In this study a device has been developed to simulate the capillary water and salt migration. The pore solution migration in the desalinated and pure Dengban soils can be monitored under the temperature-, humidity- and water head-controlled conditions. Before and after the capillary test, thermal conductivity and pore radius of the specimens are determined to estimate the contribution of external and internal salts to the salt damage of mural from the microscopic viewpoint. The experimental results show that solution migration rate decreases in the order of mixed KCl and Na2SO4 solution ? KCl solution ? Na2SO4 solution ? distilled water. The capillary rise rate decreases gradually with the specimen height, and the capillary rise rate in desalination Dengban soils is greater than the pure Dengban soils. The electric conductivity of soil solution increases vertically with the soil specimen height, and the conductivity of pure Dengban soil is greater than that of desalinated Dengban soil. The thermal conductivity of the soil specimens shows an overall tendency that increase first and then decrease with the rising specimen height. Compared with the pure Dengban soil, the pore-size distribution density of the desalinated Dengban soil reduces to a small aperture. In the case that the mixed salt solution is transported in soils, and pore sizes are concentrated in a more narrow area than that if single salt solution is used. It is shown that the salts transported by capillary, especially mixed sulfate and chloride, induce more severe damage than the salts contained in the soil, indicating that it is important to prevent rain water from leaching into upper caves for mural conservation.

Taking comprehensive influences of intermediate principal stress, brittle softening, dilation characteristic, Young’s modulus and elastic strains in the plastic zone into account, some basic assumptions are proposed for the elastoplastic calculation of a deep circular tunnel. On the basis of these assumptions, the new solutions of rock plastic displacement and ground response curve are derived; and then feasibility analysis and comparative verification of the new solutions are carried out. Parametric studies are discussed in detail. The rock plastic displacement and ground response curve presented by this study are theoretical analytical solutions without the need for any numerical method and have broad applicability as well as good comparability. It is found that the strength theory effect of a tunnel, i.e. the intermediate principal stress effect, is significant and the corresponding weak support or light support can be used; the ignorance of brittle softening or dilation characteristic will underestimate actual tunnel deformation, and hence less safety in engineering design; the rock deformation and ground response curve according to the radius-dependent Young’s modulus are between the upper and lower bounds, which can reflect gradual change of the tunnel excavation unloading effect with distance; the third definition for elastic strains in the plastic zone is preferred for its reasonability and precision, and there are interactions between different factors.

The thermal properties of saturated red clay subjected to progressively heating and cooling processes are experimentally studied under the combining impact of temperature and confining pressure. Two kinds of tests with and without drainage between temperature stages at four confining pressures (50, 100, 150 and 200 kPa respectively) are performed. The temperature is applied sequentially as 20 ℃→40 ℃→50 ℃→60 ℃→70 ℃→80 ℃→70 ℃→60 ℃…→20 ℃. In this study, the evolutions of temperature, pore pressure, volumetric strain, and the related thermodynamical mechanisms are analyzed. It is shown that during progressively heating/cooling with drainage between temperature stages, the volumetric strain due to the pore pressure dissipation associated with heating are greater than that due to the latter compression induced by negative pore pressure after progressively cooling, and eventually induces an irreversible shrinkage deformation. On the other hand, during progressively heating/cooling without drainage between temperature stages, the pore pressure induced by heating can even reach the applied confining pressure, and then maintains a significant residual pore pressure when the temperature of specimens falls to the initial temperature, revealing also an irreversible thermodynamic processes.

Particle breakage is one of the major factors to affect the strength and deformation characteristics of rockfill materials. Particle breakage of rockfill materials is more obvious at low stress level in comparison with sands, Thus, the effects of particle breakage on mechanical characteristics and constitutive model of rockfill materials need to be considered. Since the void ratio of rockfill materials is not constant during the loading process, the traditional constitutive model, which cannot use a set of parameters to simulate the same rockfill with different void ratios, is not suitable. Therefore, it is necessary to develop a hypoplastic constitutive model of rockfill materials, which incorporates the representative void ratio expressions and particle breakage. Meanwhile, this model is based on Gudehus-Bauer hypoplastic constitutive model, which can consider the different characteristics of the void ratio between rockfill and sand. Particularly, the hypoplastic constitutive theory has the least assumption. Under different representative stress paths, the strain increments of granular materials are various. However, under different representative stress paths even with the same level of stress, the representative void ratio of rockfill is difficult to satisfy the equal proportion rule that is successfully used in the Gudehus-Bauer hypoplastic constitutive model for sand. Therefore, a series of the representative void ratio expressions incorporating particle breakage is proposed based on both the particle breakage critical state theory and experimental results. Subsequently, these expressions are introduced into Gudehus-Bauer hypoplastic constitutive model to establish a hypoplastic constitutive model of rockfill materials incorporating particle breakage. In addition, the method to determine the parameters of model is provided. Finally, the constitutive model is applied to simulate the test results of rockfill materials in the literature, which indicates that the proposed constitutive model can well simulate the stress-strain behaviuors of rockfill materials.

