International Journal of Geomechanics

This study investigates stress responses associated with earth pressure balance (EPB) shield tunneling in dry granular ground. A numerical model using the discrete-element method (DEM) was employed to simulate both the advancement of EPB shield and the discharge of muck from the chamber. Some critical ground stress variations associated with tunneling were thus effectively captured. Numerical results show that the responses of the principal stress in dry granular ground, namely changes of magnitude and rotation angle, significantly depend on both the discharge rate of muck and the distance from tunnel face. The average value and distribution features of stress at the tunnel face are also subjected to the discharge rates of muck in the shield chamber. To be specific, in the case without muck discharge out of the chamber, the average stress at the tunnel face first increases with increases in advancement distance and subsequently tends toward a stable value. But for the case of discharging muck with an upper limit rate, the stress first decreases rapidly to a minimum value and subsequently increases to a stable value. Further, a significant stress release area is also found in the proximity of the tunnel face, and the changes in the contact forces among soil particles are far less than those in the case without muck discharge.

An exponential equation is introduced to predict the nonlinear variation of shear strength with matric suction for unsaturated soils. The proposed equation involves three constant parameters, two of which are effective shear strength parameters (i.e., ϕ′ and c′). The third parameter is the maximum capillary cohesion, ${c″}_{\text{max}}$, which is the maximum possible increase in shear strength due to matric suction. A procedure for the determination of ${c″}_{\text{max}}$ from the soil-water characteristic curve (SWCC) is devised. The proposed equation is validated through a series of constant-suction consolidated drained triaxial tests conducted on specimens reconstituted by isotropic consolidation from the slurry state. In addition, the validity of the equation is investigated by applying it to the test results of five other soils that were available in the literature for the low-suction range (i.e., up to 1,500 kPa). A comparative study on the prediction of shear strength was carried out between the proposed equation and six other shear strength equations found in the literature. The results show that the proposed equation provides reliable predictions of the shear strength of unsaturated soils when the shear strength converges to an asymptotic value at the residual water content.

This article outlines the long-term dynamic behavior of the monopile-supported offshore wind turbine (OWT) in clay. A three-dimensional (3D) finite-element (FE) model was developed that uses viscoelastic material constitutive models for soils in conjunction with stiffness-degradation functions to examine the long-term behavior of monopile-supported OWTs subjected to transient loading. The proposed numerical model was validated by experimental results and recorded data from real offshore wind turbines. Effects of amplitudes of wind and wave loads and their frequency, monopile length and diameter, and rotor-nacelle assemble (RNA) mass on the long-term dynamic behavior of a 5-MW OWT due to different load cycles were studied. The results show that the effect of wind load governs the design and that monopile diameter has a significant role in the long-term behavior of the structure.

The increase of significant hydrological events, such as heavy rainfall and flooding, in recent years has adversely impacted the performance of many structures, particularly the foundations that transfer the superstructure load to the ground. The shear strength and settlement of soils and foundations are all influenced by the degree of saturation of the soil, which varies with hydrological parameters, such as rainfall, and subsurface parameters, such as the water table depth. Therefore, in addition to the traditional geotechnical parameters, the hydrological parameters must be incorporated in the design to obtain the optimum design and to understand the impacts of hydrological parameters on the bearing capacity and settlement of shallow foundation for a particular location. In this study, a novel procedure, which considers the site-specific historical rainfall and water table data in the shallow foundation design, is presented with sample applications to two sites in the US. The degree of saturation of partially saturated soil within the influence zone of the foundation was modeled using the one-dimensional Richards’ equation considering infiltration rate and water table location as the top and bottom boundary conditions, respectively. The historical rainfall data and water table locations for the two study areas of Victorville, California, and Levelland, Texas, were obtained from the USGS and National Climatic Data Center (NCDC) data repositories. The results from the Victorville site indicated a near 230% increase in the ultimate bearing capacity and a corresponding 87% decrease in the elastic settlement from those calculated assuming the fully saturated condition. On the other hand, there was only a small increase in bearing capacity at the Levelland site with a corresponding decrease in settlement of 49%. The results clearly indicated that the performance of shallow foundation can be better predicted if the hydrological parameters are incorporated into the current design procedure.

