An RLW-2000 micro-computer servo-control rock triaxial testing machine was used to carry out triaxial cyclic loading and unloading experiments of saturated and natural shales under different confining pressures. The mechanical properties and hysteresis effects of two hydrated shales were analyzed. The evolution law of the damping ratio was discussed based on the energy principle. The experimental results show that as the cycle number increases, the cumulative residual strain increases gradually. The relative residual strain decreases first and then tends to a stable region, and finally increases rapidly prior to the rock damage. The deformation modulus of the saturated shale is smaller than that of the natural shale during the loading and unloading process. Besides, the loading deformation modulus is smaller than that of the unloading deformation modulus. In the loading and unloading stages, it is observed that the strain always lags the stress. The hysteresis effect of saturated shale is more obvious than that of natural shale. An energy calculation method is proposed considering the hysteresis effect with a smaller error of the calculated energy compared with that of the previous energy calculation. Finally, the calculation formula of the damping ratio is revised based on the energy principle. It is found that the damping ratio of saturated shale is larger than that of natural shale. The change of damping ratio reflects the shale damage mechanism, which can be used as an important indicator to predict the instability of shale.

The rock crushing process involves the variation of multiple variables (stress, strain, porosity, etc.) and generation, expansion and accumulation of cracks. It is an effective way to study the mechanism of rock consumption. In order to improve the rock granule model in the existing literature, the internal characteristics of rock materials should be considered. In order to characterize non-uniform distribution and accumulation of particles inside the studied rock, rock axial crushing experiment and rock lithofacies analysis were carried out. On this basis, a multi-scale cohesive particle model conforming to the internal characteristics of the real rock was constructed. According to the BPM theory of discrete elements, the mechanical relationship between the bonds of particles with different grain sizes in the multi-scale cohesive particle model was solved. It is found that the bond breaking criterion formed by the secondary particles is ≥ 2 GPa, and this criterion formed between the tertiary particles is ≥ 6 GPa. On this basis, an evolution model for simulating fragmentation of the particle model is established. By simulating the axial crushing experiment, the fracture evolution model can predict (i) the real-time changes of the bearing force on bond between the particles in the particle model from the mesoscopic point of view and (ii) the fracture sequence of the bond from the outside to the inside during the rock crushing process. It has a V shape extending from both ends of the upper surface and intersecting the middle of the rock. The rock crack characteristics simulated by the proposed method are found to be in a good agreement with the experimental results from rock axial crushing tests. The reliability of the model is verified, and the rock crushing process is analyzed from the microscopic and macroscopic perspectives.

To investigate the unloading deformation, failure and acoustic emission characteristics during excavation of rectangular roadway under different initial geostresses, excavation unloading model tests were carried out on casted surrounding-rock specimens made of cement mortar. The deformation and failure characteristics of the roof, corner and surrounding rock as well as the evolution characteristics of AE impact count, damage variable and frequency spectrum were obtained. Experimental results showed that: 1) The failure of rectangular roadway was mainly caused by the radial tension strain of the roof and surrounding rock, and the tangential compression strain of the corner. The deformation characteristics of the surrounding rock at the roof were similar to that at the side wall. 2) Increase of the initial geostress had significant effect on the radial strain rate of the roof (or the side wall) and the tangential strain rate of the corner while not significant on others. 3) The evolution characteristics of AE impact count and damage variable revealed the phased failure process of surrounding rock from micro-fracture initiation, expansion to macro-crack development until the occurrence of main fracture. The time of main fracture occurring may be relatively delayed as the geostress increased, and the relative damage proportion after excavation unloading increased nonlinearly with the increase of the initial geostress. 4) The phenomenon that the peak frequency of AE signal was concentrated with gradually increased amplitude could be regarded as a precursor information of the main fracture occurring in the surrounding rock. The higher the initial geostress, the wider the distribution of the peak frequency concentrated section. 5) As the initial geostress increased, the inner wall of surrounding rock specimens demonstrated the evolution of damage from micro-cracks to abscission layer, and to block fracture of rock mass. The failure modes of rock mass were dominated by the tensile heave in the roof and side wall, and accompanied by the phenomenon of delamination and fragmentation. Furthermore, compression and shear failure were observed at the corner when the specimens lost the bearing capacity.

