International Journal of Geomechanics

Modeling the element behavior of expansive soil with cyclic variations of its water content in stress and time dimensions is a challenging topic. In this paper, a one-dimensional (1D) model will be proposed to provide an estimation of soil deformation. To simplify the model, the effect of suction in an expansive soil will be considered indirectly in the values of the model parameters, and the model parameters will be determined via current testing methods. This paper aims to describe the behavior of unsaturated soil in a relatively simple and pragmatic manner, meanwhile, the model parameters will be determined via methods that engineers are familiar with. The performance of the model will be demonstrated and verified against experimental data for two categories of soil behavior: (1) time-dependent swelling behavior, and (2) stress-induced compression behavior. The practical approach used in this paper could be useful and eventually help to find solutions for some complicated practical problems where the influence of both categories of soil behavior need to be taken into consideration.

Because analytical soil arching models are fundamental for the design of piled embankments, numerous soil arching models were proposed for this purpose. However, the results obtained from various soil arching models differ considerably, which may be due to the different assumptions caused by an incomplete understanding of soil arching behaviors. In this study, a series of two-dimensional (2D) discrete element method models were conducted to reveal the underlying mechanism for the mobilization of soil arching and its morphological characteristics. Then, an improved analytical soil arching model was developed, taking into account the mobilization state of soil arching, modifications to the soil arching height, and the corresponding load-uniformity coefficient of the stress acting on the subsoil. Finally, the feasibility of the present model was verified by comparing the results with the experimental data. The results showed that both the soil arching height and the load-transfer efficacy depend on the mobilization state of soil arching. The maximum soil arching height was found to be about 0.8 times that of pile clear spacing, in which the load-transfer efficacy reaches a peak. The present model in which the mobilization state of soil arching is taken into account could estimate the deformation and stress state of a piled embankment more accurately.

Soil arching exists in many earth structures, such as pile-supported embankments, tunnels, buried pipes, and retaining walls. These earth structures are sometimes subjected to surface footing loading. Limited studies showed that surface footing loading weakened active soil arching formed in backfill material; however, this effect has not been well considered in the current design. In this study, two-dimensional (2D) trapdoor tests with transparent soil were carried out to investigate the development of active soil arching under static surface footing loading. The particle image velocimetry (PIV) technique was adopted to monitor the deformations of soil particles due to multiple trapdoor movements and surface footing loadings. Soil arches with a triangular expanding pattern and a tower-shaped development pattern were observed in the cases with different fill heights. Under the surface footing load, soil arching first degraded locally (at its crown) and then gradually expanded to both sides of the arch with an increase of load magnitude until full degradation of soil arching globally. The model tests with active soil arching resulted in a larger surface settlement under a footing load than that without soil arching due to the volumetric expansion of the soil during the formation of soil arching. Two analytical methods were compared with the experimental results before local degradation and after global degradation of soil arching.

Landslide is a common and severe geohazard that may pose significant risk to adjacent structures such as buildings and buried pipelines. Identification of the failure slip surface is critical for the assessment and management of landslide risk on adjacent structures since the structures are adversely affected by large deformation of soil when landslides occur and they are not located far away from the failure slip surface and sliding soil mass. Previous studies showed that the location of the failure slip surface may change during the failure process of a slope, but the failure slip surface identified by the commonly used limit equilibrium method (LEM) only corresponds to the initiation of slope failure. In addition, soil properties vary spatially, and such variability may have a significant influence on the location or spatial distribution of the failure slip surface. In this study, a sliding probability is proposed for a soil element within a slope, and the spatial–temporal variation in the sliding probability within a cohesive slope is explored to investigate evolution of the spatial distribution of the failure slip surface during the entire process of a landslide. The investigation is performed using Monte Carlo simulations integrated with smoothed particle hydrodynamics. The soil sliding probability at each location within a slope is estimated at each stage of the failure process and used subsequently in the assessment and management of landslide risk on adjacent structures.