The onset of strain localization is indicated by the weak-discontinuity bifurcation condition—namely, the determinant of the acoustic tensor must be zero. In this study, the analytical solution of the acoustic tensors in plane-strain condition was derived for both the Cauchy-Boltzmann continuum and the micropolar continuum. A typical elastoplastic constitutive model based on the Drucker-Prager criterion was implemented in Abaqus by utilizing its user-defined material (UMAT) and user-defined element (UEL) options. The validity of the localization condition was verified and compared for the Cauchy-Boltzmann continuum and the micropolar continuum through the plane-strain numerical experiments. Results showed that the determinant of the acoustic tensor in the micropolar continuum tended to a constant value when the localization occurred rather than tending to zero as in the Cauchy-Boltzmann continuum. This indicates that the micropolar continuum can prevent the ill-posed solution when bifurcation occurs; in contrast, the onset of localization in the micropolar continuum is totally the same as that in the Cauchy-Boltzmann continuum.

The significant characteristics of the non-Darcian flow of groundwater seepage in a low-permeability rock mass have been reported, but there have been few studies on the analytical solutions for deep-buried circular tunnels comprising non-Darcian flow. In this study, based on the theory of non-Darcian flow and well pumping, an analytical solution was derived for the seepage field of rock masses surrounding a deep-buried circular tunnel. With consideration of seepage force due to non-Darcian flow as a volume pressure added to the stress field of the rock masses, an elastoplastic analytical solution was derived on the basis of the unified yield criterion. The analytical solutions were validated by comparing with those for the cases of Darcian flow and disregard for seepage, and the influences of various parameters on the plastic zone radius, radial displacement, and support characteristic curve were investigated. The results showed that under the condition of non-Darcian flow, the seepage forces had a slower increasing trend toward the tunnels. The plastic zone radius and radial displacement of the rock masses with non-Darcian flow remained between the upper limits in the condition of Darcian flow and the lower limits of disregard for seepage. With an increase in the head loss between the internal and external sections of a tunnel and a decrease in the united yield criterion parameter and the exponential parameter at low gradients, the plastic zone radius and radial displacement at the tunnel walls tended to gradually increase. As the hydraulic head loss between the internal and external sections of a tunnel increased, the influences of united yield criterion parameter, exponential parameter, and supporting pressure on the plastic zone radius and radial displacement were more significant. The analytical solution presented in this study can be used to design a tunnel excavated in a low-permeability rock mass below a water table.

The present paper is about the application of the disturbed-state concept (DSC) model and a mesh-free method called the natural-element method (NEM) for analyzing the dynamic response of pile foundations. In comparison to the finite-element method (FEM), mesh-free methods use higher-order approximate functions to find unknowns. According to the existing stress level, the DSC model divides the behavior of the materials into two states: a relatively intact (RI) state and a fully adjusted (FA) state. The behaviors of RI and FA states were characterized by the hierarchical single surface (HISS) plasticity model as the continuous yield surface and the critical-state concept, respectively. The natural-element procedure and DSC model were used to predict the field behavior of two instrumented piles embedded in saturated clay and dry sand subjected to cyclic axial loading. The predictions showed an appropriate correlation with field measurements and an improvement for the analysis results of the FEM with the DSC model involving softening behavior.

The key issue of microseismic (MS) monitoring technology is to detect MS events and accurately locate the MS source. On the basis of the traditional short-term average/long-term average (STA/LTA) algorithm, a new characteristic function has been proposed that can reflect the signal energy and dispersion. The results reveal that the improved method can accurately pick up the first arrival of MS events. A new monitoring network of the four-cell-square-array in plane algorithm is proposed, which can overcome the ill-conditioned problem of the objective equation. Taking into account the actual project, we designed the monitoring arrangement scheme for the surrounding rock around the underground cavern. The improved MS monitoring scheme can envelop the front of the cave and the dome and floor areas of the working face.

