The research on K0 behavior under different stress states is important in grasping the mechanical parameter of the soils. However, there are few apparatuses and methods applicable for measuring K0 of coarse grained soil, therefore, as for the K0 behavior of coarse grained soil in different stress states, the relevant researches are almost blank. For coarse grained, in order to study the transformation law of at-rest earth pressure coefficient (K0) during loading and unloading, a new large-size K0 test apparatus was developed and used to conduct a serial of K0 tests for sandy gravel and rockfill in different stress states. The results of K0 test show the particle morphology、stress state and grading all have certain effect on the value of K0 of coarse grained soil, and the effect of stress state is the most obvious. During loading or unloading, K0 of coarse grained soil will decrease with the increasing of vertical stress. Furthermore, based on the results, a description of the K0 considering effective vertical stress during loading is established and verified. Lastly, based on this description and the relation between K0 and over consolidation ration (OCR) developed by former researcher, an empirical equation is obtained, which can well describe the relationship between K0 and stress state under an arbitrary consolidation condition.

Loess is a special soil because of its large pore structure and vertical joints. The structure of loess is unique, so the structure of loess will be destroyed if the loess is vacuum saturated. The saturation of soil is one of the important steps in the triaxial test. It is known that the saturation degree of specimen closely correlates with the accuracy of the loess mechanical properties. Eight groups of loess samples from Lanzhou and Tianshui, Gansu province, China, were used in the triaxial tests of saturation by GDS triaxial apparatus. A total volume change sensor was used during the saturation of loess samples. Later the total volume change tested by the sensor and the quantity of water in the sample tested by the controller. The saturation degrees of loess samples were calculated by the parameters of loess. After that, the B value would be collected by the computer at the same time. The relationship between the saturation degree and the B value was obtained at the same time. The influence of different methods of saturation on the loess was discussed. The nonlinear fitting on the saturation degree and B value was also conducted so that an empirical formula with exponential form was produced.

To understand the mechanism of the settlement caused by water and sand inrush in underground engineering, a set of visual test device that can change the blow mouth diameter and simulate gushing of water and sand is designed. This device can vary the condition of different thickness-to-span ratios and different contents of fine particles in sandy soil. When the thickness-to-span ratio is more than a certain value, regardless how much content of fine particles in sand, the soil will not settle. When the thickness-to-span ratio is within a certain range and the fine particles is less than a certain value (i.e. critical fine particle content), the soil will settle; if the fine particle content is greater than the critical particle content, sand burst will occur in the soil. The critical fine particle content of quicksand increases parabolically with the thickness-span ratio at the initial stage, but remains at a level at the later stage. When the thickness-to-span ratio is less than a certain value, regardless how much content of fine particles in sand, quicksand will occur. The settlement induced by water gushing and sand gushing in circular mouths is circular in plane and quadric in profile. The settlement scope after quicksand is related to the shape damage angle of subsidence surface, which is roughly 9° larger than the natural angle of repose of the soil.

This study is aimed to investigate the regularities of the shear strength, shear deformation and other mechanical properties of joint rock mass with different roughnesses under compression-shear stress. Marble specimens were prepared with the dentate height of 0, 1 and 3 mm, respectively. Then direct shear tests were conducted on the marble specimens under different normal stresses. Experiment results show that curves of shear stress and displacement can be divided into two types: sliding failure type and peak shear type. The climbing-gnawing process of joints goes through four stages: the compaction stage, the climbing stage, the saw-tooth cutting stage and the completely crushing stage. The curves of shear strength with the normal stress of the dentate joint present more obviously three-segment line feature, and a trilinear shear strength criterion is established. The curves of shear stiffness with normal stress present a convex type, and a shear stiffness model is put forward. The dilatancy curves of dentate joint present firstly shear contraction and then shear dilation, and an empirical formula for evaluating shear dilation is proposed.

Fine particles in an internally unstable sand and gravel soil may be washed out under seepage, thus the bearing capacity of a dam foundation with those sand and gravel soil layers reduces. The determination of critical piping permeability gradient of gravel soil on dam foundation is an important step in dam design and seepage prevention safety analysis. In this paper, the prediction of the critical vertical upward hydraulic gradient of piping in sand and gravel soils were developed based on the force balance theory of a single particle or a group of particle under seepage. Based on 64 groups of vertical upward piping tests of gravel soils in literatures, Kenney-Lau stability criterion method and Kezdi particle size ratio method were used to analyze the internal stability of gravel soils. The results show that the proposed method is an effective method to access the internal stability of gravel soils using particle gradation curve. By comparing the predictions of critical permeability gradient with the experimental results of 41 groups of sand-gravel soils, it is a reasonable assumption that the permeability of particles in sand-gravel soils under seepage is proportional to the surface area of particles .

The on-bottom stability of deepwater seabed pipelines is affected by the vertical penetration depth and resistance. In this study, the coupled Eulerian-Lagrangian approach (CEL) was used to simulate the vertical penetration of pipelines in cohesive soil, and the effects of strain rate behavior and strain softening on bearing capacity of pipelines were studied by subprogram VUSDFLD. The resistance provided by the heaved-soil on both sides of the pipeline during the penetration process was analyzed and the sensitivity analysis of the resistance was implemented, then a surface heave capacity factor was proposed. Results indicated that bearing capacity factor of the pipeline increases significantly with the increase of strain rate parameter. A high sensitivity of soil leads to a fast softening rate of soil and a low bearing capacity factor of pipeline. The surface heave capacity factor mainly depends on the submerged weight of the soil. Moreover, the greater the penetration depth, the more obvious the effect of the submerged weight of soil on the bearing capacity of surface heave. An expression describring the surface heave capacity is developed by fitting the numerical results.

Introducing the fluid-mechanical interaction model based on continuous porous media theory and mixture theory, applying FISH language to program, the initial seepage field setting function, the saturated-unsaturated seepage analysis module, the special stress-correction module and shear strength correction module are established. Saturated-unsaturated seepage analysis is conducted with FLAC3D. The effective stress and shear strength of soil zone in the semi-saturated region is modified based on the saturated-unsaturated seepage field and the principle of fluid-mechanical coupling for completely saturated element and extended to unsaturated soil element in FLAC3D. Based on these analysis modules from secondary development, the water drainage experiment from a sand column conducted by Liakopoulos is numerically simulated to verify the correctness of the model and program. The effect of rainfall infiltration on the seepage field, displacement field and stability evolution of soil slope reveals the mechanism of rainfall-induced failure of slope.

