Based on the Fredlund's one-dimensional consolidation theory for unsaturated soil, one-dimensional consolidation of unsaturated soil subjected to an exponentially variable vertical load is investigated with a semi-analytical method. Elastic and visco-elastic soils with a finite thickness are considered while the soil top surface is penetrable and the bottom is impenetrable to water and air. Semi-analytical solutions are obtained for two soils with variable water and gas permeability coefficients and variable water permeability coefficient, respectively. A typical example result is given to analyze the changes of the excess pore-air pressure, excess pore-water pressure and sedimentation rate of soil under different conditions. The semi-analytical solutions are compared with those without considering the variation of permeability coefficient. It is shown that the permeability coefficient changes nonlinearly in the process of consolidation; when considering variable water permeability coefficient only, the dissipation of excess pore-air pressure is trivial, while excess pore-water pressure dissipates quickly; in the other situation, excess pore-air pressure and excess pore-water pressure dissipate significantly; and accordingly the sedimentation rate of soil also speeds up; the variation of gas permeability coefficients can substantially influence dissipation of excess pore-air pressure, with little effect on dissipation of excess pore-water pressure. The results are useful in analyzing the one-dimensional consolidation of unsaturated soils .

The perforation tunnel is the channel for transporting the hydrocarbon to the well bore. The shaped charge perforation operation, however, usually leads to some damage to the surrounding sandstone. Based on the perforation efficiency tests and numerical simulations, a new method is developed to quantitatively evaluate the porosity and permeability damage degree of the compacted zone. The kinematic-hardening plastic flow law is used to describe the sandstone deformation under high strain rate loading. Then dynamic responses of near-tunnel sandstone during perforation process are simulated by using the explicit dynamic analysis code LS-DYNA. Some key data, such as the sandstone matrix stress, plastic strain and volumetric strain are obtained. Subsequently, the damage degree of compacted zone is quantitatively analyzed by using these data together with the porosity and permeability evolution models. By comparing numerical results with the experimental results of core flow efficiency and CT scanning data from the interparticle pore sandstone targets in Shengli oil field; good agreements are achieved, the validaty of the proposed method is verified. It is shown that, during shaped charge perforation, there are two major damage mechanisms for the compacted zone, namely, the plastic squeezing and shear swelling. Near the tunnel’s inner wall, the sandstone matrix mainly suffers dilatancy actions, which enhances the porosity; meanwhile, the original structure of pore are changed due to plastic squeezing, which brings a jam to the tunnels, resulting in a reduction in permeability.

As cementing materials, cement and chemical grouts have often been used for soil improvement. However, high energy consumption, high-pollution discharge and high cost restrict their applications. A new soil improvement method, called bio-grouting reinforcement technology, has recently emerged, which is based on microbial induced calcite precipitation by injecting bacteria solution and nutrient into loose sand for the purpose of binding soil particles together and improving its physico-mechanical properties. Laboratory and field experimental studies of bio-grouting treated soils are systematically summarized. The engineering characteristics of bio-treated soils, non-destructive geophysical process monitoring method, and factors influencing the improvement effect are also discussed. The research results show that the bio-grouting process has many advantages such as small disturbance, low grouting pressure, less environmental harm, and remarkable treatment effect on improving the engineering properties of the soil (e.g. strength, stiffness, liquefaction resistance), which make it a broad engineering application prospect in soil improvement. However, the durability of bio-treated soils and economy of bio-grouting still need discussing and investigating further.

Further study of the mechanism and law of salt rock deformation and breakage has great significance for the safety and stability of the gas storage. Based on the triaxial compression test under different confining pressures, a new elastoplastic damage model with improved Mohr-Coulomb criterion is presented in the framework of the irreversible thermodynamics and nonlinear continuum mechanics, which can depict the confining pressure effect, hardening and softening behaviours of salt rock. In this model, the mechanical damage criterion and hardening function highly depend on the stress state. The solution process for the coupled model is studied using finite element method and the corresponding computer code is developed on the ABAQUS solver. The numerical simulation under various stress states shows that the numerical simulations perfectly agree with the experimental data; and the proposed model is able to effectively depict the main features of mechanical behaviors of salt rock.

Failure to prepare the rockmass sample with joints of the same surface morphology as those of natural joints has long been infesting the mechanical experiment on the structural surfaces of rock mass. With the latest development of the 3D printing technique and the rapid surface scanning technique, a new method is developed for producing the joint specimens containing the joint surfaces similar to those of natural rockmass. In the proposed method, a 3D scanner is first used to obtain the point cloud data related to the surface topography of joints of natural rockmass; then the virtual specimen model containing natural joints and related surface morphology information is reconstructed, and finally the PLA mould with the joint surfaces is produced using the 3D printing technique. The shear specimens are prepared by pouring the similar material into the PLA mould. Three types of rockmass joint surfaces are copied with the proposed procedure; and a series of measurements and shear tests are conducted on the made joint specimens. The results indicate that: (1) the proposed method can copy accurately the joint morphology from the originally natural joint; (2) among the three types of rockmass joint specimens, the displacement-force curves of the specimens are very close to each other, showing that the consistency of the made specimens containing joint surfaces is satisfied under the identical conditions; (3) by comparing the joint surface contour before shearing to that after shearing, it is found that the shear-induced failure occurs heterogeneously.

