The design specification for Chinese hydraulic retaining wall (SL_379-2007) uses the safety factor criterion in stability analysis, whose allowable values are stipulated in a range of 1.20 to 1.35, which is much smaller than those specified in the overseas design regulations. In this paper, a probabilistic analysis approach is used to establish the relationship between the target reliability index and safety factor. Results show that the factors of safety follow a log-normal distribution relationship. The related criterion for the ‘Safety Margin Ratio (SMR)’ has been updated. The study uses a target case that has a reliability index of 3.2. It has been found that the allowable factor of safety is in a range between 1.43 and 1.55. Using the SMR approach, the authors further demonstrate that the calibrated allowable factor of safety is equally allocable to the extended target cases that involve different geometric and shear strength values of this retaining wall. The method and analytical results described in this paper are useful for regulation makers.

In China, the gas storage in underground salt cavern is normally built in the lacustrine sedimentary salt rock, which is rich in muddy interlayers. Since the permeability and porosity of muddy interlayers are higher than those of salt rock, the sealing of the gas storage is determined by the permeability of muddy interlayers. In this study, a new theoretical model is established for the seepage analysis of bedded salt cavern gas storage. This model is based on the sedimentary characteristics of mudstone interlayer, which is horizontally distributed. Natural gas leakage is also calculated by using the porosity-permeability testing data and the actual operation parameters of the mudstone in a gas reservoir. According to the above results, this study analyzes the influencing factors and change regularities of gas leakage and seepage range. Furthermore, an applicability evaluation is given by considering several key parameters in siting and designing the gas storage reservoir, such as safe distance from salt cavern to salt mine boundary or faults, safe salt pillar width of adjacent storage and some others design indexes of rock salt gas storage. The analytical results show that pore pressure in the muddy interlayer firstly decreases sharply along the radial direction, and then tends to be stable. Gas seepage range and leakage at a certain time are determined by the permeability parameters of the interlayer, permeability media and the initial pore pressure together. The influencing range of seepage gradually increases with time elapsing, and eventually becomes stable as well. Hence, this study can provide theoretical and technical support for the location, design and sealing evaluation of multi-interlayers salt cavern gas storage.

Water-sand inrush is one of the most common mine disasters in western China, which is adverse impact on safe-mining. This study aims to investigate the critical criterion of water-sand inrush occurrence and characteristics of water-sand migration. Water-sand transport experiments were conducted on four aeolian-sand samples with different particle sizes from an aquifer of Yuheng mine using high-velocity water-sand seepage equipment developed by the authors. The blended aeolian-sand was also measured using the aforementioned setup. Two critical velocities of water-sand inrush were identified. The first one is aeolian-sand incipience velocity ranging between 0.38?1.26 mm/s, and the second one is water-sand inrush critical flow rate with a range from 2.48 to 3.54 mm/s. The experimental data indicate that, one necessary condition of sand swarm startup is volume expansion. The process of water-sand inrush disaster is a physical process from quantitative changes to qualitative changes, which can be divided into three stages: 1) water brings sand, 2) water and sand get mixed and flow together and 3) sand brings water. The movement behavior of sand particles always manifest as concentration wave propagates. In this situation, the energy mainly transmits in the form of particle collision. When sand swarm starts, kinetic energy is transferred from water to sand. As the initial hydraulic gradient increases, the flow rate increases exponentially, and the amount of inrush-sand per unit time increases linearly.

In order to investigate the effect of CaO in different modes of occurrence and content on ettringite stabilization/solidification of lead-contaminated soil, artificial contaminated soils were solidified using high-alumina cement for AlO2- and gypsum and phosphogypsum for SO42-. All of the additives used include Portland cement or lime, which provide CaO. The pH value, strength of solidified samples and leaching concentration were investigated to understand the influence of different binders on physical-mechanical and leaching properties of ettringite-solidified soils with contamination. The micro-structural characteristics of solidified soils with contamination were also explored. The results show that the ettringite is able to stabilize/solidify heavy metal-contaminated soil. The cementation of calcium silicate hydrate cementing soil particles cannot be replaced by the filling effect of ettringite. Lime with higher CaO content is more beneficial not only due to the formation and stability of ettringite, but also due to the effect of stabilization/solidification of Pb2+-contaminated soil. Solidified samples with cement-contaminated curing agent have higher unconfined compressive strength but slow strength growth in later stage. Gypsum and phosphogypsum have little impact on pH value, unconfined compressive strength and the solidification result of solidified samples. The results of SEM test from microstructure features also indicate that the content of CaO affects the shape and effectiveness of ettringite. Ettringite can be transformed into calcium sulphoaluminate in case of insufficient CaO content. This research advances the on-site treatment technology of heavy metal contaminated sites, and is important in theoretical research and engineering application for remediation of heavy metal contaminated sites.

This study investigated the impact of the direction of cyclic loading on the cyclic behavior of saturated silt under isotropic consolidation condition. GDS hollow cylinder torsional apparatus was employed to perform a series of undrained cyclic shear tests with various principal stress direction angle ?d0 coupling with four dynamic loads, i.e., inner pressure, outer pressure, axial load and torque. The results show that the double normalized pore pressure is independent on the direction angle of cyclic loading, but sensitive to the cyclic stress ratio CSR. The generalized shear strain is independent on the direction angle of cyclic loading during cyclic loading. The direction angle of cyclic loading during cyclic shearing has a considerable influence on undrained cyclic resistance ratio of saturated silt. Cyclic resistance ratio CRR initially decreases then increases as the direction angle of cyclic loading increase, with a minimum value for ?d0=45°. Meanwhile, equations for pore water pressure and deformation to quantify the influence of ?d0 and CSR are established and the corresponding expression for CRR is given.

It is key to grasp mechanical response of shoulder sheet-pile wall in steep ground along a high-speed railway for control of lateral deformation of embankment. In this study, the sheet-pile walls in steep ground along Guiyang-Guangzhou high-speed railway were selected as monitored objects. Long-term observations for the earth pressure behind the wall, the internal force and deformation of the pile were made for more than 900 days. The layered filling-induced differential deformation between sheet pile walls was intensively discussed, and the effect of roller compaction-induced residual stress on distribution of earth pressure was discussed as well. The results indicate that the lateral deformation of embankment is effectively limited by sheet-pile walls. The lateral deformation of the pile top is about 0.94‰ height of the wall and tends to be stable in 1?1.5 years later, this is about 30% of the total displacement after embankment is filled. The roller-compacted coarse-granular filler behind the wall can still reach active earth pressure state even in small lateral deformation. The lateral deformation of fills caused by the layered filling presents a shape of small at top and large at bottom and is opposite with the displacement distribution of sheet-pile wall. The lateral deformation of fills is the main factor for parabolic distribution of earth pressure behind the wall. The recommended bending moment formula about pile according to concrete stress and steel bar strain in field test is consistent with theoretical calculation value.

