International Journal of Geomechanics

This paper will investigate the vertical vibration of a pile group connected to a rigid cap in multilayered poroelastic soils with compressible constituents. Individual piles in the pile group are modeled as one-dimensional bars and discretized into uniform elements based on the finite element method (FEM). Considering the compatibility condition between the pile group and the layered poroelastic soils with compressible constituents, solutions for the dynamic interaction problem will be derived. The resulting numerical solutions agreed closely with the existing solutions. Selected numerical examples will be provided to demonstrate the effects of the pile–soil stiffness ratio, pile spacing, compressibility parameters, and stratified property of the poroelastic soils on the vertical dynamic response of a pile group.

This paper investigates the stability of unlined rectangular tunnels in rock masses subjected to surcharge loading using adaptive finite element limit analysis based on the generalized Hoek-Brown failure criterion. Both upper and lower bound solutions are employed to calculate the stability number of the problem, and the relative errors are found to be within ±7%. A sensitivity study relevant to the Hoek-Brown parameters (σ_ci, GSI, and m_i), the normalized cover depth of the tunnel (C/H), and the width to height ratio of tunnels (W/H) are conducted. At a given value of C/H, the stability number is found to be increasing continuously with decreasing W/H and with increasing GSI and m_i. The computational results are summarized in the form of design tables and charts for practitioners to use, and potential collapse mechanisms are provided and discussed in detail.

Large deformation of the surrounding rock is a common geological hazard when excavating a new tunnel to be driven through soft rock, and this occurrence can severely interrupt the construction schedule and affect the safety of tunnel engineering. Therefore, it is important to study the risk level prediction for large deformation of the soft rock tunnel. Thus, this paper developed and examined a real-time (immediately) updated model. The model can parameterize for use in predicting levels of risk of large deformation of soft rock tunnel projects based on fuzzy multiattribute decision-making (FMADM). Two real field-based deformation indicators and 3 first-level evaluation indicators, including the 14 second-level indicators, were collected for data. Each indicator was quantitatively graded into four levels of risk of large deformation based upon the expert evaluation system for classification. Furthermore, a real-time updated comprehensive weight of each assessment indicator was presented from results using the analytic hierarchy process (AHP) and the entropy weight method (EWM). After that, a real-time updated risk assessment model was established by the use of the development of fuzzy membership functions. This risk assessment model was applied to five soft rock tunnel projects, and the corresponding model-based evaluated results were consistent with the real deformation degree based upon field measurement. This model could be used as new and well-defined sets of guidelines for improving the understanding of risk-based assessments of the potential for large deformation of the soft rock tunnel.

The in situ earth pressure coefficient at rest (K₀) for clay has been widely discussed in the literature. In engineering practice, the empirical relationships between K₀, the overconsolidation ratio (OCR), and the normally consolidated value ($K_0^{NC}$), is often used. Where $K_0^{NC}$ is as a function of friction angle (φ). These relationships do not distinguish between an increase in OCR that is due to unloading or the creep of the material. Although there is a significant amount of literature on the measurement of the change in K₀ during unloading, there is a lack of data on the evolution of K₀ due to creep. The few existing in situ measurements of K₀ are highly uncertain and are difficult to use to investigate the time evolution of K₀. Therefore, there is no clear consensus on the time evolution of K₀ within the geotechnical community. Over the last 20 years, several creep models for clay have been developed within the framework of elastoviscoplasticity. One common feature of many of these models is that they only predict a minor change in K₀ with time, because K₀ is given by one unique position on the potential surface. This contrasts with the unproven opinion of many practitioners who think that K₀ increases with time (even toward unity). To broaden the perspective of the discussion, this paper addresses the time evolution of K₀ in the framework of hyper-viscoplasticity. This framework offers a possibility for an increase in K₀ (even toward unity under certain conditions).

Japanese wood building damage was concentrated on the first story of buildings in previous large earthquakes. The main reason was attributed to an undesirable layout (lack of shear walls) in the first story. However, it is questionable whether or not the undesirable layout of the first story is the only reason for this. In this study, we will investigate the influence of soil–foundation connections on the typical damage pattern. This study is the first attempt, to the best of the author’s knowledge, to evaluate the effects of soil–foundation movement on the responses of a wooden building. Therefore, a three-dimensional nonlinear method to examine the soil–building interaction based on an input seismic wavefield, with the introduction of hypothetical soil piles, will be applied to a 2-story wooden building in the reclaimed zone of Tokyo Bay in Tokyo, Japan where liquefaction occurs. This revealed that a loose soil–foundation connection could explain the typical damage pattern. This inferred that, in addition to the undesirable layout, the soil–foundation movement caused the damage to the first story.