The high-speed railway bridge and tunnel junction is considered as a complex engineering project. At the event of strong earthquake, serious damage or even completely destruction of this junction would occur under complex topography or poor geological conditions. However, there is limited information on the seismic damage mechanism and dynamic response of interactions between different structures and locations in the complex bridge-tunnel-rock system. Therefore, large-scale shaking table tests were carried out on soft surrounding rock. The seismic dynamic response characteristics of the bridge-tunnel junction and the influence mechanism of the excitation direction were analyzed as well. The experimental results showed that the effect of strong earthquake is detrimental to the safety and stability of the bridge-tunnel-soft rock system. There were significant differences on variation laws of dynamical responses such as the acceleration, displacement and strain parts at the tunnel entrance and the end of the bridge under different earthquake excitations. Relatively large values of seismic responses could occur at the overlapped parts or sections of the bridge-tunnel junction (arch waist and arch foot at lining expansion section, arch foot at the tunnel standard section, the bridge beam end and top of the abutment inside the tunnel portal, and the soil above the cave roof etc.), which required special reinforcement or treatment. The obtained results provide useful references for the further study of structural seismic design and seismic response analysis of bridge-tunnel connecting structures in soft surrounding rock conditions.

In order to reveal the cohesive failure mechanism of the second interface of anchor system in earthen sites, two groups of 30 soil-grout interface samples are designed to carry push-out test in laboratory with different bore wall roughnesses and mixture ratios of grouting materials. The second interface properties are studied relying on the independently-developed pressure testing machine and sample preparation device. The results show that the load-slip relation curve has obvious peak strength. The 30 cm anchorage length can provide shear stresses of 0.08～0.26 MPa on smooth hole wall and 0.26～0.46 MPa on thread hole wall. The shear strength of the thread interface has increased by 4.21 times in maximam. The shear strength of samples in two groups is almost the same by the increase ratio of the calcined ginger nuts (CGN). The character of smooth/thread interface is strain hardening/softening in shear-slide curves. According to bond stress-slip relationship model, the fitting curves present good agreement with the test results. The research lays a theoretical foundation for the internal mechanical process of anchor system in earthen sites.

Grouting is an effective method to seal against water inrush in fractured rock. It is important to study the mechanism of fissure grouting to design the grouting and implement the project successfully. In this paper, firstly, the influence of the rheological properties of the slurry, diffusion path and grouting methods on fracture grouting was analyzed deeply. An improved step-wise calculation method was proposed to describe the fissures grouting process under different grouting diffusion conditions. This method can calculate key grouting parameters, such as the slurry front position, injection pressure, slurry viscosity and pressure distribution in the grouted zone, and grouting rate at any time. Secondly, the improved method was utilized to study the constant-parameters such as constant pressure grouting and constant rate grouting during the grouting process. The reliability and applicability of the approach were verified by comparing with previous results in the literature. Finally, the approach was adopted to investigate the dynamic fracture grouting characterized by staged descending rate and staged increasing pressure in practical engineering conditions. The change laws of the key factors, such as the injection pressure, grouting rate and diffusion distance, were analyzed. Therefore, this study revealed the diffusion mechanism of the dynamic fracture grouting to some extent and provided a meaningful guide for grouting design in practice.

To study the strength deterioration rules of saturated sandstone under freeze-thaw cycles, laboratory freeze-thaw tests were carried out on two groups of sandstone with different initial porosities under water saturation. On the basis of measuring the porosity variation, static peak strength and dynamic peak strength, the evolution law of porosity variation and loss rate of peak strength were analyzed within the whole freeze-thaw cycles. The extent of freeze-thaw damage was compared between the two groups sandstone by using the porosity variation to measure freeze-thaw damage. A relative residual peak strength deterioration model of saturated sandstone under freeze-thaw cycles was established based on the porosity variation. The same function as the strength deterioration model of saturated sandstone was selected to fit the test data，showing a good correlation. Within the whole freeze-thaw cycles, the porosity variation and loss rate of peak strength increased with the freeze-thaw cycles for two groups of sandstone. Freeze-thaw damage of group B was higher than that of group A, but its loss rate of peak strength was lower than group A, which indicates that the peak strength of saturated sandstone is affected by freeze-thaw cycles and strain rate and this two factors have completely opposite effect mechanism. There exists a good fitting relationship between the relative residual peak strength and the porosity variation of saturated sandstone under freeze-thaw cycles, which shows that the porosity variation before and after freezing-thawing is suitable for evaluating the relative residual peak strength of saturated sandstone after freeze-thaw cycles.

Large-diameter single pile foundation is a typical foundation type for offshore wind turbines. Model tests of large-diameter single pile are conducted using a simple mechanical load rig designed in-house in sand. The cumulative deformation response of a single pile under long-term lateral cyclic loading is discussed. The test results indicate that the cumulative displacement of the pile head presents two phase characteristics with number of cycles, and the short-term effect for lateral cyclic loading is greater than the long-term effect. Sand density and cyclic loading paths greatly influence the displacement characteristics of the pile head. The lateral secant stiffness of the pile head increases with cycling under different initial conditions. The change of horizontal static bearing capacity of single pile after cyclic loading in sands of different densities is not completely consistent with hard-soft mechanism for loose-dense sand subjected to shear. The prediction model parameters for lateral cyclic cumulative displacement of single pile are obtained by use of the empirical formula method. Comparison between the calculation results and the test results show that this model can effectively predict the cumulative displacement of single pile under different cyclic paths.

