With the development of deep rock mass engineering, the problem of rotation of principal stress axes due to excavation activity has been of great concern. Firstly, this paper analyzes the problem of rotation of principal stress axes in rock mass engineering and its effect on the stability of surrounding rock mass. Secondly, the developing of the soil hollow cylinder apparatus and its characteristic are introduced, and the critical problem of design of rock hollow cylinder apparatus is pointed out from two aspects, load method and specimen geometry. In addition, compared with soil hollow cylinder apparatus, a new torque applied technology of rock hollow cylinder apparatus and specimen geometry has been proposed. Finally, the research results of the constitutive model of soil under rotation of principal stress axes are summarized, the constitutive relation of rock under rotation of principal stress axes is also prospected. This paper provides reference and guidance for the design of hollow cylinder apparatus and relevant theoretical study.

To explore the mechanisms of jacked pile and CPTU penetration, centrifugal model tests are conducted to investigate the variations of earth pressure, excess pore pressure and penetration resistance in the process of jacked pile and CPTU penetration. Meanwhile the installation process of jacked pile and CPTU is regarded as a series of continuous sphere cavity penetration. The cone resistance, the cone side resistance and the excess pore pressure are predicted based on spherical cavity expansion theory. The comparisons between centrifugal model test data and theoretical predictions show that the excess pore pressure and the soil stress around the pile increase with the increase of penetration depth. The excess pore pressure and the soil pressure reach the maximum when the pile tip arrives the measured point. The cone resistance, the cone side friction and the excess pore pressure show linear relationship with the cone penetration depth in saturated clays. The calculated results show a good agreement with the test results, which presents the mechanisms of jacked pile and CPTU penetration well.

Based on a self-developed shear-seepage coupling testing device for coal-rock, experimental tests were conducted on coal-rock specimens in the process of CO2 injection to investigate shear-seepage of coupled properties. Meanwhile, the sheared fracture surfaces from different normal stresses were scanned, and their characteristics were analysed through two-dimensional (2D) and three-dimensional (3D) statistical parameters. The results show that the shear failure of coal-rock specimen is mainly plastic damage under the condition of constant gas pressure. With the increase of normal stress, the peak shear stress of coal-rock increases as well, and the peak shear stress and normal stress of coal-rock approximate a linear relationship. The normal stress is one of the influence factors of sheared fracture surface characteristics. Under the condition of constant gas pressure, when the normal stress increased, 2D and 3D statistical parameters generally showed a trend of decrease, in other words, the sheared fracture roughness and prominence of the coal-rock decreased with the increase of normal stress. Through analysis on coal-rock shearing process, the normal stress can affect the shape and seepage characteristics of sheared fracture surface of the specimen by influencing the extension area of the shear crack.

The drilling with pre-stressed concrete pile cased (DPC) pile was newly developed as engineering foundations, and engineers have accepted the designing method in practice. However, several key problems for ultimate shaft frictional resistance still need to be specifically investigated due to insufficient information in the literature. First, the pile-soil interface failure mode is unclear due to its multiple structural layers. Second, the grouting reinforcement mechanism needs to be specified. And last, the values of ultimate shaft frictional resistance need to be determined. This paper presents an experimental investigation on the physical mechanism and failure model on DPC pile-soil interface, which are limited thickness and three structural layers. For this purpose, a series of large-scale direct shear tests were performed. Fourteen large-scale direct shear tests on interfaces were carried out, including eight non-grouted DPC pile-soil interfaces and six grouted DPC pile-soil interfaces. The silt-clay and moderate dense sand were tested as the typical stratum around the DPC pile. The experimental results indicates the shear surface located on the interface between the grouting concrete and the boundary soil. Using the linear Mohr-Coulomb envelopes, the friction angles, and cohesion are obtained for the non-grouted DPC pile-soil interfaces and the grouted DPC pile-soil interfaces. Moreover, the ultimate shaft frictional resistance determined for the tested specimens are obtained. Values of ultimate shaft frictional resistance determined using grouted DPC pile-soil interfaces are compared with those determined non-grouted DPC pile-soil interfaces. Finally, the grouting reinforcement mechanism on DPC pile-soil interface was specified. The findings of this study should lead to a better understanding of the designing and construction of DPC pile at in-situ sites.

To investigate the mechanism of soil plug during jacked pipe pile in clay, coupled Eulerian- Lagrangian(CEL) approach is used to simulate the large deformation penetration process. On the basis of verifying the influence of mesh density on the accuracy of numerical calculation, the evolution and mechanism of soil plug is presented with soil velocity fields inside and outside pipe pile, stress field, height of soil plug and incremental filling ratio. The influence of friction coefficient, the soft or hard soil interlayer on the formation of soil plug is discussed. The numerical simulation results are verified by centrifuge experiments and theoretical calculations. The computational results show that the evolution of soil plug is mainly divided into three stages including the upwelling stage, the transition stage and the downward sliding stage during the penetration of pipe pile. As the pile jacking, the continuous concave plastic arch forms at the axis under the pile tip, where the vertical stress reaches 7-8 times the undrained shear strength indicates the initial formation of soil plug. With the increase of friction coefficient between pile and soil and so as the pile diameter, the plugging effect enhances. The soft or hard soil interlayer can significantly affect the plugging effect. Pipe pile is prone to a fully plugged mode when the hard soil is over the soft soil, while the hard soil is prone to squeeze into pile when the soft soil is over the hard soil.

The solution of a rectangular thin plate model of working face is directly affected by its boundary conditions. Its analytical solution can be solved only by a semi-analytical method under the conditions of two opposite edges simply supported without a unified and reasonable theoretical method. Due to the advance of working face, the thin plate bending model can generate three boundaries, i.e. fully clamped, two opposite edges simply supported and fully simply supported. From the governing equations of the thin plate bending problem, the basic mechanical parameters are used as the dual variables to construct the Hamilton system, giving the symplectic solutions rationally using the symplectic geometry method. This approach can deal with the above mentioned different boundary conditions with a unified analytical solution. With the characteristics of symplectic orthogonality for all eigenvectors, the proposed algorithm is proved to be stable and convergent fast. The case study shows that the calculation is correct, which provides a new way for the main roof weighting calculation.

