It is of great significance for engineering construction and environmental protection to improve saline soil by curing technology to make it an excellent roadbed filler. In this paper, taking light solidified saline soil as the research object, laboratory tests were carried out by using true triaxial apparatus. The effects of the ratio of intermediate principal stress, salt content and confining pressure on the stress-strain relationship, failure strength and shear strength index were analyzed. The strength failure mode of light solidified saline soil was revealed and the calculation method of strength index was established. The test results showed that the slope of the stress-strain curve increases with the increase in the ratio of the intermediate principal stress and the confining pressure, but the salt content has little effect on it. The salt content has a significant effect on the failure stress. When the salt content is at 4%, the failure stress reaches the maximum. Under different conditions, the light solidified saline soil could produce different strength failure modes of lateral expansion, taper, oblique crack, and diagonal. Through analysis and calculation, the prediction formula of failure strength and shear strength index of light solidified saline soil considering the ratio of intermediate principal stress and salt content was established. The comparison showed that the Mohr-Columb strength criterion is conservative in describing the strength of light solidified saline soil and it tends to underestimate the actual strength of the soil. The research results can provide references for engineering practice.

In this paper, a series of triaxial tests under coupled cyclic-seepage loads was carried out for saturated soft clay in Tianjin. The results show that the development of cumulative plastic strain is characterized by three stages: initial instantaneous growth, decelerated increase and stable / linear development, and seepage can enlarge the dynamic deformation up to 1?2 times of that compared to cyclic load only. Larger seepage force induces greater cumulative plastic deformation. The lower frequency or greater cyclic stress amplitude induces larger strain. The prediction model of cumulative plastic deformation of soft clay is established under the condition of dynamic-seepage coupling. The presence of seepage induces greater inclination to strain axis of hysteric curve at initial vibration. The dynamic elastic modulus of soft clay increases firstly and then decreases, and the larger seepage force induces lower modulus; the mathematical relationship between dynamic elastic modulus and cumulative plastic strain is revealed: under seepage condition, a prediction model of dynamic modulus was proposed considering the influences of seepage force and frequency. The damp ratio decreases to a constant value with increasing number of cycles. The larger the seepage force is, the larger the damping ratio attenuation amplitude is, and the damping ratio is approximately 0.02?0.04 at the end of the vibration. The results could provide guidance on the numerical simulation of dynamic characteristics of soft clay ground under seepage condition.

Skirted suction caisson is a new type of suction foundation for offshore wind power engineering. Model tests were carried out to investigate the variation of cumulative rotation angle and its influencing factors of the regular suction caisson (RSC) and skirted suction caisson (SSC) under horizontal cyclic loading. The cyclic loading mode includes one-way cyclic loading and variable-amplitude cyclic loading. The experimental results showed that the cumulative rotation angle of the RSC and SSC under horizontal cyclic loading mainly occurs within the first 200 loading cycles. The cumulative rotation angle was found to increase with the increase of cyclic loading amplitude and the number of loading cycles. However, the increasing rate of cumulative rotation angle decreases with the increase of loading cycle number. The relationship between the cumulative rotation angle and the number of loading cycles can be well fitted by a power function. By using Leblanc method and Miner’s rule, the cumulative rotation angle of the suction caisson foundation under long-term variable-amplitude cyclic loading was transformed into that under constant amplitude cyclic loading and then can be estimated. It was found that when the loading amplitude increases step by step, the predicted cumulative rotation angle of the RSC and SSC obtained by using the Leblanc method and Miner’s law is slightly higher than the measured model test results, indicating that the sequence and amplitude of cyclic loading have a certain influence on the cumulative rotation angle of the suction caisson foundation. In addition, the influence of cyclic loading on the ultimate bearing capacity of suction caisson foundation was studied and it was found that the ultimate bearing capacities of the RSC and SSC after cyclic loading are higher than those before cyclic loading. The research results can provide a basis for the design of the suction caisson foundation of offshore wind power.

Controlling of surrounding rock is very important for roadway rockburst prevention. Based on occurring conditions of rockbursts and mechanical properties of bolt and aluminum foam, a new support technology using combined bolt-aluminum foam support is proposed to prevent roadway rockburst. To study the supporting mechanism of combined bolt-aluminum foam in rockburst prevention, the mechanical properties of combined bolt-aluminum foam supporting samples were experimentally studied based on the similar simulation theory. The research results suggest that: the uniaxial compressive strength of samples is directly proportional to the bolt supporting density, and the relationship is positively exponential with the supporting thickness of aluminum foam; both bolt supporting density and aluminum foam supporting thickness relationships are linear positive with the consumed energy of complete samples destruction, negatively linear with remaining energy coefficient of samples, and negative logarithmic with elastic energy index and impact energy index of the samples. Therefore, impact failure of the samples is reduced by the increase of bolt supporting density and aluminum foam supporting thickness. The research results provide a theoretical basis for roadway rockburst prevention with combining bolt and aluminum foam.

