Aiming at the problems of ground settlement induced by the underground pipeline damage in water-rich sand layer, a set of visualization experimental device was designed. For these cases of 11 kinds of sand samples with particle size of skeleton d90=1.45?8.45 mm and 5 kinds of full pipe flow velocity, the law of ground settlement induced by seepage erosion was studied. Research show that: 1) There are three modes of seepage erosion induced by pipeline damage: only water inrush without settlement, soil arching formation with settlement and sand crushing with settlement; 2) The particle size of soil skeleton , damaged mouth size and thick-span ratio are the main factors to determine the seepage erosion mode of soil; 3) When the soil arching or sand crushing is formed in the soil above the damaged mouth of pipeline, the relationship between the particle size of skeleton d90 and the thick-span ratio r is that: when 8.0≥r≥4.2, the d90 decreases parabolically with the increase of r; when 12.5≥r≥8.0, d90 remains unchanged; 4) When the soil arching or sand crushing is formed in the soil above the damaged mouth of pipeline, the initial settlement radius and settlement depth with the flow velocity equal to 0 are determined by the ratio (D/d50) of the damaged mouth diameter D and the average particle size of soil d50; the settlement radius and depth increase linearly with the increase of full pipe flow velocity; when the soil arching is formed, the expansion velocity （VL、VH） of settlement radius and settlement depth with the increase of flow velocity is logarithmic to D/d50. When the sand crushing is formed, the expansion velocity VL of settlement radius with the increase of flow velocity is that, when 23.0≥D/d50≥6.0, it increases linearly with the increase of D/d50; when 42.0≥D/d50≥23.0, it decreases logarithmically with the increase of D/d50.

To further investigate the consolidation mechanism of saturated clay, the unified-hardening (UH) constitutive model considering the time effect was introduced to describe the elastic visco-plasticity of saturated clay soil as well as the Hansbo’s equation to describe the non-Darcy flow in the consolidation process. Accordingly, Terzaghi’s one-dimensional consolidation equation of saturated clay was modified, and the numerical analysis was conducted by using the finite difference method. In order to verify the applicability of the present method and the UH model, the numerical solutions provided by the present method for the cases that the flow of pore water obeys Darcy’s law were compared with the theoretical and experimental results based on one-dimensional rheological consolidation theory in the literature. Then the effects of Hansbo’s flow parameters and the UH model parameters on the rheological consolidation process were investigated. The calculation results show that the viscous effect of soil leads to an increase of pore water pressure near the impervious boundary at the early stage of consolidation, and the behaviour of the non-Darcy flow enhances this phenomenon. These two characteristics of soil also control the overall dissipation rate of excess pore water in the saturated clay layer, thereby reduce the settlement rates of the soil layer during the middle and late stage of consolidation. In addition, these aforementioned consolidation behaviours become more obvious with the increase of swelling index and overconsolidation parameter of soil. But the parameters of Hansbo’s flow and the permeability index have no effect on the ultimate settlement of the soil layer.

The physical and engineering mechanical behaviors between coral sand and terrigenous sandy soils are considerably different. To study these behaviours, a series of undrained multistage strain-controlled cyclic triaxial tests was conducted on saturated coral sand from Nansha Islands, South China Sea. The influence of effective confining pressure and relative density Dr on the dynamic shear modulus and damping ratio of coral sand was studied. Compared with the test results of coral sand and terrigenous sandy soil and gravel, significant differences were found in the maximum shear modulus Gmax, the shapes and the upper and lower boundaries of shear modulus ratio G/Gmax curves, reference shear strains , the shapes and the upper and lower boundaries of damping ratio ? curves. The maximum shear modulus Gmax of coral sand is higher than that of terrigenous sandy soil and gravel, and the Gmax of coral sand predicted by empirical equations of terrigenous sandy soil is underestimated by 30%. The nonlinearity of coral sand is slightly weaker than that of terrigenous sandy soil and gravel. The empirical formulas for predicting G/Gmax and ? of terrigenous sandy soil and gravel are not applicable for Nansha coral sand. The empirical formulas for predicting G/Gmax and ? of coral sand are proposed.

Triaxial seepage tests were conducted on the core-wall clay of a high rock-fill dam to investigate the change of hydraulic conductivity with axial strain. It was found that the compaction density and confining pressure were the two main factors influencing the change of hydraulic conductivity during triaxial compression. When the current confining pressure was greater than the pre-consolidation pressure of a compacted specimen, the specimen kept being compressed and became denser during the compression process, resulting in a decreasing trend in its hydraulic conductivity until it eventually reached a stable state. On the other hand, if the current confining pressure was far less than the pre-consolidation pressure of the specimen, the specimen deformed in a localised shear band which weakened the impermeability of the specimen, and as the shear band continued to dilate, an increasing trend in the overall hydraulic conductivity was observed. This study highlighted an important fact that heavily compacted clay under low confining pressures had a high susceptibility to localised shear bands of high permeability, which could be used to explain many historical leakage problems observed in clay-core dams.

The temperature effect on thermal conductivity of compacted Gaomiaozi (GMZ07) and Wyoming (MX80) bentonites was investigated using a thermal probe method. Dry densities and water contents of compacted bentonite specimens were kept constant under constant volume conditions, and then measurements of thermal conductivity were conducted over a wide range of temperatures (5°C to 90°C) by KD2 Pro thermal characteristic analyzer. Meanwhile, the mercury intrusion porosimetry (MIP) tests were also performed to observe the pore-size distributions of compacted specimens. The test results show that at the same water content and dry density, the thermal conductivities of GMZ07 and MX80 bentonites increase with increasing the temperature. When the temperature was 90°C, they can reach 1.2 to 1.5 times of the value at 5°C, due to the enhanced latent heat transfer (LHT) of vapor. At specimen temperature is higher than 60°C, the effect of temperature on the thermal conductivity is more significant than that when temperature is under 60°C. For unsaturated specimens, the temperature effect on the thermal conductivity decreases with increasing the dry density. For a dry specimen, the thermal conductivity hardly changes with the temperature, which is related to the enhancement mechanism of the LHT of vapor. The temperature effect on the thermal conductivity can be explained that the more water vapor and heat transfer paths can be used for LHT, the more obvious the temperature effect on the thermal conductivity.

