Coral sand is the only material for island land reclamation. Due to its special marine biogenesis and porosity structure, coral sand particles are easily crushed at normal level of engineering stress. In this study, drained triaxial shear tests were carried out on coral sand retrieved from a reclamation reef in the South China Sea to study the evolution law of the strength, deformation and particle breakage properties of coral sand with different levels of density and confining pressure. The comparative analysis of shear strength index is made between current study and previous investigations. The results show that strain softening and dilatation tendency of coral sand gradually weaken with an increase in confining pressure and decrease in compactness. Within the normal confining pressure range, the values of the peak and critical state frictional angles of coral sand were 33o?58o and 28o?47o, respectively, both of which decreased with an increase in the confining pressure. The relationship among the secant modulus E50, relative density and effective confining pressure for coral sand was established. The relationship between the peak friction angle of coral sand and the modified relative breakage index (Br*) can be fitted by a power function equation with a negative index. When the extent of particle breakage is large, the decreasing trend of peak friction angle with an increase in the modified relative breakage index slows down. The correlation between the modified relative breakage index and plastic work for coral sand could be simulated using hyperbolic curve, and it is hardly affected by density. The research results can provide useful reference and technical support for island reclamation and infrastructure construction in the South China Sea.

Weak expansive soil after treatment with waterproof and moisturizing can be used as the highway subgrade filling, and studies on its dynamic response characteristics are few. For a thorough analysis of the dynamic properties of weak expansive soil in Hefei, a series of cyclic load triaxial tests was conducted. Considering the influence of confining pressure, consolidation ratio and dynamic stress amplitude on the weak expansive soil dynamic property, the tests obtained the development curves of the accumulated deformation and dynamic strength under different test conditions. Based on the accumulated deformation of samples under different confining pressures, consolidation ratios and dynamic stress amplitudes, a prediction model for accumulated deformation of weak expansive soils was established. The accumulated strain curves of expansive soils can be divided into plastically stable, critical and destructive types according to the strain rate, and the criteria for judging the cumulative strain of weak expansive soil was given. The dynamic strength of weak expansive soil under different confining pressures and consolidation ratios was analyzed. It was found that the dynamic strength of the soil decreased a lot when the consolidation ratio was over than 1.5; the dynamic cohesion cd decreased with the increase of failure vibration, and the dynamic internal friction angle ?d decreased slightly with the increase of failure vibration. Results obtained can provide basic data and reference for relevant projects.

A series of indoor model tests was conducted to study the sand deformation characteristics around uplift circular plate anchors using digital image correlation technology, with emphasis on the effects of plate diameter, embedment ratio and relative density of sand. The experimental results show that when a same embedment ratio is reached, the peak uplift force and the displacement for the appearance of peak uplift force both increase with increasing plate diameter, yet the uplift bearing capacity coefficient decreases as plate diameter increases. The shape of influence zone concerning soil deformation is not affected by the variation of plate diameter, and this regularity is independent of soil density. For the dense sand case, as embedment ratio increases, the shape of influence zone evolves from a reversed trapezoid to a U shape, and a linear shear failure surface can be captured with the inclination angle of 1/4?p to the vertical, where ?p is the peak friction angle of soil. The surrounded soil exhibits a significant volume expansion due to the uplift of plate anchor in dense sand. For the loose sand case, as embedment ratio increases, the shape of influence zone gradually develops from a rectangle extended into soil surface, to a shell shape embedded within the soil. As to a shallow anchor in loose sand, the shear failure surface of soil perpendicularly progresses from the edge of anchor plate to soil surface. When a deep anchor in loose sand is discussed, an inside oblique shear failure surface can be observed and the inclination angle to the horizontal is about 45°+1/2?p. Irrespective of the embedment ratio, due to the uplift of plate anchor, a small volume expansion zone is observed above the anchor plate, upon which is a large volume contraction zone with the value of volume contraction increasing with the increase of anchor’s embedment ratio.

In order to accurately characterize the deformation and failure modes of prefabricated jointed rocks with different angles under uniaxial compression, a joint model based on 3D printing technology was used to simulate the structural surface in the rock mass. Rock specimens with precast joints with different angles were obtained by pouring cement mortar, and a uniaxial compression test was performed. At the same time, digital image correlation (DIC) technology was used to observe and analyze the process of crack formation, propagation, and penetration in the test specimen. The results showed that as the angle of prefabricated joints increased from 0° to 90°, a decrease followed by an increase in the strength and peak strain of the test piece was observed. Additionally, the elastic modulus of the test piece with angles of 0° and 45° decreased compared to the complete test piece. Based on the DIC test results, the cracks of specimens with angles of 0°, 30°, 45°, and 60° all started from the tip of the prefabricated joint. The crack initiation stress of the specimens with different angles was all consistent with the strength change of the specimens. Under shear stress, the cracks started in the form of shear wing cracks. The cracks of 0° and 45° specimens changed from shear to tensile cracks during the propagation process, and the shear cracks were observed in the specimens of 30° and 60° throughout this process. The crack of the specimen of 90° started from the bottom and tensile failure was eventually witnessed. In this study, an obvious linear positive correlation was found between the cracking angle θ2 of the lower wing and the cracking angle θ1 of the upper wing, and it can be expressed as θ2= 0.828 6θ1 +12.185. As the joint angle increased, the cracking stress decreased first and then increased, which was consistent with the peak stress.