Based on Biot’s porous media theory of saturated fluid, the wave function expansion method and Graf addition theorem are used to solve the vibration isolation problem of plane SV wave by periodically arranged pipe piles in saturated soil. In this method, the scattering problem of SV wave by infinite periodic piles in saturated soil is transformed into solving only one periodic element by using the characteristic that the scattering wave field of different periodic elements in saturated soil is only different in frequency domain. Therefore, it can solve the problem of SV wave scattering by infinitely periodic pipe piles in saturated soil into solving only one periodic element limit. On the basis of verifying the correctness of the method, the influence of the number of piles, the parameters of saturated soil, the thickness of pipe wall, the number of piles and the arrangement mode on the vibration isolation capacity are discussed. The results show that: (1) the vibration isolation capacity of periodically arranged pipe piles in saturated soil is better than that in elastic soil, and the vibration isolation ability is stronger in saturated soil with poor permeability; (2) the thickness of pipe wall should not be too thin, and when the ratio of inner diameter to outer diameter is about 0.6, the vibration isolation capacity of periodically arranged piles can be stable at most frequencies; (3) increasing the number of piles can significantly improve the vibration isolation capacity, which can be increased by about 20% in the low frequency range of dimensionless frequency 0?0.6, and 60% in the medium and high frequency range of dimensionless frequency 0.6?3.0; (4) the vibration isolation effect of rectangular arrangement of piles is better than that of plum shaped arrangement in the low frequency range of dimensionless frequency 0?1.0, and the other two arrangements have good shielding effect.

To study the uniaxial compressive strength calculation method of rock-steel fiber reinforced concrete (R-SFRC) composite layer, uniaxial compression test was carried out on rock, steel fiber reinforced concrete and R-SFRC composite layer specimens. The influence of concrete strength grades (C30, C40 and C50) and fiber contents (0, 40, 60 and 80 kg/m3) on the uniaxial compressive strength of steel fiber reinforced concrete and composite layers was analyzed. RFPA2D was utilized to simulate the damage process and stress-strain curve of the composite layer under uniaxial compression. The compressive strength prediction model of R-SFRC composite layer was established based on Mohr-Coulomb yield criterion. The results showed that the uniaxial compressive strength for composite layer specimens was between the compressive strength of rock and concrete. The mutual restriction of rock and concrete interface in the composite layer changes the stress state of each layer. The strength of rock in the composite layer decreases while the strength of concrete increases. The ultimate compressive strength of composite layer is the strength of concrete in the composite layer. The compressive strength of composite layer specimen increases with increasing concrete matrix strength and steel fiber content, and effect of concrete matrix strength was more significant. For the uniaxial compressive strength of composite layers of different materials, the error ranges of the numerical simulation value and theoretical calculation value relative to the experimental value are ?5.41%～?0.69% and ?8.67%～?1.21% respectively. Numerical simulation and theoretical calculation models can be used for uniaxial compressive strength prediction of composite layers.

To solve the issues of insufficient early strength of cement solidified muddy clay and high pollution, high energy consumption and high cost induced by the application of cement binder, the one-part slag-fly ash (SL-FA) based geopolymer was adopted as the binder to solidify the muddy clay. The effects of the ratio of slag to fly ash in silicon-aluminum (Si-Al) raw materials, the ratio of solid activator to Si-Al raw material and the ratio of water to binder on the mechanical properties of geopolymer solidified muddy clay were studied through the unconfined compression test. The micro-analysis (i.e., scanning electron microscope and energy dispersive spectrometer) was also conducted to investigate the development of the microstructure of the muddy clay solidified by the one-part SL-FA based geopolymer. The experimental results showed that when the Si-Al raw material was composed of 90% slag and 10% fly ash by mass, the mass ratio of alkali activator to Si-Al raw material was 0.15, and the mass ratio of water to binder was 0.7, the 14-day compressive strength of the geopolymer solidified muddy clay could reach 1.5 MPa. It was also found that the main hydration products of the slag-fly ash based geopolymer were amorphous hydrated calcium silicate hydrates and hydrated calcium aluminate hydrates, which is the main reason for the improvement of the compressive strength of solidified clay. The micro-structure of geopolymer solidified muddy clay was connected by hydrated calcium silicate hydrates and calcium aluminate hydrates. The results provide a certain theoretical basis for the practical application of one-part slag-fly ash based geopolymer as the soil binder in the improvement of muddy clay.