When cohesive soil behind a retaining wall is in active earth pressure state, cracks will appear behind the top of the retaining wall, which leads to a wide region of zero pressure, namely cracking depth. The problem of the cracking depth behind a retaining wall has not been solved well. In this study, an example is taken to illustrate that cracks can appear on the surface of filling, because the cracks are not taken into account in the variational method that is used to solve the active soil pressure of cohesive soil behind the retaining wall. The Mohr-Coulomb strength envelope is simplified by a broken line and the calculation formula of the cracking depth of the soil behind the wall is derived by the actual soil tensile strength. According to the stress boundary state and geometric boundary conditions of the upper points on the slip surface, the variational method to calculate the active earth pressure variation has been improved, and the uncertainly model of the active earth pressure is transformed into a deterministic issue. Also, the influence of internal friction angle, cohesion and tensile strength on the crack depth is analyzed. With the increase of internal friction angle and cohesion, the crack depth of soil increases and the soil pressure gradually decreases, and the slip surface shifts towards the wall back. With the increase of the tensile strength of the soil, the cracking depth and the soil pressure both gradually decrease. When the tensile strength is strong enough to resist the tension destruction of the soil, the cracking depth of the soil behind the wall becomes zero, and the soil pressure is no longer affected by the tensile strength.

In order to reasonably analyze seismic stability of filling slopes retained by gravity walls anchored horizontally with flexible reinforcements, the quasi-static method combined with the horizontally and obliquely slicing limit equilibrium methods are adopted to derive the safety factor of the wall-slope overall stability based on the hypothesis of interslice forces proposed by the Fellenius and the simplified Bishop methods. Considering the tensile rupture and the potential pulling out failure mechanism of the reinforcements, the proposed method is verified through numerical simulations under various horizontal seismic effect coefficients. Then the influences of the internal friction angle, the cohesion of the backfill and foundation soil, the vertical spacing, the ultimate tensile force, and the length of the reinforcements on the seismic wall-slope overall stability are studied, respectively. The results show that the safety factor calculated by the proposed method with the simplified Bishop’s hypothesis is fairly identical with the numerical results, which is about 6% higher than that calculated by the proposed method with the Fellenius hypothesis. Besides, the safety factor increases approximately linearly with the increase of shear strength of the filling and foundation soil, increases nonlinearly with the increase of the ultimate tensile force and the length of the top reinforcement, but decreases nonlinearly with the increase of the vertical spacing of the reinforcements. Additionally, potential slip surfaces of the retained slope by the proposed methods are close to the numerical results.

A series of cyclic triaxial tests is conducted by GDS apparatus to investigate the effects of confining pressure, loading frequency, consolidation stress ratio on the hysteretic curves of the remolded weak expansive soil under multi-step loadings and single-step loadings respectively. Morphological characteristics of hysteretic curves are quantitatively described by three parameters, namely, the area S of the hysteretic curve, the slope of curve k, center offset distance d of adjacent hysteretic curves, and the degree of closure εp. The results show that the S, d and εp of the hysteretic curve of weak expansive soils increase nonlinearly with the increase of dynamic stress amplitude, while decrease with the increase of confining pressure, frequency and consolidation stress ratio. The k decreases logarithmically with the increase of dynamic stress amplitude, and increases with the increase of confining pressure, loading frequency and consolidation stress ratio. Hysteretic curves of weak expansive soil are approximately parallel under single-step loading. The parameters of εp, d and S decrease nonlinearly with the increase of loading cycle, which reflects the phenomenon of cyclic creep. The multi-step loading history has a slight influence on the hysteretic curve of weak expansive soil.

Large-diameter single pile foundations of offshore wind turbines are always subjected to complex scour and disturbance of currents, as well as the vertical load (V) and horizontal load (H). Based on the Swipe loading method, the test of V-H combined loaded piles in sand is designed and completed in flume. The load-displacement curves and the bending moment of the pile shaft under both static and flowing states with different vertical loads are obtained. By dimensionless processing and fitting, the envelope of the V-H combined bearing capacity with consideration of the current effects is obtained. The results show that the total friction resistance of the pile side decreases due to the scouring action caused by the currents, and thus leading to decrease in the initial stiffness of the vertical load-displacement curve. When merely subjecting to the currents, the horizontal displacement and bending moment of the pile shaft are relative small, but they show significant increments when simultaneously subjecting to the combined action of horizontal force and the currents. An increase in the vertical force V on the pile head helps to improve the horizontal bearing capacity of the pile when the horizontal displacement of the pile is small. But when the horizontal displacement is large and V approaches to the limitation of the vertical bearing capacity Vu, the horizontal bearing capacity of the pile will be decreased due to the second order effects of gravity(P-Δ). The position of the maximum bending moment of the pile shaft is not significant changed with the flow state and V, which generally appears at the locations of 2–3 times the pile diameter below the mud line.