Since the rockbolt is in the state of high bearing stress after the deformation of surrounding rock, it is easily to fail due to blasting vibration, earthquake and other dynamic loads in the deep rock mass engineering. Therefore, it is urgent to study the mechanical response mechanism of the bolt under dynamic disturbance. Based on the SHPB system, a set of testing equipment was developed for studying the dynamic response of bolts. Then the mechanical response characteristics of full-length bonded bolts were investigated under dynamic disturbance. The slippage of bolt increased with the increase of the incident energy at initial dynamic load, and the peak value of stress wave decreased with the increase of propagation distance. When the stress wave propagated to the farthest end of bolt, the peak value of stress wave decreased greatly. After the second dynamic load, the peak difference of stress wave between strain gauge SG1 and strain gauge SG2 was obviously smaller than that at the first time. This result indicates that the anchorage interface begins to damage from the outside of the anchor under the dynamic load. The failure of the anchor is related to the damage of the anchor interface. The anchorage interface is damaged after the first time loading. Then the damaged anchor bolt will be further deteriorated by the external load (such as secondary impact and rock mass extrusion). As a result, it can not resist the deformation of surrounding rock and fail. The research results provide new ideas for revealing the failure behavior of the supporting bolt and adopting the reasonable design and construction.

In the landslide control and slope protection project, the circular cross-section anti-sliding pile has been widely used in geotechnical engineering practice because of its advantages of mechanized pile construction. However, compared with the traditional rectangular cross-section anti-sliding pile, there is less research on the circular cross-section anti-sliding pile, and the design and theoretical systems are not perfect enough. There is no unified conclusion on anti-sliding effect yet. In this study, the model experiments were conduct to investigate the force and deformation characteristics of circular section and rectangular section anti-slide piles, comparing their anti-slide abilities. On the premise of satisfying the similarity relation of each test, the pile-free push test, circular section anti-sliding piles test and rectangular section anti-sliding piles test were carried out respectively to observe the whole process of the deformation of soil and model piles. The experimental results were compared. Eearth pressure cells, reinforcement stress gauges, and dial indicators were arranged to monitor the earth pressure, reinforcement stress and displacement of pile top. The mechanical characteristics of the model pile were analyzed. The performance of circular cross-section anti-sliding piles is basically the same as that of rectangular cross-section anti-sliding piles under the same conditions. The force transfer is uniform in the process of loading, the requirement for the strength of the fixed end of the pile is low, and the anti-sliding bearing capacity is good. It is worth popularizing and applying in engineering practice.

Concentrated rainfall and fluctuation of groundwater level are important factors to induce the sapping and collapse of silty soil earthen archaeological site in Central Plains. Six groups of soil column models with layered sampling function were subjected to different dry-wet cycles respectively for silty soil from earthen archaeological site at Guchengzhai in Xinmi. The samples prepared for consolidation drainage shear test, nuclear magnetic resonance (NMR) test and scanning electron microscope (SEM) test were taken from the soil column, and then the correlation between the mechanical properties and pore structure influenced by dry-wet cycles was studied. The results showed that the stress-strain curve shape of the silty soil under the influence of dry-wet cycles was related to confined stress. The shear strength and cohesion increased after the first dry-wet cycle, and then gradually decreased to be stable. The stable values were lower than those of the initial sample not experiencing dry-wet process. Also, the internal friction angle slightly changed. It was recommended that mechanical parameters of dry-wet circulation after stabilization should be used for the damage restoration project in earthen archaeological sites for areas frequently affected by water. Combined with the NMR and SEM test results, the main reason for the change of macroscopic mechanical properties was that with the increasing of dry-wet cycles, the intermediate pore volume of soil firstly decreased and then slowly increased to a steady state, while the small pore volume of soil gradually increased to a steady state. The total pore volume of soil firstly decreased and then gradually increased to a steady state due to the adjustment of the two types of pore. It is believed that the intrinsic mechanism lies in the uniform contraction of the clay “net frame” after one dry and wet cycle and the destruction of the clay “net frame” in the subsequent cycle caused by the special grain grading of silty soil.

As the anti-seepage structure of the rock-fill dam, the force safety of asphalt concrete core is well concerned by the engineering field. So it is important to quickly and easily study the force and deformation of the asphalt concrete core. Firstly, the deflection deformation function of the rectangular thin plate with three sides clamped and one side free (CCCF) on the Winkler elastic foundation under the hydrostatic pressure is solved via the small deflection bending theory of the thin plate and Ritz method. Moreover, the deformation, internal force and stress calculation formulae of the CCCF rectangular thin plate are derivated. Secondly, the one-to-one mapping relation between the CCCF rectangular thin plate and the asphalt concrete core is established. The deformation, internal force and stress of the CCCF rectangular thin plate are transformed to the actual valley section core. Thus, a simplified analytical force analysis model of the asphalt concrete core is established. Then the reliability of the model is verified via the FEM. Finally, the simplified analytical force analysis model is applied to an engineering example. This model helps to quickly estimate the force and deformation of the asphalt concrete core. At the same time, the research idea expands the calculation method for the force and deformation of the asphalt concrete core.

Dense sand normally exhibits shear dilatancy softening under normal pressure and shear shrinkage hardening under high pressure. Therefore, unified model of sand and its numerical implementation is an important scientific problem for describing the loading response of dilatancy softening material, and it will also promote the progressive failure theory. In this study, the unified model of sand which reflects the mechanical properties of dense sand under normal and high pressure is firstly described, and then the key problem of numerical calculation for the model is analyzed. By using the explicit method, the secondary development of stress integration program is carried out. In the process of stress integration, loading-unloading criterion within strain space and smoothed failure surface are used for satisfying the judgment of loading condition and determination of plastic strain direction. An example is given to prove that the explicit method can overcome the difficulties in calculation for dilatancy and softening. This method shows a good way to analyze the mechanical behavior for materials with dilatancy softening and shrinkage hardening characteristics.

The freeze-thaw cycle represents one of the most common physical weathering processes in cold regions, which can significantly affect the deformation characteristics of soils. An improved constitutive model with double yield surfaces, which can take the freeze-thaw effect into consideration, was developed based on the traditional elliptic-parabolic yield surfaces model in this study. First, a set of freeze-thaw cycling tests and consolidated-drained(CD) triaxial tests were conducted on silty clay from Qinghai-Tibet Plateau. The test results showed that the stress-strain behaviors exhibited strain hardening and shear-shrinkage characteristics. The volumetric strain increased while the failure strength decreased during the shearing process. Then, the elliptic equation and parabolic equation were adopted to fit the volumetric yield surface and the shear one in p-q plane, respectively. The incremental deviatoric stress-axial strain relation and deviatoric stress-volumetric strain relation were subsequently established by following the associated flow rule. The parameters of above relations were valued according to the tested deviatoric stress-axial strain-volumetric strain relationship, and were all found varying by the number of freeze-thaw cycles with certain regularity. The c、φ、h、K、M1、M2 and a all kept decreasing with increasing the number of freeze-thaw cycles, while the t and n kept increasing, and they can all be fitted by the Logistic function. At last, the expressions of above parameters with number of freeze-thaw cycle as variable were included into the incremental constitutive equations, thus an improved model considering freeze-thaw effect was established. The deviatoric stress-axial strain curves and volumetric strain-axial strain curves predicted by the proposed model agreed basically with the experimental results.

