To explore the bearing mechanism and failure mode for rock layer of pile tip of bridge pile groups with underground karst cave, laboratory model tests of single pile and pile groups with different numbers of piles were carried out. The ultimate bearing capacity and failure modes for rock layer of pile tip of pile groups with different numbers of piles were captured. The failure surface was divided into two parts according to the characteristics of the failure mode for the rock layer with underground karst cave, and a calculation method for ultimate bearing capacity was developed by combining with limit analysis method. The theoretical calculation values match well with the experimental values, which verifies the rationality of the proposed method. Meanwhile, the relationship between the ultimate bearing capacity for the rock layer has been analyzed and it can provide reference for bridge pile foundation construction in karst area. The experimental and theoretical calculation results indicate that: (1) When the rock layer of pile tip of pile groups with underground karst cave is overall destroyed, the whole destroyed body can be regarded as a large pier foundation  similar to destroyed body of single pile. (2) When the pile spacing is smaller, the ultimate bearing capacity for the rock layer increases with the increase of the length of the outer envelope line of outer foundation pile. Furthermore, when the length of outer envelope lines are the same, the arrangement of internal foundation piles has no effect on the ultimate bearing capacity for the rock layer. (3) The coefficient of pile group effect increases with increasing the pile spacing, and the critical pile spacing is 5d−6d (d is the pile diameter).

In this paper, the experimental studies on the water retention and strength properties of unsaturated weakly expansive soil in a wide suction range were carried out, and a strength model of unsaturated soil over a wide suction range was proposed. The results show that in the wide suction range, the stress-strain curve increases with the increase of suction. The strain hardening occurs in the low suction range, while the strain softening appears in the high suction range. In the low suction range, the sample presents shear deformation, while in the high suction range, the sample begins to show a shear dilation trend at the axial strain of 2%. In addition, a soil-water retention curve model was proposed to distinguish adsorbed water from capillary water, assuming that the strength increment caused by suction is mainly determined by capillary water. The degree of saturation of capillary water was used to replace the effective stress coefficient and then it was substituted into the Bishop’s unsaturated soil strength formula. The proposed model was verified by test data of various soil types and compared with the models in the literature. The results show that the proposed model can well describe the strength of unsaturated soil in a wide suction range.

In order to realize the resource utilization of rice husk ash, rice husk ash was incorporated into the soil for improving the soil strength. Large-scale dynamic triaxial tests were carried out on rice husk ash modified soil. The effects of rice husk ash dosage, dynamic stress amplitude and reinforcement method on the cumulative axial deformation, dynamic stress-strain curve and dynamic elastic modulus of the rice husk ash modified soil were analyzed. The results showed that with the increase of rice husk ash content, the maximum dry density of rice husk ash modified soil decreases and the optimal moisture content increases. The content of rice husk ash has a significant influence on the performance of soil. Under the content of 5%−10% rice husk ash, the cumulative axial deformation of rice husk ash modified soil is low, and the soil strength is better. The dynamic stress amplitude and reinforcement method obviously influence the cumulative axial deformation of rice husk ash modified soil. As the number of cycles increases, the hysteresis loop area of the sample decreases and the dynamic elastic modulus increases gradually. Reinforcement materials can improve the performance of soil. For single-layer reinforcement scenario, grid reinforced soil samples surpass other samples in terms of performances.

In this study, one-dimensional consolidation creep tests were conducted to study the influence of sand content on the consolidation creep behavior of sand-fines mixtures in front of the check dams in the Loess Plateau of central Ningxia, and the influence mechanism of sand content on its creep properties was investigated according to the mercury intrusion tests. Drawing on the theoretical framework of the Yin-Graham isochronic elastic-viscoplastic model, an empirical creep model, which took into account the finiteness of creep under one-dimensional conditions and the attenuation of the creep coefficient, was developed to capture the stress-strain-time relationship of sand-fines mixtures. The results show that when sand content is less than 28.57%, there is a negative correlation between sand content and secondary consolidation coefficient and secondary consolidation deformation, and the creep behaviors of sand-fines mixtures are controlled by fine particles; when sand content is greater than 28.57%, the secondary consolidation coefficient and secondary consolidation deformation gradually increase with the increase in sand content, and the creep properties of sand-fines mixtures are similar to that of sandy soil. The predicted curves of the proposed model are in good agreement with the strain-time curves of one-dimensional creep tests, indicating that it is suitable for characterizing the creep characteristics of sand-fines mixtures under one-dimensional conditions

