculation model for predicting the additional stresses in the composite foundation soil is given. By combining Mindlin-Geddes and Boussinesq solutions, the equation of the additional stresses for a single PTC capped-pile foundation is derived considering the influence radius. A case history involving the installation of sparse PTC capped-piles for soil reinforcement under highway embankment is introduced, with a maximum settlement of 428 mm after construction. The predicted settlements of the composite foundation, resulting from the proposed method, are in good agreement with the observed data. Additionally, the additional stress field calculated shows a little change because the pile spacing is eight times larger than the pile diameter. Thus, the proposed method can optimize the design of the foundation soil reinforcement resulting from PTC piles.

A series of oedometer tests was performed on binary sand–bentonite mixtures considering both the effect of the sand mass fraction and the initial water content of the bentonite matrix. The experimental data reveal that the influence of the initial water content of the bentonite matrix on the overall creep behavior of the mixture is negligible. However, the reference time line (corresponding to 24 h of consolidation) is significantly affected by both the initial water content and the sand mass fraction. The local creep parameter of the bentonite matrix is quite close to that of pure bentonite for a mixture with a sand mass fraction of 50%. However, it decreases with an additional increase in the sand mass fraction due to the increasing heterogeneity of the binary mixtures and the formation of clay bridges between adjacent sand inclusions. An equivalent local creep parameter is defined, and a new structure variable is introduced, which could be approximated by the structure variable responsible for the intergranular structure evolution. Finally, a creep model is formulated using mixture theory. The proposed model has five parameters: one structure parameter that incorporates the intergranular structure effect and four that are dependent on the intrinsic behavior of pure bentonite. Only two conventional oedometer tests need to be done for calibrating the parameters. The model prediction is then compared with experimental data, revealing a satisfactory performance of the proposed model.

Resonant column (RC) testing is a widely used laboratory technique to determine the stiffness characteristics of soils under small- to medium-strain perturbations. Solid or hollow cylindrical soil specimens are set into motion in either torsional or longitudinal modes of vibration by an electromagnetic loading system whose frequency is changing until the first-mode resonant condition is reached, and then the shear modulus of soil is back-calculated from the fundamental frequency, the geometry of the specimen, and the end-restraint conditions. However, the outcomes of this test are largely affected by both the driving apparatus used for the specimen vibration and the motion-monitoring instruments lumped into a mass that oscillates with the specimen. This study presents the results pertaining to three-dimensional (3D) finite-differences (FD) simulations of RC tests on soil samples undergoing both torsional and longitudinal modes of vibration. The prime objective of the study was to examine the influence of the driving mass, the geometry of the specimen, the mode of vibration, and the boundary conditions on the RC test results. The numerical results show that the attachment of the instrumentation on the sample is the driving factor contributing to the error in the estimation of the soil dynamic characteristics, and typical equations for the calculation of the shear modulus from the resonant frequency in the longitudinal mode of vibration cannot be directly applied to their torsional-mode counterparts.

In this paper, the spectral element method (SEM) is presented to analyze the dynamic response of a multilayered half medium subjected to a harmonic moving load. In SEM, the spectral stiffness matrix of one-node and two-node elements in the wave-number domain can be derived based on the Fourier transformation and principle of wave superposition. Then, the spectral global matrix of the model can be obtained by assembling the nodded stiffness matrices of all layers. The corresponding solution in the wave-number and frequency domain is further recovered by the Fourier series and fast Fourier transform (FFT) algorithm. At last, several numerical examples were given to validate the accuracy of the proposed method and illustrate the influence of load speed and frequency, as well as the interlayer condition of pavement, on the simulation results.

The preshearing history of soil is a prominent factor governing its mechanical behavior at various loading conditions. An experimental program was developed to study the effects of preshearing at drained compression and extension conditions on the undrained behavior of saturated sand. Triaxial compression and extension tests on both loose and medium-dense sand showed distinctly different behaviors: the response in compression was dominated by stable and dilative behavior, whereas instability occurred in extension with strain softening. The initial inclination of the effective stress path varying with the preloading history is related to the fabric anisotropy of the soils, characterized by their stiffness values at very small strains, whereas at intermediate to large strains, the undrained stiffness tends to decrease in triaxial compression but grows in triaxial extension with the preshearing stress ratio. Moreover, the instability behavior of sand characterized by the reduction of shear strength is governed by the magnitude and direction of the preshearing stress. The stress ratio at which the instability occurs is not unique and can be correlated with the proposed state index accounting for the combined influences of the preshearing stress and relative density of the sand. In contrast, a unique phase-transformation line exists that is independent of the preshearing stress conditions. This is probably due to the gradual diminishing of the stress-induced anisotropy that is created by the static preshearing during the subsequent undrained loading.