Time-dependent deformation of rock mass is obviously found under high stress condition during excavation unloading process. To investigate the unloading creep characteristics of fractured rock, specimens were prepared by marbles from Jinping I hydropower station. Then unloading creep experiments were performed in the laboratory under constant axial pressure and stepwise unloading lateral stress. The results showed that when the specimens fail, the corresponding smallest stress difference is found on the specimen with the angle of 30°, the middle one is specimen with the angle of 60 °, and the largest one is the specimen with the angle of 90°. Overall, shear failure occurs in the intact specimen and the specimen with the angle of 90°, while the cracks initiate from internal tips of pre-existing flaw of specimens with angles of 30° and 60°. The Burgers model is also employed to describe deformation characteristics of all intact rock and fractured rock without the accelerated creep stage. The values of parameters are used to distinguish the deformation characteristics of specimens with different fissures and under different stress conditions. The obtained results are consistent with the views of Lemaitre equivalent strain and Sidoroff energy equivalent principle on the correlations between damaged materials and undamaged materials. According to above results, damage creep constitutive models are put forward for the fractured rock mass. Moreover, the relationship between the intact rock and fractured rock masses is established using the proposed models. Therefore, this study can provide references for further study on time-dependent deformation characteristics of fractured rock mass.

Natural rivers are curved. Bend circulation contributes to turbulence and causes erosion in concave bank, resulting in frequent flood damage on retaining wall along the river-side road. To investigate the trend of water movement at the river bend, force of retaining wall was analyzed in the condition of the flood. Based on the mechanical model of stability analysis of retaining wall, the calculating formula of the bend river retaining wall stability coefficient was deduced and calculation equation of limit punching depth under the wall was given. Shock calculation of rationality was verified in the gentle bend flood conditions and the stability of wall attenuation factors was analyzed by a numerical example. The results show that the impact load on the wall increases and the stability of wall is improved in the period of flood level rising and no erosion occurrence. However, the stability of retaining wall decreases when erosion occurs, and wall bottom scour may lead to overturning. The stability of wall decreases with the increase of water level difference between the front and back of the retaining wall, and the anti-sliding stability coefficient of the wall attenuates faster than that of the anti-overturning stability coefficient. The instability possibility of the wall is relatively high in the period of flood steeping. The calculation method and results of retaining stability coefficient are consistent with the site conditions, providing a theoretical reference for similar design and research of retaining walls along rivers.

The effects of over consolidation ratio (OCR) on thermal consolidation characteristics of hollow cylindrical saturated silty clay with different heating-cooling paths are discussed by laboratory tests. Given the same thermal loading, pore water pressures of samples with different OCR values caused by heating are almost the same, while the absolute value of negative pore water pressure caused by cooling increases with the increase of OCR value. The consolidation volumetric strain induced by heating or cooling decreases with the increase of OCR value, and the decrease can be described by exponential function. The amplitudes of the change increase with the increase of thermal loading. Besides, after a heating-cooling process circle, the final volumetric strain also decreases as OCR value increases. The volumetric strain is different under different temperature paths. Generally speaking, the more stages during heating or cooling to reach the final temperature, the larger the volumetric strain is. The difference in volumetric strain between different heating or cooling stages to the final temperature decreases with the increase of OCR value, while the OCR value has relatively small influence on the final volumetric strain after a heating-cooling process circle.

Freeze-thaw damage and rupture failure of fractured rock mass under low temperature are the key scientific problems for the construction of rock engineering in cold regions. Frost heaving pressure produced in saturated crack is a controlling factor that induces crack frost propagation. However, the magnitude and evolution law of frost heaving pressure are controversial due to the difficulty in conducting tests. By prefabricating macroscopic crack with different geometries in rock-like material, a laboratory test of frost heaving pressure in saturated cracks is conducted. A thin film pressure sensor is used to test the evolution process of frost heaving pressure in cracks and the temperature on the surface of rock is continuously monitored by several temperature sensors. Based on the results, the space-time evolution curves of frost heaving pressure in cracks are derived. The experimental results show that: 1) In freezing process, the initiation of frost heaving pressure in crack suddenly occurs, and the entire evolution process of this pressure can be divided into embryonic stage, outburst stage, falling stage and balance stage. 2) In frozen process, the ice in crack has been extruded and new visible propagation cracks are produced in tips of the cracks. 3) In melting process, the frost heaving pressure falls quickly but lags behind the freezing point. 4) For a half open fracture with a width ranging between 2?5 mm, when water starts freezing at the open end, the maximum frost heaving pressure linearly increases with the increasing of crack width. The pressure is more than 7.2 MPa in saturated crack with a width of 5 mm. Research results presented in this paper can provide reference for understanding the evolution mechanism of frost heaving pressure and studying the crack propagation caused by frost heave.

Based on the characteristics of stress axisymmetric of the anchorage body, an experimental device of anchor-soil interface is developed. By means of 2D digital speckle correlation method, the displacement field and the strain field of soil surrounding anchorage body are obtained. Experimental results show that, the deformation of the surrounding soil caused by the displacement of the anchor is mainly in a small thin layer that is near the anchorage body, and its deformation is characterized by stable range, continuous, linear and shear dilation. On this basis, the constitutive relationship of the anchorage body interface layer is established according to the flow rule associated with Coulomb yield condition, and then the expressions of axial force and shear stress distribution under the condition of drawing load are deduced. The result indicates that the degree of uniformity of shear stress on anchorage body is closely related to soil compactness. The lower the soil compaction, the smaller the resistance coefficient and the angle of internal friction, and the shear stress distribution is more uniform on the anchor. While for the soil with larger resistance coefficient and internal friction, the shear stress distribution is similar to the theory and measured results of the rock anchor solid.