Reinforced soil retaining walls showed excellent seismic performance in a large number of earthquakes. In the design of reinforced earth retaining wall under seismic loading, the dynamic load can be regarded as equivalent to the static load according to the quasi-static method. A shaking table test of reinforced soil retaining wall with full-height rigid (FHR) facing was carried out to compare the difference between the response obtained by pseudo-static method and the actual dynamic response. The facing displacement, response acceleration, dynamic earth pressure, and the reinforcement load were measured and analyzed. The following conclusions have been drawn: the distribution of the response acceleration in the reinforced soil retaining wall is non-uniform, and the accelerations in the facing and the reinforced zone are larger than those in the retained zone; the response accelerations measured are larger than those calculated from the current design guidelines; although the location of the dynamic earth pressure at the back of the facing is higher than the design value, the resultant force is only 15%-20% of that calculated by the Mononobe-Okabe method; as a result of the effects of acceleration amplification along the facing and the higher location of the dynamic earth pressure, the predominant deformation mode of the FHR facing is rotation; the distribution of the reinforcement tensile force along the wall facing is nonlinear, and the measured force is larger than that calculated from the code in China.

Large hydro-fluctuation and periodical water level changes in the reservoir have caused the slip zone soil of the banks to experience drying-wetting cycles, which has a great influence on strength characteristics of soil. In this paper, a series of drying-wetting cycle tests on the slip zone soil from an ancient landslide is carried out. The soil samples after different runs of drying-wetting tests are then used to conduct the ring shear tests, scanning electron microscope (SEM) tests and nuclear magnetic resonance (NMR) tests. The results of these tests can be used to analyze the changes of soil microstructure and strength properties during drying-wetting cycles, and to investigate the microscopic mechanism of changes in strength properties. The research indicates that the deterioration of the residual strength of soil is obvious in drying-wetting cycles. The first three drying-wetting cycles result in a larger attenuation of soil strength, then the decay trend weakens and the soil strength gradually stabilizes. At the same time, the deterioration effect of cohesion is greater than the internal friction angle. Besides, the results also show that the clump-like aggregates are gradually disintegrated with the increase of drying-wetting cycles, and the way of connection of soil particles changes from plane-plane to plane-edge and plane-angle. Meanwhile, under the drying-wetting cycles, the number of pores in the soil increases, the morphology of the soil changes, the distance between the grains increases, and the small pores gradually change to the large pores in soil. The pore and cement microstructure changes of soil particles caused by clay mineral swelling and shrinkage are the causes of deterioration of the residual strength of slip soil under the drying-wetting cycles.

This paper is aimed to study the influence of aggregate size and inclination angle of joints on the mechanical properties of rock mass. Mortars with two aggregate sizes were used to prepared joint specimens with different inclination angles. Meanwhile, the loading process was monitored by the acoustic emission system. The experimental results showed that the peak shear strength of the specimens increased with the increase of the aggregate size. The peak shear stress of the coarse particle specimen increased firstly and then decreased. However, the peak shear stress of the fine particle specimen exhibited an increasing-decreasing-increasing-decreasing trend. The acoustic emission changed with the loading of the specimen, and reached the maximum value after the peak shear stress. Rock shear fracture criterion was established according to Mohr-Coulomb failure criterion. The stress intensity factor ratio at crack tip ( ) was found to be only related to the internal friction angle of the rock-like material, and the value of increased with the increase of the internal friction angle.

A mathematical model describing the gradation for coarse-grained soils under arbitrary stress state is proposed in this paper. The model is to express the grain size distribution of soil by a gradation equation firstly, on the one hand, the relationships between the two breakage indicators BW and Bg and gradation parameter b and m were derived. On the other hand, the empirical relationships between the two breakage indicators BW and Bg and stress or strain were suggested based on laboratory test data. Therefore, the relationship between the gradation, breakage indicators and stress or strain of soils were mathematically established. Meanwhile, the methods for the determination of parameters of the proposed model are discussed. Then, a series of consolidated drainage (CD) tests was conducted under confining pressures of 0.2, 0.5, 1.0, 1.5 and 2.0 MPa, respectively. The experimental data at confining pressure of 0.2, 0.5, 1.0 and 1.5 MPa are used to obtain the model parameters, which are substituted into the model to predict the particle breaking indicators and gradation distribution when the confining pressure is 2.0 MPa. The results show that the predictive value agrees well with the experimental data and proves the applicability of the model.

As known, natural geotechnical materials always have the characteristics of structure and anisotropy. How to express their effects rationally in the constitutive model is the key point for evaluating the advantages and disadvantages of the constitutive model. On the basis of the modified Cam-clay model, a structural parameter was used to express structural characteristic of natural original soil in plain. An anisotropic parameter was used by rotating yield surface to express anisotropy effect. The anisotropic parameter also be used in the process of deducting the hardening rule. A modified Cam-clay model considering structural and anisotropic properties was established. Combined with the associated flow rule and consistency condition, the incremental stress-strain relationship from the constitutive model was derived. The model consisted of 8 parameters which can all be obtained by the conventional laboratory tests. At the last, the triaxial shear tests was carried out using natural original loess, the comparison of experimental results and model calculation shows that the modified Cam-clay model considering soil structure and anisotropy could simulate the mechanical characteristic of natural original loess correctly.