The coarse-grained soil is widely used as the filling in the railway subgrade, which is directly subjected to the long-term effect of the dynamic cyclic train loads transmitted from the track structure. Investigation of its dynamic behavior and plastic deformation characteristics under the dynamic train loads may provide ideas for the subgrade performance evaluation and settlement control. GDS dynamic triaxial tests were used to investigate the dynamic response of coarse-grained soil in railway subgrade. By introducing the plastic strain rate and shakedown theory, the evolution law of axial plastic strain of subgrade under different frequencies, confining pressures and cyclic dynamic stress ratios is divided into three types: plastic shakedown, plastic creep and incremental plastic failure, based on which the critical dynamic stress levels of plastic shakedown and plastic creep state are determined. Results show that the critical cyclic stress ratio of coarse-grained soil increases with the increase of confining pressure, while decreases with the increase of loading frequency. Based on the fitting analysis of the experimental results, an empirical formula of critical dynamic stress with confining pressure as variable is proposed, which can provide theoretical basis for the dynamic stability evaluation of subgrade at any depth.

Backfill mining operations are increasingly being performed in cold regions or even permafrost areas. The water available for the preparation of mine backfill is often naturally saline. Further, in some extremely cold regions, anti-freezing agents are often used to avoid the freezing of the paste during the pipeline transportation. This study, the time-dependent evolution of uniaxial compressive strength and initial elastic modulus of cemented paste backfill (CPB) with different saline concentrations (NaCl) at ?6 ℃ was experimentally investigated. Uniaxial compressive tests were conducted on various CPB samples for different curing times (7, 28 and 90 days). The results show that the strength of CPB decreases with the addition of NaCl. All of samples experience an increase in strength with increasing curing time, due to the ongoing cement hydration. Moreover, there exists an obvious linear relationship between the uniaxial compressive strength and the initial elastic modulus, regardless of salinity, curing time and binder type. The results of this investigation could provide some technical information for the implications for backfill practices in cold regions.

Deep buried sand is a granular structure system, and accurate determination of its original mechanical properties is a difficult problem in geotechnical engineering. Pressuremeter test (PMT) is an ideal in-situ deep-seated testing method for deep buried sand, but there is a lack of quantitative evaluation criteria for the depth effect. In order to solve the problem of depth effect of PMT, a self-developed physical model testing system was applied to simulate the stress state of deep buried sand. By applying different overburden pressures to simulate the PMT at different depths, the variation rules of PMT modulus EM and PMT plastic pressure Pf as a function of overburden pressure are obtained, and a method is suggested to revise the characteristic value of the bearing capacity of foundation fak based on the PMT plastic pressure Pf. The test results show that the depth effect does exist in the PMT. With the increase of test depth, the burial overburden pressure increases, and the change of EM is not obvious. Also, Pf is linearly correlated with . This study provides reference and basis for the determination of physical and mechanical properties of deep buried sand.

To clarify the bearing characteristics of bolt-grouting support in shallow fractured surrounding rock of roadway, comparative bolt-grouting bearing capacity test was carried out with different rock particle sizes, lithologies, and number of bolts. The results show that: 1) the bearing capacity of the bolt-grouting increases first and then decreases with increasing the particle size. Compared with the non-anchored specimens under the same conditions, the peak stress of the anchored specimens increases by 53.38% on average, and the peak strain decreases by 46.43% on average. 2) Under anchored condition, the free surface opposite to the supporting surface and its vicinity are the preferred area for macro-fracture development. The failure of supporting surface generally lags behind the free surface, and the specimens gradually transit from mainly tensile failure to tensile-shear mixed failure; tensile failure is the main failure mode of specimens under the condition of no anchor. 3) With the increase of bolt number, the bearing capacity of bolt-grouting increases gradually, but the growth rate slows down gradually. The peak strain slows down, which is consistent with the inflection point of peak stress growth. 4) The specimens have three characteristics of progressive destruction: firstly, the cracks generated before peak stress continue to propagate after peak stress; secondly, the surface material of support surface spalls with the crack stretching; thirdly, the cracks gradually develop from surface to inside of the specimens, the rock fragments fall off first, and then the macro-damage develops to the anchor area, resulting in the loss of overall specimen bearing capacity.

Microstructure change of structured clay under different stress paths is the key in understanding the intrinsic mechanism of structure variation of structured clay in K0 consolidated tests. Based on the triaxial shear tests under different stress paths for the natural clay in Hengshui area, the micro-structure of soil under different stress paths were compared and the mechanism of micro-structure variation of structured clay was studied from the microscopic perspective. The results show that variation of pore size, orientation of particles and pores is large, while the variation of particle and pore shape is small under different stress paths. The increase of spherical stress leads to the consolidation of soil particles, and the soil pores are compressed, resulting in the decrease of soil volume. The soil particles loose, the pore size and soil volume increase when the spherical stress decreases. The increase or decrease in deviatoric stress has similar impact on soil microstructures, which results in the compression of soil skeleton before the failure of soil structure. After the failure of soil structure, the soil particles are misaligned. They rolled and connected to each other, which leading to the expansion of inter-grain pores and dilatancy of soil occurs. The spherical stress and deviated stress caused by irregular shaped soil particle form mutual effects on volumetric strain and shear strain. The results reveal the microscopic mechanism of stress-strain characteristics of structured clay under different stress paths.