This study explores the effect of carbonation on leaching properties of cement stabilized/solidified lead contaminated soil using the semi-dynamic leaching tests. Artificially contaminated soils with three different lead contents are solidified with different amounts of cement. The solidified samples are curing under the standard conditions for 28 days, and part samples are selected to conduct accelerated carbonation tests, while the residual samples continued to cure. The environmental safety of cement stabilized/solidified lead contaminated soils are assessed; and the leaching mechanism of lead is discussed using the diffusion coefficient of lead and leaching index of solidified/ stabilized lead contaminated soils. The results indicate that the rate of lead leaching decreases with increasing the leaching time. The results also show that a linear function adapts approximately the cumulative amount of lead leaching with square root of leaching time. The cumulative amount of lead leaching of the carbonated samples are 0.68 to 0.91 times that of the standard curing samples, indicating that lead is solidified/ stabilized with carbonation. The diffusion coefficient of lead of solidified/ stabilized soils ranges between 2.12×10-12 cm2/s and 1.39×10-9 cm2/s, and the leaching index changes from 8.86 to 11.7. The carbonation curing results in a decease in diffusion coefficient of lead and a slight increase in leaching index. The leaching mechanism of lead from the solidified/ stabilized soils with low lead concentration is controlled by surface wash-off. However, the leaching mechanism of lead from the solidified/stabilized soils with high lead concentration is a combined result of surface wash-off and diffusion, and the diffusion controls the process.

To investigate the cementation mechanism of mixed red mud, the variations of the chemical composition, mineral composition and microstructure of mixed red mud are analyzed by means of XRD, XRF and SEM tests during the dehydrating process; and the variation of cementation strength during dehydrating is also discussed. Based on this, the triaxial shear tests under consolidation drained condition are carried out to determine the mechanical properties of five mixed red mud samples of different dehydration ages, and to reveal the influence of formation of cemented strength. The results show that the cementation strength mainly derives from the hydrating reactions of hydraulic materials like cement and carburizing reagents of activating oxide in mix red mud, and the significant growth of the strength appear within dehydration age of 70 days. It is also shown that the shear strength of mixed red mud have direct correlation with cementation strength, especially the cohesive strength. To investigate the impact of the structure on the deformation and strength of the mixed red mud, we introduce a structural parameter, which is defined as the ratio of principal stress difference of dry sample with different dehydration ages under triaxial compression to that of original sample. A good correlation between the structure and deformation and strength is found based on the structural parameter.

Revealing the evolution law of soil mechanical characteristics under wetting-drying cycles is highly significant for preventing the disaster of expansive soil. In this study, a method of super-mini-penetration (SMP) test for describing soil internal mechanical characteristics is proposed. A series of penetration tests is conducted on the remolded expansive soil specimens under cyclic wetting-drying conditions. The evolution law of penetration resistance with penetration depth and moisture content during three sequences of drying process are obtained. The results show that the proposed SMP test is a simple, quick and effective method to describe the spatiotemporal evolution characteristics of mechanical properties of expansive soil as subjected to drying-wetting cycles. The penetration resistance of the specimen generally increases with increasing drying time. Especially, the penetration resistance of the surface soil relative to the deep soil is more sensitive to drying. The effect of wetting-drying cycles on the spatiotemporal evolution characteristics of the mechanical properties of expansive soil is significant. With an increase in cycles, the penetration resistance of expansive soil decreases. The penetration curves gradually convert from typical mono-peak structure to multi-peak structure; and the spatial difference of penetration resistance becomes more evident, which is more obvious in the relatively low moisture content range. In addition, the relevant mechanisms of the penetration mechanical behaviour of expansive soil under drying and wetting-drying cycles are analyzed by drawing knowledge from soil mechanics and soil structure. During drying, the increase of the penetration resistance of expansive soil is mainly due to the decrease of porosity, the increase of soil density, inter-particle contacts and soil suction induced by shrinkage. The loose structure and cracks resulted by wetting-drying cycles are the main reasons responsible for the weakening of the overall mechanical properties of expansive soil.

