International Journal of Geomechanics

In this study, a nonlinear one-dimensional consolidation equation in terms of excess pore pressure is developed by considering the effects of variable permeability and non-Darcian flow condition. Uncoupled analysis for the determination of mobilization of shaft resistance is performed using a nonlinear load-transfer method. Finite-difference technique is used to predict the dissipation of excess pore pressure and settlement of piles. The predicted results are compared with reported numerical and experimental results and are within an error of ±7%. The consolidation behavior of Cordemais clay (exhibiting non-Darcian flow) in terms of dissipation of excess pore pressure and the subsidence of soil surface with time is studied. The development of downdrag and the distribution of shear stress around the shaft of end-bearing and floating piles are also investigated. The dissipation of excess pore pressure is about 75 and 40% for double drainage and single drainage conditions, respectively, after 25 years of consolidation. The total downdrag is about 85 and 115 mm for end-bearing and floating piles, respectively, after 25 years of consolidation.

Matric suction tends to exert important effects on the stability of geosynthetic-reinforced soil slopes; however, it is usually neglected in the previously published literature. To quantitatively evaluate the increment in soil cohesions due to the existence of soil suction in unsaturated soils, five nonlinear estimation formulas relating unsaturated shear strength to soil-water characteristic curve (SWCC) models were adopted in this study. Based on the upper-bound theorem of limit analysis, this study investigated the stability of a three-dimensional geosynthetic-reinforced slope in unsaturated soils. The influences of three different reinforcement distribution patterns, reinforcement strength, and the nonlinearity of unsaturated shear strength under SWCCs were considered. The results showed that SWCC models with various unsaturated properties can lead to highly divergent performance of the slope. The applicability of different nonlinear shear strength equations is discussed. It was found that the factor of safety (FS) increased almost linearly as the reinforcement strength increased. Among the three different reinforcement distribution patterns, downwardly strong triangular distribution was the most effective means to enhance the safety of the three-dimensional geosynthetic-reinforced slope.

The seismic displacement estimation of soil slopes was generally focused on the plane strain assumption with a linear Mohr-Coulomb failure criterion. In engineering, slope stability often becomes more of a three-dimensional (3D) issue when the reinforcement effect of antislide piles is considered. In this study, an approach was proposed for seismic displacement prediction of 3D slopes reinforced by piles with nonlinear failure criterion based on the upper bound theorem of limit analysis. To depict practical strength characteristics of soil mass, the nonlinear failure criterion was introduced by means of the generalized tangent technique. The formula for resistant force acting on piles was derived in the light of plastic mechanics. The rates of external work and internal energy dissipation were deduced, in which variation of the reinforcement effect of piles along the sliding surface was considered. The analytical expression for yield acceleration coefficient was presented, and the seismic displacement was then calculated based on the Newmark method. Compared with existing research, the validity of obtained solutions was shown. Some examples were then discussed, with the soils following different failure criteria. It was found that stability assessment of slopes on the basis of a nonlinear failure criterion is more critical in the most cases. Parametric analysis was conducted and the yield acceleration was found to be more sensitive to the initial cohesion.

In general, prefabricated vertical drains (PVDs) used in practice are usually manufactured in band shapes. To analyze its consolidation in cylindrical coordinate system, the band-shaped drain was usually converted into an equivalent circular drain surrounded by soil forming a cylindrical unit cell. However, when the band-shaped drain is converted into a circular drain, either its perimeter or its cross-sectional area or even both of them will change. This will lead to a problem that the water exchange between the soil and the drain is consequently changed, because the water quantity flowing into the band-shaped drain along the horizontal direction is controlled by its perimeter, whereas the water flowing out of it along the vertical direction is controlled by its cross-sectional area. Therefore, the effect of the actual shape of the band-shaped drain could not be modeled precisely based on the equivalent circular drain as usual. In this context, this paper attempts to solve this problem by converting the band-shaped drain into an equivalent annular drain with the same outer perimeter and cross-sectional area before and after conversion. Based on the new equivalent annular drain, the analytical solutions are further obtained for four loading schemes, including instantaneously loading, ramp loading, multistage instantaneously loading, and multistage ramp loading. Then the effect of the actual shape of the band-shaped drain is investigated by comparing the results predicted by the present method to the methods with an equivalent circular drain reported in the literature. The results show that the average degree of consolidation predicted by the present method closely matches the result based on equal cross-sectional area, which indicates that the cross-sectional area of the band-shaped drain, not its perimeter, is the key for estimation of the quantity of flow through the drain. Finally, the present method is validated by several laboratory and field tests about the soil improvement technique with a band-shaped PVD. The predicted settlements by the present method and other methods with circular drains are compared to the measured settlement. Good agreements could be observed between the results by the present method and the measured data.