Underground space supplement by excavation beneath constructed basement usually encounters the situation that existing and newly-created retaining piles form a double layered retaining structure. This is a new topic that traditional excavation design methods have not involved. This study uses large-scale model tests to investigate the influence of row spacing between the two pile layers on the pile displacement, the pile force and the earth pressure acting on the front-row piles. The analyses show that as the row spacing decreases, both the displacement and the bending moment decrease in the existing piles but increase in the supplementary piles, making the supplementary piles be the principal bearing elements. After the existing retaining piles end service, an abrupt increase in the displacement and bending moment of the supplementary piles is detected while the location of the maximum bending moment is shifted to be deeper. The unit earth pressure imposed on the existing piles gradually decreases as excavation proceeds. When the active limit state is reached, the active earth pressure approaches the distribution mode of the Rankine active earth pressure and its magnitude increases with increasing row spacing. Through the analysis and fitting of the active earth pressure data of each test, a prediction method for the active earth pressure on the existing piles is established with considerations of the spacing of pile rows, the soil friction angle, the pile-soil interface friction angle as well as the soil arching effect.

To study the influence of the embedding action of diaphragm wall below the bottom of diaphragm wall-gravity anchorage composite foundation on its bearing capacity, two sets of physical model tests of the underground diaphragm wall-anchor combined foundation and the conventional anchor foundation without considering the embedding effect of diaphragm wall-were carried out under the same conditions, relying on the design scheme of new diaphragm wall anchorage at Qingyuan side of Xijiang Super Bridge. The relationship between load and displacement, the distribution of soil pressure and the load distribution are analyzed. The results show that the vertical displacement and horizontal displacement of the combined anchor foundation are obviously smaller than that of the conventional anchor foundation due to the consolidation and fixed effect of the diaphragm wall under the vertical and horizontal loads. Compared with the conventional anchorage, diaphragm wall-gravity type composite anchorage foundation can make full use of diaphragm strength embedded into rock and bearing capacity of deep rock mass. The horizontal displacement can be reduced by about 75% under the design of cable force, which effectively improve the bearing and safety performance. The vertical load can transfer from the ground wall structure to the deep bedrock, which can share the soil pressure on the foundation base, and reduce additional stress and settlement, according to the soil pressure and load distribution law. The comprehensive analysis shows that the contribution for bearing capacity of ground diaphragm wall designed as the anchorage foundation of the retaining structure can be considered in design, when the bottom of underground diaphragm wall is embedded into bedrock.

To accurately obtain the shear strength of undrained clay by pressuremeter test (PMT), three types of evaluation method for shear strength for undrained clay were contrastively studied based on cavity expansion theory in this study. Among them, Gibson and Anderson's direct traditional method neglected the influence of over-consolidation ratio of soil and the actual length of side pressure probe; Cao's analytical method based on modified Cambridge model considered the influence of over-consolidation ratio of soil; and the finite element method based on modified Cambridge model considered the influence of both over-consolidation ratio(OCR) of soil and the actual length of the pressuremter probe. A series of pressuremeter tests was carried out on slightly overconsolidated and overconsolidated clays in two test sites. The above three methods were applied to study the influence of OCR of clay and the infinity of the pressuremeter probe on the undrained shear strength. The results show that the undrained shear strength of clay estimated from the direct traditional method is lower than that estimated from the analytical method when the effect of OCR is considered alone, and difference between them is increased with the increase of OCR. The undrained shear strength obtained from the analytical method is higher than that obtained from the FEM when the effect of the infinity of the pressuremeter probe is considered alone, and the effect of infinite length of pressuremeter probe decreases rapidly with the increase of OCR. When PMT is conducted on clay with a given pressuremeter length, there is a critical value of OCR. Under this critical value, the effect of infinite length of pressuremeter probe is greater than the OCR; above this critical value, the influences of the infinite length effect of the pressuremeter probe and the over-consolidation ratio cancel each other out. The critical OCR value is predicted to be about 23 for the silty clay in case 1; the critical OCR value is predicted to be 34 for the silty clay in case 2. In practical engineering, only finite element method can consider the effect of the actual length of the pressuremter probe and overconsolidation ratio on shear strength at the same time, which is the best method. When the clay is in a state of severe overconsolidated (the overconsolidation ratio is greater than the critical consolidation ratio), the results of direct method and the finite element method are similar. The direct traditional method can be used to calculate the shear strength of clay.

Pile foundation could be severely damaged by active fault during an earthquake, resulting damage and even collapse of superstructures. The pile failure mechanism and setback distance are not yet investigated systematically. Centrifuge and numerical modeling of pile group foundation in sand subjected to normal faulting were conducted. The influences of pile location relative to the bedrock fault on failure mechanism of pile group were investigated. Centrifuge and numerical results consistently show that when the pile group crosses the bedrock fault, the pile tilts to the hanging wall after faulting and is also bent to the hanging wall. Loading redistribution between piles results in two different failure patterns of tension and compression. Numerical parametric study of pile locations show that five characteristic zones could be classified based on different pile responses. For a soil stratum with thickness of 20.0 m, the setback distances of pile group at the side of footwall and hanging wall are 15.9 m and 23.5 m, respectively. The area within 7.9 m to the fault line at the footwall and 4.1 m to the fault line at the hanging wall is the most critical setback zone.

Undisturbed loess is a type of under-consolidated soils with structural strength, and its confined compressive deformation characteristic is distinct from those of ordinary soils. To study the general character of confined compression deformation of undisturbed loess, confined compression tests were carried out on Q3 and Q2 undisturbed loess with different water contents sampled from six fields in Xi'an and Lanzhou. The compression curves of undisturbed loess at different water contents from corresponding sites were obtained. The shape of the compression curve was analyzed. Through the compressive yield stress and the initial void ratio corresponding to the curve, the compression curves of tested loess at different water contents were normalized into a same curve, and the positions of the normalized curves of loess with the same sedimentary period were the same. The normalized curves of Q3 and Q2 loess were respectively described by composite power-exponent model. The initial void ratio and compressive yield stress of the loess in the literature were substituted into the expression of normalized curve to calculate the compression curves and compressibility indexes. It was found that the curves tested in literature was close to the compression compressibility indexes, demonstrating the normalized characteristic of loess compression curve and its application value. The test loess compression curve has the characteristics that can be normalized by its initial void ratio and compressive yield stress, which provides a convenient way for the analysis of confined deformation properties of loess.

