It is significant to reveal the evolution characteristics of rock deformation field at various loading rates for safe construction and disaster warning in rock engineering. In this study, uniaxial compressive experiments were conducted on red sandstone at four different loading rates (0.01, 0.05, 0.1 and 0.5 mm/min), and then the evolution of deformation field was analyzed by the digital speckle correlation method. Furthermore, the relationships between the loading rate and rock deformation localization, energy accumulation and release and Poisson's ratio were studied, respectively. The results show that the peak strength and the brittleness of rock increases with the increase of loading rate. It is found that with the higher loading rate, the starting position of rock deformation localization is closer to the value of peak strength. Moreover, the corresponding stress value in this position increases generally. The Poisson's ratios of rock specimen are obvious different at various loading rates. The overall trend is that with increasing loading rate, the Poisson's ratio decreases before the deformation localization, while it increases during the stage of deformation localization. With the increase of loading rate, the density of rock deformation energy increases at the energy accumulation stage, and the energy release rate increases in the process of energy release.

Recently, a new type tunnel emerges with these main feathers of super-large, cross-section and ultra-shallow. However, there are no relevant technical standards for the design and construction of the station tunnel at present. Taken Hongqihegou Station tunnel in Chongqing Metro 3rd line as the engineering background, a novel excavating method named the cross rock pillar method was developed by considering the advantages and disadvantages of the former methods of tunnel construction. A geomechanical model test was used to monitor stress and deformation of surrounding rock mass of the station tunnel during the process of excavation. The stress and displacement of surrounding rock were comprehensively analyzed during each step excavation. According to the geological parameters of the tunnel, a numerical analysis model was established based on ABAQUS software. By comparing monitoring results with numerical results, it indicates that the variation laws of stress and displacement are essentially the same. It is proved that the model test effectively reveals the variation laws of the stress and displacement of the tunnel. Therefore, this study provides valuable experience and useful guidance for the construction of similar projects in the future.

With the help of 3D scanning technology and 3D printing technology, several specimens with the same natural joint but different joint wall compressive strengths were first made, and shearing tests under different constant normal stresses were conducted on them. The direct shear test results show that the strength characteristics on both sides of soft-hard joints affect the shear expansion and shear strength, that is, with the increase of wall strength coefficient , the shear dilatation increases but the shear strength decreases. Moreover, the study also indicates that the damages in the joints surfaces occur on both sides of the soft-hard joints during the shearing process, and the equivalent joint wall strength JCSeq is affected by the wall strength coefficient . Compared with the hard joint side, the equivalent wall strength JCSeq is improved to a greater extent by the soft joint side, and the trend is strengthened with the increase of the wall strength coefficient . That is why JRC-JCS strength formula has some limitations in predicting the strength of soft-hard joints. Based on the above analysis, JCSsoft is the basis for establishing the modified relational function f( ) of soft-hard joints equivalent wall strength JCSeq with respect to wall strength coefficient . And an improved JRC-JCS shear strength formula considering different wall compressive strength combination types of soft-hard joints is proposed, which is the complement of current JRC-JCS strength formula.

A state-dependent constitutive model for unsaturated clays is proposed based on the double plastic mechanisms (i.e., shear sliding and loading collapse mechanisms), and the coupling between hydraulic hysteresis and stress-strain behavior. In the proposed model, the stress integration at the intersection of two solid phase yield surfaces is addressed using an implicit integration algorithm. The consistent tangent matrix of constitutive model is explicitly derived by considering the relationship between stress and suction. Finally, the numerical simulations of triaxial tests on unsaturated clay are carried out to demonstrate the abilities of the constitutive model. The calculated results of strain-controlled tests on unsaturated soils with different step sizes are compared to demonstrate the convergence and accuracy of the implicit algorithm. The two-dimensional ground consolidation of unsaturated soils is analyzed using the U_DYSAC2 program to demonstrate the validity of the consistent tangent matrix.

To understand creep mechanism of soft clay, one-dimensional consolidation creep tests, bound water measurement tests, and scanning electron microscope (SEM) tests on Shanghai mucky clay are conducted. The variations of bound water and microstructure in the process of consolidation-creep are quantitatively analyzed. The results show that, free water discharged under condition of p?k. Meanwhile, the creep strain and steady-state creep strain rate increase significantly. The creep process is dominated by osmotic bound water. The microstructure is solidified gradually with bound water discharging and the water content decreases exponentially with increasing loads. Deformation of soft clay is caused by consolidation-creep coupling effects. The original loose skeleton-flocculated structure translates into dense reunion-flocculated structure. The directionality of particles decreases under the long-term creep adjustment. The quantity of fine pores increases massively after creep tests and the pores with diameters less than 1 μm in dominancy. The response mechanism among loads, bound water and particles can explain engineering phenomenon well.