Biocementation of soil using microbial-induced carbonate precipitation has become a new approach for soil treatment. The biocemented sand behaves differently from uncemented sand. One of the major differences is the stress-dilatancy behavior. In this paper, the experimental data obtained from isotropically consolidated drained triaxial tests are presented. The data showed that the biocemented sand exhibited much higher dilatancy than uncemented sand of the same density in drained triaxial tests. The higher the calcium carbonate content, the higher the dilation. Higher dilatancy was also related to higher shear strength, which was mainly due to higher effective cohesion. The variation in effective cohesion with the calcium carbonate content is discussed. The stress-dilatancy relationship of biocemented sand can be interpreted under the framework of Rowe’s stress-dilatancy theory: the biocemented sand behaved like dense sand in terms of the stress-dilatancy relationship.

From the perspective of the mechanical mechanism of slope failure, cohesion c usually plays a different role with friction φ in sliding resistance, which indicates the distinct weights of their reduction factors Fs_c and Fs_φ to the overall safety factor of slope Fs in the double strength reduction method (DSRM). This study primarily introduced the equivalent influence angle of slope θ_e, for which c and φ share identical contributions to the safety factor, and investigated its influencing factors, which turns out that θ_e shows an increasing trend with an increase of slope weight H or slope height γ, but the increasing rate is decreasing. Subsequently, an equivalent influence angle chart was plotted as an effective and advantageous approach to locate the specific value of θ_e for a given slope. On this basis, the shortest reduction path method was utilized to calculate Fs_c and Fs_φ, and the stability contributions of c and φ were quantified by the different weight coefficients of Fs_c and Fs_φ to Fs. The specific weight coefficient could be solved by the correlation between slope angle θ and corresponding θ_e, thereby developing a new definition of Fs. Ultimately, a modified double strength reduction method was achieved, which was compared with other DSRMs and verified by existing slope examples.

Bridging the mechanical response of rockfill materials with laboratory grading curves and field grading curves is of obvious importance in rockfill dam engineering and has been a great challenge for geotechnical researchers for decades. In this study, a novel approach was proposed to consider the scale effect on the volumetric behavior of rockfill materials. The proposed scale effect model was incorporated into an existing volume–stress equation and was validated by both the experimental data in the literature and the one-dimensional compression test results performed by the authors. Furthermore, it was demonstrated that the single scale-related parameter in the model, namely, the Weibull modulus m, obtained experimentally from the particulate scale, can also be applied for granular assemblies. This work provides not only a theoretical framework to understand the scale effect on the volumetric behavior of rockfill materials but also a very promising application in rockfill dam engineering: one can effectively estimate the volumetric behavior of a rockfill material with the field grading curve using simply experimentally obtained volume–stress data of a downscaled material and the single-particle crushing tests data of the composing rockfill particles.

With wind farm development, a new type of foundation, the piled beam–slab foundation, was invented to support wind turbines on land. Although this foundation enables a lower-cost investment, its structure is complex. Wind turbine foundations are usually subject to combined vertical, horizontal, and moment loads. To investigate the coupled effects on the piled beam–slab foundation, scaled model tests were conducted. The results show that the presence of a vertical load increases the horizontal bearing capacity of the piled beam–slab foundation. The axial forces of the piles are mostly caused by the vertical load and moment acting on the foundation. The vertical load carried by the piles increases when the horizontal load acting on the foundation increases or the moment acting on the foundation decreases. Several equations are proposed to calculate the responses of the foundation under a combined load based on the responses of the foundation under a pure horizontal or moment load, which can be calculated by the traditional design method. Finite-element analysis on a prototype piled beam–slab foundation is performed to prove that the proposed equations are reasonable for actual foundations.