Unfrozen water contents of saturated soil with different soil types and different concentrations of NaCl are measured by low-field nuclear magnetic resonance (NMR) in freezing-thawing cycles. The distribution of unfrozen water in soil is discussed combining with T2 distribution curve from the microscopic viewpoint during freezing and thawing process. It is concluded from the experimental results that the freezing process can be divided into three stages: super-cooling stage, rapid decline stage and stable stage; while the thawing process can only be divided into two stages: stable stage and rapid melting stage, and there is no overheating phenomenon in thawing process. During the freezing process, water in large pore freezes firstly. In a sharp contrast, the pore ice in small pores melts firstly. Thermodynamic potential difference of pore water results in the sequence of phase change of pore water during the freezing-thawing process. In addition, the influence of soil type and ion concentration on unfrozen water and hysteresis during freezing-thawing process is analyzed and possible mechanisms for the hysteresis are discussed.

Traditionally dynamic consolidation tests were performed under low-energy impact, which is not sufficient to trigger significant mechanical response of soft soil so that the laboratory tests cannot properly simulate the actual energy transfer process in the field. In this study, static and dynamic consolidation tests are performed on muck soil, using a self-developed multi-directional high-energy electromagnetic impact testing system, from which the energy transfer processes in horizontal and vertical directions are characterized under repeated high-energy impacts. The experimental results include (1) The pressure increase in shallow soil is always the largest under impact; and the pressure increment in deep soil increases with the impact number; after first impact, the compression amount of shallow soil is the largest, and then the compression amount of deeper soils exceeds that of the shallow soil; and the difference becomes larger and larger as the impact number increasing, indicating that the main compression zone moves down gradually. (2) In both static and dynamic consolidation tests, the impact energy initially acts on the shallow soil, and then transfers to deeper soils, resulting in the improvement of the mechanical properties of the muck soil in depth. (3) Under impact loading, the vertical particle velocity is the maximum, followed by the radial direction, and ring direction is minimal at the same position of the surface layer of soil, and particle velocity approaches the same stable value at a certain distance; vibration acceleration also shows the same variation law; the region under the influence of vibration observed in the model test is consistent with those of the actual engineering projects.

Fluid flow tests are conducted on an artificial crossed fracture model. The relations between flux and pressure are obtained under different combinations of inlets and outlets. For these connected fractures, indeterminate equations are utilized to calculate the hydraulic aperture of each fracture element with the least squares method. The geometry of the fracture intersection is reproduced with topology method. The effect of fracture intersection on hydraulic characteristics is researched by solving Navier-Stokes equations. The results show that the fracture aperture for each fracture element can be accurately calculated based on flow test results and equations of hydraulic aperture. Within the fracture intersection, static flow and vortexes are observed respectively at lower Reynolds numbers and higher Reynolds numbers(Re≥100). It is shown that the inertial force is far larger than the viscous force and the cubic law is no longer applicative. For the cases with one inlet and two outlets, the normalized flow rate exhibits a quadratic relationship with the Reynolds number, especially when Reynolds number is large. The flow rate difference between the two outlets can be as much as 15%. The ratio of hydraulic aperture e to mechanical aperture E0, i.e. e/E0, varies with the increase of Reynolds number at the inlet. The value of e/E0 shows linear, little nonlinear and strong nonlinear relations respectively when Re10. A nonlinear modified cubic law is obtained with the relation between e/E0 and Re. This modified cubic law can be utilized to solve the nonlinear hydraulic characteristics caused by fracture intersections.

The evolution and spatial distribution of fractures are key factors in the sealability of mudstone caprock. To investigate the evolution of crack initiation, propagation, coalescence during the process of deformation and failure for mudstone, uniaxial compression test and real-time CT scanning have been carried out for mudstone specimens of Longmaxi Formation. The crack evolution under different load levels is discussed; and the relationship between crack evolution and macroscopic deformation is revealed from the mesoscopic scale based on the variations of crack area and fractal characteristics of crack structure. The results show that: the evolution of mudstone cracks can be reflected with the changes of grey frequency and crack area. When the load applied on the specimen is at a level of 0-20%, even though crack area increases rapidly, crack initiation is not active due to the closing effect of cracks in compression process. When the load reaches a level of 20%-65%, crack area changes little owing to the generating effect and closing effect of cracks contributing equally; thus mudstone is in a stable stage. When the load is up to a level of 65%-85%, crack area increases sharply as the uniaxial effect of crack extension and coalescence speeds up. The sudden increase of crack area at the load level of 65% exhibits the burstiness and instantaneity of crack transformation from initiation, closing to extension. The complicated and disordered internal structures and cracks evolving from specimen surface to interior are clearly observed with the further increase of load. The curve of crack fractal dimension appears the “up―down―up―down” trend, which shows the evolution of initiation, closure, extension and coalescence of cracks.