Geometry and mechanical characteristics of joints are key factors affecting the shear behaviors and fracture patterns of jointed rock mass. Using 3D printing technique, experimental jointed rock specimens containing rough joints with varied joint roughness coefficient (JRC), different types of geometrical joints, and fractrues network were established respectively. Then direct shear tests were conducted on the jointed rock models to study the shear strength and failure patterns. The results show that the values of shear strength of jointed rock models fluctuate greatly with varied JRC curves. The peak shear displacement decreases with increasing amplitude of fluctuation. The lowest and highest values of shear strength are found in plane and rectangular joints, respectively. Moreover, the value of shear strength of sinusoidal type is found similar to that of triangular type. The shear strength of the 3D-printed fractured rock specimens are apparently lower than that of the intact rock specimens. When considering joint roughness, the peak shear stresses of rough discrete fractures network (RDFN) models are higher than those of the discrete fractures network (DFN) models. The solid rock specimens exhibits typical brittle shear failure. The fracture patterns of DFN model and RDFN model are relatively complex and the main shear fractures are found along the shear direction. Meanwhile, certain intersection points between joints and certain surfaces have significant influence over the shear failures. The present work provides references for the application of 3D printing technique in studying the shear behaviors of fractured rock masses.

In order to study the shear-flow coupling characteristics of joints under constant normal stiffness (CNS) boundary condition, the shear-flow coupling tests under three different stiffness and seepage pressure settings were carried out for duplicate joint specimens with three different joint roughnesses. Meanwhile, the effects of normal stiffness, seepage pressure and joint roughness on the mechanical properties and seepage characteristics in joint shearing process were systematically analyzed. The test results indicate that the peak shear strength of joint increases with the increase of normal stiffness, while the flow rate, equivalent hydraulic aperture and transmissivity decrease with the increase of normal stiffness; and the flow rate of seepage through joint surfaces during shearing process decreases with the increase of joint roughness. A three-stage change rule of flow rate that is similar to the joint dilatancy is shown in the shearing process: rapid growth stage, slow growth stage, and stable stage. The same rule has also been observed in equivalent hydraulic aperture and transmissivity. During the stable stage, the flow rate is approximately linear with the variation of the normal stiffness and seepage pressure, and the joints with higher roughness present lower flow rate as the seepage pressure increases.

The unconsolidated sandstone in hydrothermal geothermal field in Jianghan basin is taken as the research object. The hydrostatic pressure is applied to a geostress equal to 12.5 MPa. After the deformation of sample is stabilised, the evolution and mechanism of the permeability of unconsolidated sandstone under the compaction of high hydrostatic pressure are studied, which can provide some suggestions for the selection of equipment operating parameters for the tailwater recharge process in the hydrothermal geothermal field. The results indicate that under high hydrostatic pressure compaction, the permeability of unconsolidated sandstone samples tends to be a constant valued of 4.0×10?3 ?m2 within the current range of 0.5 mL/min to 3.0 mL/min. The pressure difference between the two ends of the sample increases nonlinearly with time and the degree of nonlinearity gradually increases with the increase of flow rate, but eventually tends to be stabilised. In addition, the sample of unconsolidated sandstone forms a tubular erosion channel in the direction of penetration, extending to about 2/3 of the sample. Based on the stop time of particle transportation and the extension length of tubular erosion channel in the penetration direction, the average migration velocities of particles under different flow rates are determined. It is found that the particle migration velocity increases exponentially with the increase of flow rate, and the amount of microparticle migrated per unit time increases. When the pressure difference exceeds about 1/2 of the hydrostatic pressure, the sample presents erosion damage and upstream diameter shrinkage.

The dynamic response of carbonate sand-steel interface is of great significance to the safety and stability of foundations in reef islands. Based on the interface ring shear apparatus, a series of cyclic shear tests on the interface between carbonate sand and steel was carried out. The effects of normal stress, cyclic amplitude and particle size on the interface shear stiffness and damping ratio were investigated, and compared with quartz sand. The results show that normal stress and cyclic amplitude have significant effects on the interface shear stiffness and damping ratio; the increase of normal stress level increases the shear stiffness and decreases the damping ratio; the increase of cyclic amplitude leads to an approximate inverse reduction of shear stiffness and an approximate logarithmic increase of damping ratio; for carbonate sand with a uniform particle size, a critical particle size exists that results in a significantly different interface shear behavior. When the particle size of quartz sands is large, the shear stiffness and damping ratio are different from those of carbonate sands. But, if the particle size is small, the shear stiffness and damping ratio of these two sands are almost the same.

In high rockfill dam, particle breakage is one of the main factors that lead to dam deformation. However, because of the large size of the rockfill particles, the breakage degree of the prototype rockfill materials is difficult to be measured directly through laboratory test. Therefore, the common practice is to reduce the size of the prototype gradation particle to less than 60 mm before the laboratory test can be carried out. However, due to the significant difference between the prototype and the test gradation, the parameters measured in the test are often quite different from the actual parameters of the prototype rockfill materials, thus affecting the in-depth study of the mechanical properties of the prototype rockfill materials. In this paper, a new method is proposed to describe the change of particle size distribution (PSD). Firstly, based on the theory of Weibull distribution of the particle strength and fractal crushing of the particle, the calculation of PSD change of the original rockfill materials is elaborated. Then, the relevant parameters are obtained by conducting the single-particle crushing test, and by comparing with the triaxial test, the rationality of parameter selection is verified. Moreover, the effects of the discrete degree of particle strength on the shape of the PSD is analyzed. Finally, the relationship between the relative crushing parameters and stress state of the rockfill materials during loading is discussed.