Railway ballasted track stiffness is an important indicator to identify supporting condition that ensures that the facility is well designed and functioned. Although many studies have been performed on track stiffness based on experimental tests and finite-element methods, the factors influencing the track stiffness have not been completely confirmed yet, especially the influences from ballast and subgrade layers at the mesoscopic level. To address this research gap, a combination of the discrete element method and the finite difference method model was utilized to study the factors influencing the track stiffness from the particle level. Factors (related to ballast layer properties) are bulk density, thickness, and stiffness, and another factor (related to subgrade properties) is elastic modulus. Additionally, the relationship between the track stiffness and the mechanical behavior of ballast was analyzed. This study quantified the influences of track components on the track stiffness and accordingly proposed how to improve it from the ballast and subgrade layers at the mesoscopic level, which can provide guidance for railway ballasted track design and maintenance.

This research studied seismic source mechanisms of acoustic emission (AE) events generated during the failure of intact and jointed rock samples using a microscale mechanical-seismicity coupled microscale model. During rock failure, forces and displacements around microcracks were measured in the model to determine seismic moment tensor. Interpretation of AE data was carried out by decomposing the moment tensor into a double couple (DC), a compensated linear vector dipole (CLVD), and isotropic component (ISO). In this study, microscale numerical models of Lac du Bonnet (LDB) granite (intact and jointed) and a Berea sandstone sample were built to represent the mechanical behavior of hard and soft rocks. Subsequently, several numerical triaxial tests under different confining pressures were conducted to analyze source mechanisms of AE events during rock failure. Numerical analysis shows the significant contribution of positive non-DC component during the LDB granite rock failure and significant DC component during the Berea rock failure. The source mechanisms of AE move to the negative non-DC component at higher confining for both granite and sandstone samples. Joint failure causes non-DC component toward crack opening at low confining pressure and a non-DC component toward crack closure events under higher confining pressure. Simulation results, presented in this work, show how micromechanical properties of rocks and joints as well as stress conditions give rise to different types of AE source mechanisms during rock failure.

The evaluation of slope stability reliability under the condition of incomplete probability information is one of the most popular issues in geotechnical engineering. The key difficulty in the issue lies in how to effectively establish a joint probability distribution model that conforms to the distribution characteristics of shear strength parameters under the condition of incomplete probability information. In this paper, based on the incomplete probability information of the nonlinear strength parameters of rockfill, the joint distribution model of the nonlinear strength parameters will be established using the Copula function. Combined with a radial basis function network (RBFN), an intelligent response surface method for reliability analysis of rockfill dam slope stability systems will be proposed. Its feasibility and validity will be demonstrated through an example. The influences of the Copula function type and sample number on the reliability of the stability system of rockfill dam slope will be systematically studied based on the constructed intelligent response surface. The results indicate that the Copula function could reasonably describe the correlated, nonnormal distribution characteristics of the nonlinear strength parameters and effectively construct their joint distribution model. The Copula function could provide reliable, statistically significant parameter samples for reliability analysis of the stability system of rockfill dam slopes. The traditional independent normal distribution ignores the correlated, nonnormal distribution characteristics of the nonlinear strength parameters of the rockfill, which result in the smaller failure probability and overestimation on the reliability of the stability system of the dam slope. The Copula function type and the number of samples have a significant impact on the reliability of the stability system of a rockfill dam slope. Therefore, it is recommended that as much test data as possible is obtained for statistical analysis, and the optimal marginal distribution function and Copula function are selected to improve the accuracy of the failure probability analysis of the rockfill dam slope stability system, which is of great engineering significance in the safety evaluation of a dam.

The pile groups were used as a foundation for complex structures experiencing large axial, lateral, and torsional loads. As a result, they undergo vertical, lateral, and eccentric movements. Moreover, there were several studies on piles and pile groups under axial and lateral loads, but the solutions for pile groups subjected to combined axial and torsional loads were largely elusive. Therefore, a novel numerical scheme was presented to capture the nonlinear pile–soil interaction in a flow-controlled geomaterial to make allowance for the yield effects. Based on the numerical scheme, a three-dimensional finite-element analysis was performed on pile groups subjected to combined axial and torsional loads in the flow-controlled geomaterial using a computational program. The flow potential for the yield surface of the flow-controlled geomaterial is a hyperbolic function of stresses in the meridional stress plane and the smooth elliptic function in the deviatoric stress plane. The load–displacement relationship of a large diameter pile (LDP) and pile groups (1,2 and 2,2) was compared with the experimental pile load test and the numerical results reported in the literature. It was observed that the resultant displacement increases significantly with the torsional load for the LDP and pile groups. Similarly, the twist increases significantly with an increase in the axial load. The displacement and twist parameters were classified based on the plastic strain and dilation angle of the geomaterial.