The horizontal vibration of the pipe pile groups in saturated soil is studied by considering three-dimensional wave effects on the saturated soil, and the horizontal dynamic impedance of pipe pile groups is also obtained. The soil around pile and pile core soil are considered as two-phase saturated soil, and the mechanical property of the saturated soil is described by the theory of porous medium. The horizontal dynamic impedances of the saturated soil around pile and in pile core are obtained by using the mathematical physics means. Using the wave spread theory and superposition principle, the horizontal force of any pipe pile is obtained by considering boundary conditions and orthogonality of trigonometric functions. The horizontal dynamic impedances of pipe pile groups in saturated soil is obtained by considering the displacement compatibility conditions and horizontal force equilibrium conditions at pipe pile top at rigid cap. The influences of physical parameters on the dynamic characteristics of pipe pile groups in saturated soil are analyzed and discussed by using 2×2 pipe pile groups as an example. The result indicates that there are some differences between the horizontal dynamic impedances of the solid core pile groups and the pipe pile groups. It needs to consider the influence of the difference between the saturated soil around pile and in pile core on the dynamic impedance of pipe pile groups in saturated soil. Length to diameter ratio and wall thickness of pipe pile have great effect on the horizontal vibration of pipe pile groups in saturated soil, showing that the size of pipe pile should be focused in the actual engineering design.

In this study, a model for a pre-stressed bolt with two anchorage segments was established, by considering both the superiority of point-anchoring bolts and the laws of destruction partitioning and gradually destructive depth development. The model is basically used to solve the following problems. When lengthening-anchoring or fully-anchoring bolts were applied in the deep roadway, the anchorage segments of the anchor was not as uniform as the whole, and the shear capacity was low. While point-anchoring bolts could not adapt to the large deformation. Afterwards the stress characteristics and bearing capacity of two anchorage segments were studied using theoretical calculations and numerical simulations. Moreover, a comparative analysis was also provided in this study. The results show that the bolt with two anchorage segments obviously changes the shear stress curve of the lengthening-anchoring or fully-anchoring bolts, which turns the trend of the original negative exponent into the gentle. Its shear bearing capacity is also improved by the partition supporting pattern with two anchorage segments. Under the conditions of the same pre-stressed force, the load and shear stress of the anchoring section are obviously lower than that of point-anchoring bolts. Under different pre-stressed conditions, these two anchorage sections are separately used as the supporting function of point-anchoring bolts according to various periods of surrounding rock deformation, which continues to show the superiority of point-anchoring bolts.

The weathered granite soil varies significantly from Lüliang mountains to south of the Yangtze river. To comprehensively understand the mechanical properties of weathered granite and evaluate of the engineering properties of subgrade filling in mountains, a series of laboratory tests was performed in this study, including routine soil test, X-ray diffraction test and large-scale triaxial test. The experimental results show that the increase in peak deviatoric stress qf due to an increase in mean stress p is observed as nonlinear under lower confining pressure because of the existence of particle breakage. The relationship between peak stress ratio Mf and state stress ratio Mc is modified on the basis of previous research results. the elastic-plastic constitutive model of compacted granite soil in Lüliang mountains is proposed. Based on the continuity assumption of the isotropic compression-unloading test and the initial shear, combined with the Duncan-Chang model, all the parameters of the proposed constitutive model of weathered soil under the modified strength condition were obtained. After comparing the results of numerical calculation of proposed model with the experimental stress-strain data, it is found that the product of the state stress ratio and peak stress ratio was not constant but the power function of average main stress.

Heat transportation will change the mechanical properties of rock and soil materials considerably in subsurface engineering applications such as radioactive waste repositories, geothermal energy extraction and underground oil/gas storage. Thermal properties of rocks, soils and buffer materials, including thermal conductivity, thermal expansion coefficient, specific heat capacity, play a crucial role in design and evaluation of safety performance of these projects. The thermal expansion coefficient tests are conducted on sandstone with bedding planes by using the independently developed instrument for thermal expansion coefficient of rocks. The evolution of axial and radial thermal expansion coefficients of rocks are studied. The results show that for rock samples with horizontal beddings, with the change of temperature, axial thermal expansion coefficient is about 14×10?6 ℃?1, while about 9×10?6 ℃?1 in radial direction under uniform expansion. The axial thermal expansion coefficient is about 1.56 times as large as the radial thermal expansion coefficient, and the rock sample shows obvious anisotropy. Compared with rock samples with horizontal beddings, the samples with inclined and vertical beddings shows no significant difference in axial thermal expansion coefficient, but their radial thermal expansion coefficients are larger. Under the same temperature, the order of radial expansion coefficient from big to small is vertical, inclined and horizontal beddings. Because of the existence of beddings, the expansion properties of sandstone and compositions of beddings are different, which leads to the difference of axial and radial expansion coefficients. The results may provide a helpful reference for better understanding the thermo-mechanical behaviors of rocks.