With respect to the swelling of anhydrite rock surrounding the tunnel under the effect of pressured water, a multifunctional swelling test apparatus was designed and manufactured. Axial swelling strains and the lateral swelling stresses of anhydrite samples under different hydraulic pressures were tested by this test apparatus and the MTS815 rock mechanics testing system. Then the water-bearing state and mechanical performance of these samples were tested. The test results showed that when the hydraulic pressure became higher, the maximal axial swelling strain and lateral swelling stress became higher, meanwhile the activation time of swelling became shorter. With the increase of hydraulic pressure, water absorption by anhydrite increased, and the water in crystalline anhydrite phase gradually increased to maximum water absorption, while the free water gradually decreased and finally dropped to near zero. The elasticity modulus and peak strength increased with the increase of water hydraulic pressure. These indicated that the hydraulic pressure was able to increase the swelling behavior, promote the activation of swelling, influence the water conditions, and strengthen the mechanical performance. Based on the test results, the water absorption function with consideration of pressured water was established, and the swelling constitutive model in humidity stress field theory was improved.

Model tests in laboratory were conducted to investigate the behaviors of reinforced earth retaining wall of block face under the loading on the top of the retaining wall. In the model test, four conditions were simulated. The results show that, the vertical earth pressure on the upper part of retaining wall increases rapidly and the maximum vertical earth pressure of each layer moves from the lower part of the loading point to the face of wall. When the load at the top of wall exceeds 130 kPa, due to uneven settlement, the wall surface corresponding to fifth layer of reinforcement tends to shrink inward, the maximum horizontal displacement of wall surface is located in 7/10 of the wall height. In the whole loading stage, the overall strain value of the reinforcement does not increase significantly and the actual strain value of the reinforcement is much smaller than the maximum design strain of the reinforcement. Both increasing the number of the reinforced layers in the retaining wall and increasing the length of reinforcement can improve the performance of the retaining walls. However, the effect of increasing the number of the reinforced layers is better than that of increasing the length of reinforcement. The use of waste tire instead of unidirectional geogrid can effectively improve the overall performance of retaining walls, disperse additional stress caused by overloading, reduce the horizontal displacement of the face of wall and the vertical displacement at the top of the wall.

The deformation measurement system was built with a CCD camera and a high-speed camera. Experimental tests were conducted on rectangular granite specimens with pre-cracks of mode-I under different loading rates. The speckle images captured in the process of experiments were analysed by the digital speckle correlation method (DSCM). We studied the speed of crack growth, crack-tip opening displacement, crack tip opening angle and crack stress intensity factor (SIF) of granite specimens with pre-cracks of mode-I under different loading rates. The initial speed of the crack growth of granite specimens increases linearly with the loading rate. With the increase of loading rate, the maximum speed crack increased rapidly at first and then slowly. At different loading rates, the crack-tip opening displacement of granite specimens showed three stages of evolution: nonlinear slow growth，rapid growth, and linear growth. In the linear growth stage, the slope of the curve increased with the increase of loading rate. With the increase of loading rate, the initial crack opening angle of granite specimen decreases gradually and then stabilises to 0.1°～0.13°. Before the initial fracture of granite specimens, the SIF value increased exponentially with the increase of loading rate.

Flows of pore water in saturated soft soils would deviate from Darcy’s flow law under low hydraulic gradient, which is non Darcy seepage mode. Supposing that the velocity of flow is exponentially related to the hydraulic gradient, the analytical solutions of non-Darcy seepage field in shallow single-hole and double-hole circular tunnels are derived by using the principle of mirror method. The analytical solutions of non-Darcian seepage and Darcian seepage for shallow circular tunnels are compared and verified by case analysis. The influence of non-Darcy seepage index and the seepage coefficient ratio of soil to lining on tunnel seepage field is also discussed. The results show that the non-Darcy seepage index and the ratio of seepage coefficient have great influence on tunnel seepage and pore pressure of surrounding soil. With the increase of seepage index, the loss of water head in soil is accelerated, and the pore pressure and seepage flow around the tunnel are gradually reduced. With the increase of the ratio of seepage coefficient between soil and lining, the drainage capacity of lining is enhanced, the seepage flow of tunnel is gradually increased, and the pore pressure around the tunnel decreases more.

The slab failure of rock mass is one of the common destruction phenomena in underground engineering. Based upon equivalent rock mass (ERM) technique, the fracture (parallel to the free surface) and rock block are represented by the smooth-joint model and bonded-particle model, respectively. The strength, crack distribution, crack initiation and evolution, and slab failure formation mechanism of fractured rock mass under one side restriction (OSR) compression are studied by comparing experimental results and numerical modelling. With the increase in the number of fracture, the OSR compressive strength of rock mass exhibits linear decreasing. Under OSR loading, five types of cracks generated in rock mass can be identified. The type-Ⅰ and type-Ⅱ (shear) cracks are caused by the loading end effect, while the type-Ⅲ, type-Ⅳ and type-Ⅴ cracks are caused by slab failure of rock mass. In the process of slab failure, type-Ⅴ cracks which causes rock mass spalling at the unconfined boundary first occurs. Then type-Ⅳ cracks result in the propagation of original fractures generate. Finally, the type-Ⅲ cracks generate, which indicates that the slab failure phenomenon is developing towards the interior of rock mass. In various specimens under OSR loading, the relationship between the acoustic emission (AE) quantities and crack magnitude approximates Gaussian function distribution.