A series of 1D secondary consolidation tests and undrained triaxial creep tests are conducted to investigate the creep characteristics of soft dredger fill in the estuary of eastern Shantou. In 1D secondary consolidation tests，the stress-stain relationship of the soft soil shows significant nonlinear characteristics. The deformations of primary compression and secondary compression in overconsolidation state is smaller than normal consolidation case; and preloading can increase the yield strength. Pore pressure develops over time in the undrained triaxial creep tests, i.e., effective stress decreases with time. When the deviatoric stress is smaller, the deformation of soil tends to be stable, and the strain rate decreases with time; however, once the deviatoric stess reaches the critical value, the deformation develops with time rapidly, and the soil sample is damaged due to the accelerated creep. The concept of fractional order derivative is introduced to establish the fractional derivative Merchant model; and the ranges of parameters are determined. By comparison with the fitting results of traditional Merchant model, it is found that the fractional Merchant model can simulate the creep deformation better. The fractional Merchant model has a simple form and a few parameters, which can be well applied to the practical engineering.

In order to solve the stress of the surrounding rock with tubular double-ellipse caves under internal high water pressure, a plane stress model for double-ellipse caves under mixed boundary condition is employed in this study. Based on the plane elastic complex variable theory, two analytical functions of the plane stress containing internal water pressure with one elliptical cave are solved by means of conformal mapping, Cauchy integral and residue theorem. Using the Schwarz’s alternating method, the analytical expressions of component stresses in unlimited plane with two elliptical caves are obtained. According to these obtained expressions, Matlab software is applied to analyze four examples of stress in unlimited planes with two circular caves or two elliptical caves. The influences of internal pressure and the relative locations between two caves on the stress of surrounding rock are investigated. The results show that when two elliptical caves without internal pressure are arranged horizontally symmetrical, the maximum tangential orifice stress is found at 0° of left cave (or 180° of right cave). When internal pressure increases, both the maximum and minimum tangential orifice stresses decrease; if only one of these two caves containing internal pressure, when this pressure increases, the change of tangential orifice stress depends on its location relative to cave wall.

There was a heavy rain in Nanjiang County on September 16, 2011, causing a great number of debris landslides mainly sliding along the interface of soil and bedrocks. In order to study the formation mechanism of the shallow subsurface debris landslides induced by the heavy rainfall, a rainfall simulator is used to produce the rainfall during the test. The develop process of the landslide are simulated under rainfall by the centrifugal model tests and characteristic parameters of deformation and fracture are obtained. The analysis of the centrifugal model test results is performed to investigate the sliding mechanism of the landslide. The results indicate that: 1) The soil pressure and pore water pressure gradually increases with a greater acceleration, and the moisture content in the slope slowly reduces while the moisture content in the soil of slip zone increasing at 0 g ~ 50 g loading; 2) The soil pressure values from the earth pressure gauges SP1, SP2 continue to increase while the soil pressure values from SP3, SP4 slightly change, which indicates that the leading edge of the sliding body and the central body have been subjected to the thrust from the rear of the slide. 4 pore water pressure values gradually increase to the maximum and all of them reduce after the rain, which indicates that during the rainfall, pore water gradually converges to interface of soil and bedrocks , after the rain , pore water in the vicinity of interface of soil and bedrocks gradually dissipate. The value of moisture sensor MC3 in sliding body is observed to first become larger, and after 20 s, the value of moisture sensor MC2 imbedded in the slip zone soil also gradually increases during the rainfall. 3) 4 earth pressure gauges and 4 pore water pressure gauges are found to change a lot in value, and measured values of two moisture content sensors decrease rapidly, which suggests that the slope have a whole slide at this time. After that, earth pressure gauges SP1, SP2 and SP3 increase slowly to their maximum while SP4 has been declining; four pore water pressures gradually increases to the maximum, and after the second rain , four soil pressures, four pore water pressures and two moisture contents gradually reduce. 4) Finally, by comparing the model test with the prototype test, the sliding mechanism of the fill slope can be recognized as the slip-crack- whole sliding mode.