Most classical approaches for evaluating the stability of soil slopes with tension cracks are performed under dry and/or saturated conditions. The suction-induced effect of unsaturated soils is generally neglected in assessing slope stability. This article presents an analytical framework for assessing the stability of unsaturated soil slopes with tension cracks under one-dimensional (1D) steady-infiltration conditions using the limit-analysis method. The boundary-value problem was formulated based on the log-spiral failure mechanism by taking the effects of suction stress, effective unit weight of soil, and tension cracks into account. A simplified method was developed to calculate the external work rate of unsaturated soil weight. Detailed investigations of the effects of suction, Poisson’s ratio, steady infiltration, and water in tension cracks on the slope stability are presented.

In this paper, the antislip stability assessment of a gravity retaining wall is analyzed by considering the spatial variability of soils based on the theory of random fields. The effects of spatial correlation lengths of the soil/wall base friction coefficient and cohesion on the antislip safety factor of a gravity retaining wall were investigated. The numerical results indicated that the effects of spatial correlation length of the soil/wall base friction coefficient on the probability of failure (P_f) were greater than those of cohesion on the probability of slip failure. Moreover, when comparing different levels of failure probability, the antislip factor of safety corresponded to different constants when values of normalized spatial correlation length were relatively small. In other words, the antislip factor of safety increases with a decreasing failure probability, whereas the antislip factor of safety is essentially equal to 1 when the values of normalized spatial correlation length are relatively large. In addition, the numerical results obtained using the proposed method were in good agreement with those obtained using Monte Carlo simulation.

The failure criterion of rock material has been used in geotechnical engineering to assess the failure behavior of rocks under complicated stress conditions. In the present study, a new failure criterion is proposed; it takes into consideration the influence of intermediate principal stress for rock materials. The general characteristics of failure surface were analyzed, showing that the failure envelopes are convex and smooth everywhere except at the triaxial compression and triaxial tension conditions. A new shape function is presented to approach the failure criterion in the deviatoric plane, and the smoothness and convexity of the criterion is discussed. The new failure criterion is compared with some classic criteria by cross-sectional failure envelopes in the π plane and meridian failure envelopes in the meridian plane, and the advantages of the new three-dimensional (3D) failure criterion are presented. The results show that the new criterion not only inherits the strength characteristic of the Hoek-Brown failure criterion at the triaxial compression condition, but it also predicts failure stress better than other 3D failure criteria for rocks under complicated stress states.

This paper presents a numerical study on capacity envelopes of strip foundations placed on the top and the face of two typical soil slopes at different offset distances and subjected to earthquake effects using the pseudostatic method. The capacity is estimated using nonlinear two-dimensional (2D) finite-element limit analysis (FELA). Modified swipe and probe analyses are performed to develop vertical force-moment (V-M) and vertical force-shear force (V-H) capacity envelopes. Characteristic features of these capacity envelopes and critical failure modes of foundations on slopes are identified and compared with the foundations on flat ground. Relative influence of the soil and structure inertia on the capacity envelope of the foundation is also explored. The critical failure mode of a foundation on slope, subjected to gravity and seismic action, depends on the effective column height of the structure. A comparison of the capacity envelopes of typical foundations with the corresponding RC columns indicates that the foundation design methods of the current building codes cannot avoid premature failure of foundations on slopes, prior to columns.

This study presents a full-scale test of high-speed railway embankments to investigate the performance of cement-fly ash-gravel (CFG) pile-supported embankments over completely decomposed granite (CDG) soil foundations. To compare the applicability of different numerical models to the performance of CFG pile-supported embankments over CDG soils, a numerical study was conducted to analyze the embankments in terms of settlements, horizontal displacements, and load transfer from soils to piles using the finite-element software PLAXIS 2D. Furthermore, the numerical results were verified by the field measured results. First, the effectiveness of use of an equivalent area (EA) method, interface elements, and constitutive models is studied. The results show that the CFG pile-supported embankments modeled by the EA method combining with interface elements is adaptive to analyze the behavior of the improved embankments. In this case, the modified Cam-clay (MCC) model yields much better results than the Mohr-Coulomb (MC) model. Afterward, parametric analysis was conducted to study the influences of pile length and area replacement ratio on the deformability of CFG pile-supported embankment. The influences of these factors are compared and rated in terms of their degree of importance.