The theory of hypoplasticity provides an effective mean for modeling the nonlinear behavior of granular materials. In particular the CLoE hypoplastic model has salient advantages in modeling the strain localization problems. In simulating the cyclic stress-strain loops with small amplitude, however, the hypoplastic model shows the ratcheting effect. To resolve this issue, the intergranular strain tensor is incorporated into the CLoE hypoplastic model so that it can effectively describe the behavior of dense sand under cyclic loading. In addition, to reserve the consistency of the modified model in describing the sand behavior under monotonic loading conditions, the definitions of intergranular strain rate and the maximum intergranular strain are improved. The numerical results show that (1) the modified model reserves the advantage of overcoming ratcheting effect; (2) the modified model can account for the unloading stiffness of different cyclic amplitudes; (3) for large cyclic amplitudes, the area of the cyclic loop increases with cyclic number; (4) the consistency of the modified CLoE model and CLoE model can also be guaranteed under the same monotonous loading condition; (5) the modified model well represents the mechanism of the fatigue-induced failure of materials.

Forecasting of landslide failure is a significant topic in the field of geological engineering; but it is difficult to accurately determine the model parameters in the practical application. The ladder-shaped displacement curves are generally observed in actual landslide monitoring; and the changing points on the curve represent the mutation points of landslide deformation. To investigate the deformation characteristics at the mutation points in landslide, a series of rheological tests is carried out on natural samples under different loading levels and on immersion samples under different loading levels and with different water contents. The curves of accumulative deformation vs. time and deformation velocity vs. time are obtained. Based on the Qin's locking section theory and the landslide early warning model with inverse-velocity method, the experimental results are analyzed. The results show that the model parameters are independent of the loading level and water content, and are the functions of material properties. The deformation characteristics of mutation points in deformation process are similar to those at failure, and the model parameters of the landslide early warning model based on the inverse-velocity method at mutation point are identical to those at failure point. Hence the model parameters determined by the early mutation point in landslide monitoring curve can be utilized in the early warning for landslide failures.

Nonwoven geotextile is used in engineering as filter materials. When the geotextile is subjected to biaxial tension, the pore size of geotextile can be changed. Based on a 3D pore network model for the nonwoven geotextile, the method for calculating the number of structural layers in Rawal’s formulation is improved and a theoretical solution is proposed for the pore size distribution of nonwovens under equal biaxial tension. Two geotextiles are uniformly stretched in bi-directions to strains of 3%, 5% and 10%, respectively, using a self-developed bi-directionally stretching machine, and the pore-size distributions of the stretched geotextiles are determined under strain-controlled condition using the dry sieving method. By comparing the theoretical solutions and the measurements, it is shown that the theoretical method overestimates the pore sizes, and yields more reasonable results for the thinner geotextiles. In addition, the pore sizes increase with the biaxial tensile strain，and the characteristic opening size O95, O50, and O30 increase approximately linearly with the tensile strain.

Mohr circle is extensively used in strength theory of geotechnical materials, stress path and bearing capacity of foundations. The pole point is a special point on the Mohr circle. It is unique and can help to determine stresses acting on any plane. In particular, the pole point method is capable of solving complex stress problems. Based on the parallel line and normal line methods stated in textbooks of soil mechanics and material mechanics, a new method to find the pole point, i.e. the ray method, is put forward. The validity and the uniqueness of the ray method are proved by using the stress equilibrium method of the isolated element. It is concluded that the parallel line method and the normal line method are special cases of the ray method. The pole points obtained by using the parallel line method and the normal line method separately are not in the same position. But the desired stress state for any plane by using the two methods are the same. However, the parallel line method is relatively easy to use. Although the ray method is relatively complex to use, it can help to understand the concept of the pole point thoroughly.

A series of laboratory consolidated drained triaxial creep tests is performed with triaxial creep apparatus SR-6 to investigate the behaviors of the remolded loess from a loess slope of Jingyang in Shaanxi province. The tests focus on studying the creep characteristics of remolded loess with a 90% compaction degree under different water contents and confining pressures. Based on the experimental data, two different empirical models are developed. The results indicate that the loess in this area has obvious rheological properties, general showing a nonlinear attenuation creep. The creep behaviors of the loess are significantly influenced by confining pressure and water content. When the sample is under the conditions of smaller confining pressure and higher moisture, it exhibits notable creep companied by the bigger creep strain. A sudden increase in strain increment caused by equal load increment indicates that this level of load exerted on sample has exceeded the yield stress value of soil, and soil sample will produce sticky plastic deformation. Both models can aptly describe creep properties of loess under the partial stress levels between 20% and 65% and also have fewer parameters and easier method of obtaining parameters.

The conventional triaxial compression test is conducted on coal specimens, then the destroyed coal specimens are taken to perform CT scanning test, and thus the spatial distributions of fracture within the coal specimens are obtained by importing processed CT images into Mimics10.01. In addition, the body fractal dimension of coal specimens is preliminarily analyzed by self-compiled Matlab program. The results show that under triaxial compression condition, compression and shear failures prevail in coal specimens, and coal specimens exhibit a specific mechanical mechanism with varying confining pressure. The spatial distribution of coal three-dimensional fractures reconstructed by Mimics10.01 indicates that the connectivity rate and bulk density of shear failure plane of coal specimens tend to increase with the confining pressure. And the calculated results of body fractal dimension suggest that the more serious damage inside coal rock, the lower the fractal dimension value.