In this study, laboratory experiments are conducted to simulate the rockburst of granitic roadway. The obtained parameters of acoustic field and infrared temperature field are normalized based on linear function transformation. The evolution characteristics of each parameter on the same scale is also analyzed according to the original curves. Thus the precursor of granite rockburst around roadway is investigated comprehensively. The results show that there are four suitable parameters, i,e,, the acoustic emission event rate, acoustic emission energy rate, the highest temperature and the lowest temperature of infrared radiation, used to predict rockburst. When all parameters during the rockburst are normalized, the precursor can be identified more quickly and efficiently under the unified scale. An early warning sign of rockburst occurring is referred to the accelerated release of acoustic emission energy. The relative calm period of acoustic emission is a critical period, in which the methods to control rockburst-induced disaster are taken. When the lowest radiation temperature suddenly decreases and the highest radiation temperature sharply increases, as the sign of rockburst, the rockburst may occur at any time. Experimental results provide a new method and basis for predicting rockburst disaster.

The aim of this study is to investigate the damage recovery effect of salt rock under confining pressure conditions. Thus the damage recovery experiments have been performed on the salt rock specimens with different initial damage degrees under different confining pressures and holding time. Based the controlling theory of strain, the effects of the above three factors on the strain and damage recovery are analyzed under compression. The results indicate that the self-recovery ability of salt rock can be changed by confining pressure. The strain recovery process can be divided into three phases: fast recovery, retard recovery and slow recovery which are depending on the confining pressure, degree of growing cracks and holding time . Firstly, it is shown that the strain recovery process can be approximately fitted by a negative exponential function. Secondly, the increases of both holding time and confining pressure are beneficial to the effects of strain and damage recovery to a certain degree. Especially, the increase of confining pressure within the holding time can not only raises the whole recovery rate, but also prolongs the retard recovery stage. Finally, the ranking of damage magnitudes of salt rock specimens remains unchanged, even though the specimen with a higher damage degree has a high recovery rate in the pressure holding time. The high confining pressure can promote the strain and damage recovery of salt rock, but a higher confining pressure may cause the phenomenon of false recovery.

In dealing with geotechnical engineering problems with the traditional soil mechanics, the elastic mechanics approach is often used to calculate the stress distribution in soil. The calculation of the stresses in soil is usually based on simple Boussinesq’s solutions which are derived on the assumption that the loading acts on the ground surface. However, building foundations are generally buried in a certain depth beneath ground surface. In this case, the Mindlin solution is more suitable to calculate the stress distribution. The equations for calculating stresses in a semi-infinite elastic solid subjecting to a vertical rectangular and strip uniform load are significant for many geotechnical problems, such as the calculation of foundation settlement and the analysis of the influence of excavation size on its stability and deformation. Although all the equations have been presented in Reference[1-2] by Yuan Ju-yun, several errors are found in these equations. Based on the Mindlin’s formulas, the calculation equations for calculating stress in the soil subjecting to a vertical rectangular uniform load beneath the surface are derived by the integral method again. Furthermore, the equations for strip load in similar case are deduced, both of which are different from those given in Reference[1-2]. Eventually, the correctness of two groups of formulations is verified by the numerical integration.

The mechanical properties of pipe pile and solid pile are significantly different due to the plugging effect and the existence of pile core soil during pile driving process. Hence it is necessary to study the influence of the difference between the soil around the pile and pile core soil on the vibration of pipe pile in soil. By using the axisymmetric model and pile-soil contact surface continuity conditions, the solutions to vertical vibration of pipe pile in soil are obtained with the soil around the pile, pile core soil and pipe pile as a whole. For comparison, the Euler model-based solutions are also derived. Comparisons are performed among the results obtained by various methods. The results indicate that the complex stiffness at pile head obtained by using axisymmetric model is smaller than the complex stiffness obtained with Euler model, and there exists a certain gap between the results by the two models. The dynamic stiffness and dynamic damping becomes smaller when the shear modulus of soil around pile is larger than that of the pile core soil. The complex stiffness at pile head is smaller when the wall thickness is smaller. Hence the wall of pipe pile should not be too thin. In the design of pipe pile, it is necessary to consider the influence of soil properties on the vibration of pipe pile in soil, and properly select the interior radius, outer radius and length of the pipe pile.

The deformation and strength of rocks are significantly influenced by the crack closure behavior. In this study, the crack closure effect and its quantitative characterization are studied. First, based on the widely used effective medium theory, the axial strain is divided into the matrix axial strain and the crack axial strain. Three sets of test data are then used to establish a negative exponential model for the relationship between the crack axial strain and the axial stress. It is found that the crack closure behavior is well captured by the proposed model. The crack closure strain can be used to some extent to represent the microcrack quantity (or microcrack density) in the rock sample. The variation of the microcrack density under uniaxial and triaxial compressions is well described by using the crack closure strain.