When the top of a pile in soft ground is subjected to a large horizontal load, the traditional m method would underestimate the bending moment and deformation of the pile element. Therefore, it is necessary to propose relevant calculation methods for this problem. Simplifying the soft ground to an ideal elastoplastic body, and assume that there exists a critical point at a certain depth of the ground where the soil mass above the critical point behaves plastically and the soil mass below the point behaves elastically. The deflection differential control equation of the pile is established and a simplified elastoplastic calculation method for the horizontally loaded single pile is proposed. The results of field measurements and parametric study show that the maximum bending moment calculated by the simplified elastoplastic calculation method is 38.1% more accurate to field measurement in comparison with the traditional m method, and the calculation accuracy of the maximum horizontal displacement of the pile is improved by 22.3%. It indicates that the boundary conditions at pile top have significant impact on the distribution of lateral deflection and bending moment along the pile. Moreover, parameters of soil limit resistance and soil shape have remarkable influence on the maximum bending moment and lateral deflection and it is recommended to choose the lower values in pile design.

Under the condition of vertical load, the soil on the top of pile foundation after complete liquefaction is regarded as fluid, and the pile foundation is equivalent to the Euler-Bernoulli beam model. The vibration impedance of the pile top with embedded pile bottom is discussed. The expressions of relationships between displacement and internal force of pile’s top and pile’s bottom are obtained by means of the matrix transfer method after simulating liquefied soil using fluid dynamic equation and simulating lower non-liquefied soil layer with Winkler foundation, as well as considering the continuous conditions of displacement, rotation angle and internal force at the interface between the liquefied soil and the non-liquefied soil. Finally, according to the embedded conditions at the bottom of the end bearing pile, the expression of the impedance of the pile top is obtained. Compared with the existing literature, the correctness of the analysis process in this paper is verified. The parametric analysis of the conditions affecting the impedance shows that the liquefaction depth, the axial force exerted on the top of the pile and the fluid density have different effects on the impedance of the pile top.

Fracture mechanics is one of the basic disciplines in various industries. The semi-circular bending (SCB) test is a classic test used in the study of solid mechanics. At present, the research on SCB mainly focuses on surface cracks such as bottom notch or penetrating cracks. Whereas the internal cracks and internal defects that are the intrinsic properties of material have been seldomly researched employing the SCB test. Based on the 3D-ILC (3D-internal laser-engraved crack) method, an analysis was performed on the pure internal cracks with arbitrary parameters generated randomly given that surface had no influence on crack generation, internal cracks were created in SCB specimens and the specimens were assessed by the SCB test. Numerical simulation was performed to analyze the crack growth process, stress birefringence, I-II-III crack growth surface, failure pattern, and macro to micro fractography characteristics. Moreover, the distribution law of stress intensity factor and extension path were obtained that agreed with the physical test. The results show that: 1)The 3D-ILC method is useful for solving the problem of internal cracks in fracture mechanics. 2) Stress concentration is shown at the internal crack of stress nephogram, and significant variation occurs at the internal crack of SCB stress nephogram. 3) The compression-shear crack initiation occurs at the prefabricated crack of the SCB and develop into I-II-III mix-mode crack. The crack growth surfaces are divided into smooth zone and tear zone. The continuous growth is transformed into dynamic fracture and then failure, along with the mist zone and the hackle zone. 4) Based on the M-integral and maximum tensile-stress (MTS) theory, the numerical simulation of the specimens is carried out. The distribution of the stress intensity factor at the crack tip is given, and the crack growth path is simulated. The numerical simulation agrees with the experimental results. It provides experimental and theoretical basis for the study of SCB testing, internal cracks, I-II-III mix-mode fracture, simulation of crack propagation paths in the field of fracture mechanics.

Mixed sand containing carbonate, as a special geo-material, is mainly distributed in tropical coastal areas. Its unique genesis and composite structure result in different geotechnical engineering characteristics compared to terrigenous sand. Triaxial tests were carried out on the mixed sand with different contents of calcium carbonate under different confining pressures. The testing materials were prepared with coral sand from the South China Sea and silicon sand from the Changjiang River with different mass percentages. The results show that: 1) the peak strength increases approximately linearly with the increase of effective confining pressure and calcium carbonate content; 2) confining pressure and carbonate content are the main parameters influencing the dilatancy characteristics of the mixed sand. The increase of carbonate content results in the transition point between dilation and contraction response and its corresponding axial strain to increase; 3) with the increase of carbonate content, the peak value of cohesion increases linearly and the internal friction angle increase slightly; 4) the relative crushing rate of mixed sand increases with the increase of the calcium carbonate content and confining pressure.

Demonstrated by a series of triaxial compression tests, the mechanical properties of claystone are related to its stress state. The relationship between them shows that the larger the confining pressure is, the higher the claystone samples’ peak stress is and the later it develops. This is subject to the tendency of microcrack closure within specimens under high confining pressure, suggesting that high confining pressure has certain inhibitory effects on the development of microcracks. Therefore, the peak stress appears in higher magnitude and at a later stage. However, the degree of the stress drop during the softening process is not heavily affected by the variation of confining pressure. The reason is the microcracks within the samples under different confining pressures gradually connect to each other to generate shear planes of similar type, so that the mechanical properties of the specimens are weakened to the same extent. Based on the experimental study, the paper introduces a damage evolution equation considering varying stress states by taking microcrack as the damage element (because the development of microcracks is affected by stress states). Moreover, a new constitutive model is established based on the test results and numerically implemented in Abaqus and its subroutines. Comparisons between the simulation and experimental results demonstrate the model adequacy for effectively describing the mechanical properties of claystone. The study provides theoretical guidance for the engineering construction of underground infrastructure in claystone.