With the accelerated degradation of permafrost caused by global warming, the problem of slope stability of soil-rock mixture in permafrost regions has become increasingly prominent. To investigate the effects of different water contents and rock contents on the shear strength of soil-rock mixture at the freezing-thawing interface, direct shear tests were carried out on the freezing-thawing interface of soil-rock mixture, and the effects of water content (21%, 24%, 27%, 30%) and rock content (0, 10%, 20%, 30 %, 35%, 40%) on the shear strength of the interface were obtained. The test results show that the effect of water content on the shear strength of the freezing-thawing interface of soil-rock mixture can be divided into two stages: rapid decrease stage and slow decrease stage. The shear strength of the interface decreased rapidly as the water content increased from 21% to 27%, but the decreasing rate slowed down when the water content continued increasing to 30%, suggesting that the threshold value of the water content affecting the shear strength of freezing-thawing interface can be considered to be around 27%. The shear strength of freezing-thawing interface increased with rock content. The shear strength of freezing-thawing interface with the rock content of 10% significantly increased compared to that without rock, with the maximum increase of 33%. When the rock content exceeded 30%, the shear strength would increase rapidly, suggesting that the threshold value of the rock content affecting the shear strength of freezing-thawing interface can be considered to be around 30%. When the rock content was constant, the friction angle of interface gradually decreased with the increase of the water content, and the change tended to be stable after the water content reached 27%. However, the cohesive force at the interface decreased rapidly before the water content increased to 27%, and then decreased slowly. When the water content was constant, the friction angle at the interface increased with the increase of rock content especially when the rock content exceeded 30%. The cohesive force decreased first and then increased slowly with rock content ranging in 0%?30%, but the cohesive force increased rapidly and tended to be flat when the rock content exceeded 30%. When the rock content increased from 0% to 30%, the cohesion force first decreased slightly and then increased gently. After the rock content exceeded 30%, the cohesion force increased rapidly and then leveled off.

The void ratio of soil has an important influence on the soil-water characteristic curve (SWCC) of unsaturated soil. Experimental studies have shown that pores expand or contract under different hydraulic load paths, which results in the hysteresis of the soil-water characteristic curve (SWCC). Based on this finding, this paper assumes that the expansion and contraction of pores can cause the hysteresis of the soil-water characteristic curve (SWCC). In this study, axial translation technique is employed as an example to explain the expansion and contraction of soil pores under hydraulic loading. Then, an incremental equation for the soil-water characteristic curve that can describe the hysteresis is derived. In this equation, expansion and contraction of pores are calculated by assuming ?d to be a constant and combining the redefined equivalent pore radius with the Fredlund-Xing equation. The proposed soil-water characteristic curve model can be used to predict other scanning curves by simply using the model parameters obtained from the main drying curve and an arbitrary scanning curve. Finally, the applicability of the proposed model to different soil types is verified using five data sets obtained from tests. It is also shown that the proposed model is capable of predicting high-order scanning curves.

To solve the safety problem of using waste slurry to backfill discarded open pit mine, the treatment method of vacuum preloading at bottom combined with upper surcharge loading is implemented in the model test. The feasibility of using this method in the backfilling of discarded open pit mine is discussed. The test results show that the proposed method can significantly reduce the water content of the slurry and increase the shear strength of the soil. The water content of the slurry reduces from 450% to 95%–105%, and the reduction of volume reaches 73.4%. The undrained shear strength increases from zero to 9.8–13.4 kPa. In the initial self-weight consolidation stage, there is gravitational separation in the deposition of particles, and the deposition of coarse particles at the bottom is helpful to alleviate the clogging problem of vacuum preloading. During the stage of vacuum preloading, seepage direction of pore water in the slurry is not completely one-dimensional downward, and there is a hydraulic gradient in the radial direction. After the treatment, the compressibility of the soil is close to that of soft clay, while the permeability is greater than that of soft clay. Based on the experimental results and the large-strain consolidation theory, the effects of slurry thickness of a single disposal cycle on volume reduction and treatment time are analyzed using the finite difference method with consideration of the self-weight, nonlinear change of compressibility and permeability of the slurry. The process parameters for operation in site are recommended based on the results of numerical analysis.

The study of propagation law of microseismic signal in layered rock mass is of great practical significance for realizing accurate source location. In natural rock mass, the propagation velocity of microseismic signal in rock strata is not only affected by its internal factors, namely the physical properties of rock itself, but also by external factors, such as geological structural plane, fault and goaf. Through laboratory experiments, the propagation law of microseismic signals in different strata and near mined-out areas was studied. The results show that the wave velocity decreases as the propagation distance increases, and the attenuation of the wave velocity is slower with higher rock density. The attenuation ratio of propagation velocity is bigger when microseismic signal goes through more structural planes. The attenuation of energy is consistent with that of the wave velocity. It takes longer for microseismic signals to pass through a goaf with a larger cross-sectional area, whereas, the larger the values of the physical properties such as the density and elastic modulus of the rock around the goaf, the less the wave velocity attenuation when the microseismic signal passes. Marginal spectrum analysis results show that faults have a stronger obstruction to micro-seismic signals with higher frequency.