The resilient modulus has been used to characterize the stress–strain nonlinear behavior of subgrade soils and is the primary property needed for pavement design and analysis. The degrees of stress and moisture content have a significant impact on the resilient modulus of compacted cohesive soils. Due to the hot, humid climate in South China, the moisture content in embankments will increase gradually from the optimum moisture content (OMC)—the design water content for embankments—to the equilibrium moisture content, which relates to soil properties and the surrounding climate. In this process, the resilient modulus of embankments will obviously decrease. In order to predict the resilient modulus of typical compacted cohesive soils in South China, repeated triaxial tests were carried out in this study. The soil matric suction was measured by the pressure plate test and the soil–water characteristic curve (SWCC) was described using the Van Genuchten model with a relatively high coefficient of determination. Then, the effect of the moisture content, degree of compaction, and stress state including the deviator stress, confining pressure, and octahedral shear stress to the resilient modulus were analyzed. In addition, a logarithmic function was utilized to build the relationship between the matric suction and resilient modulus. Subsequently, a new resilient modulus estimation model of compacted soils in South China, which took the bulk stress, octahedral shear, and matric suction as the model variables, was developed and verified using the data of different cohesive soils from different studies. The results show that the new model matches their data well, and the coefficients of determination are high, which indicates that this new model is reasonable and widely applicable. Finally, the correlations between the physical parameters of soil samples, such as the liquid limit, dry density, the maximum dry density, plasticity index, percentage passing through a 0.075-mm sieve, and the regression coefficients of the new model, were established. The resilient modulus can be predicted much more easily with these physical parameters of compacted cohesive soils rather than conducting triaxial tests.

A multiphase, thermohydro-mechanical model of unsaturated soils is presented to assess the risk of desiccation of a geosynthetic clay liner (GCL) in a composite lining system (i.e., a high-density polyethylene geomembrane over a geosynthetic clay liner) under conditions encountered in brine or solar ponds, where temperatures can reach up to 95°C and the geosynthetic clay liner is subject to low effective stress. The model considers void ratio and temperature-dependent soil–water characteristic curves (SWCCs) for needle-punched GCLs. The model was validated against results from experimental column studies.

Pseudo-static analysis is one of the most common and simplest approaches used to assess the seismic stability of reinforced-soil retaining walls. Despite the importance of the equivalent seismic coefficient as the key parameter of the pseudo-static approach, no specified criterion has yet been provided for the selection of this coefficient. Therefore, in this study, an attempt was made to assess the horizontal seismic coefficient (k_h) in three different types of reinforced soil structure based on the physical and analytical modeling. For this purpose, first, the values of yielding acceleration (a_y), the axial force distribution along the reinforcements at the failure stage and the geometry of slip surfaces were determined in different conditions by the shaking table test. Then, by using the limit equilibrium horizontal slice method (HSM) and based on the results obtained from shaking table tests, the physical models were simulated to estimate the equivalent seismic coefficients needed to establish slip surfaces similar to those observed in the shaking table tests. Finally, according to the results obtained, the upper bound of the horizontal seismic coefficient for the steel-strip reinforced-soil wall (SSW), soil-nailed wall (SNW), and mechanically stabilized earth/soil nail hybrid retaining wall (HRW) were introduced as a function of peak ground acceleration (PGA) and was compared with existing design codes.

The present paper is framed to study analytically the propagation characteristics of the G-type wave through a functionally graded transversely isotropic substrate beneath a stratum with the consideration of couple stress. The couple stress theory concerned with size-dependent continuum mechanics, with additional parameter-characteristic length, accounts for the internal microstructure of materials, which provides deeper insight into the analysis of wave propagation aspects. Elastodynamic equations are laid down for the two media (stratum and substrate), and the frequency equation is derived by enforcing admissible boundary conditions. The solution process involves separation of variable technique and the method of infinite determinants. The frequency equation is reduced and found to fairly match with the classical Love-wave equation. The expression of group velocity of the G-type wave is also deduced in closed form. Numerical simulation and graphic illustration of the phase velocity as well as group velocity profiles of G-type waves are performed taking into account three different rocks in the functionally graded transversely isotropic substrate, i.e., granite (igneous rock), sandstone (sedimentary rock), and marble (metamorphic rock). The substantial effects of characteristic length associated with couple stress stratum, anisotropy, and functional gradedness on phase velocity is revealed and also traced out graphically. Comparative analysis of phase velocity and group velocity profiles of G-type waves for the three distinct previously mentioned rocks is one of the finest features of this study.

