Traditional theory of soil arching effect is based on the assumption of homogeneous stress distribution in loosening zone. However, because of the influence of main stress axes deflection in loosening zone, the stress distribution is inhomogeneous. Based on Terzaghi’s loosening earth pressure theory, three types of major principal stress path are assumed and then a modified analytical solution of loosening earth pressure is presented considering the influence of stress distribution. DEM is also used to compare the model. The results show that when soils arrive ultimate equilibrium state, the distributions of vertical stress and horizontal stress in loosening zone are concave and convex respectively, and lateral earth pressure coefficient equals to passive earth pressure coefficient on the axle of the loosening zone. Difference among different major principal stress paths is less than 10% and these results are all in agreement with DEM numerical results, which confirms the effectiveness of the modified analytical solution. For the convenience of calculation, it is recommended to assume an arc path of major principal stress in the calculation of relevant projects.

Synthetic transparent soil has been widely used to support technology of geotechnical visualization. However, the research on its dynamic properties is relative little, limiting its application in dynamic correlation model test. In this study, dynamic deformation and strength of transparent sand are examined by resonant column test and dynamic torsional shear test. The results are compared with those of the natural sand and Fujian standard sand. Transparent sand is manufactured by fused quartz and oil mixture with the same refraction index by Norpar? 12 and Drakeol? 15. The curves of dynamic shear modulus-strain, dynamic damping-strain, dynamic shear modulus-damping ratio, pore pressure, and dynamic strength are measured and analyzed. The dynamic properties of natural sand and Fujian standard sand are also shown for comparison. It is found that the transparent sand has similar dynamic behaviors as natural sand. Transparent sand shows a great potential as a substitute for natural sand, and it is expected to be widely used in dynamic model tests.

How to predict the travel trajectory and the bearing capacity of fixed fluke drag embedded anchor in saturated clay is a key issue to design the mooring system in deep water. From the initial state of fixed fluke drag embedded anchor embedded into the saturated seabed, the fixed fluke drag embedded anchor and the surrounding soil is regarded as a macro-element. Assuming the normal motion of the macro element along the soil yield surface under the anchor chain tension, the padeye tension is calculated by the anchor chain equation. Based on an incremental iterative method, the predicting model of fixed fluke drag embedded anchor embedded trajectory and padeye tension is presented. Then, a corresponding calculation program is developed to predict the travel trajectory and bearing capacity of Bruce Dennla MK4 fixed fluke drag embedded anchor and the drag anchors of Murff examples in saturated clay. The results indicate that the developed model and program can be well used to predict the travel trajectory, motion form and padeye tension during the whole process of embedding into the saturated soil of the fixed fluke drag anchor. Since the anchor movement pattern has an adjustment process during the travel of the fixed fluke drag embedded anchor, the rotation angle decreases firstly, then increases, and gradually tends to a stable value.

In order to understand the strength failure under complex stresses, and apply the strength theory to more materials, such as rocks and concrete, by the twin strength theory, the three-parameter twin strength criterion based on the hydrostatic pressure and the normal stress on the surface of main shear stress has been proposed, and the mathematical formula has been deduced. Then the comparative analyses between the theories under different stresses with the existing experimental data have been carried out. The results show that the ratios and under different ultimate stresses vary with different strength criteria respectively and the strength under condition of is less than that the strength under condition of . In the biaxial compressive stress , the biaxial compressive strength is larger than the uniaxial one . In triaxial compression stress, given fixed and smaller, increases gradually with the increase of , and given relatively larger, decreases gradually with the increase of . The theoretical predictions can reflect some experimental data of the granite, red sandstone, soft sandstone, concrete and other materials.

The horizontal reinforcement is a method of ground treatment for soft soil. The calculation of the ultimate bearing capacity of the reinforced grounds is important to the design of ground foundation treatment. Firstly, combined with the engineering characteristics of reinforced grounds, a changeable failure mode varying with the reinforcement parameters and the kinematically admissible velocity field are established by its failure mechanism considering the influence of the coordinate deformation between soil and the horizontal reinforcement. Secondly, with the upper limit analysis, the ultimate bearing capacity calculation was deduced by the method of energy dissipation for the level reinforcement. Thirdly, a new approach for determining the ultimate bearing capacity of the reinforced grounds is introduced by the sequential quadratic programming optimization algorithm. Finally, the comparisons with experimental results find that this proposed method is reasonable, feasible, and superior to other existing methods, also can reflect the reinforcement design parameters effect on the failure mode and its ultimate bearing capacity.

A three-dimensional (3D) roof-pillars model was established to prevent disasters caused by the sudden release of high strain energy of goaf group system in the underground mine. The stability and mutational tendency of this model were analysed qualitatively and quantitatively based on the catastrophe theory and rheological mechanics theory. The mathematical criterion and mechanical conditions of the mutation and energy release at different moments were also derived. Moreover, an analytical algorithm for the stability of goaf group system was proposed, and its validity and practicability were verified. The effect of each factor on the stability of this system was further discussed on this basis. The results showed that, with the rheological process of pillars, the tendency of system mutation decreased. However, the integrity of roof was destroyed gradually, and its boundary condition successively the following three stages: the fixed end, simple support and free end. The critical value of the effective load bearing area ratio of pillars which maintained the system stability decreased at each stage, but jumped at the starting point of the stage of simple support and free end. In addition, the average reduction rates of at three stages decreased in turn. The system stability was dominated by the numerical relationship among the roof stiffness (D), overlying strata load , the effective load bearing area ratio of pillars and the size of goaf group. Therefore, the study results provide a new idea and method for the safe excavation of mine as well as the evaluation and control of goaf group system stability.