To investigate the impact of gravel content on shear strength and dilatation of sand-gravel mixture, the large-scale simple shear tests of 21 simples were conducted using an advanced shear apparatus. Seven groups of sand-gravel mixture samples with gravel content ranging from 0% to 80% were prepared for shear tests under three different normal pressures (100, 200, 300 kPa). Relationships among the gravel content and shear strength, dilatation, shear contraction of sand-gravel mixture were analyzed based on the test results. It is shown that under the same normal stress, the internal friction angle and cohesion of the sand-gravel mixture initially increase then decrease as the gravel content rises. The shear strength achieves the maximum as the gravel content is located within the range of 40%-50%. Further analysis indicates that the shear strength of the sand-gravel mixture significantly depends on the void ratio of mixture. Meanwhile, the structure of sand-gravel mixture and dominant particles change as the gravel content increases. When the gravel content is below 20%, the sand-gravel mixture is dominated by fine component, and presents a typical suspended dense structure. In this case, the shear strength of sand-gravel mixture is close to that of the matrix. When the gravel content is between 20% and 50%, the sand-gravel mixture shows a skeleton-pore structure. As the gravel content increases, the skeleton is formed gradually, and biting force between particles increase, leading to the raise in the strength of the mixture. When the gravel content is above 50%, the mixture exhibits a typical dense skeleton structure. In this case, void ratio of the mixture increases, and the matrix is unable to fill the void as the gravel content increases, which results in lower shear strength.

The research on gas migration law during coal bed methane (CBM) exploitation is beneficial to understand the source of gas, impact of coal seam position on gas production and the law of gas pressure decay, thus providing foundation for design of exploitation time and the position and length of drilling hole. Using the multi-field coupling testing system for CBM exploitation developed by the authors, physical simulation experiments of CBM exploitation under different lengths of drilling hole in distressed zone were carried out, and relative gas flow speed and direction were analyzed. The results show that the initial stage of exploitation is the main contribution period to gas production and the surrounding area of drilling hole is the main contribution area to gas production, because gas pressure gradient is high and thus gas flows fast. Gas flow has the highest speed at the distressed zone. Existence of concentrated stress zone makes the relative flow speed attenuate quickly, and also creates a barrier to original stress zone, leading relative gas flow speed to be zero. Relative gas flow speed decreases with time while the migration direction does not change much because relatively stable seepage channels are formed at the beginning of exploitation. Due to gas pressure drop that leads gas pressure gradient and relative gas flow speed to decrease, the uncertainty of migration direction rises. With the length of drilling hole growing, the relative gas flow speed increases in the distressed zone, so it is meaningful to moderately lengthen the drilling hole.

Structured soils with interparticle bonding have higher shearing ultimate stress ratio and less free dilatancy than the corresponding reconstituted soils. To overcome this defect, a moving critical state line (MCSL) in stress space is presented based on the structured unified hardening (UH) model mainly considering soil structure collapse. By developing new yield functions and modifying dilatancy equation, the structured UH model is extended to be applicable for bonding structured soil. In stress space, MCSL parallels to the traditional static critical state line (CSL) and moves toward the CSL as structure decays. The extended model adds 1 new parameter, namely the initial bonding stress, to describe initial bond between particles. Comparisons between test data and predictions of 4 structured soils indicate that the extended model is qualified in describing the behaviors of the bonding structured soils in isotropic compression, drained and undrained triaxial compression.

The resilient modulus (MR) used for characterization of subgrade elastic support performance is an important parameter significantly affected by matric suction after traffic opening. To investigate the correlation between resilient modulus and matric suction, soil-water characteristic curve (SWCC) and repeated load triaxial (RLT) tests for three kinds of compacted clay soils were conducted, and then the prediction model of resilient modulus considering the effect of matric suction was proposed based on effective stress theory of unsaturated soils. Test results show that the resilient modulus has a good non-linear relationship with matric suction, which can be presented by an exponential function. When the gravimetric moisture content increases from 4% below optimum water content (OMC) to 4% above OMC, MR decreases by 29.1%-39.0% resulting from reduction of matric suction. Within the transition zone at SWCC, the prediction model with effective stress factor related to the degree of saturation can explain the contribution of matric suction to resilient modulus, and the model has higher predicting accuracy compared with the models with effective stress factor dependent on residual water content or air-entry value. With the same degree of saturation, the contribution of matric suction to effective stress is less significant with higher clay particle content and plastic index.

In this paper, isotropic compression tests on clayey soil with pore water of different salt concentrations were carried out. Meanwhile, the testing results were used to analyze the effect of osmotic suction on compressibility. According to the experimental results, isotropic deformation characteristics were significantly influenced by osmotic suction. As osmotic suction increases, the slope of initial elastic compression line increases and stress decreases, while the effect of osmotic suction on the slopes of elasto-plastic compression lines and swelling lines can be neglected. Based upon the experimental results, a constitutive model considering in-pore salty solution influence for clayey soil is proposed under isotropic condition. Through the verification, the model is valid for simulation of isotropic compression behavior. Additionally, the proposed model could be adopted to simulate the response of saturated soil with salty solution to the change in osmotic suction. The experimental result is of vital importance to develop a constitutive model considering chemical-mechanical coupling.

Bucket foundation is an important foundation type of offshore structures, especially for offshore wind turbines. Study on its penetration law is the key to successful application of this kind of foundation type. Previous methods for penetration suction are CPT-Based method and analytical calculation method, which are based on cone penetration test (CPT) and force balance in the penetration process, respectively. However, results from previous methods deviate from practical engineering data. In this paper, three bucket foundations with different wall thicknesses were penetrated by suction pressure. Earth pressure and pore pressure inside and outside the bucket wall, as well as the tip resistance were measured in the experiments. The influence of key parameters on the two calculation methods was analyzed and reasonable suggestions were given. According to measured pore pressure compared with numerical simulation analysis, a critical pressure calculation formula is established. Studies have shown that coefficient of friction between side wall and the soil is about 0.2. For CPT-Based method，outside-wall friction parameter kf is in the range of 0.002 6? 0.007 0；inside-wall friction parameter kf is 0.001; tip resistance coefficient kp is in the range of 0.2?0.6. For analytical calculation method, lateral earth pressure coefficient K should be between the stationary earth pressure coefficient and passive earth pressure coefficient, and tip resistance coefficient Nq is 40. The critical suction pressure formula based on protection mechanism for seepage failure can effectively increase the critical suction pressure，which is close to the actual situation.

