International Journal of Geomechanics

Piezocone is one of the most useful tools to identify soil conditions, especially when the ground is soft. The interpretation of the results is quite challenging because it requires knowledge and understanding of the cone penetration mechanism. In this study, a numerical model based on the finite-element method was used to study the soil response of a soft clay located in the Bandung basin, West Java, due to cone penetration. Geologically, this soil is identified as lacustrine that has unique characteristics. Its liquid limit (LL) and void ratio (e) are exceptionally high, that is, up to 200% and 8.0, respectively. Therefore, it is categorized as highly plastic and highly colloidal clays. Nine cone penetration simulations have been conducted, and the simulations have been performed by adopting a press-replace method (PRM). The calculated cone resistance (q_c) and pore-water pressure (u₂) are compared with the measured data. Furthermore, the undrained shear strength of the soil (s_u) was also predicted using the cone factors N_kt and N_Δu. Using the values of N_kt = 13 and N_Δu = 10, a good agreement was obtained. Moreover, investigations on the stress path behavior due to cone penetration were performed. A simulation of the dissipation test was also carried out to study the decay of excess pore pressure and to estimate the soil permeability. The predicted results show a good agreement with the measured data.

Amidst the rapid development of the metropolitan area in Taichung, Taiwan, the mass rapid transit system (MRT) exemplifies a green public transportation project that helps minimize road traffic and reduce greenhouse gas emissions. However, managing the engineering of gravelly soils during MRT construction is challenging. This study adopted the discrete-element method software PFC2D to investigate the influence of microscopic properties on the mechanical performance of gravelly soils. Triaxial test data relating to gravelly soils with pure shear stress paths were analyzed to realize the mechanical behavior. The specimens were classified into Groups A and B based on the matrix properties, with and without cohesion. The simulation adopted two types of contact models to reflect gravelly soil matrix properties, which demonstrated reasonable agreement with the test results. After validating the proposed numerical model with a series of triaxial tests of gravelly soil under different confining pressures, this study investigated the influences of the micro parameters of porosity, friction coefficient, bond strength, and normal and shear stiffness of gravels and matrix on the mechanical behavior of gravelly soils. The results indicated that the main factors for Group A were the friction coefficient and effective modulus between particles. For Group B, the friction coefficient and bonding effective modulus were the main factors. The shear strength of gravelly soils increased when the friction coefficient, effective modulus, and bonding effective modulus increased.

Different from conventional methodology, this study presents an intelligent approach to fast identify the grain-size distribution (GSD) of granular soils using a convolutional neural network (CNN) under a deep learning framework. A database including 279 images of granular soils with their GSDs is first created. Then, the framework of the CNN is tailored to identify GSD. The CNN-based model is trained to predict 11 grain sizes corresponding to 1%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, and 100% of granular soils passing (i.e., d₁, d₁₀, …, d₁₀₀) using 80% of images, followed by the model testing using the rest 20% of images. By feeding an image of a soil sample into the proposed CNN-based model, the GSD can be predicted within several seconds. The predicted GSD exhibits excellent agreement with the measured one with an average error of 2.29% on the testing sets. It can be concluded that the proposed CNN-based model successfully provides a new intelligent way to fast, accurately, and conveniently identify the GSD of granular soils through images of soils.

The fatigue behavior of water-saturated tuff plays an important role in the safety of many hydraulic engineering projects against earthquakes and blasting. In this study, laboratory tests will be conducted to investigate the mechanical properties and acoustic emission (AE) characteristics of water-saturated tuff under triaxial cyclic compression. The results showed that as the confining pressure increased, microcrack growth inside the specimen was restrained more, the strain that corresponded to the peak stress linearly increased, and the inclination angle of the shear fracture surface changed slightly. The slopes of the residual strain curves of the slightly weathered specimens slowed down first and increased later, and those of the fresh specimens increased continuously. The Hoek–Brown and Mohr–Coulomb failure criteria could forecast the triaxial fatigue strengths of the specimens, and the latter yielded better predictions. The AE characteristics revealed the damage and failure processes in the saturated tuff specimens under triaxial loading–unloading compression. In addition, the onset of the rapid damage stage of the specimens was successfully identified using the damage variables based on elastic modulus, hysteresis energy (D_U), and AE signal, respectively.