Based on similarity theory, two models of rock slope with different thicknesses of horizontal weak interlayers are maded. The sizes of slope models are: 1.80 m in height, 1.65 m in length, 1.50 m in breadth, and the slope angel is about 60°, the thicknesses of weak interlayers are 3 cm and 15 cm, respectively. Using the recorded data of sensors in the large-scale shaking table test and orthogonal design, the acceleration dynamic response characteristics of slope and influencing factors are investigated with different input wave kinds, excitation directions, frequencies and amplitudes. The results show that the dynamic acceleration amplification coefficients in the slope body increase with increasing elevation, and the amplification coefficients show nonlinearly surface effect. The horizontal amplification coefficients along slope surface increase with the increase of elevation in the form of fluctuation, especially at the middle and upper part of the slope with thin interlayer. The vertical amplification coefficients differ owing to the thickness of weak interlayer, it decreases in the local part of slope with thin interlayer and then increase to the maximum at the shoulder, but it appears at the bottom of interlayer for the slope with thick interlayer. Under the equal input seismic load, the vertical amplification coefficients are 0.75 times the horizontal amplification coefficients in the interior of slope body. On the surface of slopes, the magnitude of amplification coefficients in both horizontal and vertical directions are concerned with the elevation. Under the weak interlayer, the vertical amplification coefficients are greater than the horizontal’s, whereas above the weak interlayer, the result is reverse. The effect of weak interlayer on dynamic response of slopes also varies with the different excitation directions. The dynamic responses are amplified in horizontal direction, but it is weakened in the vertical direction. In order of importance, the factors influenting dynamic response of solpe is in turn: elevation, position of slope, thickness of weak interlayer, amplitude, wave types and excitation direction. Among the selected influencing factors, the elevation, position of slope and thickness of weak interlayer can significantly influence acceleration dynamic response of slopes.

Generally, slabbing is the typical failure mode when the hard brittle rockmass in deep tunnel is excavated under the condition of high stress. Many factors are contributed to the form and formation mechanism of slabbing, in particular the curvature radius of the excavated section. Therefore, the influence of curvature radius on the hard brittle slabbing is firstly summarized and analyzed the slabbing form and feature corresponding to curvature radius of different sections in deep tunnel of Jinping II hydropower station. Subsequently, the indoor physical model test is designed to investigate the slabbing forms of circular caves with different bore diameters and straight-wall arch caves with different sizes under plane strain condition. The results show that the slabbing mechanisms influenced by the curvature radius can be demonstrated at both scale and structural effects, which jointly affect fracture morphology and fracture mechanism of slabbing. When the sections in the same tunnel consist of different curvature radii, slabbing failure presents combinated characteristics of fracture. Finally, the mechanisms of slabbing influenced by curvature radius are analyzed and verified by numerical simulation.

The reliability analysis of a spatially variable soil slope should be considered as a system reliability problem. The multiple response surfaces method (MRSM) provides a rational vehicle to evaluate it efficiently and accurately. This study firstly determines the reasonable response surface function for safety factor with considering spatial variability based on single slip surface. Then the approximate linear relationship between the accuracy of reliability analysis and that of random field discretization is explored. Using the MRSM on a large amount of potential slip surfaces, several representative slip surfaces (RSSs) are identified with the assessment of slope system reliability. The proposed method is illustrated and verified using two spatially variable soil slopes. The results indicate that as the spatial variability becomes stronger, the slope reliability becomes more systematic, and the failure probability obtained using single slip surface significantly underestimates the system failure probability. It is possible to ensure the accuracy of reliability analysis by predefining a proper value of the accuracy of random field discretization, aiming at reducing the number of unimportant random variables. In addition, choosing a proper response surface function helps improve both efficiency and accuracy of the proposed method. In addition, MRSM can simultaneously calculate the failure probability of all potential slip surfaces and the system failure probability, and identify RSSs, providing a guidance to make slope remedy measure.