软岩的蠕变寿命是岩石流变学研究的重要内容之一，是影响岩土工程长期稳定的重要因素。根据Goodman原理，蠕变破坏点位于全应力-应变曲线的下降段（破坏段）上。当等时应力-应变曲线簇用分离变量函数表达时，证明了对于给定应变 处的等时应力-应变曲线的切线模量比和割线模量比随时间的变化规律相同。将破坏段曲线简化为直线，等时应力-应变曲线与破坏段的交点（蠕变破坏点）近似等于等时应力-应变曲线上某点 的切线与下降段直线的交点，进而推导出了该交点的蠕变寿命的表达式。 越接近峰值应变，误差越小。这样，可由短时蠕变曲线簇变换到等时应力-应变曲线簇，通过拟合方法求出给定应变 下模量比随时间变化的表达式；再由全应力-应变曲线求出峰值应力与应变、给定应变 下的应力与切线模量、下降段直线斜率等参数；即可得到软岩的蠕变寿命和长期强度。利用本文方法，对某泥岩的蠕变寿命进行了求解，得到了蠕变寿命随应力水平变化的表达式和长期强度值，理论结果与试验结果较吻合。

The effect of drilling velocity ratio, the ratio between the rotation and the penetration velocity, on the boring mechanism of auger piling was investigated with a series of model tests on both loose and dense Fujian standard sands. The variation of reaction force and rotational moment of the auger with the penetration depth, and the induced vertical and radial stresses within the soils during auger penetration were analyzed at the macro level. The variation of sand relative density and the evolution of soil fabric due to auger penetration were microscopically examined using the epoxy resin solidifying method and the digital image analyzing technique. It is found that the reaction force, the rotational moment and the induced maximum radial stress decrease with the increment of drilling velocity ratio, independent on soil initial density. In loose sand, the soils at the auger tip and around the borehole exhibits contraction behavior, whilst the dilation behavior is captured in dense sand. The extent of contact normal anisotropy on the circumferential cutting face of the solidified sand block reduces with the increment of drilling velocity ratio, yet the distribution of contact normal on the radial cutting face is not significantly affected by the drilling velocity ratio. This regularity is independent of the soil’s initial density.

A new-style sandwich reinforced soil is clay reinforced with geogrid encapsulated in thin layers of sand. To investigate the interfacial shear behavior of sandwich reinforced soil, a series of cyclic shear tests was performed through a large-scale direct shear device. The influence of thickness of sand layer, shear amplitude and normal stress on the interface shear behavior were investigated. Results indicated that the peak shear stress of the interface increased with the increase of cycle number. When the thickness of sand layers were 5, 6, 7, 8, 9 mm in the last cycle number, the corresponding peak shear stresses of the interface were 24.84, 27.4, 27.94, 26.33, 24.68 kPa, respectively, showing the maximum peak shear stress of the interface occurred with 7 mm of the thickness of thin sand layer. The ultimate vertical displacement increased with the increase of cyclic shear amplitude, and shear stiffness and damping ratio of the interface decreased with the increase of cyclic shear amplitude at the same cycle number. When the normal stresses were 30, 60, 90 kPa, in the last cycle number, the corresponding peak shear stresses of the interface were 20.4, 25.14, 32.96 kPa, respectively, showing the peak shear stress increased with the increase of normal stress, and the shear stiffness increased with the increase of normal stress at the same cycle number.

The deformation of foundation pit is considered as a typical three-dimensional (3D) space problem, but it can be simplified as a plane strain problem under specific conditions. Through 3D centrifugal model tests of a foundation pit, the deformation laws of the middle and the corner of the long side of the foundation pit were achieved under the fluctuation of the confined water level. Then 3D numerical simulation was further carried out, and the obtained results were compared with the results from the centrifugal model test. The numerical method was also used to analyse the effects of foundation pit length, excavation depth, confined water level and foundation pit area on the 3D space of foundation pit. The results showed that the corner effect was verified by the horizontal displacements at different sections of the ground wall and the surface settlement outside the foundation pit. When the length of the foundation pit was more than six times the excavation depth in Yangtze River Terrace I, the middle section of the foundation pit was basically in the plane strain state. As the excavation depth increased, the plane strain ratio increased significantly. However, the increase rate slowed down when the depth was greater than 25 m. The corner effect was more obvious with increasing confined water level, and the plane strain ratio reached 0.46 when the level exceeded 3 m. The research results provide the basis for the design and construction of the deep and large foundation pit in the typical binary structure strata of the Yangtze River Terrace I.

The topic on the seepage-stress coupling characteristics of jointed rock mass is generally considered as an important and difficult research direction in the field of rock mechanics. The natural jointed rock mass commonly exists in the three unequal compressive stress states in water conservancy and hydropower engineering and underground engineering. There are limited studies on the seepage-stress coupling characteristics of jointed rock mass under the states of three principal stresses. The fundamental study should focus on the seepage-stress coupling characteristics of the monolithic rock mass. In this study, the effect of lateral stress and the seepage mechanism of monolithic rock mass were analysed and explored. Then the theoretical formula of its seepage-stress coupling characteristics was deduced on the basis of the existing permeability-stress analytical model. Theoretical results were also compared with experimental results from true triaxial seepage-stress coupling tests in the laboratory. The rationality of the derived formula was verified by the global optimisation nonlinear fitting method and the block discrete element method (DEM). The similar physical quantities of analog circuits and seepage problems were obtained by using the circuit principle. The seepage-stress coupling characteristics of jointed rock mass with a regular hexagonal joint network were preliminarily investigated by introducing the equivalent seepage resistance method. The obtained results provide useful references for seepage analysis of rock mass in practical engineering.

In this study, experiments were conducted on raw coal specimens by using the self-developed triaxial servo-controlled seepage equipment for thermo-hydro-mechanical coupling of coal containing methane. Experiments were carried out when the axial pressure maintained at different stress levels but the confining pressure was unloaded. We studied the effects of unloading confining pressure on deformation properties and permeability characteristics of gas-filled coal at the post-peak. The results showed that the damage variable was defined by the unrecoverable deformations of radial strain , axial strain and volumetric strain and the response of these three parameters during the process of unloading. The interval [0, 1] of damage variable was satisfied to calculate the damage quantity of coal specimen in the process of unloading. When the axial pressure remained constant and the confining pressure was unloaded, the amount of damage quantity D increased with the decrease of and the damage degree of coal specimen increased greatly. When unloading the axial pressure to different stress levels, the permeability of coal specimen increased with the growing unloading times of confining pressure. It indicated that the influence of confining pressure on permeability was more and more important when decreased. At the same time, the pores and fractures in the coal specimen were developed, expanded and extended. In addition, the permeability k hardly increased at the beginning of unloading . As continued to unload, k began to increase and the slope became growing, indicating the severe damage of coal specimen. The radius of Mohr stress circle was equivalent to an increase under this stress path of loading and unloading, and coal specimens tended to destruction causing the increasing possibility of the occurrence of secondary failure. At the higher level of , with the unloading of , the radius of Mohr stress circle increased and the supportability of coal specimens became weaker. Hence, the secondary failure easily occurred, which manifested the abrupt change of axial strain and radial strain .