Complex network analysis methods can be applied to analyze the structural evolution of granular materials during loading and establish the relationships of mechanical parameters of these materials between micro/meso and macro scales. This articles mainly reviews our research results in recent years on the structural evolution of granular materials and the relationship between microscopic parameters and macroscopic mechanical properties using complex network methods. Firstly, the basic concepts of complex networks (average degree, clustering coefficient, average path length, and force chain) are introduced. Then, the microstructural evolution of granular materials under biaxial conditions and the network contact parameter changes during the initiation and sliding of soil slopes are analyzed. Furthermore, the microstructure evolution of granular materials considering particle shape, intermediate principal stress coefficient, and complex loading paths is discussed. Finally, the mesoscopic structural parameters of the contact network are associated with the macroscopic mechanical properties, and relationships between the state parameters and average degree, the microscopic structural parameters and soil dynamic strength, the average degree as well as shortest path length and yield function of the modified Cam clay model, and the average degree and the ground vertical pressure are established. The parameters in the unified hardening model are modified based on the maximum flow model.

The ground bearing capacity of thin layer soil under raft foundation is different from that of uniform ground, and cannot be simply determined by using the usual natural ground bearing capacity method. This article determines the strength index and deformation parameters of the thin layer soil ground through field plate loading tests, and divides the raft foundation into several grids to calculate the foundation stiffness of each grid employing the tangent modulus layerwise summation method. The raft foundation is considered as an elastic foundation plate supported by each grid, and the compressive stress at the bottom of the plate under the joint action of each grid is calculated. The compressive stress of each grid is compared with the ultimate ground bearing capacity predicted using thin layer soil ground strength parameters to judge the safety of the ground bearing capacity, and then according the settlement deformation of the raft, the allowable bearing capacity of this thin layer soil ground is determined by the strength bearing capacity of the thin layer soil ground and the deformation control of the raft foundation. Compared with the finite element method, it has been proven that this is a more suitable method for determining the ground bearing capacity of the thin layer soil under the raft foundation.

The pore pressure response of two-dimensional seepage field of a foundation pit with a sealed bottom under the condition of water level fluctuation is studied analytically. The soil around the pit is divided into several regions, and the excess pore water pressure response distribution in each region is expressed in the form of a series by applying the superposition principle and the separation variable method, and the coefficient matrix is solved by combining the inter-region boundary conditions and the continuity conditions to obtain the explicit analytical solution of the excess pore water pressure distribution of the two-dimensional seepage field in the pit with a capping layer under the fluctuation of water level. The results of the analytical solution are compared with the numerical results, and the correctness of the analytical method is verified. Based on this analytical solution, calculation formulae for the outflow ratio drop and water inflow caused by water level fluctuations are further provided. The sealing effect of the bottom layer and influences of the bottom layer thickness and permeability coefficient changes are discussed through a pit project example. The results show that the outflow ratio drop and single width seepage discharge are increased, but the rate of increase reduces as the bottom layer thickness decreases and the permeability coefficient increases.

For the advantages of light weight and high stability, the geocell-reinforced soil retaining walls will be widely used in the retaining construction in the near future. The performance of the facing-type geocell retaining wall reinforced by extending geocell layers with uniform spaces over the full height of the wall (FE-type), is better than other types of geocell retaining structures. However, the experimental researches on the FE-type geocell retaining walls are limited at present. The model test in this study was performed on an FE-type geocell retaining wall to examine the mechanical characteristics in terms of deformations and vertical stresses of the retaining wall under vertical static load. The strains of the extended geocell layers were examined and the effect mechanism was also discussed. Results show that the differential settlements of upper fillers of the retaining wall present as larger in the middle than that approaching the face-plate. The tuck net effect of the geocell layers inside the upper fillers owing to bending, and the strong lateral restriction of the geocell layers have transformed parts of vertical forces into the horizontal forces exerting on the geocell layer. Through the transformations of several geocell layers, the vertical stresses significantly decrease and the differential settlements almost disappear at the bottom of the retaining wall. Due to the warping deformation, the upper geocell layers exert downward tensions and inward contractions on the upper face-plates of the retaining wall. Consequently, the vertical displacements of the face-plates increase rapidly with increasing loads. And it also effectively limits the increasing of the horizontal displacements of the face-plate. The horizontal displacements of the face-plates ranging from 0.375H to 0.75H (where H is the height of the retaining wall) are larger than other parts, and the maximum displacement locates at the 0.375H height. The influences of the back filling height and loads on the distribution of the geocell layers strains along the horizontal direction are week. The potential failure surface of the FE-type geocell reinforced retaining wall resembles the curved shape that changes slowly along the vertical direction without penetrating the face-plates. The findings could provide valuable and practical references for the design and construction of the FE-type geocell-reinforced soil retaining wall.