The Daguangbao (DGB) landslide was the largest landslide triggered by the 2008 Wenchuan earthquake. Its shear-slip failure occurred in the pre-structural interlayer dislocation zone of the slope, which is mainly composed of mylonitized and brecciated tectonic fragments, and has obvious characteristics of groundwater occurrence and movement. In order to study the dynamic behavior of the dislocation zone and its possible influence on the start-up of Daguangbao landslide under the participation of groundwater during strong earthquakes, a series of triaxial tests including hydrostatic triaxial, unidirectional and bi-directional dynamic triaxial tests was conducted in the laboratory on the materials sampled from dislocation zone to analyze the liquefaction characteristics of the materials. The experimental results show that the material has strong potential liquefaction capacity; under unidirectional and bi-directional dynamic loads, the material can liquefy, but the liquefaction rate of bi-directional vibration is faster, and the dynamic strength decreases by about 20% than that of unidirectional vibration. Under bi-directional vibration, the liquefaction rate of material increases with the increase of cyclic deviating stress and radial cyclic stress, and the liquefaction rate under the phase difference 180ois faster than that under the phase difference 0o. The dynamic condition of bi-directional triaxial test may be closer to the actual stress state for the interlaminar staggered zone of Daguangbao landslide with inclined occurrence (N2oW/NE/30o); liquefaction of the staggered zone may be the cause of the sudden start of Daguangbao landslide during the strong earthquake.

To further investigate the overburden strata movement and deformation law of coal backfill mining, a concept of “continuous curved beam” of coal backfill mining was defined based on the overburden strata movement and deformation characteristics of coal backfill mining. On the basis of analyzing relationship between overburden strata deformation and curvature, a model was established for overburden strata curvature of coal backfill mining. The curvature was proposed to evaluate the deformation characteristics of rock strata. A formula of calculating the curvature K of overburden strata was deduced to reveal the relationship between compression ratio and overburden strata curvature. Compression ratio, thickness and key strata were found to be key factors influencing overburden strata curvature according to the curvature model, meanwhile, the support length also had some effect on overburden strata curvature. Thus, the curvature theoretical model was adopted to evaluate the effect of backfill mining. According to curvature variation features of overburden strata at different layer heights, the limit curvature was proposed as the criterion to judge the fracture location of strata under different compression ratios. The critical compression ratios of immediate roof and main roof without breaking were calculated, and were basically consistent with borehole measurement.

This study investigated the required strength to fill the cemented backfill by the two-step stage. The mechanical properties of cemented backfill in the goaf were analyzed. It was found that the cemented backfill mainly improved the stress state of the roof rock mass, which provided the support for the broken rock mass of the roof and the lateral pressure for the surrounding rock of the stope, resisted the closure of the stope and limited the flow of tailings. Therefore, a compressive strength model was established under the dangerous mechanical environment for cemented backfill (one side is exposed and the other side bears the pressure of tailings). This model was based on the collapse caused by shear failure of cemented backfill, and fully considered the factors affecting its stability. In this model, the wedge sliding theory was also used to analyze the limit equilibrium condition of the 3D wedge-shaped structure with a slip trend on the sliding surface of cemented backfill. The reasonability of the strength model has been verified by the vertical stress measurements and the stability of the 51 exploratory line cemented filling body of 48-54 panel of Dahongshan copper mine. The results provide significant references to the theoretical study of filling mining in-depth and the corresponding application in similar mines.

It is vital to foreknow the overall settlement of bridge foundation for high speed railway in order to keep control of the bridge line. However, it is questionable to calculate the overall settlement of the large caisson foundation at a great depth by using layer-wise summation method recommended in the specification. In this paper, a deep and large caisson foundation under main tower of a long span cable-stayed bridge is taken into account. The overall settlement of caisson foundation is calculated by finite element method (FEM) with Mohr-Coulomb model (M-C), hardening-soil model (HS) and small-strain hardening soil model (HSS), respectively. Compared with the field monitoring data, the numerical results show that the overall settlements with HSS model are in good agreement with the monitoring data. Furthermore, the settlements in the following construction stages are rationally predicted by FEM. For layer-wise summation method recommended in the specification, the empirical settlement coefficient is corrected based on the monitoring data, and it is found that when the empirical settlement coefficient is equal to 0.13, the settlements by summation method are consistent with the monitoring data in this case.

Main concerns in rock engineering are the influences of in-situ stress on the quality of contour surface and its corresponding distribution area after rock blasting and excavation. This paper aims at studying the reason for the occurrence of hanging feet at the corner of right bank abutment groove of Baihetan hydropower station after pre-split blasting excavation. Theoretical analysis is firstly applied to explain the influence of in-situ stress on crack propagation during the stage of explosion stress wave and detonation gas respectively, and then numerical simulation is used for verification. Results show that in-situ stress guides the direction of initial crack propagation during the stress wave stage, while the in-situ stress perpendicular to fracture face inhibits the crack length expansion during the detonation gas stage. Moreover, the higher the in-situ stress, the more significant effect on the initial crack propagation direction and final length. Since the connection direction between blast-holes is inconsistency and in-situ stress is high at the corner of the abutment groove, crack extension deviates the expected direction and is difficult to penetrate at last. Finally, the hanging feet is formed. It can be seen that the in-situ stress seriously affects the quality of the pre-splitting blasting seam, which further causes the phenomenon of blasting and hanging feet at the corner of the abutment groove.

Because of the widely adopted laying technologies, such as J-laying method or S-laying method in deep sea, untrenched pipelines often stay in shallow-embedded state. In this case, lateral displacements may occur due to loading of waves, currents or thermal buckling in practice, while seabed is in limit state subject to the combined load of pipeline self weight and horizontal expansion force. Therefore, in order to accurately evaluate the stability of shallow-buried pipelines under this combined loading condition, the envelopes of bearing capacity of shallow-buried pipelines in vertical-horizontal(V-H) load space and the main influencing factors in sand seabed are studied by finite element numerical analysis method. A series of model tests in laboratory is conducted to verify the reliability of numerical results. The results show that the failure envelopes expand significantly but their shapes remain unchanged with the increase of internal friction angle φ and relative embedment depth z/D. Based on this, fitting formulas are established to describe the changing laws of parameters Vmax and μ0 depend on φ and z/D correspondingly. In practice engineering, the parameters of seabed soil and the possible load combination can be incorporated into the envelope analysis to evaluate the on-bottom stability of pipelines.