To study the influence factors of sub peak adfreezing strength at the interface between frozen soil and structure, the typical fine sand located in the Hexi area of Nanjing subway construction site was selected as test sample. The large-scale multi-functional frozen soil-structure interface shearing instrument was optimized and improved to conduct direct shear tests of adfreezing strength considering multi-influencing factors. Based on the specific definition and the agreement of quantitative criteria for sub peak adfreezing strength, it was found that the shear stress was typical of steepness after the limit peak, and the peak post shear stress changed periodically . The relationships among sub peak adfreezing strength, interface temperature, roughness and normal stress were fitted and found to be inverse proportional linearity, quadratic polynomial, proportional linear relationship respectively. Through the database management software, it was found that the normal stress and roughness were the key influencing factors, but the interface temperature was non-critical influencing factors. The three-element sub peak adfreezing strength prediction model coupled with interface temperature, normal stress and roughness was constructed by multivariate nonlinear regression. It can provide important adfreezing strength parameters for artificial freezing reinforcement design, shield construction and underground structure design in artificial freezing reinforcement area or permafrost area.

From the point of view of the mobilized plane, material anisotropy has been proved to be the fundamental reason to cause the non-coaxiality of soils, which is consistent with a conclusion in the material mechanics. When angles between the two conjugate mobilized planes and the bedding plane are not equal, shear components of the plastic strain increment will occur in the principal stress plane. As a result, the direction of principal plastic strain increment will not be coaxial with the direction of principal stress. Based on this conclusion, numerical modeling of non-coaxiality should also be carried out according to an anisotropic constitutive model. The newly proposed anisotropic transformed stress method considering the effect of anisotropy is also capable of reflecting the non-coaxiality, because it changes the relative magnitudes of stress components and obtains an anisotropic transformed stress tensor with different principal directions from the ordinary stress tensor. Using this method, non-coaxial behaviors of soils can be described under the framework of the existing elastoplastic constitutive models. As an example, the anisotropic UH model is adopted to predict the non-coaxial deformation under different loading conditions to verify this method.

The study on transport and deposition of suspended particles in porous media is important to groundwater recharge, underground pollutant diffusion, nuclear waste disposal and oil exploitation. Most natural suspended particles are non-spherical. To study the effect of shape on particle migration, soil column tests with three-phases were carried out on spherical particles (10 ?m in median particle diameter) and rod-shaped particles (aspect ratios of 3:1 and 6:1). The breakthrough curves of different shape particles under different ionic intensities were obtained. The mechanism of the deposition and release of suspended particles was analyzed by DLVO theory, and the effect of shape on particle migration behavior was explained. When the ionic strength is low (6 mmol/L), the DLVO interaction energy profile shows higher energy barriers. So particles of various shapes is mainly retained in secondary minimum. When the ionic strength is high (150 mmol/L), the retention mechanism of spherical particles is the secondary minimum retention due to high energy barriers. The 6:1 rod-like particles are retained in the secondary minimum and the primary minimum, relating to the electrostatic properties and the preferential orientation of the rod-like particles. For rod-like particles, a side-on orientation provides a deeper attractive well and larger influential distance than an end-on orientation. A side-on configuration is a more stable deposition method and preferred orientation for particle deposition. The Derjaguin approximation method is used in the DLVO calculation. The results show that the theoretical prediction agrees well with the soil test results.

Analytical layer-element method is presented to study the thermo-mechanical coupling behavior of rock and soil medium with a buried point heat source in geotechnical engineering. Starting from the governing equations of three dimensional thermo-elasticity, analytical layer-elements of a single layer and the underlying half-space are derived in the transformed domain by Laplace-Fourier transform. A global stiffness matrix of layered media is assembled, which is further solved by considering the boundary conditions. The numerical inversion of Laplace-Fourier transform is adopted to obtain the actual solution. Numerical examples reveal that the calculated results acquired by the corresponding procedure show a good agreement with the results in existing literature, and the presented approach shows good applicability and high precision in studying the thermo-mechanical coupling response of layered half-space; in multilayered rock and soil system, the thermal diffusivity have a remarkable effects on the variation of temperature increment and the surface heave, while it shows negligible effects on the initial and steady state values of the temperature and surface heave; the influence of the stratification on the process of heat conduction and displacement variation is significant.

A new shear strength model is proposed for unsaturated soils to consider the capillary and adsorptive mechanisms. Firstly, two ideal states are defined for unsaturated soils, i.e., the ideal capillarity and adsorption states, which are only affected by capillary or adsorptive mechanisms, respectively. Then the shear strength expressions of the two ideal states are proposed respectively. The product of the effective degree of saturation and the suction considering the cavitation of water under the tension is selected as variable for the ideal capillarity part, and the maximum apparent cohesion is selected as variable for the ideal adsorption part preliminarily. Secondly, based on the concept of the binary-medium model, the unsaturated soil is regarded as a media combined with the two ideal parts by a participation function. The weight of each ideal part represented by participation function is dependent of a cavitation probability variable. Then the new unsaturated soil shear strength with capillary and adsorptive mechanisms is established for unsaturated soils, and applicable to a wide range of suction. Finally, the comparisons between the published experimental data and the predicted results of the proposed and current popular models are carried out, which illustrates the superiority of the proposed model. Thus, the results have shown that the capillary and adsorptive mechanisms should be distinguished when considering the effect of suction on the mechanical property for unsaturated soils.