A large-scale rainfall experiment was carried out to investigate the rules of rainfall infiltration and runoff erosion of vegetated slopes with different gradients. The parameters such as slope water content and soil and water loss were monitored in real-time. The effects of slope gradient on rainwater infiltration, slope runoff and soil erosion were quantitatively analyzed. The results show that under a continuous rainfall condition of 4 h and 60 mm/h, the cumulative rainwater infiltration of bare soil slopes with 1:2, 1:1.75, and 1:1.5 are 70.6%, 80.4%, and 92.3% those of vegetated slopes with the same slope gradient. Vegetation cover will accelerate the rainwater infiltration rate and increase the cumulative rainwater infiltration of the slope. The smaller the slope gradient of the vegetated slopes, the faster the rainwater infiltration rate and the greater the cumulative rainwater infiltration, while the bare soil slopes are the opposite. Vegetation cover can effectively reduce runoff rate and cumulative runoff. The smaller the slope gradient, the more significant the effect of reducing runoff. Compared with the bare soil slopes with the same gradient, the vegetated slopes of 1:2, 1:1.75, and 1:1.5 reduce runoff by 53.11%, 32.56%, and 17.73%, respectively. The sediment content of soil and water loss in vegetated slopes is less than 0.1%. Vegetation cover can effectively restrain soil erosion on the slope. The larger the bare soil gradient is, the more serious the slope erosion is. The soil erosion unit areas of 1:2, 1:1.75 and 1:1.5 bare soil slope are 3.623, 5.710, and 11.295 kg/m², and the sediment contents of soil and water loss are 3.06%, 4.29%, and 7.34%, respectively. Based on the test results, the relationship between the total amount of mud and sediment yield of bare soil slope y and slope gradient x and rainfall duration t is y=0.0028_t^-0.67x^4.11t0.0647  .

 In coal mining, the excavation of a coal-rock roadway and a thin coal seam will inevitably cause radial unloading of the coal-rock combination system. The radial unloading phenomenon is often accompanied by the rapid accumulation and release of energy, so it is necessary to investigate the energy evolution law of coal-rock combination specimens under the unloading confining pressure condition. To this end, the unloading confining pressure tests with different unloading rates were carried out for the coal-rock combination specimens. The results show that: (1) The axial loading and constant stress stages are the main energy storage stages of the combination specimens. The failure stage is mainly dominated by the release and dissipation of energy. (2) The acceleration of the unloading rate leads to the decrease of the peak elastic energy of the specimens, and the increment of the elastic energy at 0.03 MPa/s in the constant stress stage is 1.64, 2.70 and 3.50 times of that at 0.06 MPa/s, 0.09 MPa/s, and 0.12 MPa/s, respectively. (3) The increase of unloading rate will lead to the increase of post-peak dissipation energy of the specimen, and the post-peak dissipation energy is 28.17%, 49.53%, 69.55% and 92.87% of the peak elastic energy when the unloading rate increases from 0.03 MPa/s to 0.12 MPa/s, respectively. (4) The increase in unloading rate will significantly enhance the tensile failure tendency of coal-rock combination specimens, resulting in an increase in the fracture angle, an increase in the number of tensile secondary cracks, and an enhancement in the breaking strength. (5) A dissipative energy constitutive model considering the initial damage is established to reasonably explain the whole process of damage evolution of coal-rock combination specimens under the unloading confining pressure conditions. The research results are significant for understanding the energy evolution characteristics of coal-rock combination samples with unloading rate.

Microbial induced calcium carbonate precipitation (MICP) technology combined with modified jute fiber can effectively cement the uranium tailings and fill the pore between the particles, thus, improve the impermeability of uranium tailings. In this study, the effects of some parameters, such as particle gradation, concentration of cementing solution, as well as the length, mass content and hydrothermal treatment time of modified jute fiber, on the impermeability of uranium tailings cemented by microbe in coordination with modified fiber were studied, and the best values of these parameters were determined. Then, the structure type of calcium carbonate crystal produced by MICP and modified fiber was analyzed using the scanning electron microscope (SEM) and X-ray diffraction (XRD), in order to analyze the impermeability mechanism. The results showed that the modified fibers provided more spots for bacterial adhesion, due to the increased surface roughness by hydrothermal treatment. This promoted the growth, reproduction, migration and fixation of microorganisms in uranium tailings, increased the uniformity of calcium carbonate precipitation, and decreased the permeability coefficient of uranium tailings. In the case of gradation number A3, the permeability coefficient of the cemented tailings decreased sharply. When the fiber length, fiber mass content, hydrothermal treatment time of fiber and the cementing solution concentration were 20 mm, 0.5%, 2 h and 2 mol/L, respectively, the permeability coefficient of uranium tailings decreased by 99% after 11 rounds of bio-grouting, indicating these values are optimal. The calcium carbonate crystal, formed in the cemented uranium tailings by the modified fiber combined with MICP, presented calcite characteristic peaks at diffraction angles 2θ of 23º, 29.4º, 36º, 39.3º, which demonstrated that the crystal is mainly calcite.