The subtle alteration of surface geometry from a fresh surface to a sheared surface usually results in a considerable variation in the shear strength of jointed rock mass. Through profiling surfaces of the granite joints before and after the shear tests, an evaluation scheme was newly proposed by determining a desirable characteristic index and sampling interval of surface measurement in order to distinguish fresh and sheared joint surfaces quantitatively. The measured data demonstrated that although the mean Z₂ (root-mean square first derivation) values of all the profile lines were confirmed reasonable for estimating the joint roughness coefficient (JRC) value of the fresh joint surface, it could not completely evaluate the roughness of the sheared joint surfaces. Meanwhile, the distribution of slope angles, as the characteristic parameter, was proved to enable to clearly distinguish the fresh and sheared rock joint surfaces incorporating the small sampling scales (≤0.1 mm). The numerical simulations implemented in a mechanical shear model could confirm the critical effect of a slight change in surface geometry and further prove that the sampling interval of 0.1 mm could sufficiently capture the evolved waviness and unevenness of rock joint surfaces. Overall, it was confirmed that the results of our study provide new clues for evaluating the surface roughness of fresh and sheared rock joints and can be beneficial for understanding the variation of surface geometry during the shear process.

In practical engineering projects, bedrock is often encountered under a compressible soil layer. Compared with the soil layer, the hard bedrock has negligible deformation, so the displacement at the bottom of the soil layer is considered to be zero. In this case, it is difficult to use the Boussinesq solution to accurately calculate the stress and displacement of the soil layer under the action of shallow foundations. In the past, the solution has mostly been approximated by the layer-wise summation method or the finite-element method. So far, there is no analytic method to accurately calculate the stress and displacement of the elastic compressible soil layer. In this paper, an analytic method is proposed to solve the plane strain problem under the action of strip foundations. By using a mapping function to convert a soil layer with a finite thickness into the unit circle in the image plane, the stress boundary conditions at the ground surface and the displacement boundary conditions at the bottom of the soil layer are represented by a combined expression for the boundary values of the two analytic functions. A system of linear equations to find the coefficients of the analytic functions is established using a new power series method, and it is solved using the least-squares method. The stress and displacement at any point in the soil layer can be calculated quickly and efficiently using the derived analytic function. The effects of soil layer thickness on soil layer stress and ground surface settlement are discussed, and the results are compared with those obtained by the finite-element method.

Silty clay is a common embankment material whose moisture increases over time due to soil–atmosphere interactions and climate changes. The objective of this study is to investigate the unsaturated shear strength and microstructural evolution of compacted silty clay on wetting paths. Suction-controlled triaxial tests, scanning electron microscopy (SEM) observations, and mercury intrusion porosimetry (MIP) tests were performed on compacted silty clay specimens at various matric suctions and net confining pressures. The results show that both the matric suction and the net confining pressure have positive effects on the unsaturated shear strength of compacted silty clay. Nevertheless, they tend to cause different failure forms of the specimens, indicating distinct mechanisms behind their contributions to the soil shear strength on wetting paths. A change in matric suction does not lead to obvious evolution of clay particle orientation, but an increase in net confining pressure causes a weaker preferential orientation of clay particles. The shape of pore size distribution curves is slightly affected by the matric suction on wetting paths, while an increase in net confining pressure obviously reduces the proportion of large pores. The brittle failure of compacted silty clay appears due to the presence of open microcracks at high matric suctions.

This paper aims to assess the lateral limit load (H) of a pile embedded in cross-anisotropic clay deposits by a three-dimensional (3D) finite-element lower bound theorem in association with the second-order cone programming method. The lower bound solutions for a laterally loaded pile that is embedded in an anisotropic soil deposit can be found by formulating the element equilibrium, discontinuity shear and normal stresses equilibrium, boundary conditions, yield function, and optimizing the objective function using the second-order cone programming method and an iterative-based update procedure. The calculation procedure ceases when the discrepancy between the successive solutions satisfies the convergence criterion. Three different anisotropy models that include linear, sine, and cosine functions will be exploited to address the effect of cross-anisotropy. A parametric study will be conducted to capture the coupled effects of anisotropy degree (β), embedment length (L), and adhesion factor (α). The findings of this paper will be compared with those reported in the literature. The comparative analyses illustrated that the sine and cosine anisotropy functions yielded the least and most H predictions with β lower than 1, respectively. However, for soils with β higher than 1, the linear and the sine functions provided the lowest and the highest H predictions, respectively.