To investigate the mechanism of water and mud inrush in water-rich fault fracture zone, a large-scale indoor water and mud inrush test system considering mass transfer and crustal stress state is developed. The simulation test of water and mud inrush disaster evolution process in fault fracture zone under different hydraulic loading modes and medium parameters of fracture zone are carried out by using the device. Some findings are as follows. 1) Evolution of water and mud inrush disaster in fault fractured zones is a strong coupling process of seepage and erosion. Fine particles in the filling of fracture zones first migrate under the water pressure. With the continuous migration and loss of fine particles, the flow pattern changes from laminar flow to turbulent flow, which eventually leads to water and mud inrush disaster. 2) The larger initial porosity of filling in fractured zone and the higher of applied water pressure will induce the water inrush more easily. As a result, the evolution characteristics of seepage exhibit more obvious in the test, the increase of seepage field parameters such as permeability, porosity and Reynolds number are much faster, and the seepage field parameter evolution curves suddenly increase. 3) The evolution characteristics of water and mud inrush are more obvious under gradient loading than under constant water pressure loading condition, and the critical water pressure of water and mud inrush from filling is smaller. A generalized model of permeability evolution characteristics of fault is established with analysis of fluid state conversion on the relationship between water inflow rate and time (Q-t), the relationship between hydraulic gradient and water inflow rate (i-Q), and the evolution characteristics of permeability on the relationship between permeability and hydraulic gradient (k-i). The results provide guidance for evolution mechanism and prevention measures of water and mud inrush disaster in fault fractured zone.

The acquisition of a precise coefficient of earth pressure at rest (K0) is very important for earth pressure calculation, and the study of nonlinear characteristics of K0 has great practical significance. Nevertheless, the research on nonlinearity of K0 for coarse-grained soils has long been neglected. To fill the gap, this study presents a detailed analysis of the nonlinearity characteristics of K0 for coarse-grained soils. Based on the incremental stress-strain relation of Duncan-Chang model, we establish the mathematical formula of K0, and the sensitivity analysis of the K0 formula is presented. The proposed method is applied to predict the K0 of coarse-grained soil experimental data reported by others and by DEM method. Comparison results show that the calculated K0-value is very close to the experimental K0-value, particularly for coarse-grained soils. Furthermore, we compared the K0-value from Jaky formula and that from the proposed method, and discussed the influence of initial stress condition on K0 for coarse-grained soils. The research findings of this study are helpful for the understanding of nonlinearity and the precision of K0 in coarse-grained soils, and may provide reference for related earth pressure calculations.

In view of the alternating change of soft and hard rock mass in the work face, physical model test and numerical simulation were conducted to investigate the dynamic response of tunnel surrounding rock during the process of TBM excavation in composite strata at the Lanzhou water source construction project. Similar composite rock mass materials with different strength ratios of surrounding rock were prepared by performing analogous proportion experiments. The fiber grating technology was used to capture the strain evolution law of surrounding rock in composite strata during tunnel excavation, and the macroscopic fracture morphology of the surrounding rock of the tunnel was also analyzed. The physical experimental results show that the variation law of strain in composite strata during the TBM propulsion process reflects the spatial effect of the face thrust. The strain of soft rock is greater than that of hard rock, and with the increase of excavation steps, the difference of strain between two rock strata becomes greater. After the excavation, the macroscopic fracture morphology of surrounding rock indicates that the deformation and failure of the overlying soft rock is more serious and significant due to the difference in physical and mechanical properties of the composite rock mass. The phenomenon of “uncoordinated deformation” can also be found at the interface between soft and hard rock layers. The geomechanical parameters of a tunnel section along the project are selected to evaluate the stability of surrounding rock in composite strata during tunnel excavation based on the failure approach index (FAI). The numerical results show that FAI changes obviously in soft rock during excavation, and the plastic zone and failure zone are more widely distributed, while the lower hard rock is less affected by excavation disturbance, and only a small range of rock mass at the arch bottom enters the failure state during excavation. Both the model test and the numerical results indicate that there are differences in change laws in the deformation and failure of the surrounding rock during the excavation process. Therefore, the construction of TBM in the composite stratum can take measures such as monitoring and early warning of key parts as well as early corresponding measures reduce or avoid the occurrence of TBM jamming accidents.