Geogrids with triangular apertures are increasingly being used as a reinforcement element in various geotechnical structures because of their characteristics to offer uniform resistance in all directions. In the present study, the load–displacement characteristics of triaxial geogrid in longitudinal (machine direction, MD) and transverse (cross-machine direction, XMD) directions under pullout conditions were evaluated for various normal stresses. Transparent soil made of fused quartz sand and transparent fluid (mineral oils) was used to quantify the soil and geogrid displacement under pullout conditions through image analysis, so that both the global and local failure modes of the triaxial geogrid can be assessed. The geogrid strain at failure under pullout and multirib tensile test conditions was compared to understand the failure modes of the triaxial geogrid in MD and XMD. The triaxial geogrid in XMD had experienced larger geogrid displacement when compared with MD, and the difference in the displacement is prominent with the increase in the normal stress. For identical normal stress and clamp displacement, the extent of the shear zone was larger for geogrid tested in XMD than MD. The test results also reveal that the triaxial geogrid has failed by pullout under lower normal stresses but has also failed by rupture under relatively higher stresses, irrespective of geogrid orientation. Interestingly, under lower normal stresses, the triaxial geogrid oriented in XMD had a deformed in-plane, while in MD, out-plane geogrid deflection was noted. With the increase in the normal stress, the node rupture was noted when oriented in XMD but failed by rib rupture when oriented in MD.

The stability of a rock joint was often dominated by one or more major asperities along the slip surface with a large bearing capacity to resist instability. In this study, direct shear tests were performed on artificial split rock joints. The shear failure mechanism of major asperities was revealed and a physical model was developed to describe the shear instability of the rock joint, by coupling a strain-softening constitutive model (based on the Weibull distribution) with a friction model (based on the piecewise function). The effects of constant normal load (CNL) and asperities on cutting-tooth behavior and shear instability characteristics of a rock joint (e.g., sudden jump values Δu₁ of shear displacement, elastic energy release ΔU₁, and shear instability proneness K₂) were studied. The results showed that the shear instability of the rock joint was mainly affected by the CNL and major asperities. The physical model fits well with the shear stress–shear displacement curves of the rock joint. The necessary condition of shear instability (which is K₂ ≤ 1) and predicted sudden jump values Δu₁ of shear displacement were derived by using cusp catastrophe theory. The cutting-tooth effect of major asperity becomes more obvious with the increase of CNL. Here, Δu₁, ΔU₁, and K₂ all decrease with the increasing CNL. The proposed physical model and the shear instability analysis may improve the understanding of the unstable shear failure behavior of rock joint both in the laboratory and in nature.

In elastoplastic analyses, a yield function needs convexities and C1 continuities, to meet the requirement of return mapping algorithms. However, many classic criteria in geotechnical engineering have shortages: either numerical issues (nonconvexity and gradient discontinuities) or insufficient physical properties (lack of tensile strength and compressive cap). To overcome these deficiencies, a smoothed classic (SC) yield function is proposed after combining the generalized classic yield function, a robust smoothing technique, and investigations on physical meanings. It can reflect general soil properties, including basic shear strength, tensile strength, compressive cap, and the impact of intermediate principal stress. The yield function is convex and C2 continuous; therefore, it avoids many numerical issues in describing general strength behavior. In addition, it can provide a close approximation to the Mohr–Coulomb criterion to avoid discontinuous gradients. Formulas for the gradient, the Hessian, and the calibration procedure are provided.