According to the features of soil solidification, a series of unconfined compression and water stability tests is conducted to investigate the effects of the variation of potassium salt and phosphate concentrations on the early strength of sulphoaluminate cement-stabilized soil. X-ray diffraction (XRD) analysis of phase composition of the stabilized soil is performed to reveal the mechanism for the changes of the early strength of the stabilized soil. The experimental results show that there is a threshold of about 0.6% in potassium salt and phosphate contents. If the mixing amount of the salts in the stabilized soil is lower than this threshold value, the strength of the stabilized soil increases with the increase of the salt concentration; while the amount exceeds the threshold, the strength of the stabilized soil will decrease. If the salt concentration is not more than 2%, the water stability of the stabilized soil with salt is similar to that without salt; the softening coefficient of the stabilized soils after 7-day curing is larger than 70%, whereas the water stability of the stabilized soil with K2SO4 is poorer, and the softening coefficient is about 60%. XRD analysis shows that the generation of insoluble mineral crystal with high strength and expansibility is the main reason that the early strength of the stabilized soil is improved after introducing potassium salt or phosphate. If the salt concentration is too high, the early strength of stabilized soil will be drastically reduced due to mineral crystal excessive expansion effect, which destroys the structure of stabilized soil .

To determine the deformation and fracture properties of brittle shale containing natural cracks sampled from Pengshui region of Chongqing City, uniaxial tests under uniaxial cyclic loading have been conducted using the rock mechanics test system RMT–150C. The experimental results show that: (1) Under the combined effect of cyclic load and cracks, the existence of cracks can weaken the local properties, accelerate the cracks propagation and promote the failure of the shale, resulting in the reductions in yield stress, fracture stress and peak strength; namely, the peak strength reduces by 13.7%―58.3%. (2) The axial strain shows a closed “sharp-leaf” shape hysteresis loop with a “sparse-compact-sparse” arrangement, whereas the transverse strain displays a 8-shaped hysteresis loop upward opening with a “compact-sparse” arrangement. The curve of transverse strain versus cyclic number can be divided into four phases: initial deformation phase, constant speed deformation phase with a low speed rate, constant speed deformation phase with a high speed rate, and accelerating deformation phase. The onset of transverse strain within the initial phase can be referred to as the sign of the completion of the pre-existing cracks propagating and coalescing into the new-forming cracks, and then transferring to next phase; a subsequent mutation occurrence can be considered as the omen of whole failure. (3) The main fracture morphologies of shale containing pre-existing cracks are can be categorized into two distinct types: tensile-shear cut-through mode and tensile cut-through mode. Both modes contains at least one tensile crack which cut through the pre-existing crack. (4) During the elastic phase, there is a linear relationship between the effective modulus and the damage area coefficient F, and the larger the damage area coefficient , the lower the effective modulus. (5) Under the cyclic condition of low stress level, irreversible deformation develops slowly, and the curves of axial strain versus times always stay in the initial deformation phase, while the specimens remain stable; under the cyclic condition of high-stress level, after the specimens have experienced the three phases, failures can be observed in the specimens. Under the cyclic condition of a stress level close to the theoretical peak strength, deformation curves directly reaches the acceleration phase and several cycles of loading can result in fracture of the specimens

To improve the efficiency of the rotary-percussive drilling, it is necessary to understand the rock-breaking mechanism under tooth impact. Current research efforts in this regard are mainly focused on the numerical or experimental simulations of rock breaking under tooth impact. Thus it is important to investigate the rock-breaking mechanism under tooth impact at low velocity. In this paper, the crushing pit morphology resulting from single-tooth impact is first analyzed; and then the physico-mechanical models are developed for classifying different crushing areas. Considering the low-velocity and confining pressure effects in single-tooth impact, the theory of spherical cavity expansion is revised, from which the stress and strain solutions are obtained in the area around the cracked pit of the rock. Analytical solutions are obtained for different regions using Matlab; and the effects of confining pressure, particle velocity and rock mechanical parameters are analyzed on the dimensionless stress and displacement of the three different regions. The applicability and capability of the proposed model are shown by comparing the numerical results with those yielded from the LS-DYNA code.

Prestressed anchor technics is a primary measure to reinforce rock masses in geotechnical engineering. It plays an important role in mobilizing the self-bearing capacity of the rock masses and improving the strength and self-stability of the rock masses. Because of the complexity and diversity of anchoring engineering, anchoring mechanism has not been well understood, and the design theory and related computational method are insufficient for engineering applications. Two important problems remain to be addressed, which are related to the existing analytical solutions of the stress transfer along the anchored section of the prestressed anchor cable. Firstly the stress variation along the anchored section of prestressed anchor cable has not been well characterized，and secondly the stress singularity at the endpoint has not been considered in the existing analytical solutions. Through the analysis of applicability and limitation of the previous analytical solutions, it is shown that the stress distribution along the anchored section can be divided into three stages: elastic stage, plastic stage and failure stage. The stress distributions are different in different stages. Within this context, an expression is developed for calculating the transferred load in the above three stages. The shear stress and axial force distributions along the anchored section are determined. For a practical case, the critical anchorage length is determined by the proposed equation, and the result is compared to those of other methods, showing the validity and capability of the proposed procedure