Macroscopic and microscopic triaxial shearing experiments were carried out on Shanghai soft clay to study the evolution characteristics of pores under loading. The results show that local deformation occurs at the beginning of loading. With the increase of the loading, number of small aggregates increase, pores continually enlarge, interconnected pores occur, the pore orientation is obvious, and a shear band is formed. Microstructure parameters are nonlinearly positive to the shear stress ratio. The maximum pore area, void ratio, anisotropy ratio and fractal dimension of pores increase slowly at an initial loading stage, and then increase rapidly at the later stage. The maximum anisotropy ratio reaches 0.68 and void ratio is up to 1.96, when the axial stain is 8%. The microstructure parameters of shear band are larger than those outside the band. During the triaxial loading, soil microstructure deteriorates and the whole deformation process of soft clay is divided into three stages: damage initiation, damage developing and shear band formation, as well as soil failure. Microstructure degradation is closely related to macroscopic mechanical properties of soil.

To investigate the bearing behavior of a single pile in uniform sand under the combined action of horizontal harmonic load H(t) and torque T, the modified finite bar element method is used to establish the comprehensive stiffness matrix of the pile shaft with consideration of the stiffness contributed by the surrounding subsoil and the coupling effect between H(t) and T. Then, the internal forces and deformations of H(t)-T combined loaded pile are solved by MATLAB, and a detailed parametric analysis is carried out to discuss the influence of length-diameter ratio (L/D), pile-soil stiffness ratio (Ep /Es), dimensionless frequency (a0) and amplitude (Q0) of the horizontal harmonic load, and boundary conditions of the pile toe on the bearing behavior of pile shaft. The results show that as the depth of the pile increases, the bending moment along the pile shaft first increases and then decreases sharply, while the torque and deformation reduce drastically to almost zero in the end, i.e., there is an effective pile length ld. The ld, internal forces and displacements are greatly affected by Ep /Es. When Ep /Es increases from 1 500 to 3 500, ld rises from 27.3D to 44.1D, and the maximum bending moment of the pile increases by 23.5%, but the horizontal displacement and torsional angle at the ground surface decrease by 19.0% and 26.0%, respectively. Decreasing a0 or increasing Q0 can relatively improve the torsional bearing capacity of the pile shaft. The boundary conditions of the pile toe only affect the internal forces and deformations of finite-length piles.

Based on the uniaxial compression experiment of granite and decision tree model, the extraction method of key acoustic emission signals in rock fracture process is constructed. The extracted key signals are firstly classified by choosing the characteristic parameters, and the time-frequency features of the various key acoustic emission signals are analyzed, then the rock fracture mechanisms corresponding to these signals are discussed. The results show that with a signal recognition accuracy of over 90%, the method based on decision tree model can effectively extract the acoustic emission signals corresponding to the critical fracture events, which affect the stability of the whole rock structure during the rupture process. The key signals are divided into four categories according to the feature extraction rules. The signals of Class A correspond to the macroscopic cracking and expanding fracture of rock. The signals of Class B correspond to a large number of small-scale fractures and extensional cracks that occurred in the near and post-peak stages of the fracture process. The signals of Class C correspond to the shear-slip fracture before the rock experiencing macro-fracture. The signals of Class D correspond to the small-scale tension-shear composite fracture after the whole rock failed.

At present the theoretical calculation of the stope floor failure depth simplifies the floor as elastomer or plastomer for analysis. However, the application conditions for the two methods are not specified. Meanwhile, this calculation approach cannot reflect the layered structure of the stope floor. Therefore, the limit equilibrium failure mode of the floor under abutment pressure is regarded as a circular-arc sliding. The Swedish slice method is used to search for dangerous sliding surfaces and to obtain the stability coefficient as well as the maximum depth of the sliding surface. Based on the safety factor value given by the foundation design code, the homogeneous soft and hard rock floor is taken as an example for comparisons to be made between the calculation results with existing theoretical solutions and elastic-plastic numerical solutions. The article discusses the influence of several factors, such as slab combining soft and hard rocks and strata inclination on the base slab stability coefficient and sliding thickness. The research shows that the hard rock floor mainly experiences local plastic failure, and generally demonstrates high stability coefficient. Therefore, it can be solved with the elastic solution. The soft rock floor has large plastic zone, and can even involve plastic sliding phenomenon, it requires the use of limit equilibrium method to reduce errors. In slab combining soft and hard rock, when the thickness of the hard rock reaches a certain value, the base slab stability coefficient will be greatly increased and the plastic zone in the floor will be controlled. In the top region of the coal face, the floor with high dip angle can easily experience shallow sliding failure while the lower part undergoes opposite act. The application example shows that the circular sliding solution considers the non-uniformity of the strength parameters of layered floor and the floor’s failure mode, which is more consistent with the actual situation.

Slope failure frequently occurred during and after the excavation of phyllite talus at the Wenchuan-Maerkang expressway. Previous research on phyllite talus has mainly considered it as homogeneous soil, while its unique structural features, such as the orientation of the constituted phyllite fragments, is less concerned. Phyllite is a metamorphic rock with foliated structure, and phyllite fragments are characterized with their flattened shape that may cause them to stop accumulating along the slope in a certain orientation. Field investigation shows that the dip direction of the phyllite fragments is usually parallel to or forms only a small angle to the talus slope inclination, which is defined as fragment orientation in this paper. The fragment orientation is considered as an important factor contributing to the slope instability of phyllite talus. Model tests are designed to study the orientation characteristics of the phyllite fragments and its influencing factors by taking the phyllite talus at Wenchuan–Maerkang expressway as examples. Large-scale direct shearing test considering the fragment orientation are consequently performed to explore the effects of fragment orientation on slope stability of phyllite talus. The results show that the slope angle and the content of fine particles in the talus have significant influence over the fragment orientation; when the slope angle is 20°-30°, the orientation is most obvious and rises with the increase of fine particle content; the shearing strength parameters of phyllite talus for slope stability analysis significantly reduce due to the fragment orientation.