This paper employs a general and direct method to determine the time-dependent stress–strain behavior of soil using a Prony series parameter fitting. The rheological stress–strain data obtained from the relaxation test of soil are used to evaluate the Prony series parameters employing the method of nonlinear curve fitting. The optimized extraction of Prony series parameters for four different Indian soils is presented, which depicts the viscoelastic behavior of different soils considered in the study. Later, finite-element models are developed to simulate and validate the time-dependent constitutive relationship of the soil under both static and dynamic loading conditions. Using such time-dependent constitutive relationships of the soil, stress relaxation, creep, and steady-state dynamic response of the footing-soil system are explored through some example problems.

To mitigate earthquake-related damage to buildings, a simple alternative method to conventional base isolation techniques is to provide a geo-base isolation (GBI) system, composed of a scrap tire–sand mixture, between the base of the building foundation and the supporting soil medium. The GBI system should possess adequate dynamic stiffness and damping properties, as well as enough shear strength to resist both static and seismic loads. This study focused on the use of geogrid reinforcement to improve the bearing capacity, settlement, and rotational aspects of a shallow foundation resting on a GBI layer under static loading. Load tests were carried out on a model footing resting on GBI layer with and without geogrid reinforcement in a sand-bed tank setup. Finite-element-based numerical modeling of the footing on the GBI system with geogrid was also carried out, and the computed results were compared with those measured from the experiments. Parametric studies were carried out using the developed finite-element model to arrive at an optimum thickness of the GBI layer, number of geogrid layers, depth of placement of first geogrid, and length of geogrids. The results from the study indicate that the bearing capacity of the GBI layer can be increased up to three times by providing double-layered geogrid reinforcements with a substantial reduction in the settlement.

This paper presents a detailed investigation carried out for a helical pile installed in cohesionless soil by the displacement method. Laboratory experiments were conducted on models to study the various factors influencing the axial capacities of helical piles. The investigation showed that relative density of soil plays a significant role in the response of axial capacity of piles to geometrical variations in helical blade diameter and shaft diameter. Presence of a helical blade and its geometry (D_h) were found to influence the axial capacity significantly. The effect of method of installation on axial capacity of helical piles was investigated, and helical piles installed by the displacement method were found to show an improvement of axial capacity on the order of 1.1–1.4 times in compression and 1.8–3.0 times in tension from experimental observations. To augment the investigation, analyses were carried out using three-dimensional finite-element software (Plaxis 3D AE) on the numerical model calibrated using the results from the experimental investigation. Cavity expansion by volumetric strain was found to capture the effect of installation of helical piles by displacement method in cohesionless soil. The results of the parametric study accommodating the effect of installation by displacement method, considering different soil densities and geometric features of helical piles, were used to propose a set of semiempirical relations for predicting axial tensile and compressive capacities. A set of reduction factors were also proposed to accommodate the difference in end-bearing response of helical piles compared to a prismatic pile of identical toe diameter.

A closed-form solution for the undrained cavity expansion problem in an anisotropic soil mass is investigated based on the spatially mobilized plane (SMP) model in this study. Stress and strain around the cavity are subdivided into elastic and plastic regions. The small-strain theory in the elastic region and the large-strain theory in the plastic region are assumed in the closed-form solution, respectively. The closed-form solution is derived by introducing the stress anisotropy factor into the traditional solution in the undrained cavity expansion condition. Stress and strain in the elastic and plastic regions, plastic radius, effective stress, and limited expanding pressure are obtained considering the stress anisotropy and undrained condition. The validity of the presented solution is verified by comparing the prediction with the previous cases. Finally, parametric analysis is performed to investigate the effect of stress anisotropy on the plastic radius, effective stress, and strain in the plastic region.