In arid and semi-arid regions, salinization caused by high evaporation rate of soil will pose a potential threat to the durability of engineering structures. Therefore, accurate prediction of soil pores relative humidity is of great significance for guiding salt control of engineering structures. Based on the conclusion that the chemical potential of soil pores water is always in equilibrium with pores vapors, a relative humidity expression of unsaturated saline soil pores is deduced according to the principle of chemical equilibrium. By comparing the relative humidity of calculated and the measured solutions, the results can be obtained that the calculated values of relative humidity at different temperatures and concentrations are in good agreement with the experimental values, and are superior to that by the R&S model; the coincidence effect is more obvious especially when the concentration of solute is nearly saturated. Meanwhile, it is found that the numerical results are in good agreement with the experimental results by substituting the proposed relative humidity expression into the existing numerical model without changing the calculation parameters, especially in the prediction of evaporation rate and volume water content. In conclusion, the new formula for calculating pore relative humidity of unsaturated saline soil is effective in the process of water-salt migration calculation.

During installing procedures of torpedo anchors, tip-shape of the torpedo anchor affects penetration resistance and depth. In this study, the lower bound method of the plastic limit analysis theory is used to analyze the penetration resistance and penetration resistance coefficient for different aspect ratios of the elliptical-tip of torpedo anchor (ξ), embedment depth, and friction coefficient of anchor-soil. A series of results is obtained: 1) If a soil-anchor interface is smooth, the penetration resistance coefficient decreases with increasing the aspect ratio of torpedo anchor tip; for the rough interface of soil-anchor, the penetration resistance coefficient increases with increasing the aspect ratio. 2) The influence of the friction force markedly increases with increasing ξ due to a larger interfacial area. 3) With increasing embedment depth, the friction coefficient reduces gradually, which corresponding penetration resistance coefficient is minimally impacted by various ξ. 4) If the friction coefficient ? is larger than 0.75, the anchor tip with the aspect ratio less than 1 responds to smaller bearing capacity; and while the friction coefficient is less than 0.26, the smaller bearing capacity is provided by the anchor tip following aspect ratios greater than 1. The lower bound results of ξ = 0 and 1 show good consistent with the previous solutions for strip foundation and circular structure, which verify the lower bound method of elliptical-tip of the torpedo anchor is reasonable.

A series of resonant column tests was conducted on coral sands from the Nansha and Xisha Islands of the South China Sea with different grain gradations. It is found that there are upper and lower limits in the small-strain shear modulus, G0 distribution ranges of coral sand from different seas and with different gradations under the same effective confining pressure . The maximum and the minimum void ratios, emax and emin, are comprehensive state parameters for effectively characterizing the particle gradation and shape characteristics of sandy soils. The limit values of G0 are consistent with the extrapolated G0 results at emax and emin. Under the same , the lower limit values of G0 (G0min) and the upper limit values of G0 (G0max) of coral sand decrease with the increasing emax and emin, respectively. Empirical formulas for predicting the G0 limit values of coral sand are established based on the relationship between G0min and emax, and the relationship between G0max and emin under different . The G0 of coral sand with various void ratios e can be determined by nonlinear interpolation using the values of G0min, G0max and the relative density Dr. The new G0 prediction model has good universality for sands with similar morphology and mineralogy of the particle forms and different gradations. For other types of sandy soils, a correction coefficient a is introduced to consider the effect of the mineral composition on G0. The general applicability of the new G0 prediction model, superior to the common Hardin prediction model, is validated by the G0 experimental data published in the literatures.

To study the influence of the compaction test types on the compaction characteristics of EPS particles light weight soil, and to reveal the compaction mechanism of light weight soil, three kinds of light weight soil with different EPS particle contents were prepared for four different types of compaction tests, i.e., standard light-type, standard heavy-type, small light-type and small heavy-type. The volume compression ratios of the EPS particles were measured, respectively. The results show that the dry density of the light weight soil first increases and then decreases with increasing the water content for the three mixed ratios light weight soil under all the compaction tests. The curve shapes are similar to parabola, and the optimum water contents of the three mixed ratios light weight soil are approximately 31%, 35% and 39%. For light weight soil having the same ratio, the maximum dry density of the standard light-type compaction test can be taken as the standard. The absolute growth range of the maximum dry density for the other three kinds of compaction tests is from ?0.014 g/cm3 to 0.072 g/cm3, and the relative growth range is from ?2.647% to 13.611%. The maximum dry densities obtained by the four kinds of compaction tests are basically the same. Due to the influence of size effect of the compaction cylinder, the maximum dry density of the small-type compaction test is larger than that of the standard compaction test when the compaction energy is same, but the difference of them is small. For the same compaction test type, it is found that the increase in the compaction energy has no obvious effect on improving the maximum dry density of the light weight soil. Under the compaction action, the EPS particles exhibit obvious plastic compression, and the compression ratio decreases with increasing the water content and the EPS particle content. The volume compression ratio varies in the range of 0.955% to 31.174%. It is feasible to determine the optimum water content and the maximum dry density of the light weight soil by using the small-type compaction test instead of the standard compaction test, which can provide reference for engineering design and construction of light weight soil.