Laboratory tests have been carried out for simulating dredged-slurry treatment combining air-booster and conventional vacuum preloading methods. Settlement plate, micro pore water pressure transducer and miniature vane shearing instrument are adopted for monitoring the settlement, the pore pressure dissipation and the soil strength during the test. The test results demonstrate that the air-booster vacuum-preloading method can significantly improve the soil strength, pore-water pressure dissipation and settlement. In particular, the real-time responses of pore-water pressure during the air-pressurizing process have been obtained that, in combination with the numerical simulations, help derive the reinforcement mechanism of the air-booster vacuum preloading method, which involves using micro cracks generated by splitting soil between the booster tube and PVD using the pressurised air. Those cracks improve the soil permeability, and have lasting effect even after air pressuring ceased.

The stability study of shallow-buried soil tunnel has always been a key problem in tunnel engineering. The existence of pore water pressure can contribute to the collapse of shallow-buried soil tunnels that has serious influence over the safety of life and property. In this paper, the three-dimensional collapse mechanism of a shallow-buried earth tunnel with a circular top arc is constructed. Based on the nonlinear Mohr-Coulomb failure criterion and the limit analysis upper bound method, the collapse range of the shallow-buried earth tunnel is derived considering the pore water pressure. The formula for solving the optimal upper limit of support force is established. Moreover, the proposed method is compared with existing research for rationality verification of the method. The effects of different parameters on the collapse range, gravity and support force of the collapsed soil are analyzed. The results show that the pore water pressure has significant impact on the collapse range of the shallow buried soil tunnel as well as the gravity and support force of the collapsed soil. The effects of pore water pressure on the collapse range and weight of the collapsed soil are complex, while the support force increases with the increase of the pore water pressure coefficient. Moreover, the collapse range, gravity and support force of shallow soil tunnel are influenced with varying degrees by different parameters. This method can provide theoretical support for the design optimisation of shallow-buried soil tunnel.

Through acoustic emission monitoring tests of coral skeleton limestone under different confining pressures, mechanical properties of coral skeleton limestone under triaxial compression and basic AE characteristics were obtained. The study shows that the strength characteristics of coral skeleton limestone accord with Mohr-Coulomb strength criterion. The macroscopic failure characteristics of coral skeleton limestone under the action of triaxial compression show obvious confining pressure effect. In other words, uniaxial compression is shown as split tension fracture surface; with the increase of confining pressure, the “two-stage stepped” macroscopic fracture gradually turns into a single inclined shear zone. The AE time series evolution law of coral skeleton limestone under the action of triaxial compression is basically consistent with the stress-strain curve characteristics, and there is a good logarithmic function relationship between the cumulative ringing count and the cumulative energy count corresponding to the equivalent failure time and the confining pressure. The cumulative pre-peak ringing and cumulative energy counts show a stable and gentle phenomenon before ?cd, which is in line with the linear elastic deformation characteristics of the rock, and presents a sharp increase at about 90%, which is the beginning of the rapid and unstable expansion of micro-cracks in the rock. The evolution law of pre-peak energy of coral skeleton limestone under uniaxial compression is basically consistent with the characteristics of stress-strain curve, the pre-peak elastic energy and pre-peak dissipated energy also appear the turning point at about 90%.

Since gradation greatly affects its compaction characteristics, the quantitative description of the effect of gradation on compaction properties is crucial. Based on the continuous gradation equations of soil, 16 groups of experimental coarse-grained soils with different gradations were designed. Surface vibration compaction test and confined compression test were carried out to study the effect of gradation on compaction properties of coarse-grained soil. The results showed that the quadratic function relationship was found between the gradation curve area and dry density obtained by surface vibration compaction (dynamic stress) and confined compression (static stress). Besides, the gradation curve areas corresponding to the maximum values of dry density were close to 0.9. Both in surface vibration compaction test or confined compression test, there was optimal dry density of coarse-grained soil under certain gradation. Based on test results, the optimal gradation interval of coarse-grained soil with better compaction was given (upper and lower envelopes of gradation), which was further demonstrated through similar grading method. Meanwhile, the universal applicability of this optimal gradation interval was verified by the design gradation of coarse-grained soil of several earth-rock dams. This paper can provide the important basis for the grading design of coarse-grained soil for earth-rock dam.

The vertical displacement of soil in the underground pipeline caused by the construction of shield tunnel was studied based on the Loganathan function. The Pasternak model considering shear transfer in soil was used to simulate the pipeline-soil interaction. The key parameters of the Pasternak model—elastic coefficient k and shear coefficient gs were solved by using the iterative procedure in the Modified Vlasov model. The calculation results were compared with the results of existing literatures and engineering monitoring data, and the differences between the proposed model and the existing result were also explored. Furthermore, a parametric study was conducted to analyse the influence of soil shear stiffness, cross angle, soil modulus and tunnel radius on the pipeline-soil interaction. The results show that the values of spring factor k and shear stiffness factor gs solved via the iteration procedure can improve the accuracy of the Pasternak model; the influence of soil shear resistance on the calculated vertical displacement of pipeline is up to15.3%; with the decreases of cross angle between the pipeline and the tunnel, the vertical displacement of pipeline increases and bending moment decreases; the increase of soil elastic modulus and the tunnel radius will increase the vertical displacement and the bending moment of pipeline.

The establishment of soil constitutive model is often grounded on the isotropic compression model. Based on the analysis of the isotropic compression characteristics of the cohesionless soils, it is found that the value of the compression index is closely related to the current void ratio and spherical stress. On this basis, an expression of the compression index formulated as a separable function of the current void ratio and spherical stress is established. Then, a three-parameter compression model that describes the compression characteristics of the cohesionless soils is obtained. The comparisons with test data of four types of sands show that the model can properly fit the relationships between void ratios and stress of cohesionless soils with different initial void ratios. Compared with the compression models established by taking the limit compression curve as the refence line, the proposed model is more useful in fitting the relationships between void ratios and stresses of cohesionless soils in low-stress regions. The model also provides a foundation for establishing constitutive models of cohesionless soil.