To understand the mechanical behaviours of red bed soil-rock embankment reinforced with gabion, field tests are conducted to obtain soil pressure, the deformations of gabion reinforcement and gabion wall and the potential internal fracture surface. Meanwhile, based on in-situ tests and FLAC3D numerical modelling, a comprehensive analysis is performed to investigate the effects of reinforcement spacing, reinforcement length, compression modulus, strength indicators of foundation soil and compaction of embankment fill on mechanical behaviours of the embankment. Field tests indicate that the tensile strain distribution of gabion reinforcement along the cross-section of embankment is non-uniform. Besides, the internal potential failure surface of embankment is similar to ‘improved simplified double-segment rupture surface’. The deformation of gabion wall tends to be beam bending inside the back plane of gabion wall, whereas external-oriented bulging is outside the back plane of gabion wall. It is found that the vertical earth pressure in backfill at the same level behind the wall exhibits non-uniform distribution. The relationship between the measured earth pressure at the back wall and the height of the wall is non-linear. When the filling height of the embankment is greater than 7.5 m, the horizontal earth pressure along the back wall can refer Rankine active earth pressure value. From numerical simulation, it can be noticed that the effect of the reinforcement spacing and length on the lateral displacement of embankment is significantly greater than on the settlement. As a result, the vertical spacing and horizontal reinforcement length of the gabion reinforcement should be recommended as 0.5-1.0 m and 0.7H-1.0H, respectively. Since the changes in the compression modulus of foundation soil have a significant effect on the embankment settlement, the compression modulus of foundation soil approaches to be greater than 5 MPa for controlling embankment uneven settlement. In the aim of controlling the growth rate of lateral displacement of the gabion wall, soil cohesion should remain higher than 15 kPa. Finally, the compaction degree of embankment fill should not be less than 93%.

A full-chain carbon capture and storage (CCS) demonstration project was implemented in 2010 by injecting 105 tons of super-critical CO2 per annum down into the brine-saturated low-permeability sandstone and carbonate aquifers at depths of more than 1 620 m in the northeastern Ordos Basin, China. Based on the site-specific geology and the observational data, a numerical injection model is developed based on TOUGH2-MP/ECO2N to simulate the on-going injection process and hence evaluate the injection-induced behavior of the multiphase flow system in the reservoirs. The pressure build-up and the dynamics of the CO2 saturation plume are assessed. The results show that the model developed can reproduce well the performance of the storage reservoir. The CO2 plume spreads outward symmetrically in the horizontal plane. The Liujiagou sandstone aquifer is the most favorable reservoir for CO2 storage at this site. After 3 years of consecutive injection of CO2, the CO2 plume front in the Liujiagou unit is located at about 550 m in 3 years after the commencement of the injection. The pore pressure buildup due to injection is slightly less than 15 MPa. The plume in the Liujiagou unit is expected to migrate to around 700 m away from the injection well after 53 years of post-injection. The major storage reservoir is at the depth interval 1690-1699 m, which contributes over 80% of the storage capacity of the entire reservoir system. The contribution of the reservoirs to the total storage capacity descreases with the depth downward. The leakage of CO2 into the seals is negligible (<0.05%) during the 53 years of simulation period.

By means of the vector sum method and strain-softening constitutive model, a method for progressive failure analysis of strain-softening slopes is developed, which can show the processes of variations of the strain-softening slope safety factor, strength parameters, and failure state of slope and slip. Based on this, progressive failure progress of slope with strain-softening behavior can be simulated. Numerical results are compared to those of the limit equilibrium methods, showing the validity of the proposed method. It is shown that the vector sum method safety factor are in between peak and residual safety factors by the limit equilibrium method; the strength parameters of the main part of slip surface are at the peak value, and the other at the residual value; the proposed method compensates for the defect of the traditional limit equilibrium methods failing to provide a realistic safety factor of slope with strain-softening behavior. The strength weakening zone is found to be consistent with the failure zone in the slope. In the shearing process, the strain-softening behavior first occurs at the toe of the slope, and the strength parameters reduce to the residual value at the toe of the slope; then new strain-softening behaviors occur at the adjacent zone and then enter the residual state; finally the progressive failure develops from the toe of slope to the top of the slope. As the process continues, the progressive failure gradually develops and the safety factor decreases until the end of the progressive failure is reached. The safety factor increases with the increase of residual soft parameter, and the position of critical slip surface moves towards shallow, resulting in an increase of the vector sum method safety factor.