The theoretical and technical principles of self-expanding bolt with large amounts of expansion agents were first proposed, and its economic benefits and the effect of improving the ultimate pullout resistance were addressed. However, excessive expansion agent would generate cracks in the surrounding rock caused self-expanding bolt, and it is highly important to determine the limit content of the expansion agent. Therefore, two prerequisites for the application of self-expanding bolt were proposed, and experimental study were performed based on the clarification of experimental purposes from two levels. The former one is expansion cracking method. Firstly, the bursting and cracking test of surrounding rock was carried out to obtain the change law of normal stress at the interface and the failure law of surrounding rock. The evolution law of the cracking and failure of the moderately weathered argillaceous siltstone under the action of self-expanding anchor body was also explained. A calculation models of radial self-expansion stress at different positions were established. Based on the radial expansion stress of the contact surface, a prediction model of the critical content of the expansion agent of the anchor body was obtained, and the critical content of the expansion agent of the moderately weathered argillaceous siltstone was predicted to be 28.98%. The latter one is elastic analysis method. Based on the problem of uniform compression of circular holes in the infinite medium and stress concentration at the edges of the holes, a model for the expansion and failure of surrounding rocks under the action of self-expanding anchors body was established. A set of formulas for predicting the stress distribution of surrounding rock under the action of self-expanding anchors body was obtained, and a critical cracking equation for surrounding rock under radial expansion stress was derived. In order to reduce the field test cycle and cost, a formula for predicting the tensile strength of the moderately weathered argillaceous siltstone was established based on the acoustic-rebound value. Finally, the parameter correction of the critical cracking equation is completed based on the incompleteness of the surrounding rock and the softening effect of the surrounding rock soaking. Combined with the critical admixture value, the safety reserve coefficient was introduced, and the design formula for the limit content of self-expanding anchor body was established. According to the design formula, it is calculated that the limit content of the self-expanding anchor solid expansion agent for moderately weathered muddy siltstone is 26.5%. The paper can provide technical support for the application of self-expanding bolt in geotechnical, mining and water conservancy fields.

Many geotechnical problems, such as development and utilization of geothermal resources and nuclear waste disposal, need to consider the effect of temperature on the stress and strain properties of saturated and unsaturated soils. To consider the effects of temperature and suction on the preconsolidation stress change for soils, expressions of preconsolidation stress for soils at different temperatures are first introduced. After that, combining the expression with the yield surface of unified state parameter model for clay and sand, considering the influence of temperature and suction on the normal consolidation line and critical state line, a unified thermo elastoplastic constitutive model describing the temperature effect of saturated and unsaturated soils is proposed. The proposed constitutive model is divided into saturated part (suction less than the air entry value or bubbling pressure at certain temperature) and unsaturated part (suction more than the air entry value or bubbling pressure at certain temperature). The comparisons between the model predictions and experimental data indicate that the proposed model can quantitatively predict the stress and strain properties of saturated and unsaturated soils at different temperatures.

The stress concentration in diaphragm wall support system, combined with an intolerable difference between deformation of diaphragm wall occurred during the construction process and predicted by design probably result in large construction risk. The paper solved these problems by model tests in which an adjustable device was set on the support system. The compatible deformation among the support system, the diaphragm wall and the ground soil behind the diaphragm wall were investigated. The variation of the support axial force, the lateral earth pressure on the diaphragm wall, the ground surface settlement around the foundation pit and the maximum deformation of the diaphragm wall with the adjustment mode of supports were obtained. A smart regulation method for compatible deformation of the diaphragm wall support system was proposed. The results show that under a constant displacement condition, the variation of support axial force caused by the adjustment of the upper support is the largest, by the adjustment of the bottom support is secondary, and by the simultaneous adjustment of the four supports is the smallest. However, the influence range of the ground surface displacement increases with the depth of the adjusted support, but the amplitude decreases. The influence range and amplitude of the simultaneous adjustment of the four supports are larger than that of the adjustment of a single support. The lateral displacement of the diaphragm wall can be reduced by the extension of the support, but the axial force of the support will increase sharply. It can be concluded that it is not wise to strictly control the horizontal displacement of foundation pit. Instead, a reasonable adjusted displacement of support and the monitoring of the support axial force and displacement in succession are recommend.

The loess samples solidified by reactive magnesium oxide (MgO) and microbes are analyzed in this paper. Water content and unconfined compressive strength are measured, and X-ray diffraction (XRD) and scanning electron microscopy (SEM) are conducted on different samples to investigate the curing product and microstructure change with varying amount of reactive MgO, curing time duration, initial water content. The results show that the water content of solidified sample decreases with the increasing content of reactive MgO and the growing curing period. The unconfined compression strength increases with the increasing content of reactive MgO, and also increases with the growing curing period in general, but decreases slightly in the later period with the content of reactive MgO being 10% or 15%. With the increase of initial water content, the unconfined compressive strength of solidified samples decreases when the reactive MgO content is 5% or 10%. But it will first increase and then decrease while the amount of reactive MgO is 15% or 20%. The results of XRD and SEM tests show that the higher the content of reactive MgO, the more magnesium hydroxide remains. The hydrated magnesium carbonate produced by the reaction is swellable and cementitious, which can fill the gaps between the soil particles and cement the soil particles together.