This study investigated the influence of large water inflow from the excavation face on segmental lining damage when the earth-pressure balance (EPB) method was used in shield tunneling in a saturated silty sand layer. A refined numerical approach to investigate the damage mechanism and gradual failure mode of segmental lining was established by using the finite-element program Abaqus. The fluid–solid coupling was considered in simulating the tunneling process. The evolutions of permeability and elastic modulus of the grouting material were taken into account. The concrete damaged plasticity (CDP) model was adopted to model the segmental lining damage caused by large water inflow. Parametric studies were carried out to shed light on the mechanism of the structural response. The numerical analysis highlighted the effectiveness of the proposed method to simulate the lining damage. The mechanism of lining damage caused by large water inflow from the excavation face was identified, and it was found that sufficient water flow toward the tunnel compensated for the water loss from the excavation face and reduced the lining damage. It was also found that continuous grouting was an effective countermeasure to alleviate the lining damage.

With the current tendency to gradually increase the contribution of renewable energy to achieve significant proportions of the entire production, the wind farm construction rate is actually high. Because the location of the wind turbines (WTs) is dictated by factors usually independent of the foundation soil conditions, which are mostly based on energy production and consumption efficiency (e.g., average wind speeds, possibility of connecting to existing electrical networks, approval of authorities and local population, etc.), it is not uncommon for the construction sites to be unfavorable in terms of geotechnical demands. In these circumstances, the choice of the optimal foundation system is an important aspect in the design phase of a WT. The aim of this paper is to analyze, using three-dimensional (3D) numerical models and the suitability of currently available foundation solutions (based on a shallow foundation on the natural or improved ground), and compare the overall behavior with solutions based on rigid inclusions (RIs). The parametric study developed was based on a real soil profile, and all the foundation solutions are analyzed for realistic static WT loads. The assessment of the efficiency of each foundation system, as well as the subsequent comparative analysis, was performed in terms of the surface settlement on the foundation soil and the axial force and bending moment on the vertical reinforcements.

Despite its wide use as a tool in foundation design, the piezocone penetration test (CPTU) has been rarely recommended to work for design and analysis of laterally loaded piles. Obtaining lateral response of pile foundations is a complicated engineering problem, especially in nonhomogeneous soils. This paper presents a review of the relationship between the piezocone test and the bearing response of laterally loaded piles and introduces a framework for estimating p-y curves for clays directly using CPTU parameters. To validate this method, full-scale lateral load tests of bored piles with corresponding CPTU tests in Jiangsu soil deposits were conducted and data were compared with the predicted results. In addition, case histories were further considered in detail to study the application of the proposed method for different field conditions. The results predicted by the proposed CPTU-based p-y curve agreed relatively well with the measured results. The proposed method can provide a fast and effective design tool that can be applied to clayey soils with full consideration of soil profiles along the pile-embedded depth.

In this paper, a novel semianalytical meshless method as an efficient and accurate method is introduced for elastoplastic analysis of jointed rock mass. Because the proposed method combines the advantages of scaled boundary FEM (SBFEM) and boundary radial point interpolation method (BRPIM), it is called the scaled boundary radial point interpolation method (SBRPIM). Using the boundary nodes for problem modeling without a fundamental solution can be considered as the common points between the SBRPIM and SBFEM. However, unlike the SBFEM, which uses the FEM shape functions in a circumferential direction, the SBRPIM uses the BRPIM shape functions in this direction. Because the BRPIM shape functions possess the Kronecker delta function property, the boundary conditions can be enforced easily in the SBRPIM. To verify the proposed method, three numerical examples are presented in this paper. Comparing the results of the SBRPIM with those of the finite-difference method (FDM), point interpolation method (PIM), and element-free Galerkin (EFG) method, the efficiency and accuracy of the proposed method for elastic and elastoplastic analysis of the jointed rock masses are confirmed. SBRPIM has a theoretical significance and can be extended in the engineering application.

A probabilistic approach is proposed for face stability analysis of subway tunnels, which begins with establishment of multiple failure mechanisms, including active failure and passive failure. In terms of the complex conditions during subway construction, the buried depth, surcharge load imposed on the ground, and pore water pressure are introduced to account for the extrinsic factors that probably affect the face stability. The probabilistic analysis is performed by consolidating the multiple failure mechanisms into response surface method (RSM). The cohesion and internal friction angle of soils are considered as random variables, and the remaining parameters are considered as nonrandom variables due to their low variability or determinacy for specific projects. The influence of nonrandom variables and random variables on failure probability is investigated in detail. Finally, a reliability-based design chart is provided to evaluate the target supporting pressure against tunnel face according to different coefficients of variation of random variables. In short, by taking into account of all the possible face failures in one performance function, the proposed approach has the virtue of catching the global behaviors of face failure when multiple failure mechanisms occur at tunnel face.