Under nature and saturated conditions, a series of uniaxial and triaxial compression tests was conducted on gypsum rock mined from Jinghua mine in Jingmen. Mechanical properties of water-softening gypsum rock and the corresponding softening mechanism were studied by the scanning electron microscope (SEM). The results show that the strength and deformational properties of gypsum rock are tremendously weakened by water. With different concentrations of SO42- and Ca2+ in the solution, the softening coefficient of rock strength ranges from 0.6 to 0.72 and the softening coefficient of elastic modulus is in the range of 0.66-0.75. Besides, the cohesion c decreases significantly whereas the internal friction angle ? remains essentially unchanged. Since the gypsum rock has good ductility, the critical confining pressure of the elastic deformation to the ideal plastic deformation is found between 2.5 MPa and 5 MPa in nature sate, while the critical confining pressure is lower than 2.5 MPa for saturated gypsum rock. The softening mechanism of gypsum rock is closely related to the slight solubility of the calcium sulphate dehydrate, which is the major ingredient of rock. In distilled water, the rock structure is weakened by the dissolution process of gypsum mineral. In saturated CaSO4 solution, the dynamic equilibrium state reaches when the dissolution rate is equal to the re-crystallization rate. The continuous dissolution and re-crystallization processes tend to gradually loosen the original tight structure, resulting in softening mechanical properties. In conclusion, this study provides deep insights into the transformation process of the mechanical properties of the gypsum rock in accordance with different solutions and confining pressures, which is of great significance for the stability analysis of goaf area in gypsum mines.

The five-star-shaped concrete pile is a new type of pile with a special shaped cross-section. It is formed by cutting five circular arcs in a circle pile. The main parameters of the cross section include the radius of the circumcircle R and the corresponding circumcircle central angle θ, the radius of inward cutting circular arc r and the corresponding circumcircle central angle 2x. Through the optimization analysis of the section size, the maximum of the circumference of the five-star-shaped piles and a maximum ratio of perimeter to the area of five-star-shaped piles are derived. To further explore the vertical bearing capacity of five-star-shaped piles, experimental studies are carried out on the comparison of four single piles in the dry sand by the sand pouring method. These four single piles are the maximum ratio of perimeter to the area of the five-star-shaped pile F1, the maximum of the circumference of the five star-shaped pile F2, the round pile C1 with the same perimeter of F2, and round pile C2 with the same cross-sectional area of F2. The following conclusions are drawn from this study. The side surface area of F2 gets close to that of F1, and the ultimate bearing capacity is approximately the same. The concrete dosage of F2 is 0.75 times that of F1, the bearing capacity per concrete is higher, and thus F2 is optimal cross-sectional pile. Compared with C2, the side surface area of F2 is 1.53 times that of C2, and the ultimate bearing capacity of F2 is 2.4 times that of C2. It shows that F2 has better bearing capacity than that the same amount of concrete, and the expansion effect of cross-section is obvious. Compared with C1, F2 pile cross section area is 0.44 times that of C1, the ultimate bearing capacity is 0.96 times that of C1, and the unit volume concrete bearing capacity of F2 pile is 2.21 times that of C1, which exhibits higher load cost performance. At all levels of load, the ratio of the lateral friction of the five star-shaped piles is more than 80%, especially F2, and the proportion of side friction resistance is above 90%, which shows the characteristics of friction pile. This study provides theoretical support for the engineering application of five-star-shaped piles.

This study aims to investigate deformation characteristics of natural structural planes with a certain thickness under normal cyclic loading using laboratory tests and theoretical analysis. The studied natural structural planes are flat and slightly rough with a thickness of 2-3 mm. The host rock is limestone and the infillings are primarily rock fragments with a small amount of mud. Firstly, the variation process of normal stress applied on structural planes under engineering effect is summarized. The loading path for experiments is then designed based on the characteristics of normal stress versus closure deformation curve. Then, we present the sampling operation, experimental results of physical and mechanical properties of samples, and the preparation of testing specimens. The specimens are applied to conduct single cycle and multiple cycles of normal loading and unloading experiments. In addition, we summarize the testing curves and the normal deformation mechanisms of structural planes and study the constitutive relation of structural planes under normal cyclic loading conditions. It is found that, when the normal stress is within 20 MPa, the hyperbolic model and the improved one with the consideration of a modification coefficient can be used to well simulate the relation of normal stress versus closure deformation at various loading stages. Meanwhile, when the normal stress increases up to a high-stress level (i.e., the stress magnitude above 20 MPa is defined as high stress, but below that it is recognised as the range of mediate to low stress.), new characteristics were observed from testing data. The discrepancies between predicted values and monitored results become increasingly greater. Finally, the probable reasons for new characteristics are analyzed. It is suggested that it is necessary to use natural samples to investigate normal deformation characteristics of structural planes under high-stress conditions.