Ground response curves (GRCs) can be used to characterize the evolution law of stress and displacement during tunnel excavation. In this study, an attempt is made to obtain the GRC and find the stress field for a noncircular tunnel using a complex variable method. The analytical solution of an elastic GRC for a noncircular tunnel with a complex geometry configuration is proposed, and the stress and displacement of the rock mass caused by tunnel excavation are provided. The proposed analytical solution for a noncircular tunnel is compared with the known solutions of circular openings by degenerating the proposed model. The proposed solution is compared with a quasi-two-dimensional calculation model using FLAC finite-difference code, as well as with the known solutions of circular openings by degenerating the proposed model. The results indicated good agreement with the analytical and numerical solutions, verifying the correctness of the derivation. Finally, the GRCs are investigated based on three key points. It illustrates that the stresses and displacements of the three key points are linear and that the predicted GRCs of three key points by both models are in general agreement, if the difference in gravity field between the two methods is not considered.

Precast pile caps have been widely used in the accelerated construction of cross-sea bridges in recent years. Precast caps are connected to steel pipe composite piles through shear connection keys and reserved postpouring concrete, and the load-bearing system is transformed by pouring the postpouring concrete in hole-by-hole or layer-by-layer fashion. The process of system transformation is complicated, and it must bear the combined effects of flowing water pressure and wave force, which have a significant impact on the safety of system transformation. In this study, the finite-element method was used to analyze the influence of the hole-by-hole pouring scheme and layer-by-layer pouring scheme on the mechanical properties of shear keys and piles. It was determined that both pouring schemes are feasible, but the stress distributions of the shear keys and piles differ significantly. For the hole-by-hole pouring scheme, the stress on each pile and its shear keys varies significantly, and the maximum shear key von Mises stresses at the corresponding positions of different piles are 6.1 times the minimum von Mises stresses. In contrast to the hole-by-hole pouring scheme, the stresses of each pile and the corresponding shear keys are similar in the layer-by-layer pouring scheme, and the maximum shear key von Mises stress is only 21.7% of that in the hole-by-hole pouring scheme. Therefore, the layer-by-layer pouring scheme is more desirable.

The plane-strain critical-state friction angle is an important soil parameter in the design of geotechnical projects. The static angle of repose, which is the same as the plane-strain critical-state friction angle for normally consolidated cohesionless soil, was measured in the laboratory in this investigation. The experimental results were used to validate the relationship between the static angle of repose and the interparticle sliding friction. This relationship accounts for the dilatancy developed in the plane-strain critical state and its corresponding pore pressure coefficient at constant volume. In order to develop this relationship, the law of conservation of energy and the limit-equilibrium analysis were employed in a bidimensional plane-strain micromechanical model for the granular media in the critical state. The framework described in this paper accounts for the evolution of shearing resistance with porosity and establishes theoretical porosity thresholds for the contractive, dilative, and collapsible behavior. The results produced by the theory developed herein compared well with the present experimental results and those available in the literature.

Estimating pavement internal forces due to the swelling or shrinking of expansive soils is of vital importance for planning, design, and construction. Such efforts can minimize some possible damage due to expansive soils during pavement services. This study extends an analytical model on the single-layer Euler–Bernoulli beam resting on the Winkler foundation to estimate pavement internal forces in the multilayer pavement structure. Besides, an improved efficient algorithm is proposed in this paper for computing the value of virtual load parameters. The proposed closed-form solution and iterative algorithm are then verified by comparing the degraded solution with the single-layer beam model and the existing iterative algorithm. The influences of the beam model (single layer or multilayer) and foundation parameters on the pavement internal forces have also been investigated through parametric studies. Furthermore, the computational efficiency is compared between the iterative algorithm adopted by the previous researcher and the iterative algorithm proposed in this paper. It can be seen that a single-layer beam model may underestimate pavement internal forces induced by expansive soils. Foundation parameters will affect the value of virtual load parameters. Also, it can be found that the iterative algorithm proposed in this paper can reduce the iterative calculation time considerably. Moreover, it can be observed that with more fitting points and fitting terms, the fitting has a more accurate effect, and the advantages of using the iterative algorithm proposed in this paper are more pronounced.