The Brazilian splitting test is a broadly adopted testing procedure for characterizing the tensile strength of natural rock or rock-like material. However, the results of the Brazilian test on specimens with naturally existing voids and inclusions are strongly influenced by size effects and boundary conditions, and numerical modeling can assist in explaining and understanding the mechanisms. On the other hand, the potential of utilizing the Brazilian test to characterize inhomogeneous deformation of rock samples with voids and inclusions of dissimilar materials still awaits exploration. In the present study, fracture mechanisms in Brazilian disks with circular voids and filled inclusions are studied by using the phase field model (PFM). The finite element method is adopted to implement the PFM to study the influence of diameter, eccentricity, and quantity of the voids and inclusions on the fracture patterns and stress-strain curves. The phase field simulations can reproduce previous experimental phenomena and furthermore it deepens the understanding of the influence of inclusions and voids on the fracture pattern, overall strength, and deformation behavior of inhomogeneous rock. The findings in the study highlight the potential of characterizing inhomogeneous rock through combining Brazilian tests and numerical modeling.

Driven by the need to enhance the performances of the Winkler model when dealing with large-diameter monopiles, the primary objective of this paper is to propose a new p-y formulation that has the ability to realistically model the stiff behavior of monopiles. At first, works revolving about the determination of the ultimate resistance of piles in sands were examined. It was found that the American Petroleum Institute's ultimate resistance for a shallow depth was found suitable to model the soil reaction near the monopile head. However, all proposed methods for the ultimate resistance at large depths were empirical and significantly overestimate the sand stiffness with depth. Taking advantage of the monopile problem geometry that is axisymmetric subjected to nonaxisymmetric loading, an expression giving the soil reaction against a rigid moving disc was established using the semianalytical finite element approach. By extending the obtained formula to failure using some fundamental notions of basic soil mechanics, the ultimate resistance for depths greater than one diameter was rigorously quantified. A totally new p-y formulation encompassing a new p-y shape and a new initial stiffness was proposed and encoded in an existing Winkler computer program. The results of the new formulation were validated by the study of three case histories from the literature.

Overconsolidated (OC) clay has a strong dilatancy property, which has a significant effect on ultimate strength and deformation. Compared with normal consolidated (NC) clay, OC clay tends to show a higher strength–stress ratio, lower shear shrinkage, and larger dilatancy, as well as strain hardening and softening. A unified hardening (UH) model is a simple and practical model to describe the stress–strain relationship of OC soil. However, the degree of overconsolidation has a direct effect on dilatancy, which is mainly demonstrated by: (1) the degree of overconsolidation is directly related to the phase transformation stress ratio (M_c) that corresponds to the occurrence of dilatancy. The greater the degree of overconsolidation, the smaller the M_c; and (2) the greater the degree of overconsolidation, the smaller the volume shrinkage strain and the larger the volume dilatancy strain. In a UH model, the stress ratio (η) of the phase transformation is a constant and its value is simplified to be equal to the stress ratio of the critical state (M). A UH model cannot be employed to reflect the fact that the M_c varies with the degree of overconsolidation. To overcome the previous problems, the M_c in the dilatancy equation will be expressed as a power function of the overconsolidation stress ratio parameter (R). To reflect the characteristics of large volume shear shrinkage of underconsolidated (UC) soil, the nonassociative flow rule will be adopted, and the shape of the yield surface modified by state parameters is a water drop surface and the plastic potential surface is an elliptic surface. The improved model could better reflect the double influence of the degree of overconsolidation on dilatancy characteristics.

This study investigates the effect of soil–structure interaction (SSI) on seismic behavior of mid- and low-rise residential buildings considering linear and nonlinear behavior of structural members. For this purpose, four- and seven-story buildings representing a majority of existing building stock were used. The outcomes indicate that there are significant variations in displacement demand estimates depending on ground motion records, modeling approaches (linear/nonlinear structural modeling and fixed-base or SSI), and soil types. The findings in this study evidently indicate that linear fixed-base models are highly sensitive to dynamic amplification. Therefore, their use may result in inappropriate displacement demand estimates. If the soil amplification and nonlinear modeling are considered, the differences in the average displacement demand estimates of both the fixed-base and SSI models are negligible. Besides, the displacement demand estimate for stiffer soils is independent of modeling. This study underlines that the linear structural modeling includes SSI or the fixed-base model has to be used with nonlinear structural modeling in order to have acceptable demand estimates for the seismic evaluation of the reinforced concrete residential buildings.