In order to study the bearing characteristic of static drill rooted energy piles under thermo-mechanical coupling conditions, a self-designed model test system has been used to evaluate the thermal response of the model pile in the clay-sand double-layered foundation. The results show that the temperature of the pile is non-uniformly distributed along the depth, and the radius of thermally affected region around the pile is 3 times of its diameter. Accumulated settlements are observed on the pile top and the surrounding soil surface after 3 heating-cooling cycles. Driven under the load of 25% of the ultimate bearing capacity on top, the cumulative settlement of single pile reached 3.12‰ of the pile diameter, which is 1.77 times that of the pile top under non loading condition. Larger additional thermal stress performs at the pile middle than both ends, and the null point (where lies maximum additional thermal stress) moves upward with the increase of the pile top load. In fact, it has been observed that under small load on the top, local tensile stress occurs in the pile body due to cooling. Therefore, the bearing characteristic of static drill rooted energy pile is of high relevance to temperature change, load on pile top and stress boundaries of pile ends, and that requires comprehensively consideration in engineering design to guarantee safety service of pile.

To overcome the drawback that the ultimate pullout resistance of prestressed anchor cable cannot be directly obtained by the pullout test, this study proposes a pullout resistance performance evaluation method of pre-stressed anchor cable based on the pullout tests on prestressed anchor cable with pre-grounting in the free segment. Based on the interfacial debonding shear model and the theories of load transfer and elastic deformation, the length ratio of the anchor segment to the free segment was considered in the evaluation method. The results indicate that pre-grouting in free segment can significantly improve the pullout resistance capability of anchor cable. When the interfacial deformation behavior transfers from an elastic manner to debonding stage, the improvement of the pullout resistance will be more pronounced, which will tend to be stable after debonding. The shear displacement and axial force at the interface of the free segment and grouting segment is the highest. With increasing the pullout force, both the shear displacement and axial force increase but the increment declines to the two ends of anchor cable. The shear stress curve is one peak type in the elastic stage and then becomes a saddle type in the debonding stage. The opening of the saddle type curve becomes larger with the improvement of the pullout resistance. Finally, the length ratio of anchoring and free segment has insignificant effects to the ultimate pullout resistance. The reduction coefficient ?Q of the ultimate pullout resistance of pre-grouting anchor cable is in the range from 0.81 to 0.83. The empirical method in paper is feasible due to the fact that merely 6.02% error is existed between the measured ultimate pullout resistance and calculation result when Lb /La is 2.5.

The offset of load position greatly influences the ultimate bearing capacity of the underlying rock foundation with cavities. Based on double logarithmic spiral curve model, the limit analysis method is adopted to establish the underlying rock foundation with cavities damage body function equation when the load position offset is small. Then, the calculation formula of ultimate bearing capacity of hollow rock foundation is deduced. When the offset of load position is large, Prandtl failure mode is developed in the rock foundation with cavities. Furthermore, the influence of the thickness h of rock foundation, load position offset e and internal friction angle ? on the ultimate bearing capacity of rock foundation with cavities are analyzed. Finally, the model test of the ultimate bearing capacity of the rock foundation with cavities under different load position offset and thickness were carried out, and the results were compared and verified with the theory. The results show that when the offset of position is constant, the ultimate bearing capacity of rock foundation increases linearly and the foundation tends to be complete rock foundation with the increase of h. The ultimate bearing capacity of rock foundation with cavities increases non-linearly with the increase of e when h is a certain value. When the h reaches a certain value, the ultimate bearing capacity is close to the value of complete bedrock. When h and e are constant, the limit bearing capacity of the rock foundation with void increases gradually with the increase of the angle of internal friction ?.

Geothermal energy is the development trend of green building. However, it is not clear about the relationship between the problems of building foundation impairment and the geothermal system in red-layer soft rock areas. Based on an accident investigation project of building clusters in Chengdu, the engineering characteristics of mudstone and gypsum rock with or without geothermal system were studied, and the macro-mechanical and micro-structural characteristics of rock under different heating methods, different temperatures and different soaking times were studied as well. The results show that: 1) At the same site, there is little influence of geothermal system on rock mechanical properties, but when geothermal system is in place, the number of fissures and the cumulative opening has an obvious increase. 2) With the increase of soaking time, the maximum water content of mudstone can reach more than 30%, the water content of gypsum can not exceed 15%, the water softening effect of mudstone is more obvious than that of gypsum, and the mechanical index of mudstone decreases with the increase of water content. 3) In the natural state, the gypsum rock is brittle and the mudstone is ductile; the water-rock interface of the gypsum rock has a significant effect of dissolution, and the dissolution of the gypsum rock is intensified from the surface to the interior. 4) In the temperature ranges from 20 °C to 50 °C, the compressive strength of gypsum rock parabolically changes with temperature, while that of mudstone monotonously increases. 5) The loose filling between geothermal pipe and rock mass is easy to form seepage pathway and change the hydrodynamic condition. Under the action of flowing water, the rock mass at the crack of gypsum rock and the weak interlayer dissolves, forming big cavity and inducing building subsidence.