Field monitoring programs (e.g., convergence measurements and stress measurements in the support system) play an important role in following the response of the ground and of the support system during and after excavation. They contribute to the adaptation of the excavation and support installation method and the prediction of the long-term behavior. In the context of the Lyon–Turin link project, an access gallery (SMP2) was excavated between 2003 and 2010, and a survey gallery (SMP4) has been excavated since 2017. Both tunnels are excavated in Saint-Martin-la-Porte (France) across a strongly tectonized Carboniferous formation, which is highly fractured and heterogeneous. Squeezing behavior has been observed and led to the instability of the support system and of the tunnel face in various locations. Based on the knowledge gained from field data processing and numerical modeling in SMP2, new field data are analyzed for SMP4, which is excavated at a much deeper depth, with a different orientation and with different excavation methods. Numerical modeling of the various profiles of SMP4 considering the excavation stages and the specific yielding support is presented, which predicts the time-dependent response of the tunnel.

This paper assesses the stability of pile reinforced slopes considering the statistical variability of the soil geomechanical parameters and varying the reinforcement geometry. The variation in the factor of safety and the probability of failure of the possibilities under analysis are evaluated with the implementation of reinforcement in different situations. For this purpose, theoretical and numerical methods were used, employing Slide and RSpile programs, which use the limit-equilibrium method to verify slope stability and p–y curves to analyze pile strength. The conclusion is that increasing pile diameter and decreasing the spacing between piles contribute to increase slope stability and the best positioning for piles is close to the slope crest. The factor of safety also increases as pile length increases. However, there is a minimum length for the pile to mobilize stress on the slope and a maximum length beyond which no significant improvements in stability are detected. The displacements in the pile obtained by the p–y curves comply with the admissible recommendations in the literature.

Due to oriented internal structure and depositional direction, the strength of natural geomaterial is anisotropic and cannot be fully represented by the existing isotropic criteria. This paper proposes a simple and novel anisotropic fabric tensor to extend isotropic failure criteria to account for anisotropy in a general way. Both the cross-anisotropic strength and the depositional angle between the loading direction and depositional direction can be considered in the proposed method without introducing any additional parameters. Especially, the effect of the depositional direction is subtly considered by a generalized fabric tensor with the depositional angle and two friction angles under different loading directions as the basic parameters. Extensive parametric analysis indicates that the proposed method is versatile to introduce anisotropy to several existing isotropic criteria by a unified expression. Through comparisons with experimental results of five types of soil, the anisotropic criteria established by the proposed method show good performances in predicting the strength of cross-anisotropic geomaterials under general loading conditions as well as capturing the nonmonotonic effects of depositional direction.

The relationship between morphological features and mechanical properties of rock structural planes has been an important scientific problem in the field of deep rock mechanics and engineering geology. Because borehole image accurately records the hole-wall information of rock mass and its morphological features of in situ structural plane, the contour line of the structural plane in borehole image is analyzed, and the roughness analysis and anisotropic characteristics of the structural plane are described by using fractal dimension model of each section of the contour line. Results show that the transformed contour line can reflect the three-dimensional morphological features of in suit rock structural plane in the borehole, which can be used to determine the dominant antisliding direction of the rock structural plane in a more feasible way. They are used to estimate the shearing strength of in suit rock structural plane with less randomness and more reliable data. The formed rose diagram with a fractal dimension of the contour line can show the anisotropic characteristics of rock structural planes. The in suit analysis of the shearing strength’s distribution trend and the dominant sliding direction of rock structural planes can be realized in a novel way during the practical engineering application, which can solve the difficulties of obtaining related mechanical properties by using in suit structures information in the deep rock mass.