A typical sandy mudstone from the deep-buried roadway of the Kouzidong coal mine was taken as the research object. Based on the micro-mesoscopic experiments combining the acoustic emission (AE) and digital speckle correlation method (DSCM), as well as numerical simulations based on the discrete element method (DEM), both the micro-mesoscopic creep damage evolution and the failure mechanism of sandy mudstone were investigated. The experimental results show that the AE curve of sandy mudstone in the multilevel creep test is consistent with the axial strain curve. Three stages, including the decelerative, isokinetic, and accelerative creep stage, can be identified. The three-dimensional source location of AE events shows that the creep loading leads to more severe, homogeneous, and dispersed microdamage in sandy mudstone and a looser failure mode. It was also found by DSCM techniques that the microscopic deformation field evolution of sandy mudstone in the creep process well corresponds to the three stages of creep failure. A transition of microscopic deformation field from random, via symmetrical, to asymmetrical distribution, can be observed during the creep failure process of sandy mudstone. By introducing the time-dependent stress corrosion theory into the existing parallel-bonded model (PBM) in particle flow code (PFC), the microdamage evolution mechanism of sandy mudstone under creep loading was studied both theoretically and numerically, and the corresponding results revalidate our experimental findings.

Deep mixing using cementitious compounds are an in-situ ground improvement method for improving the strength and compressibility of loose sands. Low-calcium class F fly ash (FA) is a sustainable low-carbon replacement for Portland cement that was utilized for stabilization of loose sand. The unconfined compression strength (UCS) of uniform fine sand and well-graded medium sands, stabilized by alkali-activated FA, were determined and the strengths of optimized blends were compared with the minimum required strength advised by guidelines for cement-treated deep mixed soils. The decementation softening behavior of lightly stabilized zone was investigated under loose and dense conditions for both types of sands with two different types of binder under k₀-loading conditions. Both the stabilized sands, with 10% FA and the optimized alkali activator Na₂SiO₃:NaOH ratio of 70:30 satisfied the 700 kPa UCS requirement for deep mixed soil. The UCS strength has increased up to 1,400 kPa for 15% FA content. The variation of void ratio due to softening decementation of the lightly stabilized sand was in close correlation with changes in shear wave velocity. The well-graded medium sand was found to be less sensitive to decementation due to the higher coordination number of interparticular contact.

Pressure-dispersed compression anchor consists of several anchor units fitted with an end plate. Limited field tests have been conducted to investigate the performance of a pressure-dispersed compression anchor system in cohesive soil. This paper conducted a series of field pullout loading test of pressure-dispersed compression anchors in clay soils, considering the following influence factors of stressing sequence, grouting method and compensation load. The simultaneous stressing yielded less pullout capacity than the individual stressing because of the relatively large interaction between the units. With the increase of the load level, the movement ratio decreased with a gradually decreased rate, followed by a slight increase when the load level exceeded 0.7. The rebound stiffness declined as the load level increased and the individual stressing yielded smaller reduction in the rebound stiffness than the simultaneous stressing. The back calculation indicated that the effective anchor diameter was enlarged by 10% due to the post grouting with a grouting factor of 0.2.

Shear failure is a typical failure mode of rock masses in underground engineering. Conducting shear tests of rocks in the laboratory is an effective research method. In this paper, a shear box is presented that can be used for the direct shear test of hard rocks under three-dimensional stress conditions with a true triaxial compression testing apparatus. The new normal linear variable differential transformer (LVDT) sensor fixture is used to make the obtained normal deformation data more accurate. The lateral deformation measurement adopts a U-shaped sensor design, which solves the problem of the lateral deformation measurement. Under different lateral stresses, a series of direct shear tests were conducted on the sandstone specimens, and the test results verify the reliability and practicability of the novel laboratory shear box. With the increase of lateral stress, the peak shear strength and peak normal displacement of the sandstone specimens also increase, and the lateral deformation is restrained. The lateral splitting block can be effectively reduced by applying lateral stress.