To study the effect of cement content on the improvement of aeolian sand and its dynamic characteristics under different negative temperatures and frequencies, uniaxial, dynamic cyclic loading and creep tests were carried out. 7 days and 28 days uniaxial compressive strengths of improved aeolian sand with different cement contents were obtained to determine the optimal amount of cement by combining with frost heave variation with time. Hysteretic curve of different times was measured through the dynamic cycle test. Dynamic modulus attenuation, dynamic evolution damage and damping ratio evolution law of improved aeolian sand were determined. Parameters in the Nishihara model describing the rheological characteristics of hysteretic curves were determined by rheological experiments. Evolution equation before and after yielding hysteresis curve was established and time division standard was given for each loop. The electron microscope observation showed that the flocculent connection structure between the sand particles increased the bond strength between the particles and effectively inhibited the development of dynamic damage. The negative exponential model can be used to describe the dynamic modulus attenuation characteristics of the improved aeolian sand, and the damping ratio increased with the loading times, and finally tended to a certain value. Nishihara model considering the influence of was derived. The loading equation of the hysteresis curve was established based on the actual loading characteristics. The least squares principle can be used to identify rheological parameters.

Pit-in-pit usually occurs in pit engineering practices, making the soil at the bottom of the foundation pit become a limited soil. Thus, Rankine earth pressure theory based on conventional semi-infinite space is not applicable for calculating the earth pressure acting on retaining structure in case of pit-in-pit. This also makes it difficult to correctly estimate the excavation depth of the foundation pit for design. Based on the limited equilibrium theory and the hypothesis of planar sliding plane, four formulas for calculating passive earth pressure are deduced with cohesion force and different shapes of sliding soil mass taken into account, and corresponding mathematical expressions of shear failure angles are given. By example, the values and variation trends of passive earth pressures for different internal pit positions are calculated. The results show that the angle of shear failure is a variate that are related to the friction angle of soil, the cohesion of soil, the depth, the size and the position of inner pit. The inner pit will lead to decrease of the passive earth pressure acting on retaining structure, and there is a most unfavorable position for inner pit, where the value of passive earth pressure is minimum. This study provides a theoretical basis for the calculation of passive earth pressure in foundation pit design.

In consequence of froth breaking, there is a caved problem about air-foam treated lightweight soil after casting indoor and in-site. To analyze and solve the problem, a series of samples preparation experiments on sand were conducted on air-foam treated lightweight soil with three specific densities under different conditions, i.e., temperature, humidity and sand mixing or not. Meanwhile, the height and strength of samples were measured. Based on the thermodynamics theory, the effects of three factors on the preparation and strength of the samples were analyzed using experimental data. The results show that temperature difference is the decisive factor affecting the shrinkage and expansion of air-foam treated lightweight soil. The temperature difference impact greatly in sample of high bubble content. Humidity and sand mixing barely affect sample preparation while the mold unloading time of samples will be extended and the integrity of samples will be affected when ambient humidity becomes higher. Early strength of samples is influenced by temperature and temperature difference. temperature difference mainly affects the density of samples, and the temperature mainly affects the rate of hydration reaction of cement. The results of this study provide reference for the preparation and construction time of air-foam treated lightweight soil.

Temperature and pore pressure have significant effects on the mechanical behavior of hydrate-bearing sediment. It is critical to reveal and simulate the mechanical response of hydrate-bearing sediment under different temperatures and pore pressures for evaluating of stability of hydrate-bearing reservoirs. Through analyzing the influence of temperature and pore pressure on mechanical behavior, this paper introduces a new state parameter, referred to as ‘temperature and pore pressure condition parameter’, to describe the influence of temperature and pore pressure on mechanical behavior of hydrate-bearing sediment. Furthermore, a constitutive model is developed in the framework of continuum damage mechanics theory. This model assumes that the strength of micro-element of hydrate-bearing sediment follows the Weibull distribution, and the strength of micro-element can be described from the Drucker-Prager strength criterion. In addition, a parameter calibration method is proposed for the proposed model. The model is checked and validated by a series of triaxial compression test results of hydrate sediments under different temperature and pressure conditions. The comparison between predicated results and experimental data shows that the proposed model has capacity of simulating the stress-strain relationship of hydrate-bearing sediment. Moreover, the proposed model is able to address the influences of temperature and pore pressure on mechanical properties of hydrate-bearing sediments.

The self-boring pressure meter tests (SBPT) and seismic dilatometer tests (SDMT) are conducted in the granite residual soil to obtain the in-situ shear modulus and shear modulus degradation. The shear modulus decay by shear strain was measured by resonant column tests(RCT), and the approximate in-situ shear modulus decay curves were deduced from the data of RCT and SDMT to analyze the suitability of the proposed approach. The results show that complete in-situ shear modulus degradation curves can be directly derived using‘Stokeo model’ from the combination of initial shear modulus G0 measured by SDMT and G- decay curve measured by SBPT. By comparing the results of resonant column tests under different stress states, a deduction is found in laboratory process of sampling, transporting and preparing of soil, causing unrecoverable damage in the soil. By comparing the results of the forecasting method by RCT and SDMT, the initial stage of the G- decay curves are found consistent with the proposed method. However, the stiffness attenuation rates of the proposed method are lower. After a comprehensive comparison of the three methods, the proposed approach proved to show a good suitability in the stiffness analysis of structural soil. The result provides an important reference for the selection of engineering parameters in relative soil layers.

The surrounding rock mass and the lining are regarded as orthotropic and isotropic medium, respectively, with continuous, homogeneous and linear-elastic properties. The effects of support delay and water pressure during tunnel operation are also taken into consideration. Based on the above assumptions and the power series method of complex variable theory, the analytical method and solution for the arbitrary-shaped hydraulic tunnel that is buried in the orthotropic rock mass is established. Taking an inverted U-shaped tunnel for example, the analytical solution is perfectly satisfied with the stress boundary condition on the inner boundary of lining and stress boundary condition on the contact interface between lining and surrounding rock mass. Besides, the displacement continuity condition along the contact boundary can also be accurately satisfied. The numerical solution simulated by finite element software ANSYS is introduced to be compared with the analytical solution. The comparison indicates that the analytical results are in good agreement with the numerical analysis. With the presented analytical method for the problem, the influences of anisotropic degree of surrounding rock mass, the angle of elastic symmetric plane and the water pressure are discussed based on investigating the stress and displacement distributions of lining and surrounding rock mass.