In view of the difficulty that the spatial location, growth degree and development direction of cracks inside rock masses cannot be directly observed, the transparent materials are considered as the effective method to build similar models. In order to study the effect of raw material ratio on the mechanical properties of transparent rock-like material, and overcome the defect of insufficient brittleness of transparent rock-like material at room temperature, a novel method to prepare the transparent rock-like material is developed with epoxy resin as aggregate, modified anhydride as curing agent and rosin saturated solution (RSS) as brittleness adjustment. The influence of the mass ratio of epoxy resin to anhydride and the content of RSS on the physical and mechanical parameters of transparent brittle rock-like material were investigated using the comprehensive experimental design method. The results showed that the prepared transparent rock-like material possesses good brittleness at room temperature. Furthermore, RSS content has a significant impact on density, compressive strength, modulus of elasticity, peak strain, tensile strength, cohesion and internal friction angle of the rock-like material, while the mass ratio of epoxy resin to anhydride has remarkable impact on its density. With the increase of RSS content, the density, compressive strength, elastic modulus, peak strain, cohesion and internal friction angle of the rock-like material reduce, while the tensile strength shows a trend of decreasing first and then increasing. In addition, with the increase of the mass ratio of epoxy resin to anhydride, the density of rock-like material first increases and then decreases. Finally, an empirical formula between the mechanical properties of transparent brittle rock-like material and the mass ratio of epoxy resin to anhydride and RSS content is established using multiple linear regression analysis, which provides reference for preparing transparent brittle rock-like material.

The shattering deformation masses in high seismic intensity area are widely distributed. In order to understand the shattering failure mechanism of step-like bedding rock slopes under multi-stage earthquake excitations, a large-scale shaking table test is conducted based on the cut slope of Sanqing Expressway. The ratio of acceleration amplification factor (RAAF) is proposed to investigate the difference of acceleration dynamic response at different positions, while the Hilbert-Huang transform and marginal spectrum are used to identify the cumulative damage effect and failure process. Combined with the slope failure phenomenon, the shattering failure mechanism is clarified. The results show that the slope has an elevation amplification effect, and the acceleration amplification factor tends to increase first and then decrease with the increase of the input peak seismic wave. The RAAF undergoes a positive-to-negative mutation before and after the 0.6g input seismic wave peak, which is the critical value for changing the dynamic response between the two slope types. Under multi-stage earthquake excitations, the low-frequency component of the Hilbert spectrum decreases while the high-frequency component increases, indicating a filtering effect of rock mass and interlayer. Under horizontal seismic excitation, the internal corners of the step are easy to exhibit dynamic tensile stress concentration, resulting in the internal corners prone to be shattered. For the slope with uneven step width, the progressive failure processes involve the second step being shattered firstly, followed by the upper rock mass sliding along the weak interlayer, the rear edge of the slope top being pulled apart, and finally, the first step being pulled apart and separated from the slope body. For the slope with even step width, the cracks appeared at the internal corners of each step, while there was no obvious sliding surface. The shaking table test reveals the shattering failure mechanism of step-like bedding rock slopes. In survey, design, and construction, it is significant to pay attention to the deformation at the internal corners of each step and apply an arc treatment to reduce stress concentration. Additionally, anti-slip piles can be installed at the slope toe to increase the threshold for shattering failure and enhance stability of the rock slope.

The offshore wind industry needs to move towards floating offshore wind turbines (FOWTs) that are secured to the seabed with anchoring systems in order to harness the more substantial amounts of offshore wind resources available in deep waters. The design of the anchoring system for FOWTs is crucial to ensure stability and safety in challenging offshore conditions. As a simplified reliability-based design approach, the partial safety factor method remains important in current offshore foundation design practice, but its effectiveness in achieving the required target safety level still needs to be examined. To achieve this end, a reliability analysis is performed for the uplift limit state design of strip plate anchors embedded in sand subjected to vertical loads, and a wide range of load cases are considered in the analysis. The failure probabilities of strip plate anchors designed with the partial safety factor method are estimated and then compared to the target safety levels. The results show that the estimated reliability index decreases rapidly with the increasing of permanent to variable action ratio and gradually converges to a relatively steady state. In addition, the current partial safety factor method has been shown to be conservative for offshore plate anchor design over a wide range of permanent to variable action ratios. The results provide probabilistic insights into the effectiveness of the partial safety factor method and may aid in the further development of reliability analysis for offshore anchors.