Using rock cavern for compressed air energy storage is a promising method for large-scale energy storage. Ensuring the sealing performance and structural safety of the storage cavern are the core tasks in such construction. To verify the feasibility of shallow rock cavern, a lined cavern within granite stratum was constructed in an exploratory tunnel in Pingjiang pumped storage power station, Hunan, as the first compressed air energy storage in China. The cavern was subjected to 10 cycles of charge and discharge during the test. The test results show that the temperature of the compressed air exhibits a significantly uneven distribution, and the evolution of the temperature can be effectively controlled by the heat exchange system. The leakage rate of the chamber is about 3.2% under the long-term high-pressure condition, implying that good sealing performance is achieved. The maximum deformation of surrounding rock is about 0.35 mm under an internal pressure of 8.7 MPa, and the deformation affected zone caused by high internal pressure is within 10 m, indicating that the surrounding rock is safe. The results can help understand the working performance of underground high-pressure gas storage and can provide a reference for the design of compressed air energy storage in shallow rock caverns.

The stability of diversion tunnel’s carrying capacity jointly provided by the surrounding rock mass and plug is essential for the power generation and safe operation of hydropower projects. The diversion tunnels of Wudongde hydropower plant use column plugs instead of traditional expanded wedge plugs. Whether the safety of the new plug structure interacting with specific surrounding rock mass and subjecting to high water head conditions meets the engineering requirements is an important issue that needs verification. The paper analyzes different types of plugs, including straight column, curved column and wedge shape plugs under the action of high water head by using geophysical exploration, testing and other means in combination with numerical calculation that simulates the process of progressive failure of the interface between the surrounding rock and plug. The main results obtained include: 1) the calculation formula of anti-sliding force under the interaction of different types of plugs and surrounding rock is given, and the wedge-shaped plug is proposed. The ‘clamping effect’ of wedge-shaped plug and the ‘bending section effect’ of curved column plug are analysed. 2) The contact state between the surrounding rock and the concrete in the plug section of the Wudongde diversion tunnel is obtained showing that the interfaces of the plug section between the lining, shotcrete and surrounding rock are tightly contacted and well cemented without any weak filling or opening on the contact surfaces. The shear strength parameters of these contact surfaces are also suggested. 3) Under the action of high head, the deformation law of different types of plugs and surrounding rock mass interaction is basically the same, and the impact range of plug compression deformation approximately halves the plug length. The deformation range caused by water pressure is approximately 1 times the tunnel span, and the shearing effect between the concrete and the surrounding rock is not fully exerted. The effective shearing impacting area is approximately 1 time the tunnel span. 4) When the mechanical properties of the contact surface between the plug and the surrounding rock are poor, the safety ranking of plugs from high to low is wedge-shaped plug, curved column plug, straight column type. When the contact surface between the plug and the surrounding rock mass is favourable, the column plug can provide the same safety margin as the traditional wedge-shaped plug. Finally, the safety evaluation on the interaction of the plug and surrounding rock mass in the diversion tunnel of Wudongde hydropower plant during construction and operation periods is achieved, indicating that the safety load borne by the column plug structure can range between 3.1 and 7.4 times the design head, obtaining a safety factor greater than the required specification. Therefore, a column structure is feasible and can be used as a reference for similar projects.

The propagation characteristics and diffusion law of horizontal prestress from the slope surface to the embankment interior are key issues to be solved for the reinforcement of a prestressed subgrade. Based on the theory of elasticity, analytic formulas of the horizontal additional stress in a prestressed subgrade were derived, and the diffusion coefficient graphs of ?H under six typical slope ratios were obtained through the calculation of a number of distribution points. Meanwhile, the propagation characteristics and diffusion law of ?H were analyzed. The results show that: 1) the diffusion coefficient of horizontal additional stress (KH) at any point in a prestressed subgrade can be accurately obtained using three dimensionless parameters, i.e., the slope rate (1:m), the side length ratio of the lateral pressure plate (l/b) and the ratio of the calculation point’s horizontal depth to the width of the plate (h/b); 2) under different embankment slope ratios, the curves of KHD (or KHU)-h/b at the lower (or upper) corner points of the plate exhibit similar variation trends; at the lower corner points of the plate, the KHD-h/b relationship can be identified into two types determined by a critical l/b value ((l/b)c), i.e., first increasing and then decreasing or gradually decreasing; while the KHU at the upper corner points of the plate gradually decreases with increasing h/b; under the same embankment slope ration, KH increases with an increase in l/b, but the increase rate decreases gradually; 3) when the net spacing between the lateral pressure plate and the embankment shoulder is more than twice of the plate width, the additional stress diffusion law obtained by the interpolation of the graphic method is in good agreement with the results of the theoretical formulas and numerical simulations, which demonstrates the accuracy and applicability of the graphic method and theoretical formulas for calculating the horizontal additional stress in a prestressed embankment. The research findings form an efficient integration method and can provide a fast and simple reference for the design of a prestressed subgrade.

The construction of dam on the 500 m-level thick overburden with soft soil layer in a strong earthquake zone is beyond the control scope of the current design specification. Through the sensitivity analysis on indicators such as overburden thickness, peak of input seismic wave and thickness of soft soil layer, the main characteristic of the seismic response of deep overburden and the distribution regularity of foundation surface acceleration amplification factor of the dam foundation are studied. According to the research, the overburden thickness, input acceleration peak and soft soil thickness are positively correlated with the attenuation of the amplification factor of the foundation surface. When the peak acceleration of the input seismic wave is 0.5g, the earthquake responses in the 500-level ultra-thick overburden layer with weak soil layer are mainly attenuated, and the magnification factor of the foundation surface is less than 1. The basic law of the acceleration magnification of the overburden layer is first attenuation and then amplification with the elevation change. When there is a weak soil layer, the secondary attenuation of dynamic response will occur in the soil layer due to its filtering and seismic isolation effects. Based on the above analysis, an acceleration amplification factor distribution model of the attenuation of the overburden layer’s lower part, the secondary decay of weak soil in the central area, and amplification of the top area are further proposed. A table of recommended values of overburden acceleration magnification factor is compiled. The research results can be used as a reference for future engineering design of dam foundation on deep bed overburden with soft soil layer.