The discrete element method (DEM) is applied to investigate instability of frictional particulate assemblies sheared under the triaxial constant shear drained (CSD) condition. The instability in loose samples occurs in the contractive regime of the behavior prior to the complete mobilization of the critical state friction angle. In contrast, instability of dense assemblies occurs once the peak stress ratio is attained. For the loose assemblies, certain states for the onset of instability may be found by means of the intersection of the effective stress path with the instability line obtained from constant volume tests, the second-order work, nullification of volumetric strain rate, and the abrupt rise in the axial strain rate. However, only the second-order work and the sudden rise in the axial strain rate criteria can be applied to examine the instability of dense granular samples. Using the DEM data, evolutions of coordination number and contact fabric tensor in both loose and dense assemblies are investigated. Also, as a complement, a modified plasticity model based on the anisotropic critical state theory is presented that can effectively simulate the instability of loose and dense samples sheared under CSD stress paths using a single set of the model parameters.

A great significance should be given to layered slope stability, because the actual slope is often heterogeneous with a composition of multilayer soils due to natural formation and artificial filling. Previous research has theoretically investigated the limit-equilibrium (LE) stability of layered slopes using the single slip surface (SSS) with a circular curve, which may be inconsistent with engineering facts. Hence, this work established the failure mechanism of a composite slip surface (CSS) for a layered slope. The CSS was adopted with a circular curve and noncircular curve. Similar to the SSS, the CSS is also a smooth curve. To simplify the analysis process of slope stability and avoid the influence of the difference of material in the interior of the sliding body on the slope stability, a new LE stress method was proposed on the basis of the global mechanical-equilibrium conditions of the whole sliding body without the divided slice. Compared to the traditional LE methods, the LE stress method had a high calculation speed by avoiding the iterative calculation of slope factor of safety (FOS). Then, both the failed and stable slope examples were listed, and the limit-analysis method (LAM), the finite-element (FE) LAM, and the traditional LE Morgenstern-Price (M-P) method were adopted to analyze the slope stability. Thus, by comparison and analysis of the results in various examples, the feasibility of the present study was verified. Furthermore, the stability charts of a horizontally layered subgrade slope with two soil layers were drawn to facilitate the safety design of the slope. Moreover, the study demonstrated that the failure mechanism of SSS with a noncircular curve or the CSS with a circular curve is suggested to analyze the layered slope stability.

An appropriate constitutive model under cyclic shear loading must be able to characterize the behavior both at small and large strains correctly and consider the effects of degradation phenomena and the load history at the soil response. In this study, the limitations associated with the constitutive models at large and small strain under cyclic shear load, traditionally used in soil dynamics, are discussed. A new formulation based on the mathematical theory of hysteresis is presented. The concept of the normalized generating curve is introduced, and different zones are reached as the degradation progresses. The proposed formulation depends on the load history through the generating curve that evolves with the degradation produced in the soil structure and through the accumulation of the maximum and minimum amplitudes occurring in the load history and producing the model response; therefore, the soil response depends on this sequence. The stress–strain laws obtained by the performance of simple shear tests on cohesive soil samples are reproduced with precision.

Currently, the strength parameters are reduced by a uniform coefficient for the traditional strength-reduction technique in the slope stability analysis. However, observations proved that the contributions of the strength parameters on slope stability are different from each other when the slope reaches the limit equilibrium (LE) state, which means that the strength parameters should be reduced with their own reduction coefficient. Thus, based on the assumptions of stresses on slip surfaces and the overall mechanical equilibrium conditions of the slope sliding body, this work establishes the LE solution for slope stability under the double-strength reduction (DSR) technique. In the LE stress-based analysis, the loop iteration strategy is given to solve the LE stability of slope with the nonlinear strength criterion. Moreover, a new method (i.e., the square root method) is proposed to calculate the slope comprehensive factor of safety (FOS) for evaluating the slope stability with application of the DSR technique. Thus, by comparing and analyzing some slope examples, the accuracy of the present method is verified. Furthermore, from the obtained stability charts of slope under the DSR technique, the study shows that if the characteristic stability number is less than 0.04 for clay soil, then the internal friction angle would have a larger contribution to the slope stability than cohesion; whereas if the characteristic stability number is more than 1.00, then the larger contributor on the slope stability would be the cohesion.