In order to study the mechanical characteristics and the combined anti-slide mechanism of anchored slide-resistant pile for stabilizing the colluvial landslide, an indoor physical model test of a landslide reinforced by anchored slide-resistant pile under multi-stage loading was carried out. The variation curves of displacement of the pile top, soil pressure at the fore and rear of the pile, bending moment of the pile, axial force of anchor cable, deep horizontal displacement of sliding mass, and the characteristics of deformation and failure of the landslide are analyzed. Moreover, numerical simulation is adopted to compare with the model test and the results conform to each other. The results show that: 1) The horizontal displacement of the pile top, the bending moment of the pile and the axial force of anchor cable all show obvious three-stage characteristics under the variation of thrust load, and the load sharing ratio of the pile over the cable increases at first, then decreases, and finally tends to be stable. 2) The sliding mass resistance force at the fore of the pile is parabolic, and the value of the resistance force is small. The sliding mass thrust load at the rear of the pile is parabolic, and the resultant force point is 0.5h1 above the sliding surface (h1 is the length of the loaded segment). 3) The maximum bending moment point of the piles is always 2 cm below the sliding surface, indicating that the sliding bed at the fore of the pile has not been damaged. 4) The load distribution of the piles calculated through the measured bending moment of the piles indicates that the resistance force of the piles is mainly provided by the sliding bed below the sliding surface at the fore of the pile, and the resistance force is triangular. 5) The setting of the anchor cable can effectively limit the pile deformation. Meanwhile, the anchor cable failure problem should be paid more attention to in engineering practice. The research results can provide experimental basis for the rational design of anchored slide-resistant pile in the treatment of colluvial landslide.

The JRC-JCS model, which is widely used in the field of geotechnical engineering, has the following defects: 1) the two-dimensional joint roughness coefficient (JRC) can’t comprehensively represent the anisotropy of joint surface morphology; 2) the joint compressive strength (JCS) of joint wall can’t entirely reflect the influence of material properties on joint shear mechanical behavior. In this paper, artificial joint samples with natural joint morphology were poured based on three-dimensional laser scanning and three-dimensional printing technology, and shear tests under constant normal stress were carried out. The shear strength model with three-dimensional morphology parameters and tensile strength parameters was established based on the analysis of experiments and theoretical derivations. The impacts of normal stress and three-dimensional morphology of fracture surface on shear strength and dilatancy angle of rock joints were analyzed through laboratory tests and model comparison. The results show that the shear failure of joints is dominated by tension rather than compression. The different three-dimensional morphology causes different initial dilatancy angles. The peak dilatancy angle decreases with the increase of the normal stress. The shear strength of rock joints can be calculated by studying the changes of the peak dilatancy angle.

In order to reveal the meso-mechanical characteristics during the fracture process of the cemented waste rock-tailings backfill (CWRB) of metal ore and the synergistic mechanism of the tail-sludge cementation, uniaxial compression real-time CT scanning mechanical test was carried out to visualize and digitize the damage evolution process of the filling body with waste block proportions (WBP) of 0% (full tailings cemented backfill), 30%, 50% and 70%. The mesoscopic mechanism of mesoscopic damage and cracking and fracture evolution of CWRB was revealed. The results show that the WBP in the CWRB could affect the stress-strain response. As the WBP increases, the strength of the filling body increases. The main reason for the increase in strength is attributed to the propagation of the curved fracture surface in the sample. Related geomechanical effects, the crack morphology after the cracking of the filling body was affected by the shape, size and distribution of the waste rock. The matrix-block interface is the weakest part of the filling body. The formation and propagation of cracks eventually lead to the stress dilatancy behavior of CWRB. The damage and cracking on the interface control the crack propagation path and strength characteristics. The strength effect of CWRB depends on the content of block stone. The interaction between waste rock and waste sand controls the increase of sample strength, and the interlocking effect between blocks and rocks has an important impact on improving the overall stiffness of the sample. The research results have theoretical significance for the green disposal of solid waste and sustainable development of mineral resources

Wedge failure is an important form of instability in rock slopes. Previous research mainly used the block theory to analyze the stability from the mechanical point of view, and to a certain extent ignored the evolution process of geological action. To comprehensively analyze the hazard incubation mechanism and deformation characteristics of the wedge failure of rock slopes, a layered wedge rock slope generalized centrifugal model with two sets of structural planes was designed based on the geological prototype of Guanjiazui Zhongqiao wedge landslide in Lueyang county. With the field geological investigation and analysis, the centrifugal model test was successfully completed. The results show that: 1) Different from the traditional wedge landslide, the wedge will slide in a disintegrated manner along the intersection line of the wedge plane when the lithology of the slope is soft and the joint fissure is developed. This kind of wedge-shaped landslide is difficult to be identified geometrically due to its poor shape, and lithologic and structural conditions are its controlling factors. 2) Meanwhile, it is also found that the apparent deformation of such layered wedge-shaped soft rock slopes is not obvious. Under the action of gravity, the deformation mainly extends from the inside of the slope to the surface and bottom of the slope longitudinally, and the pressure and strain changes of the wedge-shaped sliding body are small, which is mainly manifested as tensile failure. 3) In addition, the landslide can form a multi-level wedge sliding surface, and develop multi-level instability wedges. The slope automatically searches for the "optimal" sliding surface, causing collapse and landslide disasters of different scales. At 1 260 s and 50g, a first-level wedge instability occurs, and at 2 016 s and 92g, the secondary deep wedges fails. The test reveals the deformation evolution process and disaster model of this kind of wedge landslide under the action of gravity, which provides a reference basis for the in-depth understanding of this type of landslide and the disaster prevention.