An effective approach to studying the mechanism and development of coal outburst is to conduct the outburst model tests using similar materials. A new similar material for the methane-bearing coal is developed. After mixing a pulverized with a specified particle size distribution as aggregate and a humid acid sodium aqueous solution as a cementing agent, the similar materials is made by sequentially shaping and drying the mixture. A number of orthogonal-ratio tests show that the similar material is quite similar to the raw coal with regard to bulk density and porosity under a 15 MPa forming pressure; and the material has a high compressive strength ranging from 0.5-2.8 MPa, which varies almost linearly with cementing agent concentration. Adsorption and desorption tests show that the similar material perform very well in regard to adsorption, and its adsorption isotherm agrees with that of the raw coal. The developed similar material possesses such advantages as cost-effectiveness, non-toxicity, easy manufacture, stable performance and flexible functions, and can be used to simulate the behaviors of the raw coal with different strengths, Model tests of methane outburst are carried out using the made similar material, and the phenomenon and processes of methane outburst are successfully revealed; it is shown that the similar material can simulate adsorptive and desorptive behaviors of methane-bearing coals.

In order to study the dynamic response of the large-scale underground cavern group during earthquakes，the large-scale underground cavern group of the Dagangshan hydropower is selected as the prototype to conduct the shaking table test. Being the first shaking table test on such field of research, the entire design of the experiment is given in detail. Firstly，basic simplified principles of the test，dynamic similar parameters derived by absolute similarity theory，simulation range, model dimensions and similar material ratio are given according to the objects and features of the test. Then，test boundary conditions and measurement system are designed. Polystyrene foam board is used as a flexible boundary and its material and geometric parameters are suggested. Strain brick sensors for measuring the internal stress of the surrounding rock are manufactured. Model making process, cavern forming and excavation are presented. Finally，input waveforms are elected and adjusted according to the similarity theory, and loading subsequence of the model test is given. Test data demonstrate that simplification of test conditions is reasonable, and the design and production of test model are successful, it can provide a guide for similar model designing and making.

This paper aims at studying the cyclic softening of strength and stiffness of natural sedimentary structural clay in long-term dynamic loading. The artificial structured soils with different inter-particle bonding strengths and different initial void ratios are prepared, using cement as bonding material and sugar to mimic the large pores. A series of dynamic triaxial tests is performed on the structured clay and corresponding remolded soil to investigate the influence of bonding strength, initial void ratio, confining pressure and dynamic stress amplitude on the accumulated deformation and dynamic strength. Three distinct relationships between accumulative plastic strain and cyclic number can be identified according to the critical cyclic stress ratio(CSR): plastically stable, critical and destructive types. The critical cyclic stress increases with the increase of inter-particle bonding and the decrease of initial void ratio. The structured clay with higher bonding strength behaves in more obvious brittle failure, and has a smaller strain associated with the inflection point of the accumulated strain curve, which is suitable to be regarded as the strain failure criteria for dynamic strength. Moreover, the dynamic strength increases with the increase of inter-particle bonding and the decrease of initial void ratio, and the dynamic cohesion cd decreases with the increment of cyclic failure number; while the dynamic angle of internal friction ?d basically remains unchanged. This study results can provide useful reference for controlling the dynamic catastrophes of soft ground.

The governing equation for particle accelerated transport is developed with considering adsorption-desorption effects, and then analytical solutions for particle instantaneous injection and periodic injection with point and surface sources are obtained using Laplace and Fourier transforms. Experiments on the particle instantaneous injection for point source are carried out. Comparisons between the experimental and theoretical results show that the proposed procedure can simulate the particle transport reasonably well. In addition, the transport parameters are analyzed for the case of particle instantaneous injection for the point source. It is shown that the peak concentration of particles decreases with the increase of adsorption coefficient. The effect of desorption coefficient on the curve part on the right of peak concentration is trivial; however, for the curve part on the left of the peak concentration, desorption coefficient affects both the particle concentration and the particle transport time. Furthermore, the shapes of the concentration contours in the x-y plane are approximately elliptic, and the range of the concentration contours increases with the increase of desorption coefficient. The contour gradients at upper and lower sides of the peak concentration decrease with the increase of dispersion coefficient. The study results provide a theoretical basis for various engineering items such as the underground pollutant treatment, exploration of groundwater, nuclear waste disposal and land filling of municipal solid wastes, and so on.

According to the characteristic curve of the surrounding rock via support in tunnel structures, the strength and the installation timing of the early supports can significantly influence the deformation and stress distribution of surrounding rock. Consequently, as an important part of the early supports in tunneling, the shotcrete support plays an important role in the deformation and stress distribution of the surrounding rock, depending upon the strength growth and hardening rate of the shotcrete after injection. To explore the effects of the strength growth and hardening rate of shotcrete on its early supporting performance, a series of shotcrete field tests was carried out on the tunneling sites. The variations of the strength and elastic modulus of shotcrete were investigated; and the effects of the hardening rate of shotcrete on its early supporting performance were determined. It is concluded that (1) The strength of the shotcrete develops fast and stably in the beginning; the average compressive strength after 24 hours reaches 6.5 MPa; after 48 hours, it increases up to 11.5 MPa. (2) In the surrounding weak rock, it is important to improve the early strength of shotcrete so that the surrounding rock deformation can be controlled. The greater the early strength of shotcrete, the more effective the early support. (3) The shotcrete hardening rate is an important factor controlling the settlement of the tunnel vault. Compared with slowly hardening shotcrete, the rapidly hardening shotcrete can reduce the settlements at the vault and the arch foot by 10.4% and 17%，respectively, and the horizontal displacement at the sidewall by 30.7%. (4) The effect of the hardening rates on the development of the plastic zone in the surrounding rock is insignificant. (5) The strength hardening process of shotcrete should be properly taken into account, and failure to consider the hardening rate of shotcrete may overestimate the supporting performances of the support and significantly underestimate the deformation of the supporting structure.