During the construction of deep-buried diversion tunnels, many uncertainties exist in the variables governing the safety of the support structure, which will affect the safety of the tunnel. Therefore, the probability analysis of the deep tunnel based on Monte Carlo simulation is analyzed. For that a case study of diversion tunnels at Jinping II hydropower station, southwestern China is selected and analyzed. First, the dispersions of geostress and marble mechanical parameters are studied based on the field monitoring of the bolt stress and the laboratory testing of the Jinping marble. Subsequently, the statistical characteristics of the radial subgrade modulus are obtained considering the 100,000-fold sampling. Finally, the influence of the main uncertain factors on the internal force and deformation of the supporting structure is examined through the establishment of a numerical model. The maximum bending moment and axial force appear in the vault and invert of the tunnel, respectively. When the design requires the reliability of the tunnel support structure to achieve 90% in the case of Jinping II hydropower station, the value of the bending moment and the axial force cannot be higher than 0.57 MPa and 8.0 MN, respectively. The factors those having a greater impact on the safety of the tunnel support structure are the vertical load and the radial subgrade modulus.

In this study, a hybrid method was developed to predict optimum dimensions of reinforced concrete (RC) cantilever-type retaining walls. The metaheuristic-based optimization process of RC retaining walls needs iterative analysis, including the consideration of two different design limit states. These limit states defined as design constraints are geotechnical and structural limit states. Since the optimization process is long, it is aimed to generate a prediction model for optimum dimensions of RC cantilever retaining walls. For this purpose, artificial intelligence and machine-learning methods can be combined with metaheuristic algorithms used in the optimization of the problem. The method uses artificial neural networks (ANNs) for prediction results without iterative process and flower pollination algorithm (FPA) to obtain training data for machine learning. The proposed hybrid model is called flower pollination algorithm-based artificial neural network (FPA-ANN). To verify the prediction model, mean absolute error (MAE), mean squared error (MSE), and root mean squared error (RMSE) values are calculated for the optimum data set. Because this is the first proposed prediction model for a real structural engineering application related to RC cantilever-type retaining walls, the model was tested on different input values that are not included in the machine-learning process. Also, the relations between height, unit soil weight, internal friction angle, and surcharge load are investigated via the prediction model.

With the popularization of unmanned aerial vehicles, low-cost, high-precision photogrammetry has become a useful tool for mapping and obtaining information in areas that are difficult to access. It has been applied in geomechanics because it makes it possible to obtain information from rock masses for which there is no alternative way to obtain information by traditional mapping methods, increases the frequency of analysis and reduces the exposure of professionals to unsafe conditions. In this study, we used the technique called structure from motion to acquire 2D images and create a three-dimensional point cloud. Then, using artificial neural network techniques, we implemented a semiautomatic classification of the rock mass using the Geological Strength Index (GSI). The routine was validated using five datasets with different geological characteristics, choosing the neural network with the best performance, presenting results with confidence intervals above 90%, 100% hit rates, and a low mean squared error. The procedure allows a standardized interpretation, intended to reduce the bias generated by the interpretation through a rapid and repeatable analysis, in addition to the creation of a systematic record report.

Ground improvement using a joint technology combined with in situ shallow stabilization of soft clay and soil–cement column-reinforced foundations is an effective technique for the treatment of soft soil foundations. This paper presents a methodology to study the performances of a new joint composite foundation under embankments with an emphasis on the load transfer in the stabilized layer. The deformations of the columns, the surrounding soil, and the overlying stabilized layer are compatible by applying the stress continuity and volume settlement continuity at the bottom of the stabilized layer. Two assumptions of the stabilized layer were considered: a perfectly rigid situation and a deformable one. The corresponding uniform vertical strain between the columns and the surrounding soils in the case of a rigid stabilized layer and the load transfer in the embankment, a stabilized layer, and a column-reinforced foundation in the case of a deformable stabilized layer were analyzed. The settlement and stress concentration ratios of the column-reinforced foundation were evaluated through a parameter study. To give full play to the bearing function of the columns under working conditions in this paper, a reasonable stabilized depth was assumed to be 1.8–2.5 m.