The load of marine environment is complex and unpredictable. The soil around the foundation of offshore structure is the main bearing body, which is greatly affected by the earthquake. There are a large number of sand layers in the coastal areas which have been built or are planned to be built in China. Compared with the desert Gobi area, the dynamic characteristics of the soil around the piles are significantly different. The existing studies mainly consider the evolution of soil mechanical properties under a single earthquake, and little attention is paid to the influence of earthquake sequence and earthquake history on the dynamic characteristics of soil around the foundation. With the use of the ZJU-400 centrifuge shaking table, this paper carried out a centrifuge modeling test in dry sand and saturated sand grounds, and compared the dynamic response of soil around the pile in two grounds under the earthquake. It is found that the natural frequency of saturated foundation is obviously affected by the historical effect of earthquake, while that of the dry sand ground is not. The strong pile-soil interaction can accelerate the development of excess pore pressure around the pile. Due to the historical effect of earthquake, the dilatancy characteristics of soil are gradually strengthened, the accumulation and development process of excess pore pressure around the pile is gradually slowed down, and the dissipation process becomes faster. Saturated sand ground has the characteristics of low frequency amplification and high frequency weakening. The historical effect of earthquake has no significant effect on the amplification factor of soil in dry sand ground, but for saturated ground, the amplification coefficient of soil shows an obvious increase. In the saturated sand ground, the shear modulus of soil around the pile is more affected by the excess pore pressure than the shear strain. During the shaking, the modulus gradually decreases without the phenomenon of gradual recovery in the dry sand ground.

In this paper, a user material subroutine (UMAT) of time-dependent UH model is written and embedded into the finite element software ABAQUS. Then a numerical back-analysis platform is established by combining multi-island genetic algorithm. In actual projects, factors such as rainfall, snowmelt and temporary loading will disturb the settlement law of normal creep. During this period of disturbance, the settlement rate is accelerated to form abrupt settlement, which has a great impact on the accuracy of numerical calculation. By introducing the concept of abnormal creep and analyzing the mechanism of settlement acceleration during this period, a numerical prediction method considering abnormal creep is proposed. When the fill body is lack of zoning basis and the settlement data of multiple monitoring points is taken as the back-analysis target, the distribution function of back-analysis parameters is introduced to replace the zoning of geometric model. On this basis, the influence of abnormal creep during the period of disturbance is further considered, and a numerical prediction method of settlement considering abnormal creep at multiple monitoring points is proposed. Finally, the accuracy and practicability of the numerical prediction method considering the abnormal creep under the condition of multiple monitoring points are verified by the settlement data of Chengde airport high fill.

The design philosophy of suspension bridge tunnel-type anchorage is that anchorage and rock bear the bridge load together. As the cooperative bearing mechanism of anchorage and rock, its bearing capacity is much higher than that of gravity anchorage with the same volume. However, due to the insufficient understanding of the synergy of surrounding rock, the squeezing effect between the anchorage and rock mass is still conservatively ignored in the design of tunnel anchorage. In order to understand the mechanism of coordinated bearing between anchorage and rock mass, and reveal the essence of improving the bearing capacity of tunnel anchor, a simplified mechanical model of tunnel anchor was established by analyzing the whole process from construction to completion of the bridge. Mindlin stress solution was used to analyze the law of load transmission along the anchorage axis and the distribution of compressive stress between anchorage and rock mass caused by load. A simplified method for estimating the ultimate bearing capacity of tunnel type anchorage was proposed. Then the recommend estimation method was successfully applied to Wujiagang suspension bridge project. The main conclusions are as follows: the interface force between anchorage and rock mass is mainly produced by the self weight of anchorage and the mutual extrusion of anchorage and rock mass; the additional stress at the interface between anchorage and rock mass increases first and then decreases from the rear anchor face to the front, and reaches the peak stress at about 1/3L away from the rear anchor surface; the ultimate bearing capacity of the tunnel anchorage of Wujiagang Yangtze River Bridge calculated by the allowable shear strength is 3 504 MN, about 16 times of the design load, which is basically consistent with the laboratory model test value.

Accurate prediction of surrounding rock deformation is one of the important prerequisites for scientific design and safe construction of large underground caverns. Existing prediction methods for surrounding rock deformation of underground cavern are mainly based on the monitoring data of the constructed locations surrounding rock deformation to predict the deformation trend of unconstructed locations. This causes difficulties in meeting the requirement of accurately predicting the total surrounding rock deformation of in engineering survey and design stage. Based on the statistical analysis of measured data from 31 large underground caverns in China, a method for predicting the maximum convergence deformation of underground powerhouse side wall based on multiple indexes was proposed. Firstly, it is found that the ratio of saturated uniaxial compressive strength to ground stress R/?, geological strength index (GSI) and material constant of intact rock mi have great influence on the surrounding rock deformation in the 31 cases. Their calculation methods are also given. Meantime, the ratio of maximum convergence deformation to cavern height (relative deformation value U/H) is used to evaluate the value of surrounding rock deformation. Secondly, through many statistical analyses, the prediction formula between the relative deformation value (U/H) of underground cavern and the three indexes is established. Finally, the method for predicting maximum convergence deformation of underground powerhouse side wall is verified by an engineering example. The results show that the calculated results by this method are very close to the actual results, which indicates the rationality of the proposed method.