The changes in water and temperature are typical weathering factors for the deterioration of physical and mechanical properties of engineering rock. In this study, uniaxial compression tests, splitting tensile tests and angle-changed shear tests were carried out on red-sandstone specimens under different water contents and freeze-thaw (F-T) cycles. Experimental results showed that main mechanical parameters such as uniaxial compressive strength, deformation modulus, splitting tensile strength, cohesion and friction angle decreased with the increase of water content and freeze-thaw cycles, but the laws of reduction were different. It was found that the softening effect of water on rock was obvious, and the damaging effect of F-T cycles on rock was more significant. The microstructure of rock specimens was analysed by scanning electron microscopy (SEM) under different water contents and freezing-thawing cycles. During the processes of water absorption and freezing-thawing cycle, the water softening on red sandstone plays the dissolution and medium roles. Meanwhile, the effect of temperature change is reflected in the disharmony of thermal deformation and phase transformation. An attenuation model of mechanical properties of red sandstone under freezing-thawing cycles was developed, and the attenuation rate and half-life of each mechanical parameter were analysed accordingly. The research results provide significant guidance for rock engineering in the water environment and cold regions.

The current macro-mechanical models of salt rock are usually phenomenological models, which cannot fully explain the true physical mechanism of mechanical deformation and failure of salt rock. Salt rock is a polycrystalline aggregate, which mainly consists of NaCl and a few amounts of impurities formed in the geological lithification process. Its deformation mechanism is largely controlled by mechanical properties of the grain and grain boundary. The microstructure of the grain in salt rock was obtained by scanning electron microscopy (SEM), and the micro/meso mechanical parameters of the grain and grain boundary were determined by molecular dynamics (MD) and nanoindentation technologies. Based on the numerical Voronoi polygon technology, a microscopic numerical model of salt rock was established by regarding grains of salt rock as deformable blocks. The discrete element method (DEM) was adopted to perform numerical simulation on the macroscopic behaviour of salt rock specimens under uniaxial compression and direct shear. Numerical results are in good agreement with macro-mechanical experimental results, which indicates that the method proposed on the micro/meso grains of salt rock and the DEM can describe macroscopic mechanical properties well based on physical microstructure of salt rock.

Calcareous sands deposited on coral reefs are different from the quartz sands in their physical and mechanical properties. In this paper, we conducted ring shear tests with a single cycle on the calcareous sand sampled from the South China Sea. The effects of relative density and vertical stress were compared with the quartz sand under the same particle size and test conditions. The results indicate that shear stress and displacement curves of calcareous sand in forward ring shear tests tend to soften with a clear residual strength, while that tend to harden in reversed ring shear tests. The strength in forward and reversed ring shear tests are equal. The curves of the shear stress with displacement of quartz sand in forward and reversed ring shear tests tend to soften. The ratio of residual strength to peak strength of calcareous sand ranges from 0.75 to 0.93, while that of quartz sand ranges from 0.89 to 0.96. Residual strength of calcareous sand is higher than that of quartz sand with same particle size distribution and test conditions and the ratio ranges from 1.05 to 1.3. The particle of calcareous sand, in the size of 0.5 mm to 2 mm, occurs breakage under the vertical loading of 100 kPa and 200 kPa, and breakage percentages of particle are 4% and 6% respectively.

A new model of the low skirted suction caisson was proposed to improve its bearing capacity and make it easy to penetrate and protect the soil around the barrel from the ocean current. Penetrating the caisson with a double barrel of vacuum pressure can not only improve the penetration but also effectively reduce the height of soil plug. A new method of ultra-low underneath vacuum preloading was put forward to consolidate soft soil and shorten the period quickly. The soil reinforced by this method can better simulate the foundation soil with increased strength along the depth. The suction caisson models with different specifications by the vacuum method were penetrated into soil, and then the feasibility of the low skirted suction caisson was comparatively studied. The penetration resistance was analysed, and then a corresponding formula was deduced by fitting experimental data. The bearing capacity of the suction caisson was studied by the 45°pullout test combined with the finite element analysis of PLAXIS 3D software. Finally, the rationality of the low skirted suction caisson was verified.

This study aims to investigate the mechanism of multi-phase flow migration and the evolution rule of fluid pressure in the reservoir more scientifically and precisely, and to improve the accuracy of analytical calculation and analysis. The flow field was divided into three regions firstly and then inverted the saturation of each phase fluid in the mixed fluid flow region of two-phase fluid based on the conservation equation of seepage volume in the flow field. Subsequently, a generalised Darcy’s formula suited for the two-phase flow was obtained by directly introducing the total mobility into Darcy’s formula. Thus, a more accurate analytical model was derived for characterising the fluid pressure evolution in the reservoir. At last, this analytical model was applied to calculate a CO2 injection example, and the obtained results were compared with the explicit integral solution in existing references and the simulated results of TOUGH2/ECO2N. The compared results indicated that the reliability of the analytical model was verified and the superiority was reflected in the aspect of calculation accuracy. In addition, the calculation results showed that the full process of the actual fluid pressure evolution in the reservoir could be described well by the analytical model of this work, though it was obtained under the assumption of steady flow. The main reason is contributed to that the determination method of saturation in this model is more scientific and accurate. Hence, this model can be applied in practical engineering.

The bedding plane has an important effect on mechanical properties of shale and its accumulation and dissipation of strain energy. In this study, the scanning electron microscope (SEM) tests and uniaxial compression tests were carried out on Longmaxi Formation shale with different bedding plane angles. Then, this study investigated anisotropy characteristics of elastic modulus, Poisson's ratio and the stress-strain of initial cracking, critical dilatancy and peak characteristic points. The evolutions of input strain energy, releasable elastic strain energy and dissipated strain energy during the deformation and failure of shale were analysed, and the relationship among the input strain energy, bedding angles and compressive strength was revealed. The results showed that the brittle mineral content of Longmaxi Formation shale was found to be as high as 72.58%, and its microstructure was distinctly anisotropic. With the increase of bedding plane angle, the stress and strain of initial cracking, critical dilatancy and peak characteristic points were all decreased first and then increased. A minimum value was found at 30°, and the curves generally presented U-shape. With the bedding plane angle increasing, the input strain energy, releasable strain energy and dissipated strain energy of initial cracking, critical dilatancy and peak characteristic points were also decreased first and then increased, reaching a minimum value at 30°. Stress-strain and strain energy of each feature point showed significant anisotropy sensitivity, and the anisotropic sensitivities at 0°≤ ≤30°and 30°≤ ≤60°were greater than that at 60°≤ ≤90°. The peak stress had correlations linearly with both the initial cracking stress and critical dilatancy stress, while there was a corresponding quadratic nonlinear relationship between the input strain energy and compressive strength. This study provides references for shale gas drilling, reservoir fracturing and wellbore stability prediction and warning.