The geotechnical engineering numerical analysis method is a powerful method for researching and analyzing complex geotechnical media and variable construction measures. It has irreplaceable advantages over traditional analytical methods, indoor test methods, simulation test methods and field test methods. According to the historical development, function and development trend of geotechnical engineering numerical methods, the numerical analysis can be divided into four levels: a high-level calculator, which directly contributes to the design of a specific geotechnical engineering, and calculates and analyzes the safety and stability of the geotechnical engineering project under various extreme design conditions; a powerful and unparalleled analog analyzer, which can simulate and analyze the interactions and coupling effects of various unfavorable factors in complex geotechnical engineering projects under complex working conditions; a cost-free and repeatable multi-functional testing machine, which can explore the stability mechanism of a specific geotechnical engineering project or the reinforcement mechanism of a certain engineering measure; the ultimate goal of the numerical analysis is to re-develop the numerical analysis method——to develop a new numerical analysis tool which is intelligent, fast and simple. This article discusses these four levels and gives examples of each level.

To obtain the boundary conditions for coupling thermo-mechanical simulation of underground caverns for compressed gas storage, based on the mass and energy conservation equations of compressed air, a calculation method for the solution of thermodynamic process analysis of compressed air and determination of boundary conditions of coupling thermo-mechanical simulation was proposed in software FLAC3D. The calculation method for thermodynamic process analysis of compressed gas was also validated by an example cited in the paper. The variation characteristics of boundary conditions acting on the cavern wall under the situation of charging and discharging were explored with numerical simulation method. The influences of the initial temperature, the convective heat transfer coefficient and the thermal conductivity of the surrounding rock on the boundary conditions of the cavern wall were also analyzed. The research results show that the boundary of mechanics and heat transfer analysis of underground gas storage cavern has a remarkable dynamic change during operation, which is closely related to the state of compressed air, the heat exchange characteristics of cavern wall surface and the heat conduction characteristics of surrounding rock.

The physical properties in actual developed reservoirs change under the stimulation of low-frequency vibration. However, no change in porosity and pressure is usually assumed in classic consolidation model. The analysis of solid deformation is often based on a one-dimensional physical model with constant pressure (or pressure gradient) condition. Therefore, it is usually inadequate to simulate the effect of seismic production technology near the wellbore in actual developing reservoirs, which are with varying flow velocities and pressure gradients, with the classic mathematic model and one-dimensional physical model. The control equations of consolidation model for low permeability porous media are re-derived from the continuity equations of fluid and solid. Considering different assumptions of variation extents of porosity and pressure, three consolidation models are given. Numerical simulation is then carried out with one-dimensional (with constant pressure gradient) or radial (with changing pressure gradient) physical model. The effect of seismic production technology as well as its sensitivity under different fluid and vibration parameters are evaluated with different consolidation models. Because of the influences of the inertia effect of initial flow and the strong stress sensitivity in low permeability reservoir, the increases of pressure, flow velocity, and porosity are found to be lower under vibration than the case without vibration in one-dimensional model. However, the wave-induced effect behaves differently in radial physical model. The increases of pressure and porosity are both higher under vibration than the case without vibration, and the increase of flow velocity becomes lower under vibration. As the vibration parameter increases, the volatility of values representing the wave-induced effect becomes stronger when simulated with radial physical model and different consolidation models. The results reflect that it is necessary to carry out a dynamic analysis on the complex effects of artificial seismic technology in actual developing reservoirs.

It is of great important significance to study the mechanism of initiation, propagation and coalescence in engineering rock mass and the subsequent sliding mechanism along the weak structural plane for revealing the law of deformation and failure of rock mass. Discontinuous deformation analysis (DDA) method owns an inherent advantage in simulating the sliding deformation of discrete block system, which is formed by the intersections of various macroscopic structural planes. But it is slightly inadequate in the simulation of the evolution process from continuum to discontinuum. DDA has strengthened its ability in the simulation of continuous characteristics of rock mass to a certain extent by introducing virtual joint technology to discretize the continuous area into sub blocks and then setting strength of virtual joint as that of the rock itself. However, this treatment only considers the bonding effect of virtual joints before reaching the tensile strength and ignores the strength of the strain softening stage in the complete stress-strain curve of rock. Therefore, the above deficiencies have been improved by inserting a so-called strain softening cohesive element between the sub blocks which can describe the strain softening stage in the stress-strain curve of rock. This further strengthens the ability of DDA in the simulation of continuous attributes of materials and also retains its inherent advantages in the simulation of discontinuous deformation. Finally, it is applied to solve several typical crack problems. The results show that the simulated fracture paths are in good accordance with the reference solutions, proving the validity and correctness of the improved DDA method.

Combing with the PFC3D software, the conventional triaxial loading and unloading tests were carried out on marble specimens in the laboratory. Then the results of unloading rate were compared between the numerical simulation and lab test. According to the crack propagation in mesoscale, the mechanical properties and failure mechanism of specimens under different unloading rates were discussed. The results show that the number of tensile cracks on specimens was obviously higher than that of shear cracks at the loading stage in the loading and unloading process. The crack propagated at a certain rate after the conventional triaxial test damage stress, while the crack developed suddenly after the unloading test damage stress. Under different stress paths, the difference of damage failure between the damage stress and the peak stress was mainly formed in the stage from damage stress to peak stress. The damage degree and failure form of the specimen tended to be consistent until the unloading rate beyond 6 MPa/s. With the increase of the confining pressure, rock failure showed a transition from tensile failure to shear failure at different unloading rates.

A new contact detection algorithm is proposed for the system of greatly differing particle sizes. The time complexity is linear with the particle number and the memory consumption is low. The algorithm has significant advantages to tackle the problems which have a large number of particles with arbitrary sizes and high density. Improved based on the NBS (no binary search) algorithm, the algorithm is more applicable to the polydisperse system with a multistep strategy than the NBS algorithm. With the particles divided into several groups and the detection progress divided into different steps, the particles in the same group can be considered as uniform and the size of grid can be set separately for different steps. As a result, the accuracy of the neighbour searching is increased. The number of the geometry resolution is decreased by setting the primary checking of the bounding boxes when the number of potential contact particles is more than 4. The number of groups is determined by a parameter previously chosen and is easily adjusted with different size gradations, so the algorithm has applicability to a wide range of particle systems. The properties of the algorithm are tested in several examples with different size gradation and the results show clear advantages over NBS algorithm.