Gravel was widely applied as the cushion materials for most foundations of the immersed tunnels, while the pebble as an alternative material has not been used till now. The differences in physical characteristics such as surface smoothness, alignment, contact methods and natural porosity between the gravel and pebble will affect its mechanical performance. In this paper, the mechanical deformation characteristics of pebble and gravel cushion were comparatively studied through physical modelling tests and numerical simulations. The results show that: 1) The compression curves of the two material cushions both show a two-phase anti-bending trend. The compression of pebble cushions is higher than that of the gravel under the same loading condition, whereas the overall secant modulus of pebble is approximately 30% lower than that of the gravel. 2) When the thickness of the cushion increases from 0.8 m to 1 m, a 13% increase in the secant modulus for the pebble cushion is observed, while 2.2% for the gravel. The mechanical deformation properties of the pebble cushion are more prone to be influenced by the cushion thickness than the gravel cushion. 3) When the preload increases from 52.5 kPa to 84 kPa, the secant modulus of the pebble cushion increases by 23.5%, while the modulus of the gravel cushion increases by merely 7.6%. A larger preload load leads to an earlier inflection point for the pebble cushion so as to show more stable mechanical properties. The increase in the preload load can more significantly improve the overall mechanical deformation performance for the pebble cushion than for the gravel cushion. 4) As the ditch width increases, the secant moduli of both cushions decrease. The pebble cushion is less sensitive to the changes in ditch size than the gravel cushion in the full load range. 5) The performance of the overall mechanical properties of the gravel cushion are better than the pebble cushion, but the cushion materials have limited effects on the structural settlement and stress. The pebble material can be used as an alternative cushion material for the immersed tunnels after a full analysis of the sensitivity to construction deviation. 6) Investigations on the optimal pebble grading will be the priority in the follow-up study.

Through the in-situ wave velocity test carried out after the construction of subgrade, the dynamic characteristic parameters of a subgrade-culvert transition zone in high-speed railway were obtained. The longitudinal distributive regularities of dynamic response amplitudes in the subgrade-culvert transition zone were studied by conducting two in-situ tests, respectively during both the comprehensive commissioning test and commercial operation. The comprehensive effects of train speed, running direction and driving load in the adjacent line on the dynamic responses were analyzed. Besides, the distributive regularities of vibration velocity of these two dynamic response tests were compared. Based on these test results, the effective values of vibration velocity and dynamic shear strain were calculated. The results of our study indicate that the train running direction has a significantly effect on the vibration velocity. The increment in the vibration velocity induced by driving load in the adjacent line is not negligible, and the vibration velocity shows positive linear correlation with the train speed. No significant change in vibration velocity between these two tests is found. The dynamic shear strains are less than the one fifth of the average value of the volume shear strain threshold. It can be concluded that the long-term dynamic stability of the subgrade-culvert transition zone can be guaranteed. The analytical approach and related parameters provided in this paper can serve as a valuable reference for railway engineering.

Influenced by the continuous cumulative rainfall of 350.6 mm in the previous 14 days, a large-scale reactivation of a giant basalt ancient landslide in Yanyuan Boli Village occurred on July 19, 2018. The resurrected volume was 1 390×104 m3 and 186 houses were damaged, causing significant economic losses. Based on field investigation, aerial survey of drone, drilling, physical and mechanical experiments, and numerical simulation, the factors and reactivated mechanism of the landslide were identified on the basis of geological structure and characteristics of the landslide. The results show that fractured rock-soil mass and topography are the fundamental causes for reactivation, and continuous rainfall and the rising of groundwater level are the triggering factors of reactivation. The landslide can be divided into two areas with different failure modes: main slide area and lateral collapse area. The seepage field in the landslide has significantly changed due to rainfall, and the pore water pressure has increased. As a result, the ancient landslide began to deform on July 13, and the stability factor gradually decreased. Constrained by microtopography, the main slide area is characterized by multiple stages and retrogressive failure. The front edge of the lateral slope is controlled by the main slide at the toe of the slope. During the movement of the main slide, the slope displacement and maximum shear strain increment in the lateral collapse area gradually increase, and the plastic zone expands, showing a certain relevance and lag with the main slide. This analysis indicates that due to the long-term influence of groundwater, the strength of the sliding mass gradually weakens. Rainfall leads to intensification of slope seepage and a decrease in shear strength, which induces landslide reactivation.

To reveal the relationship between the acceleration response and the fracture expansion of the soft and hard interbedded steep slope due to strong seismic deformation action, centrifuge table tests were conducted and the Voronoi joint division method in universal distinct element code (UDEC) was used to replicate the deformation and failure process. The tension and shear fracture distributions in the slope affected by different amplitudes were traced based on Fish programming, and then the relationship between the fracture evolution and mechanical response was established on the basis of the marginal spectral entropy concept. The results show that: 1) The deformation and failure process of the soft and hard interbedded slope with steep inclination is as follows: top tension crack, bottom-slope-foot shear and whole failure. 2) At the initial stage of horizontal vibration load, the crack development is mainly affected by the tension, and with the increase in amplitude, the crack expansion is mainly caused by the combined tension and shear actions. 3) The variation of the marginal spectral entropy of the monitoring points in the slope body is closely related to the damage degree of the corresponding position. The marginal spectral entropy value increases when the damage of the slope is slight, whereas the marginal spectral entropy decreases with severer damage of the slope. 4) The deformation failure process of the slope body and the depth of the sliding surface can be determined based on the variation in the marginal spectral entropy value of each monitoring point.