Based on cavity expansion theory, a method for calculating compression modulus of sand compaction pile is developed by considering the change of sand pile volume in the compaction process. The yield zone of the soil around the pile is calculated by Mohr-Coulomb yield criterion. Using the e-p soil compression curve, a formula for average compression modulus is deduced. A method considering the influence of buried depth is proposed for calculating pile-soil stress ratio of sand compaction pile. The Honai-Haiphong highway project in Vietnam is selected as a case study. Analysis indicates that as the casing pipe diameter increases from 0.5 m to 0.6 m, the compression modulus of sand compaction piles and soil around pile decrease, and the average pile-soil stress ratio increases from 2.56 to 2.72. This confirms the statement in Technical Code for Composite Foundation that the pile-soil stress ratio adopts the low value for the soil around pile with high strength, or vice versa. The theoretical calculation result of the pile-soil stress ratio is between 2 and 3 suggested by specification, demonstrating the theoretical method is reasonable and reliable, and greatly improves the sand compaction pile design.

It is well known that shear strength of soil plays a key role in analyzing the stability of slope. To clarify the variations of shear strength and parameters of the loess after leaching for different times, a series of direct shearing tests is conducted on samples from Lanzhou loess. Results show that, shear strength and cohesion of unsaturated loess decreases significantly nonlinearly with the decrease of soluble salts, and inner friction angle is slightly changed. Specifically, the reduction reaches the maximum value after leaching 1 time and trends zero after leaching 5 times. It is also found that the reduced rate is large when soluble salt content is greater than 2 620 mg/100 g and then reduces. The decrease of loess’s shear strength comes primarily from the decomposition of soluble salts and calcium carbonate at the initial stage, and it relies on the dissolution of soluble salts and weakening of matric suction afterwards.

The thermal properties of loess are important in the study of genetic mechanism of freezing-thawing hazard and thermal engineering calculation. However, temperature greatly impacts the thermal behaviors of loess. In this study, the thermal properties of loess samples with different water contents and dry densities are tested using Test Protocol Hot Disk TPS 2500S under different temperatures. The evolution of thermal properties of loess with temperature is discussed. The study reveals the following results. The thermal conductivity development of samples as temperature decreases can be divided into three stages: slowly increasing stage, rapidly increasing stage and stable stage. As temperature decreasing, the specific heat capacity of samples increases initially and then decreases, finally stabilizes. The thermal diffusivity of samples decreases at first and then increases with decreasing temperature. When the temperature is below ?20℃, the thermal diffusivity keeps stable. The larger the water content, the greater the influence of temperature on the thermal properties of samples. There exists large difference of thermal properties between unfrozen soil and frozen soil. Dry density impacts n the thermal properties mainly through the minerals and free water content of soil. In applications of freezing-thawing hazard and thermal engineering, thermal properties can be selected in accordance with soil temperature, water content and dry density. Determination of whether the soil is frozen or not is the key to determine the thermal properties.

In the deposition process, the calcareous sand, mostly retaining tiny pores of the protozoan skeleton, shows hierarchical porosity, irregular shape, high angular edge, low strength and fragile, cemented property of particles. Therefore calcareous sand are different from general terrestrial and marine sediments in its engineering properties especially the permeability due to the hierarchical porosity and highly-irregular shape. In this study, the double-ring permeation experiments are carried out to analyze the effect of density and particle size on the permeability of calcareous sand. Test results show that the seepage rate of calcareous sand increases gradually and then stabilizes within a range of small fluctuation. The permeability coefficient of calcareous sand is negatively correlated with the nonuniformity coefficient Cu. As the inflection points of particle-size distribution curve become smaller than effective diameter d10 appear near the weight-accumulative 4% percent diameter d4 of soil, the ratio of d4 to constrained diameter d60 is adopted to analyze the relationship between the dry density and the permeability of calcareous sand. An exponential function is fitted between permeability coefficient with dry density and compaction degree . The fitting exponential function of the calcareous sand permeability coefficient is or . If the values of Cu≤5, 5

The uniaxial and conventional triaxial compression tests were conducted on granite from Sanjiang region of Yunnan-Tibet railway in the laboratory. The deformation characteristics and the damage evolution law of rocks were examined under different confining pressures. The results show that the internal cracks experience five stages during the compression deformation process, including no crack, crack initiation and stable propagation, crack accelerating propagation, crack decelerating propagation and crack termination. In addition, confining pressure significantly influences deformation characteristics of rock. Based on the above analysis and the damage model of micro units, a damage constitutive model is developed under the triaxial compression condition. The proposed model has a simple function form and distinct physical meaning parameters, which can be used to describe both the pre-peak and post-peak deformation characteristics of rocks. Furthermore, theoretical results obtained by this model are in good agreement with experimental results in the literature. The rationality of the model is verified by the parameter analysis.

To explore the salt expansion behavior of coarse saline soils containing sulphate under drying, sets of laboratory experiments were conducted to investigate the impact of void ratio, salt content, initial water content and temperature on salt expansion rate. The deformation behavior as water content decreases are the compete of two types of soil-water interactions: shrinkage of soil due to drying, and the swelling of crystallized salts dissolved in pore water. The decreasing rate of water content shows positive correlation with void ratio, temperature and initial water content but negative correlation with the salt content. The test results show that the salt expansion rate of some sample reaches 9%, suggesting that the salt expansion of sulphate saline soil with decreasing water content should not be ignored. It is expected to provide guidance for evaluation on the salt expansion of foundations of residual sulphate saline soil with high salt content in the northwest region.