Diatomaceous soil is a kind of natural sedimentary soil that formed in lacustrine or marine environment, mainly composed of clay minerals and diatom remains. The light weight and high porosity of diatom results in a significant difference of physical properties between diatomaceous soil and common clayey soil without diatom, such as low density, high porosity and high water content. These physical properties cannot be predicted by empirical equations of conventional soil mechanics and still need to be explored. This study measured typical physical indices of particle size composition, specific gravity, specific surface area and Atterberg limits of diatom-kaolin mixtures at different diatom contents to enhance systematical comprehension of physical properties of diatomaceous soil. Also, the scanning electron microscope tests were performed to reveal the microscopic mechanism of these physical properties of diatomaceous soil. The results indicate that increasing diatom content causes increases in silt fraction, specific surface area and cation exchange capacity, and a decrease in specific gravity. As the diatom content increases, the Atterberg limits show ascending tendency with pore fluid being NaCl solutions of different concentrations. Despite both the liquid and plastic limits increase, the plasticity index remains almost unchanged. The microscopic investigation suggests that the abovementioned physical properties of diatomaceous soil mainly depend on the large inner cells and high water-retention capacity of diatoms. The tests also show that diatom-kaolin mixtures with diatom content higher than 80% possess low plasticity or nearly non-plastic during the test despite of the high Atterberg limits of pure diatom. This phenomenon means that the Atterberg limits cannot reflect the plasticity of diatomaceous soil, and the current classification methods for fine-grained soils according to Atterberg limits are inappropriate for diatomaceous soil. This study can provide data reference and theoretical support for research on engineering behaviors of diatomaceous soil.

The development, utilization and storage of thermal energy in deep rock masses generally involve fracture seepage and heat transfer problems, and the geometry of rough fractures has an important effect on the seepage and heat transfer characteristics. The geometric models of rough fractures with different fractal dimensions and standard deviations were constructed by successive random accumulation method. Combined with Navier-Stokes equations, momentum and energy conservation equations, the dynamic process of seepage and heat transfer in rough fractures was simulated. The rock fracture seepage and heat transfer test system was developed and the granite fracture seepage and heat transfer test was performed to verify the reliability of the numerical method. Then the variations of seepage and heat transfer characteristics in rough fracture were analyzed under different fractal dimensions and standard deviation conditions. The results show that the water-rock heat exchange effect gradually decreases with time and enters the thermal equilibrium stage after the heat front penetration. As the fractal dimension increases and the standard deviation decreases, the average convective heat transfer coefficient gradually decreases due to the significant enhancement of the dominant flow effect and the weakening of the overall heat transfer characteristics. The local convective heat transfer coefficient is positively correlated with the average fracture opening and the maximum elevation difference of the center line. The local Nusselt number increases with the increase of fractal dimension and reaches its peak at the exit.

Permeability is the capacity of a soil for transmitting a fluid, e.g., a salt solution, through soil pores. The permeability coefficient K is an important measurement of the soil permeability, which is affected by various factors. The existing research on the permeability of sandy soil is basically carried out in the freshwater environment. However, the calcareous sand is a typical marine sedimentary sand, which is deposited in the seawater environment with certain salt concentrations. In order to investigate the influence of void ratio, particle size and concentration of the salt solution on the permeability of calcareous sand in a salt solution environment, constant head and falling head permeability tests were carried out using the meter KAST - soil saturated hydraulic conductivity. The microscopic characteristics were investigated through Zeta potential and contact angle tests. The results showed that the mean particle size has the greatest influence on the permeability coefficient of a calcareous sand, and the change of the mean particle size will even lead to the difference in the order of magnitude of the permeability coefficient. The permeability coefficient (K) is negatively correlated with the salinity of the transmitting solution (P), and positively correlated with the porosity (n) and the mean particle size (d_a). A model for predicting the permeability coefficient of calcareous sand taking into account the salt content of transmitting solution has been proposed based on the current experimental studies on the influencing factors. The proposed model can provide a tool for assessing the permeability of artificially reclaimed islands and reefs in the South China Sea and analyzing the evolution of ground freshwater.