To explore the feasibility of resistivity method in monitoring the drying shrinkage of heavy metal contaminated purple soil, zinc, which is the most common and most abundant contaminant, was selected based on two-electrode method. In this study, the influence of zinc content on evaporation, shrinkage and resistivity during drying shrinkage was analyzed, and a resistivity model was deduced. The results showed that according to the evaporation amount, the drying shrinkage process can be divided into three stages: constant rate, variable rate, and residual. In the stage of variable rate stage, the evaporation rate firstly increases and then decreases, and its peak value increases with the increase of zinc content. The higher the zinc content is, the slower the shrinkage increases in the early stage; the earlier it stabilizes in the later stage, the smaller the final shrinkage is. The resistivity increase slowly at first and then rapidly. The higher the zinc content, the slower the resistivity increases in the early stage, but the faster it increases in the later stage. The resistivity model is expressed by evaporation and shrinkage, and it is related to zinc content within a certain range. The zinc content affects the evaporation rate at different stages, inhibiting the shrinkage deformation, and has an increasingly stronger effect on the resistivity. The model illustrates the relationship between them, which provides a basis for measuring the drying shrinkage deformation of purple soil through the resistivity method.

In order to study the effect of rockfall shape and size on the coefficient of restitution (COR), ten test blocks with different typical shapes and five test blocks with different sizes were adopted to conduct rockfall impact tests by a specially designed device. Based on the impact dynamics theory, the analysis of test results show that the normal COR (Rn) is very sensitive to the change of the block shape, and the Rn values of the non-angular shaped blocks are significantly higher than that of the angular shaped blocks. The Rn values of five non-angular shaped blocks are spherical > ellipsoid > cone with spherical base > cylinder > disk, and those of five angular shaped blocks are icosahedron > cube > tetrahedron > cuboid > triangle. Considering the geometric configuration and surface area distribution characteristics of the blocks, a new comprehensive shape quantification index-shape factor (?) is introduced, which has strong linear positive correlations with the Rn values of non-angular and angular shaped blocks respectively. Compared with Rn, the effect of the block shape on the tangential COR (Rt) is relatively weak, and the overall Rt values of the angular shaped blocks are slightly larger than that of the non-angular shaped blocks. Moreover, the size of the block has a significant impact on Rn and Rt. There are good linear negative correlations between the test block size and Rn and Rt. The rules obtained in this study provide important references for accurate prediction of rockfall disasters.

This project planed to investigate the brittleness index of coral reef limestone which is an important property of rock materials. For this purpose, physical property test, uniaxial and triaxial compression tests were performed on four types of reef limestone core samples (framestone, bindstone, rudstone and bioclastic limestone) which were sampled in a certain depth from a certain coral reef in South China Sea. The test results show that the physical and mechanical properties of reef limestone are strongly influenced by the fabric and diagenesis. The failure characteristics of reef limestone can be accurately described by the brittleness index which is calculated based on the relative magnitude and the absolute velocity of post peak stress dropping. The investigation also suggests that the brittleness index is sensitive to the confining pressure and decreases with the increase of confining pressure. Meanwhile, under uniaxial compression, the brittleness index of bioclastic limestone with low-porosity is similar to that of cement mortar, and higher than that of framestone and bindstone. Furthermore, with the increase of porosity, the brittleness indexes of rudstone and bioclastic limestone decrease, and the saturated uniaxial compressive strength is negatively correlated with porosity.

In recent years, surface tilt deformation of soil landslides has become an important indicator of landslide monitoring and early warning, but few studies have been done on the relationship between the slope surface tilt deformation and development characteristics of landslides. In this paper, a series of rainfall model tests combined with field tests was carried out to explore the relationship between the displacement and tilt deformation of slope surface in the process of landslide disaster evolution. A landslide warning method based on the curve of slope surface tilt deformation with time was established. The test results showed that the surface tilt deformation of soil landslides accelerates in the failure stage. There is a linear relationship between the reciprocal of tilt rate and the residual time of landslide before the failure, which can be used to predict the final failure time of landslides. In addition, in the process of landslide sliding, the surface tilt deformation and displacement of the landslide also show a linear correlation, and the corresponding slope is consistent with the sliding radius. The findings reveal, for the first time, the relationship between the slope surface tilt deformation and the geometric characteristics of the sliding zone, also provide the experimental basis for the prediction method of landslide sliding zone based on slope tilt surface deformation.