Due to the lack of mature theoretical system, the research progress of p-y curve is slow. With the consideration of the actual stress state of foundation soil and deployment of Vesic expansion theory, a new method for analysis of p-y curve based on stress increment is developed. This method combines the contributions of soil radial stress increment, longitudinal stress increment effect and deep rotation angle of pile to the horizontal soil resistance. It is a type of solution of p-y curve considering various factors. The method can be applied to different type of piles. The case results show that this method is correct and accurate in analyzing the behavior of laterally loaded pile. It produces better precision in large deformation analysis than Reese, API and Kim methods. The above research result is of great significance for in-depth study of the p-y curve development.

In this study, a rheological mechanical model was developed for the pillar-protective roof system in gypsum mines. The gypsum pillar was considered as a visco-elastoplastic body in the rheological model satisfying the Nishihara model, and the pre-setting protective roof on the top of the goaf was simplified as an elastic rectangular sheet. On the basis of the established model, a deflection differential equation was derived for the stabilization time of protective roof under the support of the pillar. Besides, a solution of Galerkin method was applied to the equation in accordance with the boundary conditions of the protective roof at three stages. When the stress ? on the pillar is greater than the limit friction resistance ?s , the plastic rheological deformation of the pillar gradually increases with time, which eventually results in the failure of the pillar-protective roof system. Based on a case study, the proposed mechanical model can forecast the stable time of the roof rock in the pillar-protective roof system of mined-out area by considering the rheological behaviour of pillars.

This article investigates the problem of Rayleigh (R) waves propagation in double unsaturated soils. Firstly, wave equations of three-phase porous medium are extended by introducing the tortuosity of fluid phases into the conservation of momentum, then decomposed with the aid of Helmholtz theorem and the sum of potential functions of body waves. The dispersion equations are established with the boundary condition and the continuities of soil layers. The influences of saturation degree, tortuosity and cover thickness on the velocity and attenuation are discussed. The results show that the R wave velocity decreases linearly with the saturation, and increases with the intrinsic permeability ? in the intermediate range. The influence of tortuosity of water phase on R wave velocity is significant in high frequency domain. The tortuosity of air phase is negligible. For a ground system of upper-soft and lower-hard layers, the R wave velocity becomes smaller for thicker surface layer, and approaches the velocity in overlaying soil, especially for high frequency or surface layer with low stiffness.

In this paper, experiments were conducted on jointed rock mass to study characteristics of elastic wave propagation under different loading conditions. Firstly, by comparing the amplitude and velocity of the elastic wave with the damage degree of rock mass, the macroscopic damage variable of jointed rock mass were characterized by the amplitude of elastic wave. Then, the mesoscopic damage variable was defined through the statistical strength theory. Afterwards, on the basis of the continuous damage theory, a constitutive model by coupling macroscopic and mesoscopic joints was established. Finally, the model was verified and analyzed by combining experimental data. The study shows that, for the intact rock, initiation and propagation of mesoscopic cracks have less effect on the amplitude and velocity of elastic wave. However, for jointed rock mass, the opening and closing of macroscopic joints have a great effect on them. The amplitude and velocity of elastic wave have similar variation law when jointed rock mass is subjected to external force. In comparison with the elastic wave velocity, the attenuation of elastic wave amplitude is sensitive to the changes of the damage degree of jointed rock mass. Thus, it can be used to characterize the macroscopic damage variable of rock mass. Under different loading methods, the damage characteristic of jointed rock mass is determined by macroscopic and mesoscopic joints and its stress state. Therefore, the constitutive model can well reflect damage mechanical properties of jointed rock mass under different stress paths. In triaxial compression tests, the effect of macroscopic joints on the damage of rock mass in the direction of axial compressive stress plays a significant role in the middle and later period of the experiment. While in triaxial tension and compression tests, the macroscopic joint maintains its effect on the damage of rock mass in the direction of axial tensile stress within the whole process of the experiment.

Natural Na-bentonite, from Gaomiaozi (GMZ), Inner Mongolia, was tested to study the long-term stability of buffer/backfilling material for high-level radioactive waste geological disposal, . This paper analyzed the effects of γ-rays irradiation on the strength and deformation characteristics of densely compacted Na-bentonite sample. The bentonite was irradiated by cobalt source at different cumulative irradiation doses in laboratory. The mineral composition changes into some more stable silicate and aluminosilicate minerals after irradiation. Then the triaxial shear test and swelling test show that the stability of the microstructure of montmorillonite decreases, and the shear strength and the peak shear strength of bentonite increase with the irradiation dosage under the same confining pressure, due to the change in the mineral composition. With the increase of confining pressure, the strength trend increases gradually slowly, and the effect of irradiation dose on the strength turns to be weaken. It indicates that the confining pressure inhibits the strength increase of bentonite sample. Due to reduction of montmorillonite after irradiation, the swelling deformation property is weakened. The performance is the decrease of the expansion force, free swelling strain, load swelling strain and other indicators.