According to the mechanical characteristics of a single pile, the full-size pile is proportionally minified. A series of experiments is conducted on the model piles under various vertical uplift static loads, horizontal static loads and vertical compressive static loads, with considering various pile lengths and surrounding sands. Relationship among parameters such as deflection, displacement, axial force and burial depth, grain gradation of surrounding sand is clarified. The influence of loading direction on the bearing capacity of single pile in calcareous sand is explored. Pile-sand interaction mechanism in calcareous sand is analyzed. It is shown that, the displacement and deformation of a single pile vary significantly with the loading direction, burial depth and grain gradation of the surrounding sand. To increase the burial depth is more favorable for uplift pile than for compression pile. The effect of increasing burial depth is more pronounced in the early loading stage, and gradually diminishes as the load increases. When a stepwise load is applied to the top of the model pile, the bearing form alters from side friction bearing to tip resistance bearing. Particle breakage and redistribution can result in a degradation in εmax of uplift pile within the late loading stage. The bearing capacity of a single pile is larger in a wider-grading calcareous sand and the stability of the pile is better in the singleness-size sand. A decrease in side friction of the model pile would occur due to the shear-induced particle breakage in the vicinity of the pile; but at the moment the particles at pile tip crush, an instantaneous reduction in side friction would occur. The research results can provide a guidance for pile optimization, design and construction in calcareous sand.

Based on the explicit FE method and parallel computing cluster platform of ABAQUS, a large-scale refined two-dimensional (2D) nonlinear FE model is developed in the time domain with considering the heterogeneity of ground soils, for a typical site on the estuarine basin with undulating bedrock and nonuniformly distributed soil layers. The artificial viscoelastic boundary condition is proposed for simulating the semi-infinite field, the main characteristics of basin surface ground motion are analyzed, including the motion amplitude, spectrum, duration and transfer functions, and the accumulation effect of ground motion and edge effect of basin are investigated. The results show that: (1) The amplification effect at different observation points on the basin ground surface is observed; and the peak ground acceleration(PGA)of basin ground surface displays a non-monotonic decrease with soil depth; the prominent accumulation effect is found to be at particular positions where the bedrock surface sharply undulates, and the pronounced edge effect is found to be on both sides of basin; the moderate and long-period ground motions of the basin are significantly amplified compared to the short-period motions. (2) For the earthquakes with high, moderate and low occurrence probabilities, the predominant periods of the basin are 0.35-0.65 s, 0.40-0.75 s and 0.50-1.05 s, respectively；the mean value contours of the peak ground acceleration and site predominant periods of the basin are plotted, and the recommended values of the amplification factor of ground motion acceleration response spectra for different period intervals are given. The seismic effect coefficient amax and characteristic period Tg of the design parameters of ground motion by this paper are significantly larger than those of the Chinese code for seismic design of buildings. (3) Ground motion durations at different points of basin ground surface extend over several orders, and the duration is also closely related to the bedrock motion characteristics. (4) Amplification effect and accumulating effect of the basin ground motion are more significant for the frequency band from 0.5 Hz to 2 Hz; while for the frequency band lower than 0.2 Hz or higher than 2.5 Hz, the ground motion amplification characteristics are insignificant. (5) Ground motion amplification effects in Fuzhou city downtown area and its adjacent areas of the basin are generally larger, compared to other areas. It is indicated that the large-scale two-dimensional nonlinear FE model can describe the influence of micro topographic relief and nonuniformly distributed basin soil layers on seismic wave propagation.

An investigation into the joints of the ore body and rock mass at Chengchao iron mine is performed with combination of digital panoramic borehole camera system and in-situ statistics. Based on software Surpac, a 3D model for the gob is built. By subdividing the 3D model, a 2D discrete element model based on Universal Distinct Element Code is developed. The model is utilized to simulate the law of displacement fields and the rock-fall patterns of the overlying rock at different elevations when ore bodies are mined. The variations of the rock movement angle are analyzed. It is shown that: (1)When the ore bodies at the elevations between -430m and -500m are mined using back filling method; the maximum displacement for the gob is 205.6cm, which is smaller than the 215.2cm measured with the caving method. By comparing the settlement values and the horizontal displacement values of the auxiliary shaft, it is found that the incline rate of surface is 0.56mm/m for back filling method, smaller than the 1.22mm/m for the caving method; the horizontal deformantion rate is 1.03mm/m, smaller than the 1.31mm/m for the caving method; it is shown that the back filling method has a good effect on controlling overlying rock displacement and ground movement. (2) The movement of the overlying rock behaves in the manner of intermittence and jump when using the caving method. A self-balancing arch can be formed in the process of the overlying rock falling; hence, a number of hidden gobs may occur. Based on the comparison between the practical monitoring data and the numerical simulation, it is found that when a long period mining is carried out at a given depth, the decrease rate of movement angle exhibits a nonuniform variation mode, namely “slow-acceleration” cycle, implying the occurance of the self-balancing arches and gobs.