Sequential limit analysis (SLA) has been recently introduced into geotechnical engineering, in which a large-deformation problem is discretized into a sequence of small-deformation finite element limit analyses, and has been successfully used for the analysis of soil-structure interactions in undrained soft clay under plane strain condition. SLA method possesses high computational efficiency and numerical stability, and its calculation error can be real-time evaluated based on the deviation between upper-bound and lower-bound solutions. In this paper, the existing plane strain SLA method is extended to be suitable for axisymmetric problems such as vertical loading behavior of spherical probe instruments, pipe piles and spudcans. A new version of SLA is developed on the numerical platform OPTUM to broaden its usage. Additionally, the updating method of the model geometry based on velocity fields obtained from upper-bound solution in the original SLA is improved, so that it can adapt to more extreme soil deformation. A series of classical plastic solution cases and model tests are then used to validate the accuracy and effectiveness of the extended SLA method. It proves that this method can be used to simulate large displacement / deformation problems such as the vertical and lateral interactions between undrained clay and facilities such as foundation, penetration equipment and pipelines.

In order to improve the applicability of Hoek-Brown failure criterion and reduce subjectivity in the determination of geological strength index (GSI) value for anisotropic rocks, a modified Hoek-Brown failure criterion is proposed, which considers the variation of the GSI value with the bedding angle of anisotropic rocks. Triaxial compression test data of anisotropic rocks with different bedding angles were first collected. The results show that the peak strength of anisotropic rocks exhibits a U-shaped relation with bedding angle β. Then, the rock specimen with the bedding angle β = 0º is defined as the intact rock. The uniaxial compressive strength σ_c and material parameter m_i of intact rocks are obtained from data fitting using Hoek-Brown failure criterion. The corresponding GSI values under different bedding angles are calculated. The relationship between GSI and bedding angle β is fitted using Gaussian function, and a new strength model of anisotropic rocks is established based on Hoek-Brown failure criterion. Finally, the proposed model is verified by comparing the peak strength obtained from the GSI-softening model. It is found that the modified Hoek-Brown failure criterion is suitable for predicting the strength of anisotropic rocks with different bedding angles and under various confining pressure. The physical meaning of the new parameters in the model is also discussed.

Nanoindentation experiment is an important means to study micro mechanical properties of rock. Up to now only a few studies have discussed the correlation between the micro mechanical properties of rock and various macroscopic strengths of the rock. Firstly, the nanoindentation experiments on four different kinds of marble were carried out by the continuous stiffness measurement technique to obtain the micro mechanical parameters of dolomite and calcite. Secondly, the microscopic data scale was upgraded by Mori-Tanaka method to obtain the homogenized elastic modulus and Poisson’s ratio. Finally, the correlation between the microscopic parameters and the macroscopic mechanical experiment results was analyzed, and the applicability of predicting macroscopic properties of rock using nanoindentation data was discussed. The results show that: (1) The elastic modulus of dolomite in marble is 122.5 GPa, and the hardness is 5.4 GPa. The elastic modulus of calcite in marble is 70.3 GPa, and the hardness is 2.3 GPa. The strength and deformation properties of calcite are relatively poorer compared to those of dolomite. (2) For dolomite and calcite in marble, an indention depth of approximately 800 nm is recommended for determining the elastic modulus and hardness with the continuous stiffness measurement technique. (3) The data dispersion at the grain boundary points is larger than that at the inner points of the grain, and the hardness can better reflect the defect effect of grain boundary than elastic modulus does. (4) The homogenized results obtained by Mori-Tanaka method are somewhat reliable for predicting macroscopic elastic modulus and Poisson’s ratio. (5) Nanoindentation data can reflect the influence of mineral types on rock strength. To accurately predict uniaxial compressive strength, tensile strength, and fracture toughness, it is necessary to consider other strength factors, such as rock texture and structure. These results demonstrate the application of nanoindentation experiment in rock materials and provide a reference for predicting macroscopic strength using micro mechanical parameters.