Collapse of granular soil often induces natural disasters such as debris flow and landslide. Existing studies lack the influence of particle shape on the collapse morphology of granular soils. In this paper, three-dimensional discrete element method is used to simulate the collapse failure mechanism of granular material cylinder specimens. Three typical particle shapes (spherical, tetrahedral and elongated) are considered in the numerical simulations. For each particle shape, cylinder specimens with the same size and the same grain size distribution are generated using the same method, and then collapse tests are performed. Based on experimental results, the final collapse height and runout distance of cylindrical specimens with different shapes of particles are analyzed and compared with the laboratory test. It can be concluded that the discrete element method can reproduce the collapse process of the granular columns very well. Compared with specimen of pure spherical particles, specimens composed of irregular shape particles can reduce the angular velocity of particles, maintain a larger final collapse height and reduce the final runout distance.

Due to the heterogeneity, opacity and porosity of rock, it is difficult to truly represent three-dimensional internal meso-structure of rock through physical experiments and numerical simulations. In this study, a 3D non-uniform numerical method was proposed to simulate the porous rock failure based on the parallel finite element method. The CT technology was used to scan the rock specimens and obtain the picture layers. The edge detection algorithm, filtering algorithm and 3D matrix mapping method were applied to process the scanned pictures and reconstruct the Brazilian disk finite element model of the rock. Thus, 3D numerical Brazilian disk models were constructed, and the influence of the porosity and pore distribution were analyzed numerically. The results indicated that the pore distribution and porosity had significant effects on the crack propagation and tensile strength. The number of the secondary cracks decreased when the porosity was smaller and pores were sparse, while the number of the secondary cracks increased when the porosity increased. The different porosities of numerical specimens were calculated by using the digital image processing method, and the relationship between the tensile strength and porosity followed an exponential function. The porosity of the specimen was obtained by the digital image processing technology, and the law of porosity and tensile strength was exponentially distributed. The median filtering algorithm improved the accuracy of the digital model in the image processing, and the 3D numerical simulation model reflected the 3D meso-structure of rock-like porous materials, which provided an effective method for studying the mechanical behavior of rock.

In order to study the long-term THM coupled behaviour of soft rock, a parallel-linkage triaxial testing machine for THM coupling in soft rock is developed. This machine is particularly applicable to study the mechanical properties of soft rock under different temperatures and various stress conditions. Using the technology of servo-control, this machine can automatically stabilize voltage and record experimental data. It can also achieve temperature gradient loading, pore pressure gradient loading and independent stress gradient loading. Furthermore, high-precision LVDT sensor is also implemented to measure the deformation. This machine can complete these tests including uniaxial compression test, triaxial compression test, triaxial rheological test, cyclic loading and unloading test, penetration test under steady state and transient conditions. It can simultaneously implement the mechanical tests of two samples with the same confining pressure, the same pore water pressure, different axial pressures and different temperatures, thanks to the unique design of this machine. It can save time because of this feature, especially for the long-term mechanical test. In addition, it can also reduce the system errors. The results of these tests during last year show that this machine is a convenient machine with complete functions, good stability and high precision.

Uniaxial compressive strength (UCS) is an important index for classification of the surrounding rock and determination of the supporting parameters in underground engineering. The commonly used standard UCS tests are expensive, time-consuming and difficult to quantitatively evaluate the UCS of fragmented rocks because these rocks cannot be effectively cored. To solve the above problems, a method for predicting UCS of rock mass based on digital drilling test technology is introduced in this paper. The key to implementing this method is to establish a quantitative and universal relationship between the drilling parameters and the UCS. Therefore, the digital drilling tests and standard UCS tests of rock specimens with different strength values are carried out based on the laboratory rock mass digital drilling test system developed by the authors. A relational model between the drilling parameters and the UCS is established by support vector machine (SVM). The drilling parameters as model input parameters include the rotation speed N, drilling rate V, torque M and thrust F measured by the sensors, and the unit cutting energy ηc deduced from energy analysis by the authors as well. The research shows that the UCS predicted by the relational model in validation set is close to the UCS measured by the uniaxial compression test, the coefficient of determination R2 is 0.977, and the mean absolute error (MAE) is 3.037 MPa. These results indicate that the relational model between the drilling parameters and the UCS based on SVM method is successful in the rock UCS prediction, and the digital drilling test technology could realize the effective prediction of UCS of rock mass.

The velocity model is the major factor affecting the seismic source localization algorithm which is the core of microseismic monitoring technique. Based on the complex velocity model, the fast sweeping method (FSM) is an algorithm to calculate the first arrival time of the seismic wave by using the eikonal equation. This method has been widely applied in the fields of earthquake localization and geophysical exploration. In this study, this method was introduced in the field of microseismic monitoring in geotechnical engineering. Especially for layered velocity models, a fast localization technique was established based on the fast sweeping method (FSM). Firstly, the fast sweeping method (FSM) in a 3D Cartesian coordinate system was proposed to calculate the initial travel time of the point source in the single velocity model and horizontal layered velocity model. Compared with the theoretical solution, the accuracy of the algorithm and its error distribution characteristics were analyzed. The travel time results calculated by the fast sweeping method (FSM) and theoretical solutions were compared to analyze the algorithm precision and error distribution characteristics. Secondly, a rapid seismic source localization technique was put forward for the layered velocity model based on the database of the arrival time differences which were calculated by fast sweeping method (FSM). Finally, the positioning accuracy and computational efficiency between the new localization technique and traditional method based on the isotropy velocity model were compared and analyzed. This study showed that the proposed technique for the layered velocity model presented a favorable accuracy and computational efficiency, and also reduced the location time significantly compared with traditional location methods. The technique can provide significant theoretical and technical support for the microseismic source localization in complex layer and acoustic emission monitoring in the laboratory.