In this study, a model of non-homogeneous saturated foundation is established by considering the transverse isotropy and non-homogeneity of foundation. In this model, the porosity, density, shear modulus and permeability coefficient change along the depth, and the coupling effect between parameters are also considered. Meanwhile, a parameter, non-uniform factor, is introduced to characterize the degree of non-homogeneity of foundation. The governing equation is established based on the Biot porous media theory. The differential operator method is applied to solve the control equation, and then the dispersion equation of Rayleigh wave in non-homogeneous saturated foundation is deduced. Results are verified by degrading the deduced results to homogeneous saturated foundation and single elastic foundation, respectively. Moreover, the propagation velocity, attenuation coefficient and displacement distribution of Rayleigh wave are analyzed using numerical examples. Results show that the non-homogeneity of saturated ground has significant influence on Rayleigh wave propagation velocity, attenuation, and displacement. Additionally, the particle trajectory is changed accordingly. This effect gradually decreases with an increase in the frequency. As the frequency approaches infinity, the Rayleigh wave velocity converges to the wave velocity in homogeneous elastic foundation. The non-homogeneity of foundation increases the impedance of Rayleigh wave propagation and accelerates the attenuation of Rayleigh wave displacement. In addition, the propagation depth is less than that of homogeneous saturated foundation. With the increase in non-homogeneity, the attenuation of vertical displacement of particles is more affected and the attenuation speed is faster than that of horizontal displacement. This difference leads to a small oblateness of the elliptical particle trajectory. Moreover, a smaller thickness of the non-homogeneity soil layer results in more non-homogeneous foundation, which has greater impacts on Rayleigh wave propagation.

The theoretical analysis of water infiltration in loess is very important for predicting deformation and strength weakening of loess. However, there are still some problems lying in the application of existing Green-Ampt infiltration analysis model using into homogeneous foundation of loess area. Based on the assumption of saturated-unsaturated stratification of loess, the modified infiltration model also considered the actual variation of permeability coefficient of soil with the depth of infiltration. The proposed approach modified the Green-Ampt model with an actual case for validation. Combined with the relation between Philip model and Green-Ampt model, as well as the actual permeability coefficient changed with time, the hydraulic parameters were estimated by measuring the variations of cumulative infiltration and water content with time. The results showed that compared with WGAM and Kostiakov models, the modified model was more effective in loess area. Based on the modified model, the effects of saturated permeability coefficient, saturated water content, initial water content and matrix suction on water infiltration depth were analyzed. The sensitivity of the parameters was analyzed and compared. It was found that the change of initial and saturated water contents of soil greatly influenced the calculation results of the model. The present research is of great significance for the relative investigation of infiltration in loess area.

Split Hopkinson pressure bar tests (SHPB) were carried out to investigate the strength characteristics of granite residual soil under high strain rates comparing to that under common strain rates. Stress-strain curves and strength properties under various strain rates were obtained and found to be strain-softening. The failure strain ?af increased with increasing the strain rate. The increase of ?af was more evident at high strain rates. The peak strength of granite residual soil was found to be rate-dependent, and the peak strength increased linearly with the increase of strain rate, but the fitting relationships under low and high strain rates were different. The coefficient of strain rate sensitivity m was proposed for the quantitative evaluation of the correlation. The results show that the dependence decreases with the increase of strain rate. The value of m at low strain rate is 26.694, while only 0.013 at high strain rate. Judging from the test results, the high-speed impact load is generally harmful to the soil. The research will lead to a deeper understanding of the impact damage mechanism and provide technical guidance for relevant construction.

On the basis of iron barites sand cementation material, clay was added as the plasticizer and the weight ratio of clay to aggregates plus clay, the weight ratio of iron ore powder plus barite powder to aggregates, the weight ratio of iron ore powder to iron ore powder plus barite powder, and the weight ratio of gypsum to mixture were selected as four factors with five variable levels. Twenty five schemes of the material mixture ratio were designed using orthogonal design method, and related laboratory tests were conducted to obtain physical and mechanical parameters of similar materials such as density, compressive strength, tensile strength, elastic modulus, Poisson’s ratio, internal friction angle and cohesion. Moreover, variations in concerned parameters were analyzed for different material mixture ratios. Based on the analysis above, similar materials of two different real rocks were prepared and two related base friction physical model tests were also carried out. The test results reveal that, 1) the physical and mechanical parameters of the similar materials fabricated according to the similarity ratio proposed in this study vary considerably, which can satisfy the requirement of different kinds of rocks; 2) the influence of each factor on the physical and mechanical parameters of similar material shows good regularity, which facilitates identifications of the optimum mixture ratios for different rock materials; 3) the physical and mechanical parameters of two different materials are consistent with those of the real rocks to a great extent; 4) these two different rock similar materials can be successfully applied in the base friction physical model tests.