Heated pipe transportation is one of the most common used ways of offshore oil and gas transportation, and the mechanism of pipe-soil foundation interaction is the key to the controllable thermal buckling design of heated pipes. The change of pipe temperature not only causes the strong thermo-hydro-mechanical coupling response of the surrounding seabed saturated soil foundation, but also induces the migration of pore water under the temperature gradient, showing a significant thermo-osmosis effect. In this paper, a numerical study is presented to study the heated pipe-saturated soil foundation interaction considering thermo-osmosis (T-O) effect based on the OpenGeoSys finite element platform. Based on the mass equation and energy equation, the theoretical expression of the influence factors of thermo-osmosis effect was derived and the critical influence factors were proposed, which determined the critical values of T-O coefficient for some common soil and water parameters. Through the secondary development in OpenGeoSys embedding the T-O effect, a heated pipe-saturated soil interaction model, which can consider the thermo- hydro-mechanical coupling behavior and the T-O effect of soils surrounding the pipe, was established. The validity of the model was verified by comparing with the analytical results, and then the effect of T-O on the pipe-soil interaction was discussed based on the influence factors of T-O effect and numerical results, including the evolutions of pore water pressure and vertical displacement of seabed soils around the pipe with different oil transportation temperature and different initial buried depth during the operation period. Results showed that the critical value of the T-O coefficient is 4.3×10?12 m2/(s·K), which provides a basis for evaluating whether the T-O effect is considered in practical engineering. The T-O effect has little influence on the peak value of pore pressure of soil, but it can cause significant stable-state negative pore pressure. The proportional relationship between the amplitude and the peak value of the negative pore pressure is determined by the proposed influence factor of T-O effect. During pipeline operation, temperature often induces soil weakening and negative pore pressure, which have adverse effects on pipe stability. This adverse effect is enhanced with the increase of oil transportation temperature and depth within 3 D (D is the diameter of the pipe), but there is no significant difference in this adverse effect caused by T-O with depth above 3 D.

The roughness of joint surface has a significant influence on the shear behavior and hydraulic behavior of rock mass. In order to acquire a three-dimensional roughness metric which can capture the anisotropic characteristics of rough surface and reflect the shear mechanism of rock joint, morphological scanning tests were performed on four groups of sandstone joints by using a three-dimensional laser scanner. Based on Grasselli’s roughness model, a new three-dimensional roughness metric was proposed. Comparisons between another three models (Grasselli’s model, Liu’s model and Tian’s model) and the new model show that the new model is more consistent with the test results. In addition, a new concept that characterizes the density of the apparent dip angle facing the shear direction is established in this study, and it is found that Liu’s model conforms to the Weibull distribution. Compared to the theoretical density function of Grasselli’s model, Liu’s model, and Tian’s model, the new model has a better agreement with the experimental data, which verifies the rationality of the new model. The new roughness metric is simple and clear physically since it only takes the apparent dip angle facing the shear direction into account. The new roughness metric is positively correlated with the roughness of joints. Furthermore, the anisotropy of surface roughness can also be well characterized by the new roughness metric.

Saline soil contains a large amount of soluble salts, and the change of water content and suction has an important effect on its engineering properties. At present, few studies focus on the water retention characteristics of saline soil, and there is no uniform and clear soil water characteristic curve of saline soil in the full suction range. In this paper, the pressure plate method, vapor equilibrium method, and chilled-mirror dew-point technique were used to measure the matric suction and total suction of saline soils. The influences of compaction degree and salt contents on the water retention characteristics of saline soils were studied by experiments. The experimental results showed that in the low suction range, with the increasing of compaction degree, the saturated volumetric water content of the soil sample decreases and the value of air-entry suction increases. The water retention capacity increases with the increase of compaction degree and pore solution concentration, and the pore solution concentration has an obvious influence on the matric suction especially for the soil with low compaction degree. In the high suction range, for the soil with high salt content, the compaction degree has a significant influence on the soil-water characteristic curve, and the water retention capacity of soil increases with the increasing of salt content.