Magnesium phosphate cement (MPC) was used to stabilize or solidify lead-contaminated soils. Unconfined compressive strength test, permeability test and leaching test were conducted to investigate the variation of mechanical properties of lead-contaminated soils treated by MPC under different conditions of curing time and lead concentration. Results show that the unconfined compressive strength increases significantly with curing time, while the hydraulic conductivity and leaching concentration change oppositely. The unconfined compressive strength and leaching concentration under 7-day curing time are satisfied by the regulations. There is a threshold value 500 mg/kg for lead concentration in terms of its influence on unconfined compressive strength and hydraulic conductivity. The unconfined compressive strength and leaching concentration increase with the lead concentration before they reach the threshold value, while the hydraulic conductivity changed oppositely. MIP test results show that the total void volume reduces as the curing time, the total void volume of soil reduces as the lead concentration before they reach the threshold value. SEM test results show that a larger aggregation formation, with a smaller void space in soils as the curing time increases. More obvious aggregation and cementation occurs in soils space as the lead concentration less than the threshold value. MgKPO4·6H2O (MKP) decreases the pore volume of soil pores with the diameter larger than 0.1 ?m, so as to affect the permeability of soil.

To investigate the water-induced ageing effect on gypsum rock, different water environment conditions were simulated in the laboratory, including different air humidity conditions and saturated groundwater conditions which may exist in the gypsum mines. Mechanical parameters of gypsum rock specimens were obtained under the pre-set conditions for different times. Meanwhile, the variation of pore structure in gypsum rock specimens was measured by nuclear magnetic resonance equipment. Then, the water-induced ageing mechanism of the gypsum rock was revealed. The results showed that the effect of water on the ageing of gypsum rock was obvious. Firstly, the ageing extent of gypsum rock increased with time, whereas mechanical parameters including uniaxial compressive strength, Brazilian tensile strength and elastic modulus decreased with time in a negative exponential relation. However, the Poisson’s ratio of gypsum rock had no clear relation with time. Secondly, the ageing extent and ageing rate of gypsum rock were closely related to the state of water. The higher the relative humidity was, the more pronounced the ageing extent was and the faster the ageing rate was. In particular, the ageing extent of gypsum rock was the most significant and its ageing rate was also the fastest when fully submerged in water. Generally, the water-induced ageing mechanism of gypsum rock was caused by the coupling of physical and chemical effects of water, and the chemical effect was the primary cause for the ageing effect. When the gypsum rock was exposed to water, the continuous dissolution and recrystallisation of the gypsum rock changed the mineral composition of the gypsum rock. As a result, the rock structure became loose, the porosity increased, and mechanical properties gradually weakened. The research results provide references for designing the gypsum mining and the evaluation of the long-term stability of gypsum goaf.

Soil-bentonite (SB) cutoff wall has been widely applied in contaminated soil remediation. Its service performance is largely influenced by stress state and consolidation deformation. Two test sections of soil-bentonite cutoff wall were built for the investigation on the consolidation behavior of SB cutoff walls. The stress state and deformation of the walls were monitored for 15 months using earth pressure load cells, piezometers and inclinometer. Results show that the major consolidation process of SB cutoff wall lasts several months. The dominant form of the wall deformation is lateral deformation. The total stress decreases in the first month, and barely changes thereafter, while the effective stress increases with the dissipation of excess pore pressure. The horizontal strain and the maximum horizontal effective stress are both correlated with the depth of the wall. At the bottom of the wall, both the total stress and pore pressure decrease to the hydrostatic pressure. The effective stress stays low during the major consolidation process. According to the deformation and stress distribution, it is suggested that there may be sliding wedges forming in the adjacent ground, squeezing the backfills integrally, which makes a reasonable explanation of the consolidation behavior of the cutoff walls. At last, recommendations on design and construction of SB cutoff walls have been proposed.

To study the permeability and anisotropy of upper Shanghai clays, oedometer tests were carried out under different pressures in both horizontal and vertical directions. Permeability was determined by the square root of time method and log time method, and the relationships between coefficient of permeability and void ratio and the flow direction were discussed. The reason of the difference in the horizontal and vertical permeability coefficients is analyzed on the microscopic level by using scanning electron microscope. Results reveal that the coefficient of permeability of upper Shanghai clays increases with the increase of void ratio. For a single test, the relationship between coefficient of permeability and void ratio is linear in e-lgk coordinate system and the permeability change index is roughly Ck=0.5e0. For all tests, the relationship between coefficient of permeability and void ratio is curved in e-lgk coordinate system. Anisotropy of upper Shanghai clays under different pressure conditions is not obvious, which is related to the formed flocculent structure in the process of soil particle deposition. In other words, the difference of the pore content in the horizontal and vertical directions leads to the anisotropy in the coefficient of permeability.

The maximum frozen depth of surrounding rock is an important basic parameter in cold region tunnel design, such as frozen-preventing lining design, waterproof and drainage design, insulating layer design. To predict frozen depth of surrounding rock, various researches have been conducted, including field monitoring, numerical simulation, and derivation of practical estimation equations. In this study, a semi-analytical method is proposed to obtain the maximum frozen depth. First, the temperature acting on the lining surface is equivalent to a constant temperature according to the principle of equal annual accumulated temperature. Then, based on the quasi-steady state assumption, an analytical formula of the maximum frozen depth which considers the existing of lining, thermal insulation layer and the unfrozen water content in frozen rock is derived, using integration method. Next, the back analysis method is used to obtain the parameter which refers to the ratio of influence radius and frozen radius in the analytical formula, combined with the numerical simulation results of the temperature field. Finally, the semi-analytical solution of the maximum frozen depth is obtained. Through comparisons of the field monitoring data and the results of numerical simulation, the rationality of the proposed method is proved. Furthermore, the influence factors of the frozen depth are analyzed using the proposed semi-analytical solution, and the results show that the initial temperature and the annual average temperature have the most significant influence, and the rock conductivity coefficient and the porosity are less influential.