Limit state design method based on reliability theory is a trend of modern geotechnical design. Calibration of the partial factors is a key to determine the design strength parameters. Although there are safety factor calibration methods, first order second moment method and designed points method to calibrate the partial factor theoretically, the mechanism of how the partial factors reflect the variance of the geotechnical materials is still unknown. The standard deviation feedback calibration method is then proposed to solve this problem. The core concept of this method is to adjust the mean value of the strength parameters to offset the influence of the standard deviation adjustment. A case study is given to illustrate the method. A comparison between this method and the design points method is taken out to see their different effects on the design strength parameters. The results show that the standard deviation feedback calibration method is more reasonable and the design points method will lead to a conservative design which would raise the budget ultimately.

In view of the effect of environmental factor on the strength and stability of subgrade, the clayey soil taken from Siping- Changchun section of Beijing-Harbin expressway was investigated，and its variations of mechanical properties affected by freeze-thaw cycles and water content were thoroughly studied and analyzed in laboratory tests. The influence mechanism was also discussed. The test soil samples with different water contents went through 0 to 16 freeze-thaw cycles, and then the triaxial compression tests were carried out indoor. Results show that with the increasing of freeze-thaw cycles and water contents, the stress-strain curve of the compacted clayey soil changes from strain-softening to strain-hardening when the water content is less than the optimum water content, and the failure mode of the test sample gradually changes from brittle failure form to plastic failure form. The ultimate strength, elastic modulus and cohesion all decrease with the increasing of freeze-thaw cycles, while there is no obvious relationship between the internal friction angle and the freeze-thaw cycle. However, the aforementioned mechanical parameters of the compacted clayey soil all tend to stabilize after 8 freeze-thaw cycles. In summary, the effect of water content on mechanical properties of compacted clayey soil is significant. Ultimate strength, elastic modulus, cohesion and internal friction angle all decrease drastically with the increasing of water content. Based on the consideration of the effect of freeze-thaw cycles and water contents on mechanical properties of compacted clayey soil, the drainage work of subgrade in the seasonal frozen area should be well conducted to keep the subgrade in a lower level of water content, and the engineering design values of mechanical parameters should be those of the soil that undergoes no less than 8 freeze-thaw cycles.

To reveal breakage behavior and deformation of particle shape for calcareous sand from the South China Sea with large shearing displacement, a series of ring shear tests was performed under controlled shear displacement. After ring shear, the particle size distributions of the samples were obtained by the sieving method and laser grain-size analyzer. The particle breakage was quantitatively examined by analyzing particle size distribution. The roundness and flatness was also analyzed to examine the variation of particle shape using the image processing technology. Experimental results indicate that shearing under different vertical pressures would lead to different stable graduations. However, the stable graduation is observed under the same shearing displacement. Severe breakage occurs in the samples with particle sizes of 0.01-0.075 mm after large displacement shearing. With the increase of the shear displacement, the roundness and the flatness of particles decrease. The relative breakage is modified to account for the severe breakage of fine grains (<0.075 mm) in large displacement shearing. The modified relative breakage ratio can cover the breakage of the particles with sizes of 0.01-0.075 mm. After large displacement shear, the shape of calcareous sand tends to be more regular, more rounded and smoother.

The energy mechanism of material damage in thermodynamic theory was introduced to slope engineering. An energy conservation equation that was applicable in strength reduction period was deduced. Then, a calculation procedure for the slope energy was developed with FLAC3D, which was applied to a traditional slope example. Compared with the result from Spencer method, the slope’s energy change was closely related to its stability in the strength reduction method, hence four new slope failure criteria-energy catastrophe criteria which were theoretically in unity were proposed. By comparison, the criterion of the kinetic energy catastrophe is in correspondence with the criteria of numerical calculation non-convergence, the loss of gravitational potential energy catastrophe is in correspondence with the criteria of dramatic increase in the marked nodal displacements, and the dissipated energy catastrophe is in correspondence with the criteria of a plastic zone going through the slope. In summary, the aforementioned correspondences demonstrate the unity of the three common failure criteria. Several examples demonstrate that the different results from the three common failure criteria are because of artificial factors such as mesh generation precision and numerical convergence criteria. In essence, the different results stem from the fact that numerical calculation is a kind of approximate solution. In application, the accuracy of safety factor can be evaluated by the consistency of the results from various slope failure criteria. The better the consistency is, the higher the accuracy is. The basic method of improving the accuracy of safety factor is strict convergence standard and fine mesh precision, but achieving strict convergence standard and fine mesh precision may require extremely long numerical calculation time. Therefore, a moderate numerical calculation time should be chosen to maximize the computational efficiency.

This article presents a method for security discrimination of adjacent underground pipelines during the construction of twin shield tunnel. Based on Winkler elastic foundation beam model, a relationship is established between strain of continuous pipelines and surface settlement by considering pipe-soil interaction. Assuming that displacement of pipelines and displacement of soil are the same, an expression of relationship between joint angle of discontinuous pipelines and surface settlement is derived. Considering aging of pipelines, a reduction factor related to time is defined, and a method to judge the security of pipelines by measuring surface settlement is built. When the strain or joint angle of pipeline reaches the maximum safety allowable value, the corresponding surface settlement is the controlling value. The pipelines are in danger when the value of surface settlement exceeds the controlling value during the construction stage. Surface settlement monitoring data are used in this method to assess the state of subsurface pipelines and the state is not easy to be monitored. The reliability of this method has been indicated in the comparison between predicted value and measured value, and the horizontal spacing L value of twin tunnel has a great influence on the control value of surface settlement. When L is relatively small, the maximum value appears on the axis of two tunnels; when L is relatively large, the maximum value appears on the vicinity of the tunnel axis. With the increase of L, the maximum control value is gradually reduced.

A series of cyclic loading and unloading tests on raw coal is conducted under the conditions of axial compression and confining pressure using self-developed servo-controlled seepage equipment for thermal-hydro-mechanical coupling of coal. The mechanical properties and permeability characteristics of coal are analyzed under different axial compressions and confining pressures. The results show that the plastic deformation degree SD increases with the increase of axial compression degree. Cumulative plastic deformation ??, cumulative amount of permeability variation ?K and cumulative volumetric strain variation ??V are defined, and some conclusions are drawn with the use of those parameters. At a relatively low level of axial loading-unloading, the cumulative plastic deformation δε3 and δε1 have the same effect on the cumulative permeability variation ?K; with the increase of ?1, ??3 is more sensitive to the change of ?K than ??1. Assuming an extreme scenario, i.e., ??V is determined only by ??1 and 2??3, the radial deformation is found more sensitive than axial deformation. The axial compression remains constant, and the axial deformation decreases linearly with the change of confining pressure. Radial deformation ?3 and ?3 also has a good linear fit. ?3 presents V-shape, and the permeability exhibits inverted V-shape. When axial load is at a low level, the loading-unloading of confining pressure does not have a significant impact on the development of plastic deformation.