The stress interaction among multiple fractures in a horizontal well causes non-uniform distribution of fracturing fluid, and then affects the geometry of hydraulic fracture. Boundary element method is used to determine the rock deformation under the hydraulic pressure of fracturing fluid, and the Poiseuille flow of the power-law fluid is conducted to calculate the flow field of fracturing fluid inside the hydraulic fracture. Considering the stress interaction and fluid distribution, a fluid-solid coupled model is proposed to illustrate the hydraulic fracture propagation in multiple simultaneous fracturing of a horizontal well. This model can simulate the fracture geometry and stress interaction, and this model figures out the mechanism of fracturing fluid distribution and fracture competition. During the process of multiple fracture propagation, the fracturing fluid is not equally distributed into each fracture. The interior fracture with the lowest width contains the least amount of fracturing fluid . The interior fracture stops growth and closes after reaching certain length due to stress interaction.

The shape of grains is one of the main factors influencing the structural characteristics of pores in filled breakstone aggregate, subsequently affecting its mechanical and hydraulic properties. Two sets of limestone gravel aggregate with diameters of 2-5 mm and 5-10 mm were investigated using CT scanning technique and graphic software, from which the pore appearance images of the porous media were acquired. Two sets of glass balls that were with the same granular compositions and porosity as those of the two sets of limestone gravel aggregate were made, and the glass ball sets were compared with the limestone gravel aggregate sets in the test. The threshold value of 0.5 mm opening was applied to the segmentation of pore structure. Three section images of each sample were selected so as to quantitatively analyze pore characteristics by means of the average values of quantitative parameters as the research object. The results show that the pore shape in filled glass ball is more complex than that in filled breakstone. The connectivity degree of pore system in filled breakstone is also less than that in glass ball.

The acoustic characteristics of rock mass are closely related to the stress state and the degree of damage of rock mass. It is an effective engineering measure to analyze the stress state and evaluate the stability of rock mass through the change of acoustic properties of rock mass. The axial and transverse ultrasonic synchronous tests during the uniaxial loading tests of sandstone were carried out and the evolution laws of wave velocity with stress in different directions during loading were obtained. Experimental results showed that the wave velocity in axial direction increased with the increase of stress, while the wave velocity increased first and then decreased in transversal direction. Since the acoustic test results were different in different directions, the influence of fracture direction on ultrasonic velocity was verified by the acoustic test of gypsum specimens with pre-existing cracks with different inclinations. The wave velocity achieved the maximum value when the fissure direction was consistent with the direction of sound wave propagation, and the wave velocity was minimum when the fissure direction was perpendicular to the propagation direction. In addition, in order to analyze the correlation between the wave velocity and stress state, the relationship between wave velocity and volumetric strain was established. The results showed that the average velocity gradually increased with the increase of volumetric strain, and reached the maximum value at the peak of volumetric strain, and then began to decrease with the decline of volumetric strain. According to the relationship between stress and wave velocity during the loading process, an exponential fitting formula of stress and wave velocity was obtained. As a result, the wave velocity obtained by field test could be used to predict the stress range of the surrounding rock and evaluate the stability of the rock mass.

As a flexible retaining structure, reinforced earth retaining wall has superior seismic performance comparing to traditional gravity retaining wall. the dynamic response of the structure under earthquake and other dynamic loads is related to its natural frequency, especially its minimum fundamental frequency. In this study, the reinforced soil retaining wall with full-high rigid facing is investigated. A method is developed to determine the natural frequencies by using elastic foundation beam model and linear spring model to represent panel, fill and reinforcement. Fundamental frequencies calculated by the proposed method agree well with those calculated by the traditional Rayleigh’s method for gravity retaining walls and reinforced soil retaining walls, respectively. Parametric analyses show that laying reinforcements in the backfill can increase the fundamental frequencies. the reinforcement length and reinforcement-soil interface friction coefficient trivially influence the fundamental frequencies of reinforced soil retaining walls. the effects of reinforcement density on fundamental frequencies decrease gradually as reinforcement vertical spacing increases. the values of fundamental frequencies decrease firstly and then increase as facing width increases. fundamental frequencies tend to constant values with decreasing panel facing modulus.

The large deformation is one of tunnel disasters that cannot be completely avoided. When the tunnel is not properly treated after large deformation, it is likely to cause secondary disasters such as arches multiple replacement or tunnel collapses. With the self-designed excavation device and the internal displacement-measuring device for surrounding rock, a physical model test system was used to investigate the tunnel progressive failure, deformation and stress variation during tunnel excavation and buried depth increasing, thus revealing the key supporting parts to prevent large deformation of tunnels. Furthermore, the failure characteristics of the primary lining during large deformation treatment was also studied, and the supporting measures for large deformation treatment were defined. The results are summarised as follows. During the large deformation of the tunnel, the deformation of the vault and the arch bottom are larger than that of the arch waist and arch foot, and the difference increases gradually with the increase of buried depth. The radial and tangential stresses of the vault of tunnel decrease during the large deformation, while the tangential stress of the arch foot increases sharply. When the deformed arch frame is replaced, there may be tension failure at the lining arch near the replacement position, and shear failure at the lining arch waist. Therefore, it is necessary to retain the temporary steel support when the large deformation is observed, and the bottom horizontal support or the temporary inverted arch should be added. This study is helpful to obtain the key prevention and control location during and after the large deformation, which provides guidance and reference for the prevention and safe treatment of large deformation.