A new tension-compression dispersive anchor cable was recently adopted in projects, which is based on the combination of tension, compression and load dispersive anchor cables. However, there is a limited study on the mechanism and design method of the anchor cable. Based on the Kelvin solution, we derived analytical solutions of the shear stress between surrounding rock (soil) and the anchoring segments of the tension-dispersive and compression-dispersive anchor cables, respectively. According to the superposition principle, a simplified calculation method was proposed to obtain the shear stress between anchoring segment and surrounding rock (soil). On the basis of theoretical results, the new anchor cable was applied in practical engineering, and the basic performance test of the anchor cable and the comparison with the prestress loss of the tension-dispersed anchor cable were performed, respectively. Theoretical results indicated that the values of shear stress at both ends of the unit anchoring segment were relatively higher, while the stress reached the lowest in the middle. The distribution characteristics of shear stress were consistent with testing results in the literature and numerical simulation results. According to testing results, i.e., the bearing capacity of the anchor cable, the peak values of shear stress in tension and compression sections were obtained using the simplified method, which agreed well with peak values of shear stress by means of the uniaxial compressive strength of rock in tension and compression anchor cables. The comparison results and the long-term monitoring data reveal that the effect of engineering application is good, and the load transfer of the anchor cable is stable and reliable.

According to the characteristics of prestressed anchor cables in foundation pit, the principles of selecting the locking values of the tensile anchor cable, pressure anchor cable, scattered-load anchor cable and multi-row anchor cable were discussed through the methods of theoretical derivation, analytical calculation, numerical simulation and example validation. Due to the difference in the stiffness of the anchor head, the results show that according to the specification range, the small locking value is adopted for the tensile anchor cable, but the pressure anchor cable requires the large locking value. Scattered-load anchor cables should be tensioned and locked by groups, and the locking value of steel strand in each group is related to the length of the free segment, dispersion distance and displacement of anchor head. The locking value of the segment bearing body increases with the enlargement of the distance away from the orifice. Different locking values are applied for multi-row anchor cables considering the cable tension locking sequence. Based on the difference of the real anchor head displacement after each row anchor cable is locked, the smaller locking value should be adopted for the upper anchor cables, but the larger locking value is for the lower anchor cables. Therefore, the obtained relevant laws can provide useful references for the stretch locking construction of anchor cables in foundation pit.

Considering the effect of joint mass on the propagation of stress waves, this study derived formulas for the transmission and reflection coefficients of harmonic waves through the thick viscoelastic joints based on the displacement function of waves. The correlation coefficient of waves was used to describe the waveform changes of the wavelet through viscoelastic joints. Then, we discussed the applicable conditions of a simplified displacement discontinuity model with thick viscoelastic joints. When the correlation coefficient of transmission wave was 0.9, which was calculated by the thick viscoelastic joint model and the displacement discontinuity model, the corresponding thickness of the joint was defined as critical thickness. It was found that there was little impact of the impedance ratio between the rock mass and joint on the critical thickness. The relationship between the critical thickness and the centre frequency followed a negative exponential law. Meanwhile, the larger incident angle was, the lesser impact of centre frequency on critical thickness was. The field tests showed that when the thickness of the joint was 0.03 m, mechanical parameters were extremely close to the results calculated by the thick viscoelastic joint model and the displacement discontinuity model. With the increase of the joint thickness and the centre frequency of the wavelet, the deviation of calculated results using the displacement discontinuity model was larger, which was consistent with the results of theoretical analysis.

The anti-liquefaction-rigid-drainage pile is a new pile type combining the bearing capacity of rigid pile and the drain ability of gravel pile. In a pile foundation project, pile driving field tests of the rigid-drainage pile were carried out. The horizontal and vertical soil pressures around drainage piles and ordinary piles were recorded in different locations and depths using dynamic sensors during field tests. According to these test results, rigid-drainage pile can reduce the disturbance of soil pressures during pile driving. At the location 0.6 m far from the pile center in the depth -15 m, the peak horizontal soil pressure around the rigid-drainage pile is a quarter of the ordinary pile’s. The rigid-drainage pile can reduce the impact of pile driving on the effective stress in soil, which makes the liquefiable soil around rigid-drainage piles more stable. In the depth of -5 m, the effect of rigid-drainage piles on soil pressure is not obvious. In deep depths (-10, -15 m), the effective influence radius of rigid-drainage piles is about 4 times as large as the pile diameter. The data of piles tests provide other engineers with reliable evidences.

The stability of the left bank slope at Baihetan hydropower station is especially important in the construction and operation periods, due to the influences of the dislocation interfaces with a slightly-inclined angle and NW directional faults. In this study, the quasi-three-dimensional numerical model was established by using the realistic failure process analysis (RFPA3D) software with the arch axis profile as an example. Then the stress distribution field and displacement field of the left bank slope were studied during excavation unloading process, and the rock failure mechanism was revealed. In addition, this study analysed the initiation, propagation, extension and interaction mechanisms of microcracks during the progressive failure process. By combining with the distribution of microseismic events during excavation unloading and stress adjustment on the left bank slope, the potentially dangerous parts of the engineering were delineated, and the possible failure modes of the slope were explored. The results showed that the safety factor of the slope after excavation was 1.33 calculated with the centrifuge loading method. It was found that the most obvious deformation and failure of the slope surface were controlled by the inclined dislocation interfaces. Meanwhile, the wedge-shaped sliding within a small scale played a dominant role in the damaged zone. This study provides references for the later construction of the left bank slope and practical guidance for constructing similar rock slopes.

Reliability sensitivity analysis provides a rational vehicle to shed light on effects of different uncertain parameters on slope failure probability when only a limited number of data is available. This paper develops an efficient reliability sensitivity analysis method for slope stability. The proposed approach combines direct Monte Carlo simulation (DMCS) and probability density reweighting method (PDRM) to, efficiently and accurately, calculate slope failure probabilities for different cases considered in sensitivity analysis. In the proposed approach, random field theory is used to model geotechnical spatial variability. Local spatial averaging technique is applied to constructing the joint probability density function of geotechnical parameters with relatively low dimensions, which is directly used in PDRM. Finally, the proposed approach is illustrated using a design scenario of James Bay Dyke. Results show that: only one run of DMCS is needed in the proposed approach, avoiding repeated generation of random samples and performing slope stability analysis. This saves a large amount of computational efforts, and significantly improves computational efficiency for DMCS-based reliability sensitivity analysis. Using joint probability density function of geotechnical parameters derived from local spatial averaging in PDRM gives proper estimates of slope failure probability and avoids biased estimates, making PDRM feasible in reliability sensitivity analysis of slope stability in spatially variable soils.