Excavation of foundation pit induces the horizontal displacement of soil, and the excessive horizontal displacement will cause damage to adjacent underground structures. The traditional virtual mirror technology based on the assumption of uniform convergence of circular hole underestimates the horizontal displacement of shallow soil, which is unsafe in engineering practice. Most of the research in this field is based on plane strain analysis, and the spatial effect of foundation pit cannot be considered. The two-dimensional and three-dimensional semi-analytical solutions of soil horizontal displacement considering nonuniform convergence characteristic were derived using virtual mirror technology and infinitesimal method. The nonuniform convergence mode in line with the actual situation was proposed by comparing the semi-analytical solution with the measured data, the existing theoretical solution, and the finite element results. Furthermore, the spatial effect of soil horizontal displacement was analyzed in detail based on an engineering case in Hangzhou deep soft clay. The results indicate that the limit of downward convergence is the convergence mode closer to the actual situation. The plane strain solution, which is generally considered to be safe, cannot consider the horizontal displacement of soil parallel to the retaining wall and will underestimate the total horizontal displacement of shallow soil near the corner of foundation pit. The soil displacement parallel to the retaining wall that is easily ignored in engineering practice may cause unexpected harm to shallowly buried structures such as shallow foundations and underground pipelines.

The construction technology of strength composite pile is very complex. In order to understand the soil squeezing effect and load transfer mechanism in the piling process, the soil compaction effect test and static cone penetration test of strength composite pile were conducted, and the changes of pore water pressure, total stress and effective stress of soil around the strength composite pile were studied. The test results showed that the soil squeezing effect of strength composite pile decreased gradually along the diameter and increased gradually along the length. In the piling process, the effective stress of the surrounding soil layer increased by 12%−63%, and the side friction resistance of pile increased significantly. The tip resistance and side resistance increased by about 13%−84% and 8%−97%, respectively. Additionally, four load-bearing members with decreasing bearing capacity were formed along the pile diameter through the combination of a variety of strength materials, which realized a better bearing efficiency than that of conventional piles with homogeneous materials. It is consistent with the characteristics that the shear stress and compressive stress of piles caused by the pile top load along the diameter direction and length direction will gradually decay, which is also the source of the significant economic advantage of the strength composite pile.

 The existing theoretical studies on the stability of shield tunnel excavation face have not taken into account the effects of rainfall environment and groundwater level, especially the geotechnical effects induced by heavy rainfall infiltration. Firstly, a layered hypothetical Green-Ampt model is used to simulate the heavy rainfall infiltration process. Secondly, on the basis of the shear strength theory and the soil effective stress theory, the modified Mohr-Coulomb criterion and the Darcy’s law are combined to derive the apparent cohesion expression in segmental form related to rainfall intensity and duration, and then, based on the upper bound theory of limit analysis, a three-dimensional (3D) logarithmic spiral failure mechanism is constructed, and the work done by apparent cohesion is introduced into the virtual power equation of upper bound theorem, the upper limit solution of the support force of shield tunnel excavation face is obtained under the conditions of heavy rainfall infiltration and groundwater level through optimization calculation. Finally, the theoretical upper bound solution is compared with the 3D numerical results and existing experimental results, and a good agreement is obtained. In addition, the key parameters such as rainfall intensity and water table depth are analyzed. The results show that when the rainfall intensity is small (F/k_s≤1, where F is the mean rainfall intensity, and k_s is the saturated infiltration coefficient), the ultimate supporting force of tunnel excavation face accelerates with the increase of rainfall intensity, but has no obvious change with rainfall duration; the ultimate supporting force increases slowly with the increase of rainfall intensity, but significantly with the increase of rainfall duration when the rainfall intensity is large (F/k_s>1); the ultimate supporting force of tunnel excavation face decreases linearly when the normalized groundwater table depth is small (Z_w/D≤3, where Z_w is the depth of groundwater level, and D is the excavation diameter); and the rate of change gradually decreases and finally becomes stable when the normalized groundwater table depth is large (Z_w/D>3).

To investigate the mesoscopic damage accumulation and the loading-induced fracture process in freeze-thawed rocks, a method coupling water-ice particle phase transition and expansion based on the discrete element method is proposed. The rock freeze-thaw process is simulated using a particle flow program, and the reliability of the simulation results is verified through laboratory experiments. The frost heave evaluation index λ_v of pore water particles is quantitatively characterized, and a functional relationship between λ_v and the number of freeze-thaw cycles N is established. Furthermore, the fracture characteristics and the evolution of microcrack, displacement field and force chain field in freeze-thawed rocks during the loading process are evaluated. The results show that: (1) The volumetric expansion of pore water in the rock and continuous water replenishment are the essential causes of rock damage under freeze-thaw treatments. Microcracks in the samples are dominated by tensile cracks during the freeze-thaw process, exhibiting an “initially slow, then fast” evolutionary trend, with more significant displacement of rock particles on the periphery than those in the interior. (2) The number of microcracks in the samples under loading exhibits a “slow→gradual→rapid” growth trend. The numbers of freeze-thaw cycles positively correlated with the number of microcracks but negatively correlated with the microcrack initiation stress σ_i . (3) The fracture process and morphology of the samples differ significantly before and after freeze-thaw treatment. When the load approaches the peak stress σ_f , there are “abnormal signals” in the microcrack distribution, displacement field and force chain field, which can serve as a precursor to failure identification. Under the influence of freeze-thaw cycles, the spatial arrangement of microcracks inside the samples becomes more complex, and the fracture mode transitions from dominance by tensile microcracks to dominance by mixed tensile-shear microcracks. This study provides a new idea and method for exploring the failure behavior of freeze-thawed rocks.