Fissure is one of the most important factors affecting mechanical properties and failure modes of rocks. Based on experimental results, digital photographic technology and acoustic emission (AE) monitoring to investigate tensile strength and failure modes of intact, and containing different crack angles ? single, different angles between rock bridge and crack ? double filled and unfilled pre-existing fissures rock-like material Brazilian disc samples. The effects on final failure mechanisms of samples with the change of filling condition, ? and ? are also discussed. The results show that the tensile strength of single pre-existing fissure filled samples decreases firstly and then increases and finally decreases with the increase of ?. The wing cracks initiate from the tip of the pre-fissure, and then cut-through, finally the samples show brittle destruction. The tensile strength of those samples containing two pre-existing fissures decreases firstly and then increases with the increase of ?. The propagation and evolution of wing cracks initiated from two pre-existing fissures and far-field cracks lead to the the destruction of samples. The tensile strength of filled samples is higher than unfilled samples obviously, when the corresponding values of ? and ? are the same. Whether the pre-existing fissure filling has a great impact on the crack propagation time and the number of cracks. At the initial loading stage, the cumulative acoustic emission counts are relatively stable. When the surface of samples initiated cracks initially or the tensile strength increases to the peak, the acoustic emission activities become extremely active. When the corresponding values of ? and ? are the same, the fluctuation of acoustic emission counts of the filled samples is more violent than that of the unfilled.

Natural gas hydrate is anticipated to be a potential energy resource. During the mining process from methane hydrate reservoir, the decomposition of methane hydrate can cause risks threatening the safety of engineering and geology. A series of shear tests under different confining pressure conditions was carried out using self-developed low-temperature and high-pressure triaxial apparatus to investigate the mechanical properties of sediments containing methane hydrate under different conditions. The experiment combined with conventional triaxial shear method and multi-stage loading tests was focused on the hydrate decomposition process. It was found that the strength of the sediments was significantly increased due to the presence of hydrates. In the process of depressurization, the strength of sediments was found to be affected by the variations of effective confining pressure and the hydrate saturation. At the early stage, due to the decrease of pore pressure, the effective confining pressure significantly increased, and the shear strength increased as well. In the later stage, the shear strength of samples decreased under high effective confining pressure due to the decrease of hydrate content. The effective confining pressure had a great effect on volumetric strain of the hydrate sediments, and higher effective confining pressure led to obvious shear shrinkage.

At present, the research of subgrade lateral deformation mostly focuses on field test or numerical simulation, but few on theoretical calculations and analysis. To deeply investigate the lateral deformation of ground, the analytical expressions of the lateral deformation of ground under the infinite long line load, uniform strip load, triangular distribution strip load and trapezoidal distribution strip load are derived on the basis of elasticity solution of J. V. Boussinesq and plane strain problem. Compared with the finite element numerical results, the analytical solutions are verified. On this basis, the influence of the Poisson's ratio of soil, the width of embankment and slope of embankment on the lateral deformation field are mainly analyzed. The result of calculation shows that the lateral deformation of soil on the same vertical line under the embankment load is bow-shaped along the depth and the maximum deformation is at a depth below the ground surface. The Poisson's ratio has a significant effect on the lateral deformation of ground: in the shallow soil near the slop foot, the direction of lateral deformation of ground is gradually transferred from the inside embankment to the outside because of the increase of Poisson's ratio of soil; In addition, the higher the poisson ratio is, the more the horizontal displacement of the soil will have, and the maximum displacement moves up gradually. The results can provide some theoretical support for the deformation analysis and stability monitoring of similar embankment engineering.

Moisture-temperature coupling control equations are introduced into the heat and mass balance equation for the consideration of phase change. Smoothed Particle Hydrodynamics method is used to solve the equations, where both moisture transfer driven by temperature and heat transfer by water are considered. While the phase transition process is ignored, the model can directly degrade to a generic coupled model of moisture and heat at normal temperature. Therefore it can be used to study freezing and thawing cycle of frozen soils (e. g., seasonal or daily variation in ground temperature). To validate the model, the solution of a semi-infinite unsaturated porous medium subject to a frozen wall is studied. The comparison result shows that the process of moisture migration impacts on heat transfer, with the degree intensifying as time. In addition, the model is used in an unsaturated subgrade with periodical temperature boundary. The temperature and moisture distribution of subgrade are different when considering different solar radiation in sunny and shady slope. The evolution of water content field and thermal field is given at the end. Since the algorithm can basically reflect the real physical process of phase change, it’s credible to study other problems in the field of frozen soil by SPH method.

Taking the skarn specimens extracted near the deep ore drift in Dongguashan copper mine as the research object, a modified split Hopkinson pressure bar (SHPB) system was used to investigate dynamic characteristics of deep skarn under the combined action of high axial pressure and confining pressure under frequent dynamic disturbance. The results show that internal microcracks in specimens were closed due to high pre-loaded axial pressure and confining pressure, which results in no concave phenomenon at the initial stage of dynamic stress-strain curves during the impact load, i.e. no compaction stage. After peak stress, the phenomenon of reduced strain rebound occurs when the elastic force stored in the specimen is greater than the impact load during unloading. Otherwise, the compressive strain increases until the specimen breaks. The accumulative impact disturbance times rises while with increasing confining pressure, and it reduces due to the increased axial compression. Furthermore, the growth or decreasing rate is influenced by the confining pressure and axial pressure. The dynamic deformation modulus and peak stress decrease with the increase of impact disturbance times, but the maximum strain and dynamic peak strain grow up. Moreover, the average of dynamic strength has been cut down with the rising axial pressure, but it decreases firstly and then increases with the increase of confining pressure. The dynamic deformation of rock is accompanied by elastic deformation, the strain rebound generally increases firstly and then decreases with increasing impact disturbance times on the whole.

Based on the elasto-plastic solution to the cavity expansion for jacked piles problems in natural saturated clay, an analytical solution to the excess pore water dissipation around the pile is derived with aid of the stress state of the soil immediately after pile installation. On this basis, a theoretical method for calculating the time-dependent bearing capacity of the jacked pile in natural saturated soil is presented by considering the soil relaxation effects on the soil around the pile during reconsolidation of the soil. The theoretical method is verified by the published centrifuge model test and field test respectively, and the variation of the bearing capacity of the jacked pile with consolidation time is studied in detail. The results show that the proposed theoretical method can be applied to reasonably predict the time-depended bearing capacity of the jacked pile, because the in-situ mechanical property of the soil, the installation effect, as well as the reconsolidation effect after installation are properly considered by the proposed method. The results of this paper have great theoretical significance in proper determining the bearing capacity of a jacked pile in clay.