In order to determine the evaluation method of lateral elastic foundation coefficient of the finite soil mass against the loading pile segment, based on the proportion coefficient of the foundation coefficient of the corresponding infinite soil mass, the laterally loaded beam-on-elastic-foundation method for a stabilizing pile is adopted in combination with the upper bound limit analysis method under the consideration of the internal forces and deformation continuity of the whole pile element. Thus, a calculation method for the proportion coefficient of the lateral elastic foundation coefficient of the finite soil mass is provided. The relationship between the proportion coefficient with the slope angle of the finite soil mass is determined. The proposed method is verified by numerical simulation method for a practical example. In the case of various physical and mechanical parameters of soil mass and stabilizing pile, analysis results of the example show further that the proportion coefficient decreases nonlinearly with the increase in the slope angle of the local soil. Also, the length of embedded segment (below the slip surface) and the size of cross section of stabilizing piles, cohesion, internal friction angle, and unit weight of the local soil have effects on the proportion coefficient. But the influence of the unit weight is minor, other factors have significant positive correlations with the proportion coefficient.

Accurate prediction of ground deformation and its influence range based on the elastic-plastic stress-strain relationship of soil during and after shield tunnelling has important theoretical significance and engineering application value. On the basis of deducing the scope of plastic loosening zone of surrounding soil, this paper puts forward the conclusion that the angle between plane segment and horizontal plane of composite sliding surface produced during shield tunnelling and after leaving is 45°?φ/2 and 45°+φ/2 (φ is internal friction angle), respectively. The formula of safety factor Fs for the composite sliding surface in soil under both active and passive states is proposed. Moreover, the possible sliding surface is ascertained by the random search method, and the influence range of lateral disturbance in shield tunnelling and departure stage is quantitatively determined. In shield tunnelling, the surface uplift is caused by the shear displacement of the soil on top of the shield. The shear displacement formula is deduced to predict the surface uplift by controlling stress τ1, τ2, τ3 and the shear modulus Ge, Gp1, Gp2 at different stages in the direct shear test curve, as well as controlling stress τ1, τ2, τm and corresponding distances l1, l2, lm according to the τ-l curve. After the shield is removed, the ground settlement is caused by the consolidation of the soil resulted from the dissipation of negative excess pore water pressure responding to the soil unloading on top the shield. The shear strain of the soil according to the modified Cambridge model is used to predict the settlement. Taking Wuhan River-crossing Tunnel as an engineering case, the influence range of surface deformation, heave and settlement are calculated quantitatively. It is concluded that the influence range of transverse disturbance and heave of surface at different construction stages of shield tunnelling are in good agreement with the measured values, and the predicted value of unloading settlement is in accordance with the measured value of the total settlement.

The 2008 Ms8.0 Wenchuan earthquake had caused serious damage to structures in loess tableland region far away from the epicentre, the earthquake damage and the amplification effects of ground motion on local sites were significant. Using the typical loess tableland as the object of investigation, the dynamic response characteristics of loess tableland slope under the action of strong vibration were systematically studied by applying mobile ground motion observation arrays and numerical analysis. Based on large-scale shaking table test and numerical simulation, the change rules of surface acceleration responses of loess tableland with different thicknesses under different actions of Wenchuan earthquake were analyzed, and the potential effects on the earthquake damage were studied. The results show that: 1) The phenomenon of low predominant frequency corresponding to high peak ground acceleration (PGA) amplification at the top of loess tableland slope may be related to the ratio of slope height difference to incident wave length. 2) The loess tableland with a slope gradient ranging from 60o-70o has the most significant amplification effects on ground motion. The PGA on top of slope can be amplified by 2 times, the predominant period can be increased by 5 times. 3) The amplification effects on ground motion are predominant for thick loess tableland, PGA horizontal amplification coefficient potentially subject to an increase by 2 times and seismic intensity by 1 degree. 4) Along with the increase of loess thickness, the predominant period of response spectrum increases, so that buildings with natural period ranging from 0.7 to 2.0 seconds or greater than 3 seconds will undergo obvious horizontal amplification effects.

Based on long-term and continuous ground temperature observation data, this paper analyzed the temperature, thermal state, freeze-thaw cycle, artificial permafrost table and rate of change of block-stone embankment along the Gongyu expressway located in a high-temperature permafrost region with an average altitude of 4 260 m. The influence of asphalt concrete pavement and cement concrete pavement over underlying permafrost is also studied in order to evaluate its applicability. It is found that the asphalt concrete pavement absorbed more heat and warmed up the underlying permafrost, which leaded to the rapid degradation of permafrost. During the observation period, the permafrost degradation rate in the centre of the block-stone embankment with asphalt concrete was 33.5 cm/a, almost five times that of the natural foundation. Moreover, the effect of shady-sunny slope was serious. The degradation rate of the permafrost of the sunny shoulder was 30.00 cm/a, which was significantly higher than that of shady shoulder (22.0 cm/a). Compared with asphalt concrete pavement, cement concrete pavement had higher thermal reflection and smaller thermal radiation absorption, which is conducive to raising the permafrost table or slowing down the degradation rate of the permafrost, and the cooling effect is better. However, during the observation period, it was found that the area growth rate of melting plate under asphalt concrete pavement in block-stone embankment in warm permafrost region was 12.24 m2/a, while that under cement concrete pavement was 9.28 m2/a, indicating that the area of melting plate always increases at different rates. Therefore, the existence of block-stone layer and the change of pavement type have not completely solved the heat balance problem of expressway embankment in warm temperature permafrost region. It is suggested that reinforcement measures or composite embankment structure should be adopted.