The seismic permanent displacement of cantilever retaining walls are affected by the potential slip failure characteristics of the backfill above the heel plate. In this paper, solutions of the permanent displacement are derived based on the quasi-static approach, upper bound limit analysis and Newmark sliding block model. The proposed method involves three possible wall displacement modes comprising overall rotation with log-spiral curve, horizontal translation along the wall base and rotation around the wall toe. A worked example indicates that the seismic permanent displacement obtained by the first two modes are similar and much higher than the third. Thus, the first two modes are more crucial for engineering project design. Compared with existing empirical formulas, relative error of the proposed method is below 30%. Besides, the proposed result is 7% less than that by Ambraseys-Menu’s method with probability confidence level of 0.7. For the overall rotation mode of the wall-slope system under a specified horizontal seismic action, sensitivity analysis of ten main parameters’ influence on the horizontal yield acceleration is conducted and the order of these sensitivities is recommended.

It is usually difficult to determine the spatial form of three-dimensional passive slip surfaces behind retaining walls. Based on numerical simulation, this paper assumes friction angle ratio of the wall-soil contact surface to be zero (δ/? = 0), and uses the thin-plate smoothing spline function to search for the three-dimensional slip surface at the end of retaining wall under different internal friction angles of soil mass. By analogy with the failure of foundation bearing capacity, an equation is proposed to curve fit the three-dimensional slip surface, and the equation for three-dimensional slip surface at the end of the retaining wall is summarized. For translational failure mode of rigid retaining walls with upright wall back, horizontal filling level and non-cohesive soil, δ/? = 0, the volume of sliding body behind the wall is calculated based on the three-dimensional slip surface equation. The calculation of three-dimensional passive earth pressure is deduced, and verification is carried out. The results are obtained as follows: (1)Compared with the Soubra passive earth pressure coefficient, the three-dimensional earth pressure coefficient obtained by the proposed method in this paper is closer to the numerical simulation result; (2)There is a significant difference between the three-dimensional passive earth pressure coefficient and Rankine passive earth pressure coefficient when the length-to-depth ratio of the wall is less than 4.0. As the length-to-depth ratio increases and the soil internal friction angle decreases, the three-dimensional passive earth pressure coefficient reduces to the Rankine’ value, and the position of the three-dimensional passive earth pressure resultant action point tends to be the position of the Rankine’s point.

To use seawater as the raw material in the reef foundation reinforcement based on microbial induced calcite precipitation (MICP), seawater concentration test and MICP sand reinforcement test with concentrated seawater as calcium source solution were carried out. In this work, the influence of bacterial fixation mode, bacterial injection batch, the urea concentration in cement solution and the injection velocity of cement solution on the reinforcement effect were studied. The research results showed that the maximum concentration of seawater is three times the original concentration under the condition of no precipitation of calcium ion, and the content of calcium ion is about 0.033 mol/L. The amount of urea added should be 3 times that of the content of calcium ion in concentrated seawater, so that the calcium ion in cementation liquid can be effectively used to produce precipitation. The best reinforcement effect can be obtained by using the cement injection velocity of 2 mL/min to reinforce 5 cm sand column samples. After reinforcement, the unconfined compressive strength of the sand column sample can reach 653 kPa, which takes 4.5 days. Increasing the batch of bacterial injection cannot effectively improve the reinforcement effect of the sand column.

Based on Barron’s equal strain consolidation theory of sand-drained ground and Gibson’s one-dimensional large strain consolidation theory, and considering the well resistance of drains, the variation of radial permeability coefficient and the vertical flow, a more general governing equation of large strain consolidation of sand-drained ground is established and solved by using the finite difference method. The correctness of the numerical solution is verified by comparing with the existing consolidation model and small-strain analytical solution. Using the numerical solution, the large strain consolidation behaviors of sand-drained ground are investigated. The analyses show that the well resistance of drains can reduce the consolidation rate of sand-drained ground. But when the permeability coefficient of drains increases to a certain value, the well resistance of drains can be ignored. The variation pattern of radial permeability coefficient has a great influence on the consolidation rate of sand-drained ground. The consolidation rate is faster under the parabolic pattern than that the linear pattern. The vertical flow accelerates the consolidation rate of sand-drained ground, when the radius ratio is small, and the influence of the vertical flow on the consolidation rate should be considered. The larger the ratio of compression index to permeability index is, the slower the consolidation rate of sand-drained ground will be.