The step-tapered pile foundation is a relatively new structure. For this reason, current literature on step-tapered piles is still not systematic or comprehensive. Thus, it is of the utmost importance to carry out tests and theoretical research on the deformation and load-bearing characteristics of step-tapered piles. The research presented in this paper analyzed the deformation and load-bearing characteristics of step-tapered piles under lateral load using physical model tests and theoretical analysis. A total of four model piles were adapted for the test. The tensile and compressive strengths of the piles were measured, and corresponding load and deflection curves were obtained. Meanwhile, a computational program was developed to simulate and model variable cross-section piles. The internal force and displacement curves of piles under lateral stress at different positions of variable cross-section and stress boundary conditions at the pile top were theoretically calculated using this program. The results from the model test and theoretical analysis were compared and showed good consistency. The results indicated that a step-tapered pile is highly effective at resisting lateral loads. It is most effective when the enlarged pile section is between 33.3 and 46.7% of the total pile length.

Pile foundations are inevitably subjected to combined vertical-torsional (V-T) loadings in offshore engineering contexts. This article proposes an analytical method for calculating the V-T response of a single pile in two-layered nonhomogeneous soil. First, the results predicted by this new method are validated against experimental data and found to be in good agreement. The behavior of a single pile under combined V-T loads is then subjected to a parametric study, exploring a range of dimensionless parameters related to the thickness of the top layer, the embedment depth of the foundation (or pile length), the relative pile–soil stiffness, soil profiles, and the level of vertical or torsional load mobilization. This reveals two interesting features of the behavior of a pile under combined V-T loads: (1) the loading path (loading sequence) has an effect on load response and deformation; (2) the length–diameter ratio, soil profile parameters, and relative pile–soil stiffness have a significant effect on the behavior of pile under combined V-T loads, but the top layer thickness has a minimal effect.

Predicting the lateral displacement in the diaphragm wall is of significance for the design of foundation pits. This study investigated the behavior of diaphragm walls with internal bracing under eccentric loads. The wall and the internal brace were idealized as an elastic flat plate with an elastic bar. To acquire an analytical solution, governing differential equations for lateral displacements were developed based on energy equilibrium. Closed-form solutions to lateral displacements, bending moment, and shear force were obtained by means of the Ritz method. Validation showed that the predictions were in a good agreement with observations and numerical results. Parametric studies indicated that the penetration depth, the flexural rigidity of the diaphragm wall, and the rigidity of internal brace have significant effects on the lateral response of the diaphragm wall.

The as-laid embedment is a key design element of deep-water pipelines, as it significantly affects the longitudinal and transverse pipe–soil interaction and thermal insulation. Due to the small-amplitude oscillations driven by the sea state as well as lay vessel motions, the as-laid embedment is significantly greater than the predictions derived from static penetration analyses. This paper presents a numerical study into the cyclic embedment behavior of deep-water pipelines in soft clay using a recently developed sequential limit analysis (SLA) technique. It directly models the evolution of model geometry and material properties through a consecutive series of small displacement analyses and incorporates the effect of strain softening and strain rate on soil strength. A rigid plane-strain pipe section is pushed-in-place and then subjected to small-amplitude cyclic lateral displacements, with an accumulated lateral pipe displacement of up to 60 diameters. Careful comparisons between the numerical results with published centrifuge data are provided to demonstrate the robustness of SLA, in terms of pipe invert trajectory and soil resistance, followed by an extensive parametric study investigating the effect of strain rate, strain softening, oscillation amplitude, and loading history. More detailed aspects of the cyclic loading behavior are discussed with reference to soil failure mechanisms and bearing capacity yield envelopes. A strong correlation between accumulated lateral pipe displacement and pipe embedment is identified, regardless of the varying oscillation amplitudes, and soil buoyancy is found to play the most significant role in determining the final pipe embedment. The present study aims to assist the development of a macro-element plasticity pipe–soil model that can be incorporated into simplified three-dimensional (3D) pipeline–seabed models, as well as to promote the application of SLA to model complex large-displacement pipe–soil interaction problems.