To develop offshore facilities in South China Sea, it is vital to understand the characteristics of particle shape and breakage properties of the calcareous sand. As a result, micro- and macro-mechanical behaviors of crushable calcareous sand sampled from South China Sea are addressed in this study. Firstly, calcareous sand particles are scanned by electron microscope and analyzed by an image processing software (i.e., ImageJ). Based on the processed image, two shape parameters, namely circularity and solidity, are defined and quantified. Secondly, drained triaxial tests on calcareous sand at various confining stresses are carried out, so as to investigate the effects of particle breakage on deformation, shear strength and energy dissipation of the calcareous sand. Results show that the shape of the calcareous sand with relatively large particle size (diameter greater than 2 mm) and relatively small particle size (diameter less than 0.5 mm) tend to be circular and the particle surface is relatively smooth. Comparatively, shape of calcareous sand with moderate-size diameter (grain diameter ranged between 0.5 and 2 mm) is more irregular, and sand particles have more surface edges. Particle breakage is identified to occur in triaxial tests, leading to a better graded sand packing. With the increasing initial confining pressure, the degree of particle breakage and the energy dissipation due to the breakage both increase. In the meantime, dilation of the sand is partially suppressed by the breakage. Under two conditions only considering friction dissipation and considering both friction and volume dissipation, the dissipated energy resulting from breakage in the triaxial tests with relatively high (600 kPa) initial confining stresses could account for 25% and 18% of the total plastic energy input, respectively.

This study aims to investigate the anisotropic characteristics of phyllite in Maoxian group from the northwest of Sichuan province in China. A series of uniaxial and triaxial compressive tests is conducted on phyllite samples by using the MTS815Teststar rock mechanics testing system. Experimental results show that the large deformation and partial failure generally appear before samples reach half of their peak strengths. However, the occurrences of large deformation or initial dilatancy do not indicate that rock completely loses its bearing capacity. If the targeted support or protection can be applied on surrounding rock in time, rock strength is still strong. Damage dilation is regarded as a typical deformation and failure characteristic of phyllite. Furthermore, it affects lateral deformation more easily than axial deformation. Thus, lateral deformation can reflect rock yield or weakening more sensitively. Results show that the anisotropy of rock strength decreases with increasing water content and confining pressure. Moreover, Young’s modulus and Poisson’s ratio vary with the loading direction and translated-confining pressure. From the obtained results of phyllite at different water contents, the strength of phyllite under dry condition presents the most obvious anisotropic characteristics in uniaxial compressive tests. It is found that the anisotropic ratios of compressive strength are mainly below 0.8. Especially, sericite phyllite exhibits the strongest anisotropy with a ratio of 0.48, and the following anisotropic orders from strong to weak are sericite phyllite, chlorite phyllite, quartz phyllite, and carbonaceous phyllite. Moreover, there are different failure modes of phyllite, which also change with the increase of confining pressure and water content. Shear failure occurred uniformly along the weak planes, which indicates that the anisotropic mechanical characteristics of phyllite are controlled by shear mechanism independently.

Physical similarity model test is an effective method to study complex and difficult mining technology, and the investigation of physical and mechanical properties of rock-like materials is the prerequisite for successful experiments. Based on similarity principle and mechanical properties of rock, mechanical representativeness of uniaxial compressive strength for rock-like materials is demonstrated. On the basis of the statistics and analysis of the influential factors which are closely related to uniaxial compressive strength of rock-like materials, a dimensionless calculation model has been constructed with dimensional analysis. Combined with proportioning tests of rock-like materials, the quantitative relations have been set up to correlate compressive strength with sand diameter, total amount of rock-like materials, filling materials（sand）dosage, water usage, curing approach, etc. The calculation expression of compressive strength is verified by 3 group proportioning tests of rock-like materials which are representative. The average error between computed results of compressive strength and the measured results at laboratory is 4.20%. The calculation model is suitable for investigating parameters of rock-like materials. The results provide references for predicting physical and mechanical parameters of rock-like materials and efficiently selecting similar rock-like materials for physical model test in geotechnical engineering.

The aim of study is to investigate anti-seepage performance of landfill covers with different types of vegetation. First, three field test sites were built with different types of vegetation conditions, i.e. no vegetation, Bermuda grass (fibrous root) and vetiver grass (tap root). During one year of vegetation growing under the subtropical climate, the real-time responses of volumetric water content (VWC) and matric suction were monitored at different depths of the covers from July to September when the rainfall and evaporation were the largest. Results show that the vetiver grass enhances the coefficient of permeability and reduces the integrity of the cover. The Bermuda grass decreases the coefficient of permeability and improves the performance of the cover. The reason is that the effects of different plant roots on the moisture flow path in cover are different. The monitoring results indicate that the effect of vegetation on the water content changes at 5 cm depth can be ignored. The changes of VWC and matric suction at this depth are mainly controlled by the rainfall and evaporation. The water content at 50 cm depth changes only in cover with vetiver grass. This indicates that the vetiver grass results in preferential flow in deep soils. The Bermuda grass mainly changes the moisture flow within small depth.

A series of model tests was conducted to compare the horizontal bearing capacities of longitudinal section shaped pile (belled pile and tapered pile), and traditional equal-diameter pile, which have identical volume of concrete. The distribution of the internal force and deformation of piles, the ultimate bearing capacity, and lateral soil pressure were measured under different levels of horizontal load. The distribution of the ultimate bearing capacity and lateral soil pressure of three different pile types were comparatively analyzed. A theoretical method was developed to calculate the lateral bearing capacity of longitudinal section shaped pile based on p-y curve method and by considering shape effect. The distribution of pile moments and influencing factors were discussed. The results show that the horizontal bearing capacity of tapered pile is larger than those of belled pile and equal-diameter pile with identical volume of concrete. The ultimate horizontal bearing capacity of tapered pile embedded in sand and clay is nearly 1.25 times and 1.33 times of that of equal-diameter pile. These experimental results are useful for practical engineering and analysis of longitudinal section shaped piles in different applications.