Based on the linear distribution model of the side friction of pile, the theoretical value of pile side friction will far exceed the actual value when the pile is comparatively long. Consequently, the calculation of neutral plane depth and pile-soil stress ratio of the rigid pile composite foundation is quite different from the reality and needs modification. Therefore, the distribution of pile side friction was assumed to be piecewise linear, with influence of negative friction and behavior of pile upward and downward puncture deformation taken into consideration. According to the cushion-pile-soil deformation compatibility condition, a theoretical calculation model was created to obtain the neutral plane depth and pile-soil stress ratio on the top surface and neutral surface of pile. Finally, the calculation method was verified by field measurement as well as model test results of the rigid pile composite foundation, and theoretical values agreed well with the experimental data.

To analyze the stability of a thick overburden slope under the effect of transient pore pressure, a method for calculating transient pore pressure was developed based on the formation and distribution of transient saturated area. A modification on Bishop limited equilibrium method was made with the consideration of the influence of transient pore pressure on stability, and a routine using VB was programmed. The proposed procedure was employed to analyze stability of an overburden slope along Ru-Chen expressway in two calculation conditions whether the transient water pressure was considered or not. The results show that the greater the thickness of transient saturated zone, the smaller the safety factor of slope. The lower safety factor exists in slope affected by transient water pressure under the same thickness of transient saturated zone. With the expansion of transient saturation zone, the decrease rate of slope safety factor keeps increasing, up to 39.79%. The potential sliding surface will be tangent to the wetting front and becomes deeper if the thickness of transient saturated area is over 4 meters.

To resolve the deficiencies of the calculation model for anti-sliding pile with prestressed anchor cable, a calculation model that considers prestress and landslide thrusting stage according to the actual construction process and loading conditions of anti-slide pile with anchor cable was established. The impact of prestress loss of anchor cable and displacement of pile in construction stage on displacement compatibility of pile-anchor cable is considered. Meanwhile, the method to calculate the internal forces and deformation of anti-sliding pile with prestressed anchor cable was established based on weighted residual method. A calculation procedure of anti-slide pile with anchor cable was created based on Matlab platform, which was used to analyze engineering examples. The results were compared with those obtained from the conventional calculation model. It is shown that the improved model can incorporate the effect of the special process of anti-slide pile with anchor cable bearing active load in the prestressed stage on displacement compatibility of pile-anchor cable. Also, the improved model properly takes into account the prestress loss of upper anchor cable, which is caused by the lower anchor cable in prestressed stage. Furthermore, compared with the conventional calculation model, the improved model is more in line with the engineering practice.

The key to predicting the water inrush or not from karst collapsed column is to accurately describe mechanical properties and damage depth of rock under the mining influence. A modified parallel rod model is established based on the dissipative structure theory, and the damage evolution equation is obtained according to the data of acoustic emission (AE) and stress-strain curve. By combining Drucker-Prager (D-P) yield criterion with the corresponding non-association flow law, the D-P elastic-plastic double scalar stochastic damage constitutive model was established. The damage law of WDX40 karst collapse column in 51510 working face of Chengzhuang coal mine is studied by the newly established constitutive model. The main conclusions are as follows. Rock material can be regarded as a structure which has the characteristic of non-uniform energy dissipation, showing the "elastic modulus degradation" under cyclic loading. The D-P stochastic damage model can well reflect the "nonlinear" and "random" characteristics of rock, and the rationality of the stress-strain curve is calculated in the probability sense. The depth of compressive damage in the karst collapsed column shows an asymmetric inverted saddle-shaped distribution, and the depth in the side-wall away from the starting cut is the largest, which is 1.75 times the depth of complete floor

The spatial distribution of wrinkles in a wrinkle network is the basis for hydraulic connectivity analysis and leakage evaluation. The rules describing the spatial distribution of wrinkle remain to be investigated. Based on a typical wrinkle network from the bottom of a site, the probability distribution types for the geometric parameters of wrinkle were analyzed using K-S test method, and the statistics of the geometric parameters were obtained. A method for evaluating leakage through a wrinkle network was proposed by generating stochastic wrinkle networks based on the statistics of the geometric parameters of wrinkles, and choosing a characteristic value from the calculated leakages through the stochastic wrinkle networks to represent the leakage from the actual wrinkle network. The proposed method was then applied to representative cases. The results indicate that the statistical analysis for a typical wrinkle network from the bottom of a site can provide an important reference for the determination of the distribution types and statistics of various geometric parameters of wrinkles in similar sites. The mean value of leakages calculated from the stochastically generated wrinkle network can be used as a characteristic value to conservatively represent the leakage through the actual wrinkle network. Referring to the case considered in this study, after rich experience about the statistical law of geometric parameters of wrinkles has been accumulated, the aerial image of the whole wrinkle network is no longer needed in the design process of a site, thus providing new ideas for assessing leakage through sites with wrinkle network.

The hole digging foundation in the strong weathered rock mass is often considered as the main footing form of overhead transmission line project in mountainous area. This paper focused on the shaft foundations and the spread foundations embedded in the strong weathered rock mass. The load-displacement relationships of sixteen full-size foundations were obtained through a series of full-scale pull-out test in site. The test results show that the load-displacement curves of the foundation are similar as that of on the ground surface. They can be divided into three stages: initial straight-line phase, transition curve phase and ending straight line phase. The parameter M is selected to represent the percentage of displacement in the corresponding stage. The M values at different load-displacement stages are different. M decreases as the distance away from the foundation edge increases. A normalized load-displacement curve obeying the hyperbolic relation for the shaft foundations and the spread foundations was presented, and parameters Q/QL2 and s were conducted by curve fitting. The comparative analysis for the same footing under different foundation conditions such as loess and strong weathered rock shows that the parameter b value is negatively related to the shear strength parameters of soil or rock. The uplift bearing capacity of foundation in above two foundation conditions increases successively with shear strength parameter c and ?. Finally, comparative analysis of the predict results using hyperbolic model proposed in the paper and the field test results concludes that the evaluation model to load-displacement response above was available in engineering design.