This study presents an analytical solution for predicting the vertical settlement behavior of geosynthetic-encased stone columns (GESCs) and GESC-supported embankments, taking into account the radial bulging deformation of GESCs. The analysis is performed based on the deformation pattern of a GESC-reinforced foundation. The solution is obtained by enforcing compatibility between the settlements of the column and the soil for each element of the GESC-reinforced foundation. This method has been verified via a comparison with data from the literature. Parametric studies were conducted to investigate the influence of encasement stiffness, embankment fill height, and dilatancy angle on the behaviors of the GESC. The results indicate that the theoretical approaches proposed in this study are suitable for predicting the deformation of GESC-supported embankments. In GESC-reinforced foundation design, it is safer to rely on this predicted settlement when the dilatancy angle ψ_c is also considered.

The vector-sum method (VSM) is advanced by considering the vector characteristics. As a cost of this advantage, a predefined global sliding direction is required, which has been a key issue for the application of the VSM. Although the VSM has been steadily refined since its initial version in 2008, limited progress on this issue has been achieved, and the global sliding direction is still prespecified by assumption. In the aim of solving this issue, this article proposes a rigorous analytical solution based on the principle of potential energy minimization. In addition, the VSM is improved by comparing the resisting moment with the driving moment at the moment center, which can be determined by the shape of the slip surface. Finally, this method is verified to be feasible by three classical earthen slopes with different soil composition. Compared with the solutions found by the rigorous Morgenstern-Price method with the half-sine function as the interslice force function, the calculated results demonstrate that the proposed method can accurately estimate slope stability.

This paper presents a discrete-element method (DEM) study of the scaling effect of strain localization in dense sand. We used disc particles and clump particles to generate specimens for biaxial compression tests, and we adopted a flexible boundary. To investigate the scaling effect, we used a fixed size of specimen composed of different size particles. Numerical simulation reveals that the obtained stress difference and volume strain at a certain strain decrease with R (ratio of specimen width and mean particle size d₅₀). As long as the ratio R is larger than 30, the stress difference, volume strain, peak friction angle, and maximum dilatancy angle approach stable values. The influence of R on the shear strength and volumetric strain of the specimen made of disc particles is greater than the one made of clump particles. With the increase in R, the coordination number and average and maximum contact forces decrease gradually, and the shear band width in dense sand gradually decreases and converges to a constant. The shear band inclination does not reveal obvious variation with R, and the obtained results are close to Roscoe’s formula. When R is large enough (i.e., R ≥ 40), the obtained shear band width keeps a constant, and the influence of particle size on shear band disappears. With the decrease in d₅₀, the ratio of shear band width with d₅₀ gradually decreases; it reaches a stable value when R is large enough.

The underdip toppling failure (UTF) mechanism in rock slopes with a set of discontinuities is analyzed in this study. Various perspectives from the most relevant authors are examined to understand the behavior of this mechanism. The UTF mechanism in rock slopes is explained using field observations from a case study of a mica schist slope validated through a retrospective analysis. An engineering approach is presented to assess the quantitative and qualitative influences of the main parameters on the UTF mechanism based on a numerical method that uses continuous FEM code with Phase 2. The results show that the level of the water table (WT) is the most determinant parameter among the others considered, namely, the discontinuity strength, slope height, state of the natural stresses (K₀), and discontinuity spacing. Finally, recommendations are provided to better understand the parameters that are more influential on the calculation of the factor of safety (FoS) in UTF scenarios.

It is widely acknowledged that the occurrence of seismic force and cracks will increase the active earth pressure acting on a retaining wall. In this work, a kinematic approach for the evaluation of seismic active earth pressure was established, and a rotational mechanism consisting of a logarithmic spiral curve and a vertical crack was used to calculate the rates of external work and internal energy dissipation. Although numerous cracks exist in cohesive backfill, this study only considered the most critical vertical crack that led to the maximum active earth pressure. Different crack types, including pre-existing cracks and cracks forming contemporaneously with the mechanism, are discussed in the paper. According to the quasi-static method, the concept of seismic coefficient was adopted to account for earthquake effects on the active earth pressure. In the domain of soil plasticity, explicit expressions of seismic earth thrust for different cracks types were derived. A comparison analysis was carried out to verify the effectiveness of the present calculations. The influence of different parameters on the seismic earth pressure is discussed. It was found that the presence of cracks and the seismic actions have adverse effects on the stability of soil–wall systems, and that the pre-existing cracks always yield a more conservative result than the formation cracks.