This research develops a two-dimensional logarithmic spiral dynamic failure model to solve the problem of heading face instability of shield tunneling. By discretizing the dynamic failure model, the dissipation power of the system in the failure zone can be calculated accurately, and the upper bound solution of support pressure can be derived by using the upper bound method. Based on the current solution, the failure mode of instability and the corresponding limit support pressure at collapse were obtained by optimization. Through parameter analysis, the influence of soil strength on the failure mode and limit support pressure was analyzed. The results showed that with the increase of soil strength, the failure zone becomes smaller and the failure height and support pressure decrease correspondingly. The analytical solutions, including the failure mode and limit support pressure, obtained by the two-dimensional logarithmic spiral model were compared with those obtained by the numerical simulation and the existing analytical solutions. The comparative analysis showed that the analytical solutions by the proposed model are in good agreement with those from numerical simulation and other existing theoretical models. Finally, the model test was selected as an example to verify the model and the results showed that the solutions from the proposed model are close to the previous test results. The analytical solutions from the two-dimensional logarithmic dynamic failure model can provide a theoretical basis for engineering design in predicting the failure mode and limit support pressure.

In order to investigate the dynamic properties and energy dissipation of loess under passive confining pressure condition, a Φ100 mm split Hopkinson pressure bar (SHPB) apparatus was used to conduct dynamic compression tests on loess samples with different initial densities at different impact speeds. The validity of the experiment was verified by considering the stress equilibrium and constant strain rate, the dynamic stress-strain curve and absorbed energy time history curve of the loess sample were obtained. Experimental results show that when the strain rate is in the range of 330～620 s?1, the yield strength of the loess sample with the initial density of 1.7 g/cm3 and the moisture content of 4.45% is less affected by the strain rate, and the strain rate effect is not obvious. When the strain rate is about 445 s?1, the yield strength, dynamic peak stress and specific energy absorption of the loess sample with a moisture content of 3.77% increase with the increase of the initial density of the sample. In the incident energy range of 1 700～5 500 J, the energy dissipation rate of the loess sample with the initial density of 1.7 g/cm3 and the moisture content of 4.45% is about 27%, and the specific energy absorption increases with the increase of incident energy. The test results can provide technical reference for the construction and protection of military and civil engineering construction related to loess.

A series of 1g model tests was carried out using a motor servo horizontal cyclic loading equipment to study the stiffness and cumulative displacement of the large-diameter single pile in sand under horizontal cyclic loading. The test results indicated that the residual displacement generated by one loading and unloading cycle is about 80% of the peak displacement. As the number of cycles increases, the area of the cyclic loading hysteresis curve gradually decreases, indicating that the behavior of soil around the pile changes from elastoplastic to elastic stage. The secant stiffness of the hysteretic curve increases firstly and then decreases with the increase of the number of cycles, which is caused by the trend that the soil around the pile is gradually dense, and the resistance of the soil around the pile is developing from shallow to deep. The secant stiffness of the hysteretic curve increases firstly and then decreases with the increase of the number of cycles, which is caused by the progressive compaction of the soil around the pile in the shallow layer and the tendency of the resistance of the soil around the pile to transfer from the shallow layer to the deep layer. The cumulative displacement of the pile top decreases to approximately the same extent as the pile diameter increases. However, with the increase of burial depth, the reduction of displacement also gradually decreases, revealing the existence of critical burial depth. An empirical model of cyclic cumulative displacement is proposed by linear fitting in a double logarithmic coordinate system based on the exponential model. It is found that the effect of increasing the pile diameter on the reduction of cyclic cumulative displacement is better than that of increasing the buried depth.

Based on the changes of total stress and pore water pressure in the spring model, the mechanism of vacuum preloading consolidation is studied by theoretically discussing the consolidation behavior of soil in vertical and horizontal directions with different boundary conditions. The variation of effective stress and development of deformation are analyzed for the soil under vacuum preloading under stress paths of K0 consolidation and isotropic consolidation based on the boundary conditions, respectively. Meanwhile, the theoretical earth pressure coefficient (K0MC) at rest for soft soil is deduced based on modified Cam-clay model. With engineering applications, the correctness and effectiveness of the mechanism is confirmed by the application in large area of ground improvement for soft soil and statistical analysis of some engineering cases. It is found that the drainage boundary condition and the scale of treated area have effects on deformation of soil mass. For the middle part of the treated zone under large-area vacuum preloading, the variation of effective stress is not isotropic, but following the pattern of one-dimensional compression. According to the statistic results from some site investigations, one-dimensional compression equation (Cc-Cr method) can be applied to predict the settlement at the center of the treated zone when the one dimensional condition is found in the center of treated zone as the ratio between treated breadth and length of vertical drains larger than 4?5.