In order to study failure problem of glass fiber reinforced plastics (GFRP) bolts under blasting dynamic load, this study establishes a numerical model of full-length mortar anchored GFRP bolt using FLAC^3D software. The stress and failure characteristics of the bolt, bolt-mortar interface, and mortar-rock interface under pre-tension static load and blasting dynamic load are investigated, and the influences of dynamic load intensity, surrounding rock strength and mortar strength on bolt stress are analyzed. The reliability of the research results is verified in comparison with the existing experimental results. The results show that the maximum axial stress of GFRP bolt increases linearly with the increase of dynamic load intensity and decreases with the increase of surrounding rock or mortar strength, and the axial stress distribution of GFRP bolt is more concentrated than that of metal bolt. The shear stress of the two interfaces increases rapidly to the peak along the bolt, and then decreases to zero. After that, the interface with relatively weak bonding properties is debonded at first, and the shear stress at the debonding position decreases to the residual bond strength. Meanwhile, the peak shear stress shifts to the bottom of the hole. The shear stress distribution of GFRP bolt is more concentrated than that of metal bolt, and the peak position is more prominent. The greater the dynamic load intensity, the larger the debonding length and shear stress distribution of the failure interface, and the peak position of the shear stress of the undamaged interface will be transferred to the bottom of the hole. The greater the surrounding rock strength or the smaller the mortar strength, the more likely the shear failure occurs at the bolt-mortar interface; on the contrary, it is more likely to occur at the mortar-rock interface.

In this paper, the seismic responses of inclined liquefaction site-pile-structure system under near-field pulsed and non-pulsed ground motion were studied. Firstly, a finite element model of inclined liquefaction site-pile-structure system was established based on the finite element platform OpenSees, and the reliability of the numerical model was verified by comparing with the results of large-scale shaking table test. On this basis, the finite element model of typical inclined liquefaction site-pile-structure system was established, the seismic response laws of the system under near-field pulsed and non-pulsed ground motion were compared and analyzed, and the correlation analysis between ground motion intensity index and seismic response was carried out. The results show that compared with the non-pulsed ground motion, the residual displacement of the site increases obviously under the near-field pulsed ground motion, especially at the top of the loose sand layer, which increases by about 40%. Pile bending moment and pile residual displacement increase significantly, especially at the pile top. The correlation between the ground motion intensity index and the response under non-pulsed ground motion is higher than that under pulsed ground motion. Three strength parameters with good correlation, i.e. PGV/PGA (peak ground velocity/ peak ground acceleration), AI (Arias intensity) and T_m(mean period), can indicate the influence of ground motion on the failure response of liquefaction site-pile-structure system to a certain extent.

Jacket foundations are considered as the most promising solution for offshore wind turbines (OWTs), due to its large lateral stiffness and adaptability to different marine environments. Nowadays, the study on jacket foundations mainly focuses on the bearing and deformation behavior, with significantly less attention being paid to their fatigue damage. Besides, the actual pile-soil interaction is always ignored when assessing the fatigue damage or fatigue life of the foundations for OWTs. This paper presents the development of a three-dimensional numerical model of tetrapod piled jacket foundations, with the interaction between the corner piles and the soil being implemented into it. The numerical model is validated against centrifuge tests. Then, a time-domain fatigue analysis method for OWT jacket foundations is proposed. The influences of pile-soil interaction, wind-wave load coupling effect, and lateral loading direction on the fatigue damage of the jacket are examined in detail. Main findings include: (1) Neglecting the pile-soil interaction will yield significant underestimation, i.e., approximately 40%, of the fatigue damage of OWT jackets, and this becomes even more discernible at the lower oblique support joints, exhibiting an underestimation up to 90% in certain scenarios. (2) Wind load plays a dominant role in determining fatigue damage of the jacket, and the proportion of fatigue damage caused by winds is much greater than that caused by waves. (3) The lateral wind-wave loading along the diagonal direction develops a greater displacement of the jacket than that along the orthogonal direction. However, the corresponding fatigue damage is relatively small, about 70% of that along the orthogonal direction.