The response of unsaturated soil slopes subjected to raised groundwater levels is generally governed by undrained shear strength. A series of unconsolidated undrained triaxial tests was conducted on reconstructed specimens of Xiashu loess with different water content. This can avoid the complexity involved with the measurement of matric suction. The Duncan-Chang hyperbolic model was modified to approximate the measured stress-strain responses. A correlation between water content and shear-strength parameters was established. The proposed model was then implemented in a finite-difference analysis program, against which two conventional calculations using the standard Mohr-Coulomb model with constant and varying soil properties with depth were compared. The analyses of an unsaturated soil slope subjected to raised groundwater levels demonstrate that soil properties should vary with water content; otherwise, an unsafe evaluation of slope stability will be obtained because the displacements of the slope may be underestimated if the dependency of water content is neglected.

The use of stone columns is a viable, cost-effective, and environmentally friendly ground improvement technique. Columns are made of compacted aggregate and installed in soft soil as reinforcements to increase the bearing capacity and to reduce foundation settlement. In the literature, the methods available to estimate the bearing capacity of soil reinforced with stone columns assumed bulging and general shear as the only modes of failure. However, ground reinforced with stone columns may also fail by local or punching shear mechanisms, depending on the soil, columns, and geometry of the system. This paper presents analytical models to estimate the bearing capacity of the foundation on a soft soil reinforced with stone columns that are subjected to local or punching shear failure mechanisms. The models were based on a limit equilibrium technique and the composite soil properties. The proposed theories were validated with experimental and analytical data available in the literature. Design charts are presented for practical purposes.

To simulate the onset and growth of strong discontinuities (cracks or slip planes) and the failure mechanism in soil slopes with reasonable accuracy, a program was developed based on the existing extended finite-element method (XFEM). This program consists of a new analytical algorithm for strong-discontinuity initiation and propagation in soils considering the stress concentration and redistribution around the discontinuity tip and an integration scheme for the XFEM enriched discontinuities. Two different types of interfacial contact constitutive models are used to describe the interface behavior of the cracks or slip planes. Analyses of two slope failure examples are presented to demonstrate the reliability of this program and the reference values for engineering measurements.

In the framework of limit analysis, most previous studies of nonhomogeneous slopes have only involved the spatial variation of cohesion, whereas cases involving the spatially varying friction angle have rarely been studied. The main aim of the study presented in this paper was to propose an effective approach for analyzing slopes with friction angle spatial variability. According to the upper-bound theorem of limit analysis, a new two-dimensional (2D) failure mechanism using a spatial discretization technique was studied. With this mechanism, the slip surface was composed of a series of straight lines, instead of the continuous curve used in conventional limit analysis. The mechanism-generation procedure for the slip surface is described herein, and the mathematical formulations for the weight work rate and energy dissipation are also presented. After a discussion about accuracy and efficiency, the validation of the proposed approach, conducted previous achievements and numerical simulations, is presented. The good agreement shows that the proposed mechanism is effective. Finally, the analysis of the stability of nonhomogeneous slopes, with friction angles varying linearly with depth, and the investigation into the effects of the slope angles and friction angle distributions are presented.

Modeling approaches to fully capture the behavior of fiber-reinforced clay remain lacking, although the beneficial role of discrete short fibers in reinforcing soils is an established concept. This lack results from tremendous difficulties in characterizing the complex interaction mechanism between fiber and soils. Based on a thorough analysis of the interaction mechanism in terms of stress and deformation pattern, this paper developed a modeling approach to predict the before-failure and failure behaviors of fiber-reinforced clay subjected to triaxial compression. The approach was built on the innovative combination of the superposition method with the energy-based homogenization technique by using two reasonable assumptions about the interaction mechanism: (1) the active fibers work in their elastic domains throughout the entire loading process; and (2) the active fibers’ partial slippage along the interactive course controls the composite’s before-failure behavior, whereas the failure of the composite is governed by fiber pullout. The before-failure behavior was modeled by establishing an elastoplastic stiffness matrix derived from superimposing the incremental stress contributions of the fiber phase and the clay phase described by the modified Cam-clay (MCC) model that is capable of reflecting the nonlinearity. A macroscopic failure criterion that can be used as an alert for the attainment of the failure state was developed based on the equilibrium of the rates of work dissipated on the interactive course and in the clay matrix. Good agreement of the predicted behavior with the observed response in drained triaxial compression tests verified the approach’s ability in accurately reproducing the stress–strain and volumetric behavior of fiber-reinforced clay despite some discrepancies between the predicted and experimental results. The discrepancies direct the further refinement of the proposed modeling approach, which is expected to serve as an effective tool that can be used for deformation and stability analysis of fiber-reinforced geotechnical structures.