The construction of compressed air energy storage power stations in salt caverns is one of the important means to solve the problem of wind/solar energy generation, grid connection and grid peak regulation. In recent years, in order to make full use of the bedded salt formations in China, the option of building horizontal caverns has been proposed. However, horizontal caverns bring us not only more cavern space but also larger roofs, which may become a gas leakage channel due to deformation, failure and even cracking. In this study, the argillaceous anhydrite layer above bedded salt formations is identified as the key roof which controls the stability of horizontal cavern, and its stress state and possible types of instability are analyzed. Additionally, based on the linear planning method, a feasible range of internal pressures and the depths of the cavern for compressed air energy storage are determined. Finally, stability and influent factors are analyzed in detail by numerical simulation. The results reveal that (1) the maximum operating internal pressure is the main determinant of the stability of key roof. The maximum internal pressure should be less than 75% of the original gravitational stress, but not less than 60%. (2) The increase of the cavern dimensions will enlarge the displacement and plastic zone of the key roof, and change the volume shrinkage of the cavern, which requires comprehensive optimization. (3) Increasing the thickness of the protective salt layer can reduce the subsidence and the plastic zone volume of the key roof but increase the volume shrinkage, which is also not conducive to the full utilization of bedded salt resources.

Aiming at the stability analysis of multistage homogeneous loess slopes, a systematic analysis method is proposed. Firstly, the possible failure mechanisms such as slope toe circle and midpoint circle under the overall failure mode and local failure mode of multistage loess slopes are established. Then, based on the improved limit equilibrium method, an analytical solution of slope safety factor corresponding to each failure mechanism is derived, and the optimization method is used to determine the minimum value of safety factor corresponding to each failure mechanism under the overall and the local failure modes. Finally, the flowchart of systematically performing multistage loess slope stability analysis is put forward to determine the minimum safety factor and the critical slip surface. The results of numerical examples show that the proposed method are comparative to the traditional stripe method. The relative deviation of the minimum safety factor calculated by the two methods is less than 5%, and the critical slip surfaces determined by the two methods agree to each other very well. The proposed method in this paper can be widely applied to various multistage homogeneous loess slopes and even single-stage slopes, which is promising for engineering applications.

The wall-well system could effectively decrease the discharge inside or drawdown outside the foundation pit that comprising suspended waterproof curtain and dewatering wells. Foundation pit is assumed to be a large partially penetrating well with the same area. Based on the assumption that flow into the well is equal to the vertical flow inside the waterproof curtain, an equation is established to calculate the discharge inside and drawdown outside the pit. The discharge calculated by the proposed equation is close to that by simulated results while L/M > 0.6 and r/M < 2.0. It is showed that the calculated discharge is slightly larger than those by numerical analysis or case history. This phenomenon may be attributed to the neglection of energy loss during the flow direction change. Then the influence of some factors including the ratio between horizontal and vertical permeability coefficients, the foundation area and the curtain length is studied. Parametric analyses show that the discharge inside or drawdown outside the pit can be determined by the degree of flow field approaching vertical flow. The proposed equation is suitable for preliminary analysis of decompression dewater when cutting off confined aquifer is difficult.

The hardening strain model with small strain (HSS) can appropriately consider the nonlinear and stress-dependent behavior of soils at small strain range. Thus, it is widely used for deformation predictions during pit and tunnel excavations. However, currently there is a lack of systematic study on how to reasonably and conveniently determine all the model parameters. This paper firstly introduces the definitions of all the HSS model parameters and their measurements. Based on the statistics of laboratory and field test data, the relationships between the model parameters and the void ratio of Shanghai soils are proposed, which can be conveniently used in engineering practice. Finally, the HSS model with parameters determined by the proposed method is used to analyze the deformation in four excavation cases. The results show that with the field measured small strain shear stiffness, the calculated lateral deformations of the retaining wall agree well with the measurements. It indicates that the proposed method for determining the HSS model parameters is appropriate and it can provide references for similar projects in Shanghai and similar regions.