Generally, the tunnel-type anchorage in the suspension bridge has a strong capacity to resist the uplift load due to the appearance of the wedge. In this study, two failure modes, i.e. friction damage on the anchor interface and the surrounding rock rupture owing to lateral extrusion, are investigated based on the assumption of homogeneous surrounding rock and Mohr-Coulomb strength criterion. According to the distribution of the force on interface between anchors and rock, the equilibrium conditions are established to further analyze the change of stress on rupture surface within the surrounding rocks. Besides, the wedge-effect coefficients of the tunnel-type anchorage with respect to two failure modes are derived, comparing with the pull-out capacity of the anchorage with constant section. Subsequently, the formulations about the pull-out bearing capacity of tunnel-type anchorage are obtained under these two failure modes. Research results show that the wedge-effect on the pull-out bearing capacity of the tunnel-type anchorage has obvious regularities for different classes of surrounding rock. Finally, the calculated results by the proposed procedure are compared with measured values of the built tunnel-type anchorage. It is shown that the proposed procedure is reasonable to consider the wedge-effect for the bearing capacity of the tunnel-type anchorage.

This paper aims to study the impact of the Copulas function for modeling the dependence structure between shear strength parameters on geotechnical system reliability and reveal how the dependence structure of shear strength parameters affects the system reliability. The method for modeling the dependence structure of shear strength parameters with incomplete probability information is first introduced. Then the Monte Carlo-based simulation is performed to analyze the system reliability by taking into account Copula-based bivariate distributions of shear strength parameters. Finally typical geotechnical structures such as a retaining wall and a rock wedge slope are analyzed to demonstrate the impact of Copula selection on geotechnical system reliability. The results indicate that the Copulas function provides an effective way to construct the bivariate distribution of shear strength parameters. The difference of geotechnical system probabilities of failure among various Copulas is significant and should not be ignored. In addition, the impact of Copula-based dependence structure on geotechnical system reliability consists of two processes essentially: one is the impact of dependence structure between shear strength parameters on the probability of each failure mode and the other is the impact of all failure modes on system reliability.

The stress distribution of a hinged arch, which is formed by the initially-fracturing main roof, is analyzed, and a new formulation for horizontal thrust is developed by adopting an exponential law. Based on strict geometric transformation, the amount of deformation and subsidence are obtained. The results are validated by comparing the calculated results of the proposed formulation to the numerical results and field observations. The new formulation for thrust is applied to evaluate the structural stability. It is found that for shallow super-long mining face, when the angle of rotation is less than 3°, the thrust of the block with fragmentation i >0.4 increases slightly with the length of mining face and the angle of rotation, and the block fails to destablize due to sliding when i<0.45; if i≤0.3 the thrust increases significantly; the block is prone to fail due to overhigh thrust when the angle of rotation is more than 6°; the stable range for rotation by the proposed formulation is 12%, at least larger than the traditional range.

null

Fracture surface roughness and inertial effect of fluid in fractured rock masses are important factors for characterizing the hydraulic behaviors of rock masses in many rock engineering projects. This study focuses on the influences of fracture surface roughness and inertial effect of fluid on the hydraulic behaviors of 2D discrete fracture networks (DFNs), by solving a set of nonlinear fluid flow equations using the rank-one inverse Broyden quasi-Newton method. Two different boundary conditions are applied to the DFNs; and then the permeability coefficients of the models are calculated. The results show that when the hydraulic gradient J is small (<0.5), the influences of the fracture surface roughness and the inertial effect of fluid on the permeability of DFN are negligible. When the hydraulic gradient J is relatively large (>0.5), the influences of fracture surface roughness and the inertial effect of fluid on the permeability of DFN increase with the increment of the hydraulic gradient. Under two kinds of boundary conditions, the maximum influences of fracture surface roughness and the inertial effect of fluid on the permeability of DFN can be as high as 18.1% and 27.5%, respectively, corresponding to the hydraulic gradient in the range of 0.1-10. Therefore, the fracture surface roughness and the inertial effect of fluid need to be considered when calculating the hydraulic properties of fluid flow in 2D rock fracture networks, especially when the hydraulic gradient is large.

A discrete element method (DEM) contact model for lunar soil taking rolling resistance and van der Waals forces into consideration is used to simulate the horizontal pushing test, which is a simplified machine-lunar soil interaction problem. The soil failure mechanism and the effects of excavation blade depth, inclination and speed are analyzed; and some suggestions for real lunar excavation are offered. The results show that the pushing resistance increases with pushing displacement to a peak state rapidly and then tends to be stable after a sharp declination. With the accumulation of soil heap in front of the blade, the pushing resistance increases slowly again. The pushing resistance, energy consumption and the affected area are greater under larger pushing depth while the slip surfaces are all straight. As the pushing inclination increases, the pushing resistance, energy consumption and the affected area decrease; and the slip surface remains straight. The pushing resistance, energy consumption and affected area increase with the speed. Since the pushing resistance is provided by frictional force between the machine and lunar ground, which is proportional to the weight of an excavator; it is recommended that shallow, inclined and slow excavation should be used in early-stage lunar base construction. This scheme is advantageous in its low demand for the mass of the excavation machine, thus reducing the mass launched from the earth to the moon.