Soft soil is one type of material with nonlinear rheological behavior. The existing rheological models ignore nonlinear deformation under transient loading conditions. Rheological testing parameters for soft soil are mathematical parameters rather than physical model parameters. One soil generally corresponds to only one type of material parameters. But model parameters of the existing rheological model often change irregularly under varying loading conditions. Soft soils containing organic materials as the rheological phase are tested under one-dimensional consolidation-creep condition. A rheological model containing seven components is developed with same material parameters under different loading conditions. Non-linear spring is applied to describe elastic-plastic deformation subjected to transient loading. Parallel three-element components are applied to describe viscosity, plasticity and elastic deformation with time. The results show that the rheological behavior of soft soil can be accurately reflected by seven-components rheological model with incorporating cubic nonlinear transient elastic modulus.

In order to solve serious blasting vibration problem of the traditional drilling and blasting method for tunnel construction, a new type of rock fragmentation technology—high pressure gas expansion rock fragmentation technology is provided. The values of vibration velocity and effect of rock fragmentation received from field tests by high pressure gas expansion rock fragmentation technology are compared and analyzed with that of traditional drilling and blasting method. The results show that high pressure gas expansion rock fragmentation technology has smaller values of vibration velocity than the traditional drilling and blasting method at the time of construction. The results also prove that it is feasible to use this technology in tunnel projects and the problem of high vibration risk by traditional drilling and blasting method in the tunnel could be resolved through high pressure gas expansion rock fragmentation technology. The high pressure gas expansion rock fragmentation technology provides a new way for relevant engineering projects to break rocks.

Bentonite is often used as buffer and backfill material in deep geological repository because of its high swelling potential and low permeability, but its expansion characteristics will be influenced by solution concentration of underground fluid. In response to this phenomenon, the deformation characteristics of commercial bentonite under different loads and different concentrations of NaCl solution have been studied using uniaxial oedometer apparatus. The results show that with the increase of concentration of NaCl solution, swelling potential of bentonite significantly reduces, and there is a good linear relationship between the ma2016-01-08ximum swelling strain and overburden load in the log-log plot. Using the concept and calculation method of effective pressure, volume change behavior of bentonite in NaCl solution with different concentrations can be expressed by a unique curve, which also proves the rationality of effective pressure theory.

In the study, evolution laws of the roof structure were first investigated based on the principle of gob-side entry retaining formed by roof cutting and pressure release. Subsequently, stress and displacement transfer mechanisms of the surrounding rocks around a thick-seam fast-extracted mining face were discussed. Correspondingly, an active stability control approach for the entry surroundings was proposed by building a rock-steady structure using the upper strata of the main roof, bulking gangues at the gob and cutting cantilever of the entry roof. According to the deformation characteristics of the entry surroundings, we designed a series of cooperative supports with characteristics of constant resistance and pressure release. Particularly, the roof was supported by the constant resistance and large deformation anchor cable and pier-type unit. The gangue rib was stabilized by the sliding-type side wall prevention structure, self-advancing dynamic pressure bearing device and side wall prevention bolt with multi-resistances. Finally, industrial field tests were conducted in S1201 haulage entry of Ningtiaota coal mine for the first time. The results indicated that with increasing mining height and accelerating mining speed, a small amount of rotational deformation occurred inevitably owing to the movement of the main roof. Moreover, this deformation was not suitable to be supported compulsorily, and should be better coordinated by building a rock-steady structure artificially. The motion state of the gangues was closely related to the roof pressure, and thus an enhanced presplitting helped to reduce impacts of the dynamic mining pressure. Field monitoring results showed that the roof pressure first reached the constant resistance, and then the displacement tended to be stable. Finally, the rock-steady structure of the entry surroundings was formed. It was verified that the designed support structures could effectively coordinate with the movement of the surrounding rocks, and the indexes of the retained entry met requirements of the next mining panel.

When highway tunnels are constructed in the southwest mountainous area in China, they often have to pass through the landslides and other adverse geological disasters. Especially, the tunnel orthogonally crossing the landslide is considered as the most complicated situation. In this case, two serious conditions are coupled, i.e., deep-seated landslide mass with a great thickness and exposed slip surface of landslide. Thus, the coupled action leads to strong interactions between landslide and tunnel, which also can induce the deformation and crack diseases of landslide and tunnel. In this study, a highway tunnel is investigated, which orthogonally crosses an old landslide with an over 60 m thickness in the southwest mountainous area. Due to the interaction of the tunnel-landslide system, the mechanism of the engineering diseases is deeply studied. The numerical analysis method is used to calculate and analyze the stress and deformations of landslide and tunnel as well. Finally, the calculated results are compared with those obtained by the recommended transfer coefficient method. The results verify that the numerical analysis method is more reasonable for analyzing the interaction between the landslide and the tunnel. Meanwhile, it is proven that the final control measures based on stress deformation control theory has obvious technical and economic advantages. At present, the tunnel has successfully passed through the landslide. As it is rare to easily manage this complex tunnel-landslide system, its design analysis method and comprehensive prevention measures provide helpful references for the similar projects in the future.

Currently, Zhanjiang bay tunnel is the deepest cross-sea shield tunnel in China. The tunnel segments are connected by oblique bolts with staggered arrangements. In this study, the forces in the maximum covering soil section and the maximum water depth section were calculated by a modified usage method and a beam-spring method, respectively. Then the distribution pattern of water-earth pressure was analyzed by comparing theoretical results with measured results of segment stress and water-earth pressure. The analytical results show that both of the modified usage method and the beam-spring method can well simulate the stress state of tunnel segment. Especially, the results calculated by beam-spring method is relatively closer to measured results than those by the modified usage method. It is found that segments are in flattening states and the results obtained by beam-spring method are greatly affected by the position of capping blocks. Terzaghi’s relaxed earth pressure and the linear distribution of water-earth pressure are also applicable and safe for the deeply buried submarine tunnel. In this project, the lateral load of the segment is dominated by radial water pressure, which mainly affected the axial force. Therefore, this study can provide a useful reference for the design and construction of high-head cross-sea shield tunnel segments in the future.