To accurately estimate the damage of chute under ore impact, a theoretical model is developed for describing the damage of the chute considering the influential factors of chute damage. The impulse distributions induced by ore impact on chute are obtained with MATLAB. Based on the MADIS-FLAC3D data coupling technique, a numerical model of chute is developed and used to analyze the distribution of chute damage. Combining the capacity monitoring system (CMS) and the 3D software SURPAC, the 3D information relating to chute collapse has been effectively determined and visualized, and the proposed procedure is validated. It is shown that (1) the range of impact calculated with the proposed model is at the elevations from -265.83 m to -292.28 m; and in this zone, the impact momentum increases firstly, then decreases, and reaches its peak value 13.52×103 N?S at the elevation -272.88 m; (2) the displacement occurs in the zone from the elevations -264.12 m to -294.88 m under the ore impact, where a plastic zone is formed, indicating that the stability of surrounding rock in this section has been degraded; (3) the location of the maximum measured displacement is close to the position of the maximum cross-section of collapse area from the monitored results. The shape of the most serious collapse area of chute from monitoring agrees with the plastic zone determined by the numerical analysis.

To study the influence of principal stresses on the stability of roadway intersection, five numerical models of roadway intersection are developed based on the engineering project in Wujiagou coal mine, and the rotation of the disturbance stress field due to roadway excavation is investigated. By introducing the pointwise safety factor method, the influence of principal stress and the angle of roadway intersection is analyzed on the displacement field and the disturbed broken zone around the surrounding roadway intersection. The results show that the orientation of principal stress in the vicinity of the intersection rotates after the excavation of the main roadway, and thus the stress field changes from σh type to σv type. In addition, during the excavation at the intersection, the surrounding rockmass is influenced by the σv-type stress field, and the surrounding rockmass at the intersection becomes more stable when the intersection angle (α≤90°）increases. At the same intersection angle, the intersection point on the direction of the maximum horizontal principal stress becomes more stable in a newly excavated tunnel. These results are confirmed by the monitoring data at the intersection in Wujiagou coalmine.

This paper reports the design of an auxiliary shaft at a depth of more than 1 000 m at Cixi colliery and its surrounding geological conditions, and addresses the scheme for monitoring the in-situ stress around the shaft and the principle of inverse calculation for the in-situ geostress using the “Baoshen” analytical solution. The average values and nonuniform distribution of horizontal stresses are actually obtained at the depth of 1 208 m through inverse calculation and analysis of field monitoring data, which reveals that the ratio of horizontal stress to vertical stress conforms to the dead weight stress field. The lateral pressure coefficient ranges from 0.256 to 0.278. The in-situ stress distribution around the shaft is illustrated based on finite element simulations; and thus the reliability of inverse data based on field measurements is corroborated. Meanwhile, it is shown that the nonuniform distribution of horizontal stress is related to some factors such as the shear modulus of rock and geological tectonic, etc. With combining the information on tectonic evolution in this area and geomechanical principles, the consistence among the results of in-situ test, analytical solution and numerical simulations is comprehensively analyzed. In this context, the universality of the Hoek & Brown statistic principle about the distribution of horizontal in-situ stress is questioned, and the existing analytical methods for calculating the in-situ horizontal stress are also discussed, laying a foundation for using the “Baoshen” solution to the shaft lining design.

To understand the mechanical characteristics of the three-layer support system (primary support, first liner and secondary liner) of shallow tunnel in loess region, large scale in-situ tests are conducted based on a loess tunnel. The contacting pressure on the surrounding rock, first liner and secondary liner, the axial force of steel, the strain of secondary liner and the stress of steel arch are systematically monitored by the steel string transducers. The test results indicate that: (1) The surrounding rock pressure is larger on the tunnel left hance with a maximum value of 240 kPa. (2) Under the influence of geology and construction, the rock pressure in shallow region has a large difference between the monitored and calculated values; and the pressure calculated based on the Terzaghi's rock load theory is relatively close to that monitored in the field. (3) The secondary liner pressure approximately takes on as “cat ears”. Moreover, the loading ratio is approximately 51.34%, 37.29% and 11.38% for the primary support, first liner and secondary liner, respectively. (4) The steel arch is at pressure in majority, the hance and vault bear larger pressure, where the steel arch is close to yield. (5) The primary support coupled with the first liner bears the majority of rock load. In summary, as an effective support system, the three-layer support system presented for this case study can serve as a benchmark for effective design and safe construction of similar projects.

Shield machine is a kind of typical equipment for tunneling, it is featured by heavy body, complex structure and high load. The load acting on cutterhead during tunneling is a core mechanical parameter for the equipment. The reliable calculation of the load is important in terms of positive equipment design and load control of tunnel constructions under different geological conditions. The center tool of shield cutterhead bears a large proportion of axial load in the tunneling process. To further analyze the effects of center tool with different structures on the load on shield cutterhead, the studies of the interaction between the shield cutterhead and the soil of tunnel face are conducted by means of numerical simulation and model experiment. The dynamic cutting load distribution along the normal and tangential directions of cutterhead is obtained; and the temporal variations of the total thrust and torque during the process of excavation are also achieved. The effects of the offset center tool on the total thrust and torque on cutterhead are emphatically analyzed. By comparing the loads acting on different types of cutterheads, one can see that the offset center tool produce lower total thrust and torque than the fishtail center tool does. The results of numerical simulation are in good agreement with the experimental results. The offset center tool installed on a cutterhead can reduce the total thrust effectively. From the loading point of view, the offset center tool is better than the fishtail center tool. The research results can provide some useful guidelines for the design of new shield cutterheads.