Dynamic compression tests were carried out to investigate dynamic properties of artificial frozen sand under active confining pressure. In the experiments, a variable cross-section split Hopkinson pressure bar (SHPB) apparatus with a diameter of 50 mm was employed. Frozen sand was sampled from -94.52 m level at a coalmine in Jining, Shandong province, and the ratio of length to diameter of the cylindrical frozen sand specimen was 0.5. The influence of confining pressure and strain rate on dynamic properties for artificial frozen sand was analyzed as well. The results show that under uniaxial loading condition, the stress-strain curve can be divided into four stages, i.e., elastic stage, plastic stage, viscous stage and failure stage. Moreover, the corresponding failure mode is brittle. However, under confining pressure, the plastic stage has been extended, there is no viscous stage and crack failure occurred. Dynamic compressive strength of artificial frozen sand under confining pressure is greater than that under uniaxial loading condition, and increases with the increase of confining pressure. When the strain rate is 220 s-1 and the temperature is -15 ℃, under the confining pressures of 0.5 MPa, 1.0 MPa, 1.5 MPa and 2.0 MPa, dynamic compressive strengths are 1.19, 1.40, 1.58 and 1.81 times that under uniaxial loading condition, respectively. Strain convergence phenomenon occurs under different confining pressure conditions, and the convergence position tends to uniaxial loading condition. The dynamic compressive strength increased with the increase of strain rate.

null

It is unreasonable to calculate the buoyancy force value of underground structures near riverside directly using ground elevation or river level as anti-buoyancy water level. Seepage analysis of the field is necessary for anti-buoyancy design. Based on Bennett hypothesis, a simplified algorithm for seepage analysis of double strata near riverside is presented to calculate the hydraulic head distribution in high-permeable layer. A practical method is followed to calculate the buoyancy force on the bottom slab of underground structures for anti-buoyancy design. This method can be applied for different boundary conditions, e.g., the blankets with different thicknesses and permeability on the upstream and downstream of levee , or the blanket with finite or infinite width. This algorithm only requires simple calculation parameters, but the results are accurate enough for using. After verification and comparison, a conclusion can be drawn that downstream small-scale structures in high-permeable layer have little influence on hydraulic head distribution, but the large-scale structures or the narrow-width blanket downstream will rise the head dramatically. The buoyancy in the middle of bottom slab is greater than the surrounding area when the bottom slab of basement locates in the blanket. But it changes to linear distribution when the bottom slab of basement intrudes into high-permeable layer.

The soil arching effect is closely related to the supporting force and depths during tunneling construction. The soil arching effect was studied experimentally considering various depths of embedded tunnel and densities of the sand layer. The displacement of sand particles and development of soil arching near the face of excavation were analyzed. The development of soil arching and the its failure under various depths were revealed. Comparing with the laboratory tests, the two-dimensional particle flow code (PFC2D) was employed to simulate the soil arching effect. The results show that the development of soil arching under different depths are the same, closely related to the supporting force and the ground settlement. The failure of soil shows funnel shape at low density , and strip shape at high density. The range of soil arching increases with the increase of depths. The change of particle contact force, porosity and average soil pressure are the same via PFC2D simulation for various depths. Finally, the conclusion is drawn that the soil arching effect obtained from the laboratory tests is similar to that of PFC2D simulation.

This study aims to study the damage and rupture laws of the main roofs between coal seams with a close distance. A continuum damage model was adopted to investigate the damage laws of floor caused by mining the upper coal seam. Morever, the main roofs were clearly classified based on their positions in damage area of the floor in the goaf. Thus, an initial fracture span formula was obtained for the damaged main roofs. The numerical calculation results show that the damaged area in the floor is spoon-shaped and the damage value decreases from the damage core area to the marginal area. The main roofs between coal seams can be divided into three types, including no damage, partial damage and complete damage. In addition, the influence of three types of main roofs on the behaviours of the lower strata decreases successively. The fracture length declines with the increase of the damage value, until the limit damage value occurs, which results in the fracture length approaching to zero. The height-length ratio of critical blocks formed from the damaged main roofs increases and the voussoir beam slides easily to a great extent. The laws of damaged main roofs are verified in Xiegou coal mine.