Illegal temporary surface surcharge can induce additional ground stress, which may cause adverse effects or damage to the existing shield tunnel structure. However, current methods for predicting tunnel longitudinal responses to surface surcharge generally simplify the tunnel as an Euler-Bernoulli beam resting on Winkler foundation, which ignores the shearing deformation of shield tunnel and cannot consider effect of the burial depth on the coefficient of subgrade reaction. In this study, a new analytical method is developed to evaluate the existing shield tunnel deformation subjected to surface surcharge, considering the tunnel shearing effect and the burial depth of the tunnel. In the proposed method, the shield tunnel is treated as Timoshenko beam resting on Winkler foundation and the effect of burial depth of shield tunnel on the coefficient of subgrade reaction is considered. The effectiveness and feasibility of the proposed method is verified by a three-dimensional finite element analysis and measured data from a case. Parametric analysis indicates that a shallowly buried shield tunnel suffers large settlement when the distance between the load center and tunnel axis is relatively close. Increasing the equivalent bending stiffness and the coefficient of subgrade reaction would remarkably reduce tunnel settlement. The tunnel settlement was hardly sensitive to equivalent shear stiffness. However, increasing equivalent shear stiffness would significantly reduce the dislocation between the adjacent rings. In general, the proposed method will serve to be theory support for reasonably predicting the shield tunnel responses to surface surcharge.

To further investigate the two-dimensional consolidation mechanism of elastic saturated clay layers, Hansbo’s equation is introduced to describe the non-Darcy flow in the consolidation process. Accordingly, Biot’s two-dimensional consolidation equations are modified, and their numerical analyses are conducted by the finite element method based on the weighted residual method. In order to verify its validity, the numerical solution by the present method for the one-dimensional consolidation theory with the non-Darcy flow is compared with that by the finite volume method in literature. The effects of the parameters of Hansbo’s flow on the two-dimensional consolidation process are investigated. Numerical results indicate that the behaviour of Hansbo’s flow amplifies the Mandel-Cryer effect in the early phase of consolidation, that is, the peak value of pore water pressure based on Hansbo’s flow is greater than that based on the Darcy flow, while it needs more time to reach its peak based on Hansbo’s flow. Then Hansbo’s flow delays the dissipation of pore water pressure in the entire soil layers in the middle and later phase of consolidation, thereby hinders the development of settlement. The aforementioned effects on consolidation are more remarkable with the increase of Hansbo’s flow parameters.

Coarse grained soils have been widely used in engineering construction because of the advantages of high shear strength, easy compactability, small deformation and so on. There is still few research on dynamic properties of coarse grained soils, and theoretical research lagged behind the practices. In engineering designs, coarse grained soils were usually regarded as nonliquefiable soils. However, reports of after-earthquake surveys both in China and abroad show that coarse grained soils were serious liquefied. Therefore, under a given earthquake shaking, coarse grained soils with loose to moderate density have much potential to be liquefied. In this study, a Cyclic-mobility constitutive model, which can describe the cyclic-mobility and liquefaction characteristics of soils, was employed to discover the dynamic characteristics of coarse grained soils. On the basis of model parameters calibration, the different influencing factors which affecting dynamic properties of coarse grained soils were further studied, such as the friction between loading plates and specimen, gravitational force of specimen, cyclic shear ratio, confining pressure and loading frequency. The results show that Cyclic-mobility constitutive model can well reproduce the stress-strain characteristics of coarse grained soils under undrained triaxial test, gravitational field and the friction between loading plates and specimen have some effects on the laboratory element test results of coarse grained soils. Under lower confining pressure or higher cyclic shear stress ratio, the phenomenon of larger dynamic strain and the uneven distribution of dynamic strain inside samples will be more obvious. The distributions of volumetric strain inside samples are similar under lower loading frequencies, however, the combined action of high loading frequency and cyclic shear stress ratio will cause large damage on the sample. In the triaxial test, when studying the dynamic response of soil elements at different locations by triaxial test, attention should be paid to the arrangement of strain sensors, and the results should be considered comprehensively and corrected appropriatel.

Particle size is an important factor affecting macroscopic mechanical properties and computational efficiency of particle discrete element model. To account for the uncertainty of numerical results caused by the stochastic model, we study the effect of particle size using statistical methods. The result of the global test shows that the distribution position of mechanical parameters (i.e. peak strength, elastic modulus, Poisson's ratio and peak strain) and the failure characteristic parameter (bond failure rate) is significantly affected by the characteristic length ratio L/R. The coefficient of variation (CV) of these parameters increases as L/R decreases. The further multiple comparisons show that L/R has no significant effect on the distribution of these parameters as L/R≥125. When L/R≥79, the L/R has no significant effect on the distribution position of bond failure rate within the range of three adjacent ball radius levels. With the decrease of L/R, the damage degree of the model increases gradually, and the failure mode is changed from the whole shear failure to the unstable failure caused by the local damage, which makes the model lose the simulation effectiveness to rock materials. Finally, by combining the statistical results of mechanical parameters, the failure mode of the model and the computational efficiency, the L/R=200 should be suitable for the simulation of rock materials.

The equivalent discrete lumped mass model with multi-degree of freedom for the long-distance buried pipeline is proposed in this paper to evaluate the longitudinal dynamic of the pipeline. The proposed model is able to use the non-uniform excitation as the inputs by considering the traveling wave effect along the longitudinal direction of the pipeline. Moreover, the ideal elasto-plastic constitutive relation of the soil-spring model is used to simulate the slip effect between the pipe and the soil under seismic loading. Based on the longitudinal seismic displacement input, the displacements of the lumped-mass point and the relative displacement between pipe and soil can be obtained. Moreover, the internal force and deformation of the pipeline can be obtained by analyzing the force-displacement relation between every two lumped-mass points with the structural mechanics method. Finally, the accuracy and the efficiency of the proposed method is validated by comparing with the finite element analysis results of the continuous pipeline.