Using the self-developed code based on the SPH (smoothed particle hydrodynamics) method, evolution of the shear bands and acoustic emission events during failure processes of heterogeneous rock specimens under plane strain uniaxial compression is studied. Based on the Mohr-Coulomb criterion with tensile truncation, the stress calculation method of the failed particle is deduced under the assumption that the spherical stress tensor is invariant during the stress drop. Numerical results show that the longitudinal stress-longitudinal strain curves of the specimens with lower homogeneous degrees demonstrate obvious strain hardening and strain softening stages; strain hardening stages of the specimens with higher homogeneous degrees become less obvious with an increase of the homogeneous degree, and post-peak behaviors of specimens are obviously brittle. Stress drops of particles with different cohesions are theoretically analyzed. Results show that stress drops of particles with higher cohesions are not necessarily larger than the particles with lower cohesions. Stress drops are not only related to the cohesions of the particles, but also to the stress states of the particles when they fail. Effects of homogeneous degree on distributions of failed particles are qualitatively analyzed. The analysis shows that as the homogeneous degree increases, failed particles are more likely to form a narrow shear band quickly penetrating the specimen, with almost no particles failed outside the shear band once the specimen is penetrated. Therefore, the number of final failed particles in the specimen is fewer with an increase of the homogeneous degree.

Boolean fragment operation was used to develop a simple three-dimensional block-cutting technique. The hexahedral grid that covers the whole solution domain was generated, and then each hexahedron was divided into 48 tetrahedrons to produce the mathematical cover. Each tetrahedron unit was intersected with the solution domain using the Boolean intersection operation to generate the manifold block. Based on the oriented theory of topology and the three-dimensional simplex integration theory, four kinds of oriented geometry data structure including oriented edge, oriented wire, oriented face and oriented shell, were generated in each manifold block to construct enclosed and oriented three-dimensional manifold elements. The concepts of connected inner face pair and connected oriented manifold element were proposed, and then the physical covers were generated by searching the connected inner face pair of the oriented manifold elements. Furthermore, the key points for calculation of the modified symmetric and anti-symmetric decomposition-based three-dimensional numerical manifold method were summarized. A simulation on the tension process of finite plastic deformation on three-dimensional joints was performed without the consideration of three- dimensional contact, crack propagation and singular field at crack tips. Reasonable numerical simulation results were obtained, which justified the validation of the pre-processing module and calculation algorithm.

In order to detect tunnel deformation more sensitively, it is necessary to pre-stretch the strain optical fiber between anchor points when using fixed-point method to fix the sensing optical fiber. A certain amount of pre-tensile strain is thus generated inside the optical fiber due to the pre-stretching. Previous test results indicate that have shown that the monitoring results of the fiber will be affected by different pretension degrees. In order to reduce the monitoring error of optical fiber and improve the accuracy of monitoring, a scheme of optical fiber tension test was designed according to the strain characteristics and mechanical properties of sensing optical fiber, and a test system was then constructed to carry out the optical fiber tension test under different initial pre-tension values. By calculating and analyzing the experimental data, the optimal initial pre-tension values of two special sensing fibers were obtained. Combined with the above research results, the sensing optical fiber was applied in the Dahongmen Bridge Station–Heyi Station in the third phase of Beijing Metro Line No. 8. The results of this study can provide reference for the layout of sensing optical fiber in structural health monitoring engineering.

There are many differences in the standards of bi-directional load testing, among Chinese (JGJ/T 403-2017), American (ASTM D8169/D8169M-18), and European (ISO 22477-1:2018) standards, on the loading system, the displacement measuring system, the displacement convergence criterion, and the loading termination conditions. The layout of hydraulic tubing in the embedded jack in the American and European standards are significantly different from the Chinese standard. The displacement convergence criterion of the Chinese standard is more rigorous than the American and European standards. For pile integrity testing, the anomaly near the embedded jack shall not be directly judged as defect. Model testing of annulus embedded jacks were carried out. It shows that the position of the fracture surface in the tested pile is at the lower edge of the cylinder, and the concrete near the embedded jack would not be locally crushed. Field testing of double layer embedded jacks were conducted. It reveals that the soil filling rate of the fracture surface in the test pile is 50%-60%. Grouting process of the embedded jack after-testing is suggested. When grouting as required, the working pile can be used normally.

The application of current high-strain dynamic testing method to determine the bearing capacity of single piles depends on the experiences of the testing personnel, and the solution is not unique. To overcome these shortcomings, a new direct high-strain testing method is introduced in this paper. A finite element model is developed for a cast-in-place pile to analyze the effectiveness of the direct high-strain method for the bearing capacity of single piles. It is shown that the finite element model can accurately simulate the results in the published literature. By carrying out finite element dynamic analysis, the stress field, displacement field, velocity field and acceleration field of the pile for a hammer blow are obtained. According to the finite element calculation results, the bearing capacity is determined using the direct high-strain method. The bearing capacity determined by the direct high-strain method is consistent with that from static pile load test. The direct high-strain method eliminates the problem of multiple solutions in the existing high-strain method. Therefore, a unique bearing capacity of single piles can be obtained using the new direct high-strain testing method.