In the existing theoretical analysis, laminar rock mass of sliding bed is normally equivalent to homogeneous or horizontal non-homogeneous body for calculations of the internal forces and deformation on the anti-sliding pile. However, most of slide beds in Three Gorges reservoir area are composite inclined rock mass. Firstly, numerical simulations using the 3DEC software were conducted to examine the relationships of the effective scope of the anti-sliding pile with the cross-section width and depth of pile. The upper and lower limits of the effective scopes of the pile were determined. Secondly, a formula was developed for the comprehensive foundation coefficient (KZi) of sliding bed on composite inclined rock mass within the effective scope of pile by the area equivalent. Finally, several formulas of internal forces and displacement considering KZi, were deduced on the elastic anti-sliding pile. Moreover, correspondingly MATLAB calculation programs were compiled. Taking the Majiagou landslide as an example, the internal forces and displacement of anti-sliding pile under the sliding surface were calculated by the equivalent method, the horizontal layer foundation coefficient method and KZi-based method, respectively. Comparisons between the KZi-based method and the horizontal layer foundation coefficient method show that the displacement of sliding surface by the former is greater than that of the latter (14.6%) and the difference of internal force is negligible. According to the displacement of sliding surface, the displacement of pile top calculated by the KZi-based method is similar with that of field monitoring, while it is smaller by a conventional method and dangerous for design. Thus, the KZi-based method can provide theoretical support for the area of composite inclined rock mass on the design of anti-sliding piles.

This study is to determine the width of coal pillar between roadways of the fully mechanised caving face with double roadways layout in a thick coal seam. The coal pillar between haulage gate and auxiliary haulage gate of the 4301 working face in the fourth panel was selected as a case study. Firstly, based on theoretical analysis, the pillar between roadways was along tendency divided into mining influence area, relative stability area and bolt supporting area after the first mining. The width of first mining influence area was 2.82 m determined by using the limit equilibrium theory, and the width of pillar between roadways was subsequently obtained as 7.83 m. Then, a numerical method was conducted to systematically simulate the laws of stress evolution, failure, and roadways deformation of pillars with the widths of 4, 6, 8, 10, 12, 15 and 20 m under the secondary mining influence. It is found that after the first mining, the high stress transfers from the coal wall to the pillar with the increase of pillar width. Besides, the definitions of the maximum and the minimum critical sizes are given. It is pointed out that the width of the narrow pillar should be less than the maximum critical size. By synthesising numerical results, theoretical results and principles of the pillar, the most favourable width of the pillar between roadways is suggested as 8 m. Finally, the rationality of the pillar width is verified by engineering practice. The results can provide a reference for the determination of narrow pillar between roadways of fully mechanised caving face with similar mining conditions.

At present, the studies of adjacent pile deformation caused by foundation excavation in transient case are common. Less attention has been paid to the influences of time behavior on the pile-soils interaction. The disturbed soils show the rheological properties during excavation. A two-stage method considering the soil visco-elasticity is proposed to obtain the time-domain solution for the interaction between adjacent pile and foundation pit excavation. On the first stage, based on Boltzmann visco-elastic model, a Mindlin’s time-domain solution is derived to analyze the influence of excavation on soil additional stress at the position of adjacent pile in visco-elastic foundation. On the second stage, a Pasternak’s two-parameter foundation model is adopted to simulate the interaction between pile and soils. A simplified time-domain solution of pile horizontal deformation is derived in consideration of rheological effects. Finally, the simulation results from finite difference software are compared to simplified time-domain solution with good agreements. The parameters of Boltzmann visco-elastic model (bulk modulus, shear modulus, viscosity coefficient), pile diameter, excavation depth, distance between pile and foundation pit and excavation size are analyzed to investigate the influences on pile deformation. It is shown that the presented simplified time-domain solution can reflect the influence of excavation on adjacent pile deformation and the development trend of pile deformation with time. It may provide certain theoretical basis for practical projects.

This article develops a new method using particle flow code (PFC) to simulate the roller compaction test of rockfill. Firstly, a PFC model composed of random polygonal clusters is established. By considering the grading, meso-structure and breakage phenomenon of rockfill particularly, the model can simulate the macro and meso mechanical responses of rockfill during rolling compaction progress. Then, a new technology is developed to simulate rolling compaction load and rolling compaction progress, and adopted for simulating the in-situ compaction test. At last, the roller compaction test of rockfill in Shuibuya project is simulated. The results of numerical model and physical sample are compared qualitatively and analyzed in detail, which validates this numerical method. This new numerical method will play an important role in simulating the rolling compaction of rockfill quantitatively and studying the roller compaction effect on rockfill.

Recently, field data shows that considerable wall deflection can be induced by pre-excavation dewatering. A numerical model was established and verified based on field data, and a parametric study was carried out to investigate the effect of soil permeability on wall deflection. The results indicate that the wall deflection, in a prescribed dewatering time, will be more obvious with the better soil permeability, the larger anisotropy coefficient of soil permeability within the dewatering depth. That is because, in this case, greater pore pressure decrease will appear at the soil layers, and larger total stress redistribution will appear at the soil-wall interface. Moreover, the depth of wall deflection has relation with the permeability of soil below the bottom of dewatering depth. If the soil below the bottom of dewatering depth is aquiclude and the center of the aquiclude thickness is deeper than the dewatering depth, the wall deflection will roughly appear within the dewatering depth. If the soil below the bottom of dewatering depth is well-drained soils, the wall deflection may reach the location of another aquiclude below the well-drained soils. Besides, when dewatering wells enter the well-drained soils, greater deflection increment will occur with the dewatering depth increase, and over dewatering should be avoided strictly. However, when thick and poorly drained soils appear below the bottom of dewatering depth, wall deflection increment induced by increasing dewatering depth is not apparent.