This paper presents a seismic slope stability assessment study of an inherently heterogeneous hydraulic fill dam in Texas. Extensive cone penetration soundings conducted at the dam site were interpreted to estimate different dam material properties. Eight sections distributed along the length of the dam were considered for the slope stability assessment to incorporate the effect of variations in material properties on the hydraulic fill dam. The conventional pseudostatic analysis method was modified to account for the differences in the behavior of the dam materials during static and seismic loading conditions. The current state of practice considers the same critical slip surface for both static and pseudostatic analyses. A modified pseudostatic slope stability analysis was performed to identify other possible critical slip surfaces with lower factors of safety values than the critical slip surface obtained from the static stability analysis. A reliability-based approach was then adopted to incorporate the effect of uncertainty on the estimation of shear-strength parameters of the dam materials, and the expected performance during different seismic events was evaluated. Two hypothetical earthquake events were also considered for the postearthquake deformation and serviceability assessment studies. The approach adopted in this research study facilitates identifying the critical sections of a dam that the conventional analysis method may not consider. Hence, this new approach would be integral toward developing guidelines for evaluating the seismic resilience of earthen dams.

This paper develops new models and normalized solutions to capture the response of piles subjected to a mixed translation–rotational soil movement. A flexible pile is divided into upper and lower portions, which join together by a rotational spring. The upper (rigid) portion rotates about the depth of the maximum bending moment at a constant stiffness (BiP–k_θ model), or a constant rotation ratio between the bi-portions (BiP–η model), or it is restrained by elastic lower portion (R–E model). The theoretical relationship between the stiffness (k_θ) and the rotation ratio (η) is established, along with the limits of the stiffness and the ratio. Response amplification is detected for inward rotating bi-portions, and quantified by new singularity stiffness(s). The three BiP models are adopted to capture the response of piles subjected to excavation loading in centrifuge tests. The impact of the five input parameters [i.e., limiting force per unit length (FPUL), modulus of subgrade reaction, modulus ratio of stable over sliding layers, and rotational stiffness of each portion] on the predictions is elaborated. Use of reduced values of the modulus, modulus ratio, and limiting FPUL, in particular, the rigid-pile-based 2-layer model also captures well the response of flexible piles, while a variable stress factor is used to model response profiles. Inward rotating bi-portions may amplify the pile response, and pull the maximum bending moment up into weak layer(s), and incur an unexpected failure. Centrifuge tests on anchored-piles are modeled using the BiP–η model to show the impact of inward rotation of the bi-portions. The new models and solutions are useful for the design of piles adjacent to excavation.

In land reclamation projects, the vacuum preloading method has been widely used to strengthen dredged fills by removing water. However, during the improvement process, clogging inevitably occurs in the drains and soils, hindering water drainage and causing inhomogeneous consolidation results. Therefore, it is essential to evaluate the effect of clogging on the consolidation behavior of dredged slurry at different radii. In this study, analytical solutions are derived under an uneven strain assumption to calculate the consolidation in the clogging zone and the normal zone, with time-dependent discharge capacity and clogging in the soil considered. Results calculated by the proposed solutions indicated that the clogging effect slows down the development of consolidation, reduces the final consolidation degree, and increases the difference between consolidations at different radii. It is found that the influence of the clogging effect's varies with the speed of the discharge capacity decay, the value of the initial discharge capacity of the drain, the permeability, and the radius of the clogging zone. Finally, a practical application of the proposed solution is discussed, and the proposed solution is suggested for the calculation of consolidation when treating high-water-content slurry.

During an earthquake, large movements of soil may cause the instability and uplift of buried structures in liquefied soil. Intensive studies on the uplift behavior of underground structures have mainly focused on a single tunnel. In this paper, numerical modeling is performed to investigate the uplift mechanism of twin tunnels with various configurations under earthquake loading. The excess pore pressure and uplift displacements of twin tunnels with different mutual positions and spacings are analyzed and compared with those of a single tunnel. The effects of the tunnel spacing and the mutual position on the uplift mechanisms of the tunnels are investigated. The results show that the generation of excess pore pressure and the liquefaction of soil surrounding the tunnels are prerequisites for the uplift. The interaction between twin tunnels affects their uplift behavior.

Considerable parts of worldwide railway networks cross deserts where there are frequent sandstorms, which cause contamination of ballast. One of the most important challenges for railway industries is the adverse effects of sand contamination on the mechanical properties of ballast that sometimes halt the operation of railways. Although the available studies have focused on dry sand contamination, the majority of the published reports indicate that wet sand contamination is a more important source of the problem. Due to rainwater infiltration, sands act as a lubricant in the ballast, which causes ballast aggregates to slide easily over each other, and therefore reduce the vertical stiffness and shear strengths of the ballast. To address this problem, the characteristics of railway ballast contaminated with wet sand will be investigated in this study through a comprehensive experimental investigation. Large-scale direct shear tests and plate load tests (PLTs) will be conducted on ballast samples contaminated with different percentages of sand and water content. New mathematical expressions will be derived that present ballast mechanical properties as a function of sand percentage and water content. Through the analysis of the obtained results, a new guideline will be developed for the effective maintenance of ballast in sandy areas.