Pile driving is a complex dynamic process, which induces ground movement. The generated ground vibrations have considerable potential to damage the nearby structures and could lead to possible disturbance of building occupants. The safety of underground structures during the driven process constitutes a significant concern. In this study, two different standards were adopted to estimate the security of tunnels adjacent to the driven pile site. The continuous pile driving installation from the initial position to the final depth was executed using arbitrary Lagrangian–Eulerian (ALE) adaptive mesh and element deletion methods. Through the elimination of all the simplified hypothesis and the adoption of a three-dimensional model, consistent results compared to field monitoring were obtained. The damage of the adjacent tunnels during the driven pile was evaluated in terms of three criteria, peak particle velocity and distance (Swiss code) and displacement limitations (Chinese code). The effects of tunnel depths (H) relative to pile length (L) were investigated by driving the pile near the tunnel axis and next to and below the pile toe. The results indicated that the peak particle velocity limit value criterion could be adopted for the safety evaluation of an existing structure during pile driving. However, the maximum allowable radial distance from the existing structures to the pile centerline is not always an acceptable criterion to justify the safety of existing structures neighboring the driven pile site. A peak vertical velocity is a convenient approach that can result in a reasonable evaluation of the peak particle velocity.

In this paper, the generalized interpolation material point (GIMP) method will be utilized to simulate the postfailure behavior of slopes that accounts for spatial variability. Because the spatial variability of soil has not been considered in the analysis of the postfailure behavior of slopes, the local average subdivision (LAS) will be used to model the spatial variability of soil properties. By combining these two methods with Monte Carlo simulation, a random generalized interpolation material point (RGIMP) method will be proposed. A homogeneous slope will first be analyzed to verify the correctness of the GIMP implementation used in this paper. Then, a strain-softening slope will be analyzed as an illustrative example to investigate the influence of the spatial variability of soil on the postfailure behavior using the RGIMP. The results show that the runout distance, the retrogression distance, and the sliding volume of the slope that considers spatial variability have significant variations compared with the homogeneous slope. In addition, the kinetic energy of the slope will be investigated. The maximum and average global kinetic energy of the slope that considers spatial variability show obvious differences compared with the homogeneous slope. According to the Pearson correlation coefficient, this shows that there were relatively strong correlations between the maximum (average) global kinetic energy and the runout distance. However, there was no apparent correlation between the maximum global kinetic energy and the retrogression distance (sliding volume).

A track-unsaturated ground model was proposed to study the dynamic responses of the unsaturated subgrade induced by moving trainloads. The subgrade was modeled by an unsaturated poroelastic half-space, and the railway track system comprises a rail, sleepers, and ballast. The double Fourier transform was applied to solve the governing equations of the entire system in the frequency domain. The semianalytical results in the time domain were further obtained using the fast inverse Fourier transform. The effects of saturation, train speeds, and intrinsic permeability of soil on the dynamic responses were studied in both the time and frequency domain. It was found that the effects of saturation and train speed on the responses of displacement, acceleration, and pore-water pressure are significant. For low saturation, the vertical displacement response of unsaturated half-space is similar to that of the elastic half-space, and the intrinsic permeability has little effect. While for high saturation, the amplitude of vertical displacement increases with the increase of intrinsic permeability of the soil.