Currently, locally non-differential piecewise functions, the explicit forward Euler method (EFEM), and the Newton-Raphson iteration method (NRIM) are commonly adopted plastic yield criterions for implicit finite element solution. However, the EFEM will cause stress deviation from the yield surface and lead to non-convergent result, and the NRIM needs to frequently calculate derivative of yield criterion. They are both inconvenient for non-differential functions. A method combining the semi-implicit backward Euler method and the Steffensen iterative method is proposed for these piecewise yield criterions to update stress and consistent tangent modulus. The combination method is applied to write code in ANSYS’s UserMat module for an elastic-perfect plastic constitutive model, based on a composite unified strength criterion with unified strength theory and tension cut-off. Then the custom constitutive model is applied to simulate excavation process of a deep circular tunnel, and the numerical simulation results fit well with the theoretical solutions, which verifies the correctness and practicability of the model and the combination method. The combination method can prevent stress deviation from the yield surface and does not need to solve partial derivatives of the plastic flow function. It is simple and easy for popularization and application.

To investigate the permeability characteristics of filling media and its mechanical properties and failure mechanisms under high water pressure conditions, a large-scale testing system was developed for the measurement of coupled seepage and triaxial stress. We proposed a pressure chamber especially for coupling three-dimensional (3D) seepage-stress, and designed a perfect sealing and loading system. This system has multiple functions including automatically applying triaxial stress, surrounding rock pressure and osmotic pressure, gushing water and sand acquisition and monitoring parallel multivariate information. It can be used to perform the tests of permeability and correlative mechanics measurement for large-scale and different patterns karst conduit and karst fissure under the initial confining pressure. The maximum applied lateral pressure and permeability pressure are up to 2 MPa and 5 MPa, respectively. The maximum axial loading capacity is 300 kN. The precise acquisition and analysis of seepage pressure, strain and displacement in the water environment can be achieved by the fibre technology, flexible waterproof technology of resistor elements and automatic data acquisition system. The permeability tests were conducted on filling medium under the condition of occurrence, and seepage characteristics of different kinds of filling medium were systematically analysed. The seepage failure model of filling medium was further revealed. Therefore, the developed system provides a new testing method for hydropower and tunnelling engineering.

For the borehole image obtained by digital panoramic borehole camera system, a method for recognizing structural plane is described based on characteristic function. According to the sinusoidal shaped characteristics of borehole side-wall discontinuities in the image, the method adopts standard sine function as characteristic function. The recognition of borehole structural plane is realized by changing the function parameters to match the nearby pixel grey gradient distribution and by choosing the best matching sine curve as the characteristic curve of structure plane. The method ignores the traditional process of manual selecting characteristic points and fitting characteristic curve. It provides a practical and effective way for the automatic recognition of borehole structural planes in engineering application with good stability and high accuracy.

Current methods for groundwater level testing could not accurately reflect the actual variation of the groundwater level under vacuum pre-loading, and thus their limitations were comprehensively analyzed in this paper. A new testing method, namely closed piezometric testing method-laser testing method, was described in detail. Moreover, field tests were conducted to verify the rationality of the developed method. The results show that some restrictions are found in the existing closed piezometric testing method. On the one hand, the employed devices are very complex and challenging to install. On the other hand, experimental results are affected by many factors, which indicates that the results could not truly represent the ground level under vacuum pre-loading condition. However, the proposed new testing method has merits of easy, low-cost and intuitive results, owning to its principle of laser ranging. The field testing results show that during vacuum pre-loading, the groundwater level falls with the subsidence of ground foundation. In addition, the overall decline is slightly greater than the consolidation settlement of the foundation. Therefore, the new method can accurately reflect the actual variation of the groundwater level, and is useful to evaluate and analyse the effect of vacuum pre-loading.

This study aims to improve the automation and accuracy of extracting parameters of rock mass structural planes. By considering rock mass characteristics of the structural plane in borehole images, a new method was developed for the pre-processing of structural plane image and the extraction of structural plane parameter. First, the noise in borehole images was removed with an adaptive median filtering method. An improved variance method was also proposed by comparing the gradient operator method and the maximum interclass variance method, which is more suitable for the actual borehole image segmentation. Then, a Canny operator was used to detect the edge of segmented images. Finally, the sine curve of the structural plane was obtained by the polynomial curve fitting, and thus the structural plane parameters including dip, dip angle and width were calculated. The comparison of results obtained by the proposed extraction method and the traditional manual recognition shows that the proposed method is more accurate. Therefore, it is demostrated that the proposed method is feasible and accurate, and its extraction efficiency is higher than the traditional manual method.