The flow process of grout in a rock mass fracture network is an important part to judge the effect of grouting in engineering. Till date, little research has been done on the diffusion and migration pattern of grout in a random fracture network. Based on this, herein, a visualized fracture grouting test system with constant pressure was developed, consisting of pressure supply equipment, constant-pressure pulping equipment, fracture simulation equipment, and monitoring equipment. The system can simulate the flow process of the fracture network under the effect of various parameters such as the grouting pressure, grout characteristics, and fracture aperture and achieve the real-time monitoring of grout pressure, flow velocity, and diffusion distance in a random fracture network. Also, the governing equation of grout flow in a single rough fracture was obtained. The reliability of the test system was proved by the comparison of the single fracture grouting theory with the test results. Moreover, the grouting diffusion mechanism of the random fracture network was studied. The results show that (1) the pressure at the same point in the fracture increased with an increase in the grouting pressure and a decrease in the fracture aperture and was not affected by a change in the grout viscosity. (2) After the grout was dispersed into several branches in the fracture network, the pressures dropped significantly, the flow velocity decreased, and the pressure of each branch and flow distribution coefficient were greatly affected by the angle of bifurcation (intersection) fracture. The development of a test system and study results has a certain guiding value for grouting engineering.

The small-strain shear modulus of soils is a key parameter in the design of geotechnical systems and analysis of the soil–structure response to earth and earth-supported infrastructure. However, the small-strain shear modulus is not unique to a specific soil type. The small-strain shear modulus is a complex function of the state conditions (e.g., void ratio, suction) and stress states (e.g., in situ stresses, preconsolidation stresses). For unsaturated soils, the variations of the state conditions at a given stress state are captured by the soil–water characteristic curve (SWCC). This paper presents a general method to predict the small-strain shear modulus of unsaturated soils based on an inverse relationship between the small-strain shear modulus and the SWCC. Both vary with matric suction, but the two do not exactly mirror each other. Thus, the original SWCC was modified to improve the alignment with the small-strain shear modulus. In addition, a change in the void ratio induced by changing the net normal stress led to a change in the SWCC. Therefore, the modified SWCC was further adjusted for net normal stresses different from the original net normal stress. Finally, a small-strain shear modulus prediction equation was developed in accordance with the modified SWCC. The proposed equation consisted of several fitting parameters that were estimated from the modified SWCC. The developed prediction models were shown to be applicable to the numerous case studies for various net normal stresses and over a large range of matric suctions.

Bacterial cementation is a relatively new technology for stabilization of granular systems where cementation happens incrementally with repeated injection of the cementation fluid through the pores of the granular media. The mechanical performance of the bacterially cemented systems is critically dependent on the spatial and temporal variations of the precipitated cement. Numerical models proposed hitherto for mechanical behavior of bacterial cementation have assumed simple homogenized cementation. This paper investigates mechanical characteristics of bacterially cemented granular materials with a view to developing reliable numerical models that incorporate the spatial and temporal variations in cementation. The numerical models are validated by controlled experiments. Sand columns have been cemented to various degrees using a bacterial system. The degree of cementation was estimated by scanning electron microscopy. Triaxial compression tests have been conducted by varying the hydrostatic pressure. A multiscale numerical model is developed to simulate the mechanical response of the cemented system. The numerical results show good agreement with the experimental observations.

The site earthquake often causes serious soil liquefaction, especially results in volume change and surface subsidence to cause many secondary disasters. Different from the previous research on volume strain settlement of remolded clean sand after liquefaction, this paper adopts a gel push sampler to obtain undisturbed silt samples with fine particles in Xiong’an New Area of China. Dynamic triaxial tests are carried out on undisturbed and remodeling model soil samples. After the liquefaction failure of the test body occurs, the postliquefaction volumetric strain is measured. Based on the relevant test data, the volume strain characteristics of low plastic silt with fine particles after liquefaction are compared and discussed. The results show that cyclic stress ratio (CSR) is proportional to γ_max at the time of failure. The threshold value of the maximum shear stress of the remolded sample is 3.75%, while it is about 7.5% for the undisturbed low plastic silt sample. The repeated shear stress of the undisturbed sample is higher than that of the remolded sample. When the shear strain increases to a certain value, the volume strain tends to a fixed value after liquefaction. The volume strain caused by liquefaction of low plasticity remolded and undisturbed samples is significantly higher than that of clean sand. The maximum shear strain increases with the decrease of liquefaction safety factor. It underestimates the settlement after liquefaction by remolded clean sand or remolded soil sample, and the undisturbed soil sample as the engineering design basis is more in line with the actual soil condition.