This paper reports the density effect on the aging-induced increase in soil stiffness based on discrete-element method (DEM) simulations conducted on dense, medium-dense, and loose samples for the study. Like experimental observations, among the three investigated, the medium-dense sample showed the highest aging rate in terms of the increase in the small-strain shear modulus. This finding was the result of competition between two opposite effects that coexist during aging: contact-force homogenization and unjamming events; both effects were relevant to the packing density. During aging, though a looser sample could undergo a greater homogenization of contact forces to give rise to a higher aging rate, it had a greater risk of experiencing unjamming event. This unjamming was accompanied by a sudden increase in the number of sliding contacts, which destroyed the aging-induced homogenization of the contact forces, and in turn, demolished the aging-induced stiffness increase. On the contrary, although a denser sample was less susceptible to the occurrence of unjamming events, it had less room to enable the contact forces to become more homogeneous during aging. As a result, eventually the medium-dense sample was revealed as having the greatest aging rate.

Borehole breakouts appear in drilling and production operations when rock subjected to in situ stress experience shear failure. By finite-element modeling of borehole breakouts, the authors obtained the stress state close to the borehole wall, the shape of borehole breakouts, and the mechanism of borehole breakouts in terms of initiation, propagation, and stability. It is desirable that this information on borehole breakouts be used to determine in situ stress by inverse analysis. Artificial neural network provides such a tool to establish the relationship between in situ stress and information on borehole breakouts. Determination of in situ stress by inverse analysis consisted of two steps. First, finite-element modeling provided sets of data on in situ stress and borehole breakout measures, which were used to train an artificial neural network that could eventually represent the relationship between the in situ stress and borehole breakout measures. Second, for a given measure of borehole breakouts, the trained artificial neural network was used to predict the corresponding in situ stress. Verification of the modeling of borehole breakouts was conducted, and numerical experiments were carried out. The results showed that the inverse analysis based on finite-element modeling of borehole breakouts and artificial neural network can effectively determine the in situ stress.

Fiber-reinforced calcareous sands manifest unique properties of increased shear strength and particle breakage. These features are of practical importance for some offshore engineering constructions because the strength improvement and efficient dense compaction of soils are both important. This paper presents experimental evaluations of the characteristics of shear strength and particle breakage of fiber-reinforced calcareous sands by direct shear and ring shear (RS) tests with different vertical loading stress, fiber content, and fiber length. In the tests, the mixture of fiber and sands can make the specimen a spatially interlocked and unitary coherent network with efficient stress transmission. In addition, the overall deformation of the sand specimen would increase with the fiber content due to low stiffness of fiber elements. Thus, in direct shear tests, the secant elastic modulus decreased, while the shear strength increased with the fiber content. The contribution of fiber to the shear strength of the sand specimen came mainly from the friction and tension forces exerted when they were deformed. These two forces could mobilize the additional shear resistance of sands and thus increase the overall shear strength of the sample. In the RS tests, the breakage intensity of calcareous sands increased with the vertical loading stress, fiber content, and fiber length. At low fiber content and length, the interparticle contacts and interlocking effects influenced the shear strength and particle breakage significantly, while at higher fiber content, the role of fiber friction and tension forces became dominant.

The coupled discrete-element method–finite difference method (DEM–FDM) numerical modeling scheme was proposed to simulate a single stone column in soft clay according to the Universal Distinct Element Code (UDEC). The discrete granular materials that form a stone column can be better simulated by the DEM model, which extends beyond the capability of continuum approaches. In the proposed DEM–FDM coupled numerical model, the stone column was represented by discrete rigid blocks according to the Voronoi tessellation, and the surrounding clay was represented as continuous Mohr-Coulomb material. The settlement, bulging deformation, and failure process involving complicated interactions between the stone column and the surrounding soil were captured well by the proposed numerical model. The accuracy of the proposed model was verified using experimental model test results. The consistency between the numerical simulation and the model test results indicated good potential of the proposed modeling scheme for studying further the load deformation and the failure behaviors of soft ground reinforced by stone columns. The limitations of the proposed numerical model are also discussed in this paper.