To accurately determine mechanical parameters of the rock mass, an improved method for the quantification and correction of GSI value was proposed on the basis of the comprehensive analysis on some quantification methods of GSI value. Firstly, by using the scan line method, this study obtained the average spacing of joints (d) and rock mass block rating (RBR) which was a new quantitative factor based on rock mass block index (RBI). The volumetric joint count of rock mass (Jv) and rock mass structure rating (SR) were acquired according to the three-dimensional (3D) joint networks of the rock mass. Then GSI values of rock mass were quantified by utilising these above parameters, surface condition rating (SCR) and the joint condition factor (Jc). To overcome the drawbacks of GSI system, the correction method and formula of GSI value were established by considering the effects of joint orientation and groundwater on the mechanical parameters of rock mass. The established method was applied in a lead-zinc mine as an example. According to the new quantification and correction method of GSI value and Hoek-Brown strength criterion, the mechanical parameters of the ore body, hanging wall rock and footwall rock were determined. The accuracy and the feasibility of this method were verified by the comparative analysis with in-situ deformation tests. This study provides the theoretical and practical basis to acquire mechanical parameters of jointed rock mass based on rock mechanical tests in the laboratory.

Reflection-transmission matrix method for the dynamic response of the layered transversely isotropic saturated soil (TISS) under axisymmetric deformation is established in this study. Applied the Fourier and Hankel transforms, the partial differential equations satisfied by the state vector of the TISS can be duduced to the corresponding ordinary differential equations for the state vector. Solving the ordinary differential equations yields the general solutions for the state vector of the TISS. Based on the aforementioned general solutions, the transfer matrices of the state vector and wave vector are obtained. Reflection-transmission matrices of the layered TISS are derived by the transfer matrix of the wave vector. By using the obtained reflection-transmission matrices, the boundary conditions and the continuity conditions at the interfaces of the soil layers, the transformed domain solution of the layered TISS subjected to a vertical force is obtained. Applying the inverse Hankel transform to the transformed domain solution of the TISS gives the frequency domain response of the TISS.

A lenticle existing in engineering site affacts significantly on the earthquake ground motion. There exists obvious difference between the seismic response of water-saturated poroelastic site and the corresponding dry site. However, there are few studies in literature on influence of lenticle on seismic response in deep water-saturated poroelastic soft site up to now. Using finite element and indirect boundary element coupled method (FEM-IBEM), this paper investigates the effect of width, thickness and embedded depth of a lenticle on dynamic structural response by the nonlinear dynamic saturated soil-structure interaction model, and particularly analyzes the effect of dynamic coupling effect between solid frame and pore water. The lenticle width, thickness and embedded depth show different effects on the structural response, resulting in obvious difference with those of a lenticle in dry site. The dynamic coupling effect between solid frame and pore water has significant effect on seismic response of the structure.

The macro- and micro-numerical simulations of biaxial shear tests for ideal granular samples with different initial densities were carried out using discrete element method. Loop structures, represented by Voronoi polygons, were taken as the basic meso-mechanical unit of granular material by mesh-subdivision. The evolution of numbers, geometrical morphology and mechanical characteristic for different loops were simulated. The meso-mechanical characteristic at critical state for granular media was especially analyzed. Numerical results show that the evolution of higher-order loops and lower-order loops are different. The evolution of same types of loops in granular samples with different initial densities are different before critical state reaching. The number proportions, geometrical morphology, contact forces of side particles and inner sliding rates of same kinds of loops in granular samples with different initial densities have reached critical state respectively. From mesoscopic point of view, critical state of granular media is the consequence of mutual transformation for higher-order loops and lower-order loops, and is the combined average and external manifestation of all loops. Meso-scale loop structures under a state of dynamic equilibrium at critical state reflected by the constant shear stress and development of shear deformation under constant volume on macro-scale. Numerical results also show the evolution of numbers, geometrical morphology, mechanical characteristic and contact stabilities of meso-scale loop structures, are closely related to the development of strength, dilatancy and critical state for granular material. So it can be taken as the basic unit of meso-mechanism for granular media.

The interbedding slope with soft and hard strong-weathered rock is a special high slope, and the composition and properties of its structural surfaces are complex. The complexity and mutual influence of the interaction between the slope and supporting structure are relatively high, and it is the most probable that the collapsible structure will be destroyed when exposed to air. Based on the engineering of a deep cutting slope of Lanyong highway, we conducted in-situ monitoring the anchor rod stress, inner anchor force, and the displacement of the slope during and after the support. Moreover, numerical analysis was performed on the joint rock module of geotechnical engineering using PLAXIS software. The monitoring and numerical results indicate that the potential polygonal sliding surface of rock slope is formed by joint fractures and rock layers, and there is more than one potential sliding surface with the same sliding possibility. The retaining structure has great influence on the stability of structure surface when it penetrates any part of the fold line of the slip surface, but the penetrating shale surface has a greater effect on the structural surface than penetrating fissures. The variation of prestressing force of supporting anchor cable has a certain regularity, and the process of prestressing variation has a certain effect on the displacement of the slope and internal force of retaining supporting structure. For internal forces of supporting structure, the influence of the active deformation of the slope is greater than the influence of passive deformation of the slope. However, for a sensitive degree, sliding effect of the structural surface on the displacement of slope surface is less than the influence of internal forces of supporting structure. The support of the slope is stable, and the supporting design is reasonable, which can provide corresponding advice for the design of similar supporting slopes.