Based on the theory of porous media and the generalized thermoelasticity model, the propagation characteristics of Rayleigh wave (R wave) in saturated porothermoelastic media are studied in this paper. A thermal-fluid-solid elastic wave coupled model of saturated porous medium is established by considering the fluid-solid coupling wave equations for saturated porous medium, continuity equation and the generalised thermoelastic basic equations. By introducing the potential function and combining the boundary conditions of free water permeability and adiabatic, the diffusion characteristic equation of R wave in saturated porous thermoelastic medium is derived theoretically. The effects of thermophysical parameters such as porosity, permeability coefficient, thermal expansion coefficient, initial temperature and relaxation time on the wave velocity and characteristic attenuation of R wave are analyzed by numerical examples. The results show that the increase of porosity causes the decrease of R wave velocity, but results in an increase in its characteristic attenuation. With the increase of the permeability coefficient, the R wave velocity first remains constant, then increases sharply, and finally stabilizes; and its characteristic attenuation first increases and then decreases, and finally tends to be stable. The increase of thermal expansion coefficient causes the increase of R wave velocity, but has no effect on the characteristic attenuation; the increase of initial temperature leads to slight increase of R wave velocity, but has little impact on its characteristic attenuation; the relaxation time has almost no effect on both the velocity and the characteristic attenuation of R wave.

Accurate inference of the probability distributions of geotechnical parameters is a crucial step for reliability analysis and risk assessment in geotechnical engineering. At present, the probability distributions of geotechnical parameters are mainly inferred based on in-situ and/or laboratory test data. This paper aims to propose an adaptive Bayesian updating approach for the probability distribution inference of geotechnical parameters, in which a quantitative termination strategy for subset simulation is presented. Moreover, a framework for the inference and reliability analysis of the probability distributions of geotechnical parameters is constructed. The effectiveness of the proposed approach is verified by taking the landslide on No. 3 Freeway in Taiwan and a saturated clay slope as examples. Finally, the influence of the number of samples in each subset simulation level on the inference of probability distributions is addressed in this paper. The results indicate that, in comparison with the maximum likelihood and Markov chain Monte Carlo methods, the proposed approach is more efficient in calculation, simpler in programming, and can provide an effective way to solve the problem of probability distribution inference of geotechnical parameters at low acceptance probability levels. The number of random samples in each subset simulation level has certain influence on probability distribution inference. As the number of samples in each level increases, the posterior statistics of geotechnical parameters and threshold of subset simulation gradually converge. In addition, the rationality of the established quantitative termination strategy for subset simulation can be verified according to the variation of complementary cumulative distribution function with the subset simulation threshold.

Real-time measurement of deep three-dimensional geostress field distribution and disturbance law are very important for the engineering analysis of rock mass stability. However, there lacks the mature testing technology and supporting instruments for measuring the ground stress in weak surrounding rock at depth. The article firstly introduces in details the fundamental principles and technology of the rheological stress recovery method of the ground stress test for deep soft rock. Then in-depth theoretical analysis is carried out to study the rationality of sensing units’ layout of the rock mass pressure sensor. The results show that if the three sensing surfaces of the upper hemisphere of the six-direction pressure sensor are set to be perpendicular to each other’s normal direction, the three sensing surfaces of the lower hemisphere cannot be arranged perpendicular to each other’s normal direction; if the three sensing surfaces of the upper hemisphere are set to be perpendicular to each other’s normal direction, the three sensing surfaces’ normal direction of the lower hemisphere cannot be parallel to the three sensing surfaces’ of the upper hemisphere, and two of the three sensing surfaces of the lower hemisphere cannot have their normal directions perpendicular to any of the sensing surfaces’ normal direction of the upper hemisphere, three sensing surfaces of the lower hemisphere cannot be close to perpendicular to each other, and the cosine of the normal direction and the x, y, z axes are preferably not much different. If the three sensing surfaces of the upper hemisphere are not perpendicular to each other, a layout design is considered with the three sensing surfaces of the upper hemisphere having their normal directions set at an angle of 45° to the horizontal plane and those of lower hemisphere set at an angle of 30° to the horizontal plane, and the angle of the sensing surfaces’ the normal directions of the upper and lower hemispheres projected on the Oxy horizontal plane to be 60°. The calculation results show that this arrangement is one of the arrangements that are scientifically sound.

A new multi-functional triaxial testing system for hydrate-bearing sediments was developed. The system was consisted of a gas charge/discharge module, a stress loading module, a temperature control module, a data acquisition module, and an auxiliary module. High-pressure liquid pumps were used to generate confining pressure and axial pressure under the modes of constant pressure, constant flux, gradient pressurization, and gas pressure tracking. Therefore, the syntheses of differently conditioned hydrate-bearing sediments and triaxial shearing of samples could be achieved. The function of confining pressure tracking gas pressure kept the net confining pressure unchanged during the hydrate formation process, which ensures the consistency of initial conditions of the samples. The volumetric deformations of the samples could be measured by the volume change of the confining liquid, and the gas pressure could be controlled by adjusting the back pressure regulator. Hence, the decomposition process of hydrate-bearing sediments under different pressure reduction rates could be modelled. Triaxial specimens were made of clayey fine sand and carbon dioxide. The hydrate syntheses, triaxial shearing tests, and decomposition tests were carried out to verify the reliability of the testing system.

The shape and size characteristics of underground cavity are of great significance to the stability calculation and evaluation of geological structure in the region. However, the borehole sonar currently used lacks universality and it is difficult to meet the requirements of both detection accuracy and detection range. Therefore, on the basis of traditional borehole sonar method, this paper proposes a new method of detecting cavity based on multi-frequency borehole sonar. With considerations given to the contribution of different frequencies to the actual detection object, this method synthesizes the detection data of various frequencies and effectively solves the problem of the opposition between detection scale and detection accuracy. Firstly, the optimal frequency expression of the detection system is derived to provide basic parameters to reflect the influence of different frequencies on the analysis of detection results to a greater extent, and the frequency influence factor parameter is established to describe the deviation degree between the natural frequency and the optimal frequency. Then, on the basis of superposition of the frequency influence factor, the detection data of various detection frequencies are synthesised, and detection objects are accurately measured by using the improved spectrum ranging method. Finally, after adding depth and azimuth information, the three-dimensional reconstruction of cavity shape is realized. The feasibility and accuracy of the method are verified by applying it to practical engineering projects.