The nonlinear behaviors of rock–lining and the lining–lining interfaces significantly influence the seismic response of tunnels with composite linings. Therefore, a closed-form analytical solution was developed for the dynamic response of tunnels with composite linings subjected to SH (horizontally polarized shear)-waves, and the two interfaces were modeled by the Coulomb-slip model. A case study was then conducted to investigate the effects of the two interfaces on the seismic response of tunnels. The results show that the imperfections of the two interfaces lead to decreases in the stresses of the composite linings and an increase in the deformation of the final lining. However, the two interfaces play opposite roles in influencing the deformation of the initial lining. As the shear stiffness of the shotcrete–cast concrete interface increases, the peak stresses in the composite linings increase rapidly at first and then tend to be stable. The effect of the shear strength of the interface, however, depends on the stress distribution in the interface. The mechanical parameters of the interfaces should be carefully determined for the seismic design of tunnels with composite linings.

Treating the soil within the failure zone behind the wall as the specimen in a simple shear box, and the soil transition from initial to limit states as the shearing process of soil specimen, a geometrical model was proposed to describe the relation between wall displacement and soil shear strain. Then, a unified expression for predicting earth pressure as a function of wall displacement from active to passive states was established based on the proposed geometrical model. The core parameters for predicting the displacement-dependent earth pressure curve, namely, the wall displacements at active and passive states, can be obtained by this unified expression according to the stress–strain curves of backfill soil. Additionally, analyses indicate that the initial earth pressure coefficient, the failure zone size behind the wall, and the stress–strain behavior of backfill play a decisive role in the displacement-dependent earth pressure curves. These parameters are responsible for the large wall displacement differences at the limit states among various types of backfill soil and under different densities. More importantly, a reasonable agreement between the proposed model and the experimental results in the literature is obvious.

The polymer antiseepage wall is a new technology for the reinforcement of earth dams. It has the advantages of being efficient, ultrathin, minimally invasive, and economical, which makes it even more popular and widespread in many applications. However, research on the seismic response of the polymer antiseepage wall is lacking. Because most earth dams are damaged by seismic loads and the antiseepage system is the weakest part, the related dynamic studies are of great significance for this new type of polymer antiseepage wall. In this study, two physical centrifuge models manufactured with prototype materials were used to investigate the seismic response characteristics of a prototype structure. The results of the seismic response characteristics of the earth dam with a polymer antiseepage wall are presented, along with a comparison of the response characteristics for earth dams with a concrete core. The results show that the earth dam with a polymer antiseepage wall had a good capacity for seismic damage resistance.

This paper presents an analytical solution that extends the graphical correction of the consolidation curve, proposed by Terzaghi, to account for linearly increasing load (Terzaghi, K. 1943. Theoretical soil mechanics. New York: Wiley.). Although simple, the proposed procedure can substantially improve settlement prediction. It considers the ratio of the area below the actual loading curve to the corresponding area due to an instantaneous load. As a result, any time-dependent loading can be interpreted. The accuracy of the proposed method was verified by analyzing two case histories of a large embankment constructed on compressible soils. As in Terzaghi’s graphical approach, the method provides approximate results, but it is efficient in reproducing general time-dependent loading and is easily programmed in spreadsheets.

An analytical model is proposed for the evaluation of seismic active thrust on retaining wall resting on sandy soil. Foundation and backfill soil is idealized as Kelvin-Voigt solid. The governing differential equation of vertically propagating shear wave is solved by maintaining the displacement continuity and stress compatibility at the interface between backfill and foundation soil. Nonlinear distribution of accelerations induced in the backfill soil is then used to compute the seismic inertia force. Total seismic active thrust is obtained by solving the force equilibrium equation of the triangular failure wedge. The present method computes the strain-dependent shear-wave velocity and damping ratio of sand from its modulus reduction and damping ratio curves using an iterative scheme. Amplification and deamplification of the input excitations in the foundation soil and backfill soil are compared with results of dynamic centrifuge tests reported in the literature. An increase in the magnitude of seismic active thrust is observed when the retaining wall is situated on sandy soil in place of the rigid stratum, and the percentage increase is 18% in comparison to the latest method available in the literature. The seismic active thrust values for long period input excitations are nearly 13 to 17% higher compared to short-period input excitations. The parametric study reveals that depth of backfill and foundation soil, impedance ratio, amplitude, and frequency of input motion dominate the response of the backfill soil under seismic condition.