Based on multipoint ground motion input method and an equivalent visco-elastic constitutive model for rockfill, a semi-analytical method is used to conduct dynamic analysis of high concrete face rockfill dams (CFRD) subject to non-uniform input motion. The validation of the calculation method is first conducted in the frequency domain. Under the condition that the acceleration time histories on the surface of the free field are assumed to be consistent for different modes of non-uniform and uniform seismic input, the following main conclusions can be drawn. The seismic response to the non-uniform input is generally smaller than that to the uniform input. The dynamic tensile stresses become obviously larger at the waterstops (the seepage control system of CFRD) for the non-uniform input than for the uniform one. The seismic responses of the high CFRD to the non-uniform input are characterized by the smaller in central part and the larger around the boundary. The seismic response of high CFRDs under the incidence of P, SV and SH waves are featured by the following several main facts and findings: With increasing incident angle for SH waves, the response intensity shows almost constant. There exists a critical angle for SV waves. When the incident angle is close to the critical angle, the response intensity first increases and then decreases sharply, and before that, the intensity is nearly constant, while after that, the intensity keeps decreasing. There exists a feature angle for P waves. When the incident angle is smaller than it, the response intensity is almost constant, while when larger, the intensity keeps decreasing.

The intercalation is a common geological structure, and it significantly influences the stress wave propagation and its response to rock mass under the seismic or blasting loads. In the past studies, the researches focus on vibration isolation performance and transmissivity of stress wave propagation at intercalation. However, the energy evolution of stress wave in the process of multiple refraction and reflection in intercalation are seldom investigated, and the stress response and failure of intercalation have not been well analysed. Therefore, the variation laws of energy coefficients in the propagation process of stress waves were studied by theoretical analysis. The influence of the wave impedance of rock mass and the angle of incidence of stress wave to the accumulated energy coefficients of stress wave in the intercalation are analysed, as well as the intercalation stress response and the dynamic safety coefficients of a plane failure slope. As a result, the residual energy coefficients of stress wave decreased exponentially with the times of refraction and reflection during the stress wave propagation in the intercalation. It was found that the residual energy of stress waves could be neglected after the fourth refraction and reflection. The differences in accumulated energy coefficients of stress waves in the different media increased with the relative differences of wave impedance between the intercalation and surrounding rock. Under the incident of the plane harmonic wave, the shear stress and shear strength in the intercalation exhibited a fluctuation of the same frequency as the incident wave. Compared to P-wave incident, SV-wave incident generated greater shear stress in the intercalation, and caused the greatest impact on the slope stability. When the SV-wave was incident, the safety factor of the slope was more sensitive to the dip angle of intercalation than that when P-wave incident, and the safety factor decreased more rapidly with the increase of the dip angle.

According to the simulation of centrifuge model test of spudcan penetration in various soil profiles, several influencing factors of the penetration resistance are discussed using coupled Eulerian-Lagrangian (CEL) finite element method. Five typical seabed profiles commonly encountered in the field are considered: clay, sand, clay overlying sand, sand overlying clay, and layered soils. It is shown that different ranges of Eulerian domain barely affect the calculation results. To ensure the accuracy of the numerical results of CEL, fine mesh is defined near the spudcan and coarse mesh far away. For various soil profiles, the study confirms that decreasing the size of fine mesh or increasing the range of fine mesh can ease the vibration of the penetration resistance curves. By comparison, the size of fine mesh is proposed to be set as 0.05 times the diameter of the spudcan, meanwhile, the range of fine mesh is proposed to be set as 2 times the diameter of spudcan for various soil profiles. When the displacement control is adopted, the penetration rate has a little effect on the profile of clay, while a great effect on the profile of sand overlying clay. By comparison, a penetration rate of 0.2 m/s is selected in FEM when investigating the calculated penetration resistance in various soil profiles.

The dynamic response is significant to the elastic wave problem in soil caused by the external load. This paper proposes a solution to calculate dynamic stress responses of an arbitrary point in a transverse isotropic multilayered soil subjected to a time-harmonic load. The generalized plane-strain equation is transformed from frequency-spatial domain into frequency-wave number domain by Fourier transformation in this algorithm. Combined with the introduction of the dual vector, the state equation is solved by the precise integration method. Based on the displacement response of the soil in the frequency-wave number domain, the dynamic stress response of any point is obtained by the inverse Fourier transformation. The time-harmonic load can be applied at the surface of the soil or under ground. The accuracy of the algorithm in this paper is verified by a comparison with an existing solution. An extensive parametric analysis on the influence of anisotropy, excited frequency and damping ratio on the dynamic stress response provides reliable numerical basis for engineering practice.

It is well known that rock materials have different mechanical behaviors under the compressive and tensile loading. Correspondingly, there are two types of elastic modulus; they are compressive elastic modulus Ec, and tensile elastic modulus Et, respectively. In order to distinguish which indirect test methodologies, including three-points bending test and Brazilian disc test, is more suitable to measure the tensile elastic modulus Et of rock materials, a series of experimental tests, including uniaxial compressive test (UCT), direct tensile test (DTT), three-points bending test, as well as the Brazilian disc test is performed for three typical types of rock: white marble, granite and sandstone. Based on these experimental results, comprehensive comparative investigation on the reliability of these measurement methods for tensile elastic modulus Et of rock materials is systematically conducted. Finally, it is found that the Brazilian disc test could be a suitable and reliable method to measure the tensile elastic method of rock materials, due to the excellent agreement with that measured by direct tension tests, and due to the simplicity of sample preparation, as well as test operation.

In current study, a high accuracy small strain triaxial apparatus is developed using linear variable differential transformer (LVDT) sensors. The structure of the triaxial apparatus, the specification of the sensors, and the technical details of the small-strain measurement are introduced in detail. E/P regulators are used to control the axial load and cell pressure. Meanwhile, a motor driven actuator can be used to provide axial load. The apparatus can achieve complex stress path for triaxial tests with controlled stress or strain. Constant temperature, ground connection, and high accuracy DC power to reduce the electrical noise in the signals and to obtain a precise small strain measurement. Median filtering method is applied to increase the accuracy of measured LVDT displacement data. Undrained triaxial shear tests under K0 consolidation are conducted on Shanghai soft clay. The secant modulus of Shanghai soft clay in small strain level (0.001%-0.1%) is obtained. It is confirmed that the developed triaxial apparatus using LVDT displacement sensors can obtain secant modulus in smaller strain better than other existing products.