Recently, considerable risks to the internal instability of mechanically stabilized earth (MSE) walls have been encountered from the inadequate drainage capacity of some backfill under extremely heavy rainfall. Due to its high drainage capacity, geocomposite is regarded as an appropriate material for drainage purposes in many geotechnical structures, including MSE walls. However, the installation of a geocomposite drain produces hydrologically complex boundary conditions, and unsaturated flow through the MSE wall becomes more complicated. This article reports a series of numerical simulations conducted to investigate the influences of MSE wall dimensions and drainage capacity on seepage responses inside the protected zone of the wall. The results indicated that the distance from the upstream water source to the drainage face (L) contributes most to the level of the phreatic surface inside the protected (reinforced) zone. Furthermore, a relationship existed between the permeability of the soil on the upstream side and the lowering of the phreatic surface due to increased geonet transmissivity. Results reported in this study might reinforce understanding of complex flow behaviors in MSE walls with back drain installation.

A lower bound finite element analysis in conjunction with linear programming technique was used to evaluate the effect of soil–foundation interface roughness on the seismic limit load of shallow foundations placed in the vicinity of slopes. The approach followed in this study to take into account seismic forces is pseudostatic. The roughness factor (δ/φ) at soil–foundation interface was considered in a complete range of 0 (perfectly smooth) to 1 (fully rough) stepwise with an increment of 1/3. Furthermore, the effects of superstructure seismic inertia forces, and nonlinear programming, which is called second-order cone programming, are also addressed. The formulation of the finite element lower solution was extended for the case of pseudostatic conditions. The findings of this study have been compared with solutions reported in the literature, and the consistency of the results confirmed the formulation of the extended finite element lower bound in the presence of seismic forces. Results of the current study clearly endorse the fact that the soil–foundation interface roughness exhibits more highlighted contributions when a footing is underlain by an adjacent slope and subject to seismic loading. The results of the current study are presented in the form of table design charts.

The monopile is the most common form of foundation employed in offshore wind turbines. These foundations are subjected to millions of repeated load cycles, owing to wind and wave action. In this study, a series of six cyclic lateral load tests and two monotonic tests were performed on an aluminum model pipe pile with an outer diameter of 63.5 mm and a wall thickness of 2.5 mm. A model monopile was embedded in medium dense sand (D_r = 55%) and subjected to asymmetric two-way cyclic loading. From the experimental investigations, the effects of embedded length and the asymmetric two-way cyclic loads on the lateral pile head displacements and the cyclic secant stiffness of the soil–pile system were studied. Linear regression analysis was also performed to fit the conventional degradation parameters using the minimum number of critical constraints, which included the loading conditions and the flexibility parameters of the soil–pile system. From the test results, it was observed that asymmetric two-way loading causes a reversal of accumulated displacement for a pile embedded at a greater depth (L > 1.91T) under relatively lower amplitudes (ζ_b < 0.37). The cyclic secant stiffness was observed to increase at a relatively constant rate (A_κ) with the logarithm of the number of cycles. The study also revealed that the magnitude of initial cyclic secant stiffness, in comparison with the monotonic stiffness, exhibited a critical drop near the specific load character, ζ_c = −0.38.

The accurate prediction of disc-cutter wear has been a longstanding topic that affects the tunneling efficiency and performance of tunnel boring machines, especially the prediction of flat wear extent in mixed ground. In the present study, a novel method for predicting the wear depth of the disc cutter is proposed based on energy analysis. Initially, the critical condition for the occurrence of normal and flat wear is determined. Subsequently, according to the geometry of disc cutters and the relationship between wear loss and friction work, the normal and flat wear depth of the disc cutter are obtained, respectively. By comparing calculated wear with Ren's prediction model and a case study, the effectiveness of the present model was verified. Finally, the influences of penetration depth and installation radius on cutter wear in mixed ground were analyzed. The research results indicate that the critical condition for the occurrence of flat wear is that the resistance moments acting on disc cutters exceed the driving moment. In addition, it was found that flat wear extent was much larger than the normal one. The disc cutter is prone to wear abnormally in mixed ground. Sensitivity analyses on penetration depth and installation radius show that wear from of disc cutters is determined by penetration depth. Furthermore, both normal and flat wear exert positive correlations with installation radius. The present work may provide some primary understanding about the determination of the critical condition for the occurrence of flat wear under different engineering environments.