Based on the 3D joint network simulation, the structure classification was carried out on the left bank dam foundation in Baihetan hydropower station on Jinsha River. Firstly, the dam foundation was separated into several statistical homogeneous subsections, according to the group distribution and density distribution of joints. Then, a 3D joint network model was established for each subsection by considering the statistics and probability theory. Secondly, uniform angle rotating measuring-lines were arranged on the positive, side and top surfaces of the model, respectively. Subsequently, a certain RBI value of each line and 3 rose diagrams for each orthogonal surface were obtained. Finally, the mean value of these RBIs from 54 measuring-lines was defined as 3D equivalent rock mass block index RBI3D. Moreover, the RBI3D, as an evaluation index, was used to classify the structure of dam foundation. Results showed that the integrity of No.10 bench in the left bank was relatively good. It was found that the ratios of the whole blocky structure, blocky structure and sub-blocky structure were 53.3%, 26.7% and 20.0%, respectively. The analytical results agreed well with the actual situation. Therefore, the presented method has been proven to be scientific rationality and engineering practicability, as the space anisotropy of joint distribution was fully considered and a single quantitative evaluation index was employed to determine the rock mass structure type.

The velocity model is significant for precisely locating the earthquake source in inclined strata. This study analyzed the propagation path and propagation regularity of stress waves in inclined strata, according to the fast ray-tracing technique between two points in horizontal layered media. Microseismic location algorithm was developed for inclined strata using grid searching method, which was also compared with traditional location methods based on single velocity model. The particle flow theory was exploratively used to establish the stratum numerical model. Subsequently, the validity of the proposed algorithm was verified by simulating the stress wave propagation through the interaction of particles. Results show that the traditional location algorithm, based on simplified elastic wave propagation at a constant velocity, is imprecise for inclined strata, indicating that the single velocity model cannot satisfy the locating demand for complex media., On the basis of the inclined layered velocity model and a modified computational method for stress wave propagation path, the seismic sources can be located much more accurately. Meanwhile, the location deviation was found to be proportional to the dip angle and stratum number. In addition, the location precision can be also increased by reducing the grid size in the grid searching method, suggesting that reasonable sizes and searching strategies can facilitate the application of grid searching method in practical engineering. The microseismic location method and the proposed verification process can provide important theoretical and technical support for further studies on the positioning and monitoring technology in the complex strata.

To reveal the mechanical behavior characteristics of surrounding rock of hydraulic tunnels under high temperature and hydraulic pressure conditions, this paper proposes a coupled thermo-hydro-mechanical-damage (THMD) model which considers change of hard rock strength parameters involving damage evolution based on multi-field coupling theory with permeability coefficient and heat transfer coefficient varying with rock damage. The implementation of THMD numerical simulation is conducted by the FLAC3D software. The model is proved to be feasible by validation of a physical modeling test, and then the THMD model is employed to deduce the evolution process of multi-field coupling of hydraulic tunnels with high temperature and hydraulic pressure. The evolution process is used to analyze the load-bearing characteristics influenced by different factors. The multi-field coupling effects are very significant in hydraulic tunnels with high temperature and hydraulic pressure after water-filling operation. The complex tensile stress generated by combined load of high temperature gradient and hydraulic pressure contributes to adverse impacts on safe performance of surrounding rock. The damage and crack depth of surrounding rock are larger if the temperature gradient, hydraulic pressure and thermal expansion coefficient are larger. If the lateral pressure coefficient is close to 1, more cracks around the tunnel with random directions tend to form. If the lateral pressure coefficient is less than 1/3, fewer cracks appear around the tunnel and the directions of the cracks are mostly parallel to the direction of the maximum initial stress. The general bolt-shotcrete support does not have good performance in reinforcement of hydraulic tunnels.

According to the realistic damage characteristics of slopes on the seismogenic faults as well as the fracture mechanismss of the faults in the Wenchuan earthquake, a so-called strip-shape hypocenter is employed to assess the dynamic responses of the typical slopes. Furthermore, the strip-shape hypocenter can be divided into four stages in accordance with their spatial locations from the initial break to the end break of the seismogenic fault. These stages consist of the thrust fault hypocenter, the thrust and a bit strike-slip fault hypocenter, the thrust and strike-slip fault hypocenter and the strike-slip and a bit thrust fault hypocenter. Thus, the distinct element method is applied to simulate the dynamic response of the Daguangbao landslide on the Longmenshan seismogenic fault in Sichuan province in China. Then, the dynamic formation mechanism, the triggered main controlling factors, and dynamic characteristics of damage, collapse and accumulation are revealed. The results show that the critical damage of the slope is mainly caused by the thrust and a bit strike-slip fault hypocenter, by considering the location of the front end of the fault break and the formation time of the critical damage. While before the critical damage of the slope, its damage is induced by the thrust fault hypocenter between the initial break of the faults and Wenchuan county. The following processes of ejecting, colliding, crushing and accumulating are mainly triggered by the coupled inertia force and the gravitational force, which are also influenced at a certain degree by the above two kinds of hypocenters with respect to their fracture mechanisms. Before the critical damage, the whole slope underwent a positive horizontal distance towards its free face. Then, the slide bed begins a negative horizontal distance away from its free face with a continuous positive horizontal distance of the slide mass, which results in a critical separation between them. At the ejecting, colliding and crushing stage, the slide bed continues its negative horizontal displacement to an extreme limit, then a positive horizontal one and followed by its zero horizontal displacement. The vertical displacement at this stage is relatively small. The motion characteristics of the slide bed in this process is in accordance with that of the monitoring seismic data, for example, the acceleration, the velocity and the displacement. Meanwhile, the slide mass continues its positive displacement to a constant limit with a colliding, crushing, accumulating to self-stabilization process. As far as the mechanical factors are concerned, the horizontal seismic force formed by the single thrust fault hypocenter and the thrust and a bit strike-slip fault hypocenter provides a key contribution to the whole slope including the slide bed and the slide mass before the critical damage state. At the ejecting, colliding, crushing and accumulating stage, the dynamic responses of the slide mass is mainly influenced by the inertia force and the gravitational force. However, on the basis of a special topography, the horizontal and the vertical seismic forces from the following two hypocenters give their finite influences to the slide mass.