Infinite element is an effective artificial boundary, which can be utilised to address the elastic wave propagation problem. Based on the traditional dynamic infinite elements, we proposed a novel dynamic infinite element using a directional interpolation technique. The shape function of the element has been derived in detail. The totally analytical stiffness matrix related to the element is developed to improve the computation effectiveness. By using the infinite element mentioned above, Lamb problem with line source in an elastic medium is computed. The validity of the infinite element is verified by comparing the computed results with the analytical solution of displacements on the ground. The example analysis shows that the edge size in the finite element mesh should be less than 1/8 shear wavelength. The distance between an imposed source point and the mesh boundary should be equal to 5 times the shear wavelength. The amplitude attenuation coefficient in the infinite element has little effect on the calculation results, and thus a relatively small value is recommended.

In order to study stress field of plane and cylindrical waves in viscoelastic medium, the wave motion theories were selected and the stress at the wave front was established. A 3D constitutive model of viscoelastic medium was adopted to describe the dynamic behavior of the medium. The analysis presented herein extended application of wave motion theory in viscoelastic medium. Additionally, the parameter values of viscoelastic boundary were obtained referring to the classical solution in which the behavior of the medium was assumed as linear elastic. The application of the visco-elastic artificial boundary defined by the derivation of this paper was investigated using the finite element program calculated in frequency domain. The results show that high accuracy could be achieved using the artificial boundary established herein, compared with the classical visco-elastic boundary and the viscous boundary which are deduced in elastic medium. As a result of the constitutive model and parameter setting differences between the theoretical analysis in the present paper and the time domain finite element method, a conclusion could be reached that the present results could not be used in the time domain finite element method. Last but not least, a suggestion on further development of the present study has been given, combining the characteristics of time domain finite element method.

Accurate determination of random field parameters and correlation function is the key for probabilistic characterization of inherent spatial variability (ISV) of soil properties. Based on indirect cone penetration test (CPT) data, the conventional statistical methods and Bayesian approaches are compared for their validity of estimating random field parameters and correlation function of sand effective friction angle, φ'. The reason for the difference between these two methods is presented. Effects of sampling size of indirect CPT data on the accuracy of these two methods are also investigated. The results indicate that the conventional statistical methods don't consider the model uncertainty associated with the transformation model between φ' and CPT data, leading to a low rate of correct identification of true correlation function of φ', overestimating standard deviation and underestimating scale of fluctuation. However, the Bayesian approaches can take the model uncertainty into proper consideration and reasonably determine the random field parameters and correlation function of φ'. Model uncertainty should be carefully considered when indirect test data is used to characterize ISV of soil properties. In addition, the rate of correct identification of true correlation function of φ' estimated from Bayesian approaches improves and the uncertainties of random field parameters of φ' gradually decrease with increasing sampling size of indirect CPT data. It is suggested that large numbers of site observation data should be collected for improving the accuracy of probabilistically characterizing ISV of soil properties.

Under engineering loads such as traffic and seismic loads, granular soil is often subjected to complex 3D stress paths. Most existing constitutive theories and models for granular soil have been proposed based on results from laboratory tests conducted under relatively simple loading conditions. Therefore, the existing constitutive theories and models need to be verified under more realistic stress paths if they are to be applied to engineering practice. However, due to limitations in mechanical control, conducting laboratory tests with complex 3D loading is often very difficult or even impossible. In this study, a new numerical test method is proposed to provide the means to investigate the mechanical response of granular material under complex 3D loading. Spherical numerical specimens are adopted along with high precision stress-controlled boundaries, allowing for the application of any arbitrary stress state, forming the basis for complex 3D loading. Combined with a set of stress and strain measurement techniques, most existing physical tests can be quantitatively and numerically reproduced. Then, the 3D rotation of principal stress axes is achieved, exhibiting the advantages to use this method in the investigation of the mechanical response of granular material under complex 3D loading.

This paper provides a method of random generation of soil-rock mixture using digital imaging technology and rock form database technology. By MATLAB program, the boundary extraction, boundary smoothness and regularization to rock images can be obtained. After those process, the coordinate of block nodes and shape parameters needed for database construction can be extracted. The database of rock shape can be built with massive of rock samples coordinate and shape parameters. Under some constraint conditions, the data of specific rock samples could be selected from the database. After rotation and zooming, by the random distribution rule, the rock samples can be placed in a 2D space of specific shape. These are the circulation processes of model establishment. When the block proportion reached the setting limit, the soil-rock mixture model establishing finished. As to import the model into ABAQUS, a program using PYTHON language was used to realize the process. This random distribution method can take many parameters such as particle diameter, shape parameters and block proportion in consideration. It also avoids the fiction of rock shape in general soil-rock mixture random distribution and the difficulty of model data acquisition with digital imaging technology. This method provides a good way to study the general law of soil-rock mixture physical and mechanical properties.

To study the flow deformation characteristics of liquefied sand, the experiments on the fluid characteristics of liquefied sand were conducted. Based on the boundary layer theory, a test device for apparent viscosity of liquefied sand had been developed independently. The device, mainly composed of adjustable speed motor, motor speed regulator, cylindrical rotor, torque sensor and other components, can effectively measure the apparent viscosity of liquefied sand. The apparent viscosity of saturated sand, the friction torque of cylindrical rotor under different pore pressure ratios, and rotating speeds are studied, especially the apparent viscosity of liquefied sand. The results show that both in the pore pressure development before liquefaction and the pore pressure dissipation after liquefaction, the friction torque is affected by the pore pressure ratio and rotational speed. As the pore water pressure of liquefaction sand gradually dissipates, the strength recovers, and the flow capacity decreases as the apparent viscosity increases. There is a linear correlation between apparent viscosity and pore pressure ratio of the liquefaction sand. The apparent viscosity decreases as the shear strain rate increases. There is a power function relationship between the apparent viscosity and shear strain rate, showing that the liquefaction sand has the characteristics of typical power-law shear thinning non-Newtonian fluid.