Deep excavations are commonly used in urban areas due to the extreme increase in ground worth. Stability and displacement of adjacent structures are two main design criteria in stabilized excavated walls. In this paper, one of the largest excavations in urban areas in Iran, the Farmanieh Project, is examined as a case study. The minimum and maximum depths of the excavation were about 26 and 45 m, respectively. Three different support systems including multistrand anchors, steel piles combined with multistrand anchors, and concrete piles combined with multistrand anchors were used. Owing to the limitation of lateral movement of adjacent buildings and urban facilities, the displacement of the excavated walls was controlled by precise instruments during and after the construction procedure to compare actual displacement and numerical analyses predictions. The results showed that if the construction procedure, including delays during the excavation procedure and vibration of construction machinery during drilling, the groundwater level, and soil properties were not properly taken into account, numerical analysis could be misleading in predicting wall displacement during deep excavations. Therefore, considering the factor of safety against displacement is as important as the factor of safety against failure.

This paper develops a nonlinear shear-strength reduction technique for analyzing the stability of uniform slopes with the general nonlinear failure criterion. To incorporate the nonlinear power-law failure criterion into the finite-element slope stability analysis with the shear-strength reduction method (SSR-FEM), a general tangential approach is employed to obtain the instantaneous cohesive strength and internal friction angle from the power-law failure criterion iteratively. Therefore, the shear strength of the power-law criterion can be expressed by the instantaneous Mohr–Coulomb (MC) parameters related to stress within the slope soil. By this way, it is possible to use the linear MC failure criterion for the nonlinear slope stability analysis based on the conventional SSR-FEM. The proposed approach is applied to three typical examples including a homogenous 2D slope, a homogenous 3D slope, and a high rockfill embankment slope. The results show that the nonlinear shear-strength reduction technique presented in this paper can guarantee the accuracy and feasibility to determine the factor of safety (FOS) and provide an alternative way for stability analysis of simple uniform cohesive slopes with a general power-law criterion.

For over 100 years it has been accepted that the permeability or hydraulic conductivity of a soil is controlled by the size of pores through which the fluid flows, and that this pore size should be a function of particle sizes. All well-known formulas (such as the empirical Hazen or analytical Kozeny–Carman) are based on the squared value of some characteristic particle or pore size. Recent work has established which particles control the porosity or density of a granular material, so it follows that these particles may also govern the hydraulic conductivity. In this work, a new yet simple technique was used to obtain a characteristic “smallest” particle size, which is a function of both the particle size distribution and the geometrical packing. The use of this new proposed characteristic particle size was shown to be valid both theoretically and in comparison with the characteristic particle or pore sizes used in classical predictive methods for the permeability of granular materials. A very simple fractal theory showed what the characteristic particle size that controls conductivity should be, and a simple discrete element simulation was used to confirm the result.

Slope slip surface plays a vital role in the prevention and control of landslide hazards. For homogeneous soil slopes, a three-dimensional (3D) limit analysis method was used to analyze the slope critical slip surfaces with a nonlinear strength envelope. The critical slip surfaces can be obtained once the least upper bounds on slope stability were calculated by using a numerical optimization procedure. For various clay slopes, some comparisons of critical slip surfaces associated with linear and nonlinear strength criteria were made to evaluate the effects of soil strength nonlinearity on slip surfaces. For slopes with a nonlinear criterion, the critical slip surface tends to be shallower gradually as the slope inclination increases, which is opposite to that for slopes with a linear strength criterion. Besides, the influences of width constraint on the critical slip surfaces associated with two strength criteria become more significant for steeper slopes. Compared with the solutions for slopes with the linear strength criterion, 3D effects on slope slip surfaces will be more remarkable for slopes considering soil strength nonlinearity.

In engineering, retaining walls commonly rotate about the base (RB) and narrow backfill is often compressed. Current methods cannot accurately predict the passive earth pressure of a narrow, cohesionless backfill against the retaining wall. Therefore, this study will investigate the passive failure mechanism of narrow, cohesionless backfill under the RB mode using lower-bound finite element limit analysis (FELA). The backfill area above the slip surface reached a plastic state. This type of phenomenon means that stress deflection occurs in a narrow backfill, which is affected by the boundary conditions on each side. The principal stress trajectory is assumed to be a concave circular arc. In addition, a method to determine the stress trajectory, stress deflection angle, and lateral earth pressure coefficient will be proposed through stress analysis of soil elements. Moreover, a theoretical method to calculate the passive earth pressure in a narrow, cohesionless backfill under RB mode will be proposed that uses the limit equilibrium (LE) method, which is more suitable than the current methods. The parameter analysis results showed that decreasing the retaining wall slope or the interface friction helps to reduce the earth pressure against the retaining wall. In addition, the earth pressure at the bottom of the wall increases significantly.