With the proposed FDEM-Flow (combined finite-discrete element method considering fluid-solid coupling) method as a tool, we study the effects of in-situ stress on hydraulic fracturing. Through a numerical examples of an injection hole under the state of different in-situ stresses, the influence of in-situ stress on the direction and morphology of fracturing crack is studied. The results show that both the initiation pressure and direction of fracturing fractures are closely related to in-situ stress. When the vertical in-situ stress keeps constant and the lateral pressure coefficient ?>1, the initiation pressure and breakdown pressure are decreased. When ?>1 and its value is comparatively small, horizontal cracks are mainly produced with some oblique cracks generating. However, when ?>1 and the value is larger, cracks are strictly extended along the direction of the maximum principal stress and oblique cracks no longer occur. When ?<1, fracture pattern is dominated by vertical cracks. But when ?=1, the horizontal in-situ stress and vertical in-situ stress are equal, there is no advantage direction for crack extension. Under the conditions of different lateral pressure coefficients, the direction of crack propagation always coincides with that of the maximum principal stress. The fracture initiation and propagation are mainly controlled by the maximum principal tensile stress, and fractures are initiated in the concentration zones of tensile stress. These results agree well with the available experimental results and theoretical analysis, which demonstrate the effectiveness of FDEM-Flow method to simulate hydraulic fracturing.

Activation of the far-field natural fracture by stress field induced by the hydrofracture can change the rock stress field, reduce the difficulty in formation of interconnected fracture network, such as main, branch and stress relaxation fractures, and improve shale gas reservoir stimulation efficiency, and therefore, has broad prospects of application to the shale gas development. The failure characteristics of natural fractures in the vicinity of hydrofracture are one of the core issues to the activation of far-field natural fracture by stress field induced by the hydrofracture. Displacement discontinuity method (DDM) is used to construct a mechanical model for activation of natural closed fractures in the vicinity of hydrofracture. According to Mohr-Coulomb failure criteria, the constraint conditions in different natural fracture failure states are established; the calculation methods of natural fracture characteristics are formed; and the correctness of these calculation methods are verified by comparison with the existing calculation results of semi-analytical solution. On this basis, the opening, slipping and closing behaviors of natural fractures in the vicinity of hydrofracture are simulated; the natural fracture displacement discontinuity and maximum principal stress distribution characteristics under the action of stress field induced by the hydrofracture are analyzed; and the factors influencing natural fracture failure characteristics are studied. The research results show that the maximum principal stress at the backside surface of natural fracture is in bimodal distribution. As the angle between the natural fracture and the maximum horizontal principal stress increases; the maximum principal tensile stress increases. In the vicinity of hydrofracture, the tension-type failure length of natural fracture is less than its shear-type failure length. As the angle between the natural fracture and the maximum horizontal principal stress increases; the tension-type failure length and shear-type failure length of natural fracture decrease. The natural fracture failure length increases with the increase of net pressure and increases with the decrease of the distance from hydrofracture tip to natural fracture center. The shear-type failure length decreases with the increase of friction coefficient and cohesive force. The research results can provide theoretical guidance for shale gas fracturing technology.

Effectively simulating unsaturated flow is of great significance in many areas such as the soil slope stability analysis, seepage of earth-rockfill dam, contaminant transport. Due to the strongly nonlinear characteristics of Richards equation for the unsaturated flow, numerical solution schemes such as the finite element method with an efficient iterative algorithm usually have to be employed. Picard method as a practical nonlinear iterative method is widely applied to the unsaturated flow field; but usually suffers from convergence oscillation, slow convergence rate and inaccurate solution. In order to improve the computing performance, an efficient finite element procedure based on the adaptive relaxed Picard method is developed. Through the simulations of 1D and 2D unsaturated seepage problems, accuracy, efficiency and robustness of the proposed procedure are validated by comparing with the traditional methods. It is shown that the adaptive relaxed Picard method can effectively reduce the convergence oscillation and significantly improve the convergence rate with the accuracy guaranteed. The proposed procedure provides a helpful reference for the program development and application to unsaturated flow.