Geological disasters such as landslides occur frequently in southern Jiangxi, mostly induced by rainfall and artificial slope cutting. Physical model tests, especially field model tests, are considered effective to reveal the mechanism of landslides. The case on a cutting slope featured with weathered metamorphic rock in southern Jiangxi is investigated, in which a self-designed rainfall simulation system is applied. The alignment of field instrumentation includes four boreholes for measuring water content as well as pore pressure and two monitoring points for displacement. An artificial rainfall was carried out on the site. The ground response in terms of water content of slope soil, pore water pressure, displacement from the rainfall process is investigated, and the infiltration manner of weathered metamorphic rock slope under continuous heavy rainfall is discovered. The deformation and failure mode of slope based on weathered metamorphic rock is concluded. The results show that there is a lag in the change of soil moisture content after rainfall. The moisture content performs continuously increasing with the infiltration of rainfall water, and the value decreases with depth. The log of pore pressure collected from the bore hole shows an obvious response to the rainfall condition in shallower layers, and the soil below the depth of 1 m stays unsaturated. The displacement at the monitoring point is obviously positively correlated with the rate of change of water content, and the slope deformation is mainly performed on the side of the slope where the water content changes significantly. The slope displacement shows obvious dependence on the change of the pore pressure. The dilatancy deformation of the soil inside the slope reduces the pore pressure value, resulting in the decrease of the rate of slope deformation. The deformation and failure process of a steep artificial cut-slope under heavy rainfall conditions can be divided into three stages: the scattering and slipping of the slope surface; the formation of tensile cracks on the platform; the overall collapse of the slope.

Tortuosity is an important parameter to study the permeability of soil, which reflects the degree of twists and turns of fluid flowing through porous media. Existing studies have shown that the tortuosity is not only related to porosity, but also closely related to particle arrangement of soil. Based on the new geometric tortuosity model of soil, a modified calculation model of tortuosity of soil unite is then proposed with emphasis on the coupling effect of porosity and particle arrangement on the equivalent flow length. The results showed that the proposed model agreed well with the experimental results in the literature, and the effects of particle arrangement and porosity on the tortuosity were further discussed. Furthermore, the minimum sample size needed for calculating the overall tortuosity of soil was determined under certain confidence and error limits based on the sampling reliability theory and this method has been proven to be reasonable and effective for representing the whole soil sample. Finally, the theoretical solution of soil tortuosity based on the proposed method was compared with the numerical simulation solution of soil tortuosity based on COMSOL Multiphysics, and the satisfactory agreement proved the reasonability and validity of the proposed method. The study provides an alternative approach for the calculation of soil tortuosity.

Gravels are widely found in the sliding zone soils (SZS) of ancient landslides. It is the key to determine the mobilized strength of SZS with gravels for the stability evaluation and prevention of ancient landslides. Taking the reactivation of Jiangdingya ancient landslide in Zhouqu County, Gansu Province in 2018 as an example, the mobilized strength of SZS with gravels was studied based on the test and back analysis. The results suggested that: (1) after long-distance shearing, the cementation in the SZS has been gradually lost, and the residual strength is mainly controlled by the sliding frictional resistance and occlusion between the soil particles. When the gravel content is higher, the interaction force between the soil particles is stronger, so the residual strength is relatively larger. There is a positive linear correlation between the friction coefficient and roughness of the shear surface. The gravel changes the roughness of the shear surface, which increases the friction resistance and leads to the increase of its residual strength. (2) The statistical analysis showed that the φr of SZS with gravels is controlled by both clay and gravel content, which is significantly different from the mechanism that the φr of SZS without gravels is mainly controlled by clay content. It is suggested that the ratio of gravel content to clay content should be used as an index to estimate the φr of SZS with gravels. (3) The mobilized strength of ancient landslides is generally greater than the residual strength, but slightly less than the recovery strength. Before reactivation, the strength of SZS has gradually attenuated from the recovery strength to the residual strength. At this time, the ancient landslide is in the creeping state as a whole. Under the action of external forces, the strength of SZS tends to decay sharply, which induces the accelerated sliding of the ancient landslide.

The parallel seismic method has been successfully used to detect the depth of the existing pile, but the effect of this method on detecting the pile defects is not clear. In this paper, a large-scale model pile experiment was designed and carried out to analyze the stress wave transmission corresponding to various defects, and the influence of test distance and excitation position on the stress wave transmission was further discussed. The results show that the parallel seismic method can effectively identify the serious defects when the test borehole is close to the test pile, and the increase of test distance will greatly increase the difficulty of detecting pile length and defect identification. The broken pile and the necking can cause the decrease of wave velocity and the prolongation of the first arrival time, and the abnormal characteristics of stress wave group corresponding to the slight defects are not significant. When the bearing stratum of the pile is hard bedrock, the sediment at the pile tip may lead to the second mutation of the slope for the fitting line of the first break wave. When the defect is close to the shortest transfer path of stress wave, the abnormal characteristics of first break wave are more obvious.