With a theoretical basis more rigorous than the limit equilibrium method, the upper-bound limit finite element method can be used to determine not only the safety factor of slope but also the critical slip surface so that it will have a broad prospect of application. To remove the limitation that the traditional upper-bound limit finite element method cannot address the effect of heterogeneity, a new Mohr-Coulomb yield surface linearization method is proposed herein, based on the linearized spatial discretization. Within this context, the linearized constraint equations for plastic flow are developed, which enriches the upper-bound limit method based on linear programming and lays a solid foundation for the application of linear programming technics to the upper-bound limit analysis. Two examples are analyzed, showing that the proposed method stably yields a convergent solution from above the upper-bound solution. In analyzing the stability of a slope, if the strength anisotropy is ignored, the factor of safety is overestimated, resulting in a larger factor of safety of the slope.

To resolve the deficiencies of the calculation model for anti-sliding pile with prestressed anchor cable, an improved model is presented with detailed derivations based on the finite difference theory. According to the actual construction process and loading conditions of pile and anchor cable, the model includes two stages, i.e. applying prestress and landslide thrusting. By improving the distribution of landslide thrust and revising the pile-anchor deformation compatibility equation, the actual active mechanical characteristics of anti-sliding pile with prestressed anchoring cable are properly taken into account. The proposed procedure is implemented into a Matlab code and used to analyze practical examples. The results are compared with those of the conventional theories. It is shown that the influence of prestress application on anchor-pile can be properly taken into account in the proposed procedure and its corresponding code, and it is not necessary to equate the anchor cable tension with the shear force and the bending moment at the top of pile. In addition, it can also be used in the design calculation for the cases that the sliding body with arbitrary thickness above the sliding surface exists in front of the piles and for the arbitrary complex foundations. Compared to the conventional theories, the improved procedure is more reasonable and reliable.

Based on the upper-bound limit analysis, which is theoretically rigorous, the bearing capacity of the composite foundation of upper geosynthetics-encased stone column (UGESC) in soft soil is analyzed through constructing the consistent velocity field and considering the effect of gravity and skin friction of the encased section of the stone column. An equation for calculating the ultimate bearing capacity of single UGESC is proposed based on the energy balance principle. Based on this, a stochastic optimal method is used to search for the critical sliding surface, and a more reasonable and rigorous solution for ultimate bearing capacity of UGESC is derived. A case study is performed, showing that the relative errors of the calculated results are less than 5.4% compared to the previous results in the literatures. The optimal value of upper encased depth is about eight times the pile radius. The procedure proposed in this paper can provide certain reference for engineering practice.

To calculate the transmitted and reflected velocity waveform of elastic longitudinal wave with arbitrary function form propagating through nonlinear joint by Fourier series theory. Based on the discontinuous displacement model and the momentum balance equation at the wave front, the governing equation for the propagation of an elastic longitudinal wave through the nonlinear joint is developed, in which the joint deformation is described by the Barton-Bandis model. With assuming that the minimal positive period of stress waveform function remains unchanged as stress wave propagates across the nonlinear joint, the arbitrary Fourier series solutions of transmitted and reflected waves are obtained using the Fourier series method and the periodic extension method, and the Fourier series solutions are validated. Based on Fourier series solutions, the dependence of the amplitude and phase on the order number of the harmonic waves generated by the transmitted wave at the joint is analyzed. It is shown that the relationship between the amplitude and the order number of the harmonic waves follows a negative exponential law. The attenuation index of the amplitude is a quadratic function of the order number when the order number is less than 7, whereas the attenuation index of the amplitude is a linear function of the order number when the order number is less than 7. A linear relationship exists between the phase and the order number for harmonic waves of different orders.

Seasonal rainfall and reservoir operation can cause the fluctuation of reservoir water level and the change of pore water pressure distribution in the slope body, which is detrimental to slope stability. Based on the theory of shear strength of unsaturated soils and pore water pressure field obtained with finite element method for the transient unsaturated seepage, an improved numerical procedure for critical slip field of slope is proposed. In this procedure, the combined effect of water level fluctuation and transient unsaturated seepage on the behaviors of slopes can be taken into account. First, the transient unsaturated seepage fields of a clay/ silt slope are calculated using SEEP/W module in Geostudio to obtain pore-water pressure fields, which are imported into the improved procedure of critical slip field of slope method to evaluate changes in the factor of safety with changing water level. In addition, the influence of fluctuation velocity of water level and matric suction on the slope stability are also analyzed emphatically; and the slope stability changing course with water level fluctuation is revealed, so as to show the feasibility and reliability of the improved procedure. Moreover, the results not only show that the slope stability changing course is affected by fluctuation velocity of water level, matric suction, and so on, but also reveal that the more reasonable slope stability changing rules and essences are obtained by considering matric suction.