By developing a hierarchical modeling method, using discrete element analysis software PFC3D, a discrete-element method (DEM) model can be built and solved more efficiently. In this method, smaller particles are used for the concerned part, medium-sized particles are used in the transition zone, and larger particles are used for the part that is close to the model boundary. By adjusting the micromechanical parameters of the particles through simulated triaxial tests, the macroscopic mechanical properties of the soils constituted by different-sized particles are kept consistent. This method can be applied to study the soil–structure interaction problems with the DEM method by saving computation cost. The proposed method is verified by a two-dimensional (2D) simulation of a laboratory pile load test. This method is then further applied to investigate the pile–soil–cap interaction problem focusing on the pile cap effects. A geohazard-related impact case is also analyzed using this method.

This study presents a novel three-dimensional (3D) analytical solution to the Richards equation, which takes as a basis a linear unsaturated model that was previously proposed. Specifically, the partial differential equation governing the transient, unsaturated flow phenomenon could be solved when using specific constitutive hydraulic functions that linearize the problem. The new 3D analytical solution could also be simplified to two-dimensional (2D) and one-dimensional (1D) analytical solutions, which make possible the evaluation of water flow using the constraints relevant for field and experimental settings. Two general cases of transient moisture movement are simulated using the new analytical solutions. The first case involves a wetting process, in which the flow within the soil mass is triggered by the initial presence of a specific region within the domain that had been subjected to an increased volumetric water content (e.g., because of precipitation or irrigation). In this case, water flows under unsaturated conditions from the region of increased moisture to the surrounding soil mass. The second case involves the recovery of an unconfined soil mass (e.g., an aquifer) within which a limited region had been subjected to a decreased volumetric water content (e.g., because of a localized drying process). In this case, water flow occurs from the soil mass into the region characterized by an initially low moisture content. The solutions presented in the study can be implemented to address a broad range of applications, providing insight into the complex phenomenon of soil wetting and drying. In particular, the solutions highlight the relative impact of the advective and diffusive components of unsaturated flow processes in multidimensional transient problems.

An energy-absorbing cable designated as constant resistance large deformation (CRLD) has been recently developed and applied to slopes and underground excavations. Under a static pull loading, the CRLD cable can exhibit a large deformation and a constant resistance with a significant radial expansion. The cable mainly consists of a cone, steel strands, and a thick-walled pipe, which is different from the thin-walled expanding oil-well cases. To analyze the strength-deformation mechanism, a laboratory static pull test was developed. The radial deformation of the cable and the evolution of the drawing force were obtained. An analytical model for the thick-walled cylinder was developed and applied to analyze the mechanical behavior of the CRLD cable in an elastic–plastic framework. The analytical model was verified by experimental results, and consequently, a sensitivity analysis was carried out. The relationship between the constant resistance and the geometrical and mechanical parameters was then revealed. Furthermore, a numerical simulation was established, and the corresponding parameters were calibrated in line with the experimental results. The numerical results were compared with the experimental and analytical ones, and excellent agreement was found. An additional sensitivity study was conducted in relation to the metal yield limit. Both models proved to be reliable as assistive tools to design and improve CRLD cables in the future.

Soil nailing is a technique commonly used as a temporary or permanent earth-retention system in soft soils. Habitually, the design of a soil nailing focuses on its performance at failure and computing a safety factor, thus neglecting ground deformations. In this study, an analysis and a comparison of the convenience of the use of the limit-equilibrium method and the FEM for designing a soil nailing were conducted. The assessment considered both the suitability of an easy and fast design process and the necessity to take into account such issues as ground deformations to avoid problematic consequences that can arise during the execution phase and service life. For performing the analyses, a numerical study of the Amherst wall, a full-scale soil-nailed wall built to be an experimental test in the last years of the twentieth century, was carried out. A two-step process for designing soil-nailed walls is proposed. The first step involves the use of limit-equilibrium methods to define the main parameters. The second step deals with the development of a finite-element model to consider ground deformations and determine nail forces. An approach based on the use the Mohr-Coulomb model for simulating materials more similar to granular soils and the hardening soil model for simulating materials more similar to cohesive soils is also presented as an answer for the numerical modeling of soil-nailed walls in ground situations where the soil is neither purely cohesive nor purely granular.