To study the distribution law of the energy that induced rockburst in a coal rock system, the calculation formula of pre-peak energy distribution of binary and ternary combined models was theoretically analyzed. The axial loading experiments of binary and ternary self-constructed combined bodies with different proportions were carried out. The experimental results show that with the increase of coal-rock height ratio, the pre-peak total energy increases, but the increase amplitude decreases gradually. For combined bodies with the same coal-rock height ratio, the harder the rock component is, the smaller the pre-peak total energy will be. Independent of coal-rock height ratios, the proportion of energy stored in the coal component is always higher, which is more than 50%. With the increase of coal-rock height ratio, the proportion of energy stored in the coal component increases gradually, and the proportion of energy stored in the rock component decreases gradually. For combined body with the same coal-rock height ratio, the harder the rock component is, the higher the proportion of energy stored in the coal component will be. The pre-peak energy of a combined body is mainly distributed in the coal component, followed by the gritstone, and the fine sandstone has the least accumulated energy. This shows that the energy in a coal-rock system is mainly distributed in the weak coal-rock strata, the greater the elastic modulus of a rock, the less the amount of accumulated energy. Based on this, two concepts of prevention and control of rock burst, direct release of energy and indirect release of energy, were put forward. The results are of guiding significance to the prevention and control of rock burst in the field.

In view of the poor stability of aeolian sand slope and the problems such as difficult hole-forming and limited reinforcement in commonly used reinforcement technologies, the new sodium silicate-ester slurry with binary acid ester as curing agent is proposed to reinforce aeolian sand. The gel time and the time-varying behavior of viscosity of new sodium silicate-ester slurry are systematically tested by inverted cup method and rotational viscometer respectively. The influences of different slurry filling rates on the strength and permeability of grouted-sand are studied by triaxial test. The results show that the gel time is between 20 s and 60 min. The viscosity of the slurry increases exponentially with time and the viscosity is low before gelation. When the slurry filling rate increases from 0% to 100%, the cohesion of grouted-sand increases linearly from 0 kPa to 148.4 kPa, and the internal friction angle increases from 32.1° to 37.8° according to the parabolic law, and the permeability coefficient decreases from 3.42×10?5 cm/s to 8.74×10?7 cm/s. Comparing with sodium silicate-phosphoric acid slurry and sodium silicate-sulfuric acid slurry, the compressive strength of grouted-sand reinforced by new sodium silicate-ester slurry is higher. The research results of this paper provide a new method for the reinforcement of aeolian sand slope.

In order to study the correlation between micro-seismic, electromagnetic radiation signals and crack evolution characteristics of briquette under uniaxial compression loading, the failure experiments were carried out at four different range of particle sizes including 0–0.25 mm, 0.25–0.5 mm, 0.5–1.0 mm and 1–2 mm by means of self-designed and low-noise static loading experimental system. In addition, the micro-seismic, electromagnetic radiation signals and damage videos recorded during the failure of the briquette were collected synchronously. Based on the image semantic segmentation software, a fast crack extraction method on coal mass surfaces was proposed, and the variation law of crack area of briquette was calculated. The results show that the micro-seismic, electromagnetic radiation signals and crack area generated by briquette under the uniaxial compression process have a good correlation in the time domain. The curve of crack area with time in the process of briquette failure can be divided into four stages. In the first stage, the briquette is under compaction, the stress value is small, and the crack area on the surface increases at a very slow rate. In the second stage, with the increase of stress and the accumulation of internal elastic potential energy, the rate of increase of the surface area of the briquette is significantly higher than that of the first stage, accompanied by many fine cracks. In the third stage, as the stress continues to increase and the elastic energy is accumulated inside the material, the deformation process of the sample is accelerated, resulting in a further increase in the crack area. In the last stage, the crack area of the sample reaches its peak, the bearing capacity decreases sharply, the instability occurs, and the loading process ends.

Ratio of safety margin (RSM) provides a useful tool to bridge safety criteria of deterministic design method based on safety factor and reliability-based design (RBD). Generalized reliability RSM(RRSM) overcomes the limitation on the distribution type of uncertain parameters, which restricts the application of the principle of the ratio of safety margin (RSM), and makes the RSM principle feasible in general design situations. This paper proposes a calibration method and framework for determining the allowable safety factor (FSa) based on generalized RRSM for a given target failure probability (or target reliability index). The proposed approach is applied to study the safety criteria for foundation bearing capacity. The FSa values of foundation bearing capacity are calibrated for the traditional method based on the ratio of resistance over load and strength reduction method, respectively. The proposed approach is illustrated using a design problem of foundation bearing capacity, and the relationship between RRSM and generalized RRSM is also discussed. And then, the influence of different parameters and calculation models on the allowable safety factor of foundation bearing capacity is discussed. It is shown that the feasible design domains of the deterministic design method based on FSa values calibrated from the proposed approach are identical to those of RBD under different design situations. This validates the proposed method for bearing capacity design problems.