In the coordinate system of strength parameters, utilizing the concept of strength reserve area can deduce the FOS expression of double reduction method (DRM), which is more rational than several existing expressions, and the results do not have enormous deviations from results of existing expressions. In this new method, for the same K, adopting different mathematical expressions gets different virtual initial points. There are two approaches that are based on the ratio of reduction. The first approach only adjusts tan?, and the other only adjusts friction. In theory, these two reduction paths are different. However, this research shows that the difference has no influence on results of DRM, and selecting the shortest reduction path can reduce calculation time. The reduction strategy of traditional strength reduction method (SRM) is different from double reduction method. In DRM result, the critical friction and cohesion appear in such a way that critical friction is over initial friction, or critical cohesion is over initial cohesion, which is not reasonable. By analyzing DRM calculation results given various K, and focusing on the essence of strength reduction method to get two types of results. The first type result is the result that has to conform to calculation accuracy, and the other result is that the result has to align with the fact that critical points must embody the phenomenon of parameters of the weakening.

To investigate the influence of structure properties of soil and unloading form on macro- and micro- mechanical behaviors of soil around a foundation pit, a series of plane strain simulation tests were conducted on the soil within the region affected by pit excavation using three-dimensional (3-D) distinct element method. First, a simple 3-D bond contact model, representing inter-particle bonding effect, was implemented into 3-D distinct element analysis code PFC3D. After that, conventional triaxial and plane strain tests on remolded and structured soil were simulated under four stress paths. Finally, the reloading process was simulated on the bottom soil element after unloading. The results show that the peak strength of the soil in passive region and the vertical strain corresponding to failure during unloading increase with unloading ratio, and the strength is smaller than that of the soil in active region. During reloading, the peak strength of the soil in passive region increases with unloading ratio, and is smaller than that without unloading or immediate loading. In addition, volume changes are significantly influenced by structure effect and unloading form. On the microscopic scale, increase in unloading ratio and enhancement in structure degree will increase the normal contact forces in a plane perpendicular to the direction of the maximum principal stress and thus enhance the strength of the soil.

Seepage with free surface is essentially one type of nonlinear free boundary problems. In order to solve this problem in global domain, the nodal virtual flux method deducts the virtual flux gradually in each iteration. It has the advantages of weak mesh dependency, fast convergence rate, etc. To reveal its internal theoretical foundation, this paper builds an equivalent bridge between the nodal virtual flux method and variational inequality of Signorini type through a complementarity constraint. The criterion of free surface is optimized by introducing an amplification coefficient for transition region, verified through a sand flume model test. The comparisons show that the improved algorithm has better numerical stability and higher curvature tolerance, and provides an effective approach for optimization design of seepage control structure with large-scale grid.

In-situ stress condition is one of the essential information of stability analysis and engineering design for underground caverns, and it is particularly important for safety assessment of underground engineering in deep and high-stress condition as well as disaster prevention. Taking Jinping underground laboratory in China (CJPL) buried at a depth of 2 400 m as an example, the principle and method of in-situ stress measurement under high stress based on overcoring method have been expounded at first. Furthermore, some technological improvements are given: 1) Adopting new drill and overcoring gradually to reduce the extent and scope of stress concentration at the roots of the cores; 2) Using large diameter drill to increase the thickness of the hollow cylindrical core, namely increasing the time that fracture needs to cause the rock core to break; 3) Drilling exploration hole near the measuring points to get the integrity condition and to evaluate the feasibility of the test. After theoretical explanation, plane stress state, perpendicular to borehole axis with different depths and three-dimensional stress state are given to analyze original stress field and stress redistribution led by excavation. Verified by the relationship between stress distribution and damage of surrounding rock as well as the result of numerical calculation, the result of three-dimensional stress is proven to be reliable and can be used as a basis of further scientific analysis, engineering stability evaluation and design of disaster prevention.

Geomechanics magnetic model test is inspired by the idea of centrifugal model test, in which the centrifuge force field is used to simulate the gravity field. In uniform gradient magnetic field, material containing ferromagnet is subject to uniform magnetic force. Therefore, gravity can also be simulated by magnetic force. According to the basic principle of electromagnetism, a uniform gradient magnetic field generator is designed and manufactured. There are 2 000-turn coils in the solenoid, whose current ranges from 0 A to 16 A. The magnetic force of ferromagnetic material in the model can be increased 10 to 30 times of the gravity. A geomechanics magnetic model test is carried out using the high slope at left bank of Jinping-I hydropower station as an example. Results of the test are compared with numerical simulation results, which shows that the body force of the model can be increased to the multiple required by the similar ratio under a certain current value. Results of the physical model test are consistent with results of numerical simulation. In addition, if the current continues to rise, the overload process can be simulated, and thus the factor of safety under overload condition can be determined.

In the process of rock failure, a large number of low energy weak acoustic emission signals are easy to conceal the characteristics of a few high energy key signals, which is not conducive to the analysis of precursory characteristics and disaster warning. From the rock crack and destabilization warning point of view, this paper is aimed at optimum seeking of precursory key characteristic signals of acoustic emission in rock crack and destabilization, and puts forward the identification criterion and definition of key acoustic emission precursory signals of rock fracture. A new optimum seeking method for the key feature signal is proposed based on signal energy contribution rate, and the method is verified by experiments with granite, marble, coal and siltite. Experimental results show that the proposed method can effectively extract key acoustic emission signals which affect the overall stability of rock in the process of rock fracture, and this method can be applied in a wide range of applications. We also focus on the consistency test of the characteristic signals of the selected critical events from the precursory characteristics of the acoustic emission event rate, the relative quiet period, and the cumulative energy jump. The consistency test results show that the key feature signals after optimization are highly consistent with the precursory characteristics, and can effectively improve the calculation efficiency of acoustic emission characteristic parameters in late stage, and the method is feasible and effective.

Hydraulic fracturing is the key technology for shale gas development. Brittleness is one of the most important parameters to evaluate the fracability of shales. One of the most commonly used methods in laboratory is the measurement of wave velocity of pillar core to obtain Young’s modulus and Poisson’s ratio, so as to evaluate the rock brittleness coefficient. However, it is difficult to obtain regular shale core samples. Therefore, in this study, we develop a new device and a novel method with high precision to measure the velocity of shale debris based on pulse echo. Major factors that influence the measurement precision of shale debris wave velocity are analyzed. This study also proposes a new pin-type high frequency sound wave probe and a digital signal processing data analysis method. The measurement precision is improved by increasing sampling frequency and signal-to-noise ratio as well as using the autocorrelation method and pulse echo method. The comparison between results from regular core and debris confirms the reliability of the proposed procedure.