Because of the extreme terrain limitations and heavy traffic in congested urban areas, the supporting structures for deep excavations often undergo an asymmetric loading condition at two sides. This article reports the results of an investigation based on detailed numerical modeling of the supporting scheme and mechanical performance of a deep excavation of a metro station having different elevations at two sides. The two proposed types of design schemes for support and excavation were discussed and analyzed to evaluate the suitability of using regular or irregular supporting structures. The mechanical performance of the adopted supporting scheme was then evaluated using three-dimensional numerical analysis and verified via field observations. The results show that the irregular supporting structure functions satisfactorily when subjected to asymmetric loading, and it can also realize semicovered excavation in narrow and congested downtown areas to effectively solve the problem of traffic. The solution presented in this study can provide valuable references for the design of urban deep excavations with an asymmetric side elevation.

A hyperbolic model-based normalized shear modulus reduction (G/G_max) formulation for cement treated clay is presented. This three-parameter normalized shear modulus reduction model can be constructed using maximum shear modulus and isotropic compression experiments. The model assumes that cemented clay follows the failure pattern of a modified structured Cam-clay model. The effects of cement content and confining pressures on G/G_max are modeled and validated with experimental results from resonant column and cyclic triaxial testing. In addition, the damping ratio is calculated based on the Masing rule; however, for cement treated clays, this rule overestimates the damping ratio at all ranges of shear strain. In this study, correction factors are established to propose a reliable damping ratio model. The newly proposed G/G_max and damping formulations provide reasonable estimates that match well with the experimental results. These formulations can be used in the seismic response analysis of cement treated ground.

Because of the existence of wall-soil friction behind retaining wall, the transfer direction of the principal stress will be inevitably changed. Therefore, the arching effect of soil mass is considered and the circle principal stress trajectory is simplified to a tilt straight line. The differential thin-layer unit is divided along the direction of this approximate tilt linear principal stress trajectory. The major principal stress on the tilt linear principal stress trajectory is regarded as being linearly distributed with depth. Considering the limit equilibrium conditions of the layered soil, a new analytical method for determining the active earth pressure on the inclined retaining wall under translational mode in cohesionless soil mass is proposed. Compared with the measured values, the errors of the maximum active earth pressure obtained using the proposed method are less than 6%; thus, this method achieves satisfactory results. The parameter study indicated that the inclination of the retaining wall back has a significant influence on the failure angle of slip surface, the normal active earth pressure, and the height of application of the resultant force.

Due to their low strength, high compressibility, and collapsibility, loess deposits are susceptible to large deformation and landslides during tunneling. An objective, scientific approach to stability assessment is urgently required. To reveal the sensitivity of loess's stability to its physical and mechanical properties and further establish a stability assessment model, the boreholes and monitored settlements of 30 tunnel sections were collected from the tunnel project in Yan'an, China, and analyzed by rough set theory. The results show that the stability of loess deposits during tunneling is more dependent on the strength parameter ϕ (35.295%) than its c (26.471%), and dry density (20.588%) contributes the most to the stability in physical properties, followed by water content (11.764%) and void ratio (5.882%). Then, with the physical and mechanical parameters as the index system, a stability assessment model based on extension theory was established and finally applied to the practical project for verification. This study helps in getting an insightful understanding of loess's stability and provides a reference for future studies and tunnel construction through loess deposits.

The problem of active earth pressure has historically been investigated assuming dry or saturated backfills, ignoring the fact that soils are usually unsaturated in nature. This note proposes a kinematic framework for the estimation of active earth pressure induced by unsaturated backfills with different water levels. Several formulas consisting of linear and nonlinear forms are used to represent the apparent cohesion due to matric suction. The seepage effect is included in this analysis with the water level to be located at any depth below the crest of the retained soil masses. The impact of cracks is also considered as part of the failure mechanism. An explicit function of the resultant active earth pressure is derived using the principle of work rate balance. The results show that the magnitude of unsaturated active earth pressure is influenced by the soil properties and wall geometry as well as by the models used to estimate the apparent cohesion due to matric suction.