By analyzing the field test data of blasting vibration of underground caverns in Taohuazui mining area, the characteristics of the peak particle velocity propagation attenuation on the side wall induced by blasting excavation in underground caverns were found, which presented a larger amplification effect in the vicinity of the source and the central area of the side wall. A dynamic analytical model of simply supported slabs with four sides was established. Furthermore, according to dimensional analysis method, a formula for predicting the peak particle velocity of side wall in underground cavern blasting was deduced in the case of four side constraints of side wall. Finally, the measured data were substituted into the Sadovsky formula and the formula deduced in this paper for linear calculation. The average errors between the predicted values by the Sadovsky formula and by the formula developed in this paper and the measured values were 30.7% and 5.1%, respectively. These results indicate that the proposed formula can accurately reflect the amplification effect of the peak particle velocity in the side wall of underground cavern blasting. In particular, it can accurately predict the peak velocity of blasting vibration in a region where the amplification effect is significant, and provides a reference for better control of side wall blasting vibration damage.

According to the limit state of geotechnical structures at a specific site under normal conditions of use, the reliability index of various structures is calculated by geometric reliability method that developed in recent years. At the same site, considering the discreteness of load-displacement curves of bored piles, anti-floating anchors and single CFG piles, the regression parameters of these curves show differences and can be regarded as random variables. The correlation and joint distribution characteristics of the site-specific regression parameters are discussed. Based on the joint divergence probability density contour (PDC) of these regression parameters, which means the random variables just reach the critical state of limit bearing capacity, the reliability index of geotechnical structures is calculated by the geometric reliability algorithm in the original space of random variables. The feasibility of geometric reliability algorithm is verified by comparing the geometric reliability index with the results calculated by conventional first-order reliability method. The results show that the geometric reliability evaluation model is simple to implement and can be easily accepted by engineers and technicians.

In reality, the rolling parameters are often difficult to be strictly controlled to meet the application requirements of continuous compaction control technology, which greatly limits the scope of application and application effects of the technology. In order to adapt the dynamic variation of rolling parameters to the engineering practice, this paper performs multiple regression analysis for the rolling parameters, and uses Kriging interpolation method to estimate the target value/measured value of the continuous compaction control index. A continuous compaction monitoring model with the qualification rate of compaction as the quality evaluation index is proposed. The on-site rolling test is designed and carried out, and the multivariate regression relationship between compaction index and rolling parameters is established. The backward prediction analysis of conventional quality inspection index is conducted by combining Kriging interpolation method, which verifies the validity of the model. The continuous compaction control method based on the multivariate regression of rolling parameters is applied to the control of rolling construction quality of granite residual soil of District 2-1 of Shenzhen Airport T4 expansion project. The construction quality is verified to meet the design requirements by the conventional quality inspection method, demonstrating the effectiveness of the proposed method.

To solve the problem that the influence coefficient (c) of the dynamic prediction time of mining subsidence, the subsidence value (Wv) and the time point (??) at the maximum ground subsidence speed are set as fixed values, these parameters were theoretically and experimentally analyzed using segmented Knothe function. Based on the theoretical knowledge of the probability integral method, the functions of these three parameters were deduced by considering time as the independent variable. Then, the segmented Knothe function was optimized, which is showing better generalization performance and prediction accuracy. The experimental results reveal that the optimized Knothe function eliminates the defect that c, ? and Wv of the conventional one were set as fixed values, and compensates for the shortage in similar geological conditions were used to generate a fitted empirical formula. After comparing and analyzing the prediction results from the optimized and the conventional models, it was concluded that the predicted results in the optimization model remain within the time range of the measured maximum subsidence velocity. The maximum mean square error and relative mean error of the optimized model are 0.116 m and 3.8%, respectively, which are improved by 65.1% and 51.9% compared with the conventional model. It proves the superiority of this model.

The anisotropy and size effect of a rock joint have a great influence on its mechanical properties. Considering the relationship between these two features is of great significance to the evaluation of rock mass stability. In this study, in order to investigate the anisotropic distribution of rock joints at different scales, the anisotropic variation coefficient AVC3D is proposed which considers the three-dimensional morphology parameter of orthogonal directions. The AVC3D of four natural joints at ten different sampling sizes are measured based on the progressive coverage statistical method, and the size effect law of the AVC3D is obtained. The conclusions are as follows: the AVC3D reduces to stable with the increase of the rock joint size, and there is a good negative logarithm function relationship between them. The normalized statistical average values of the AVC3D and the normalized size of the rock joints show a good linear function, indicating that the AVC3D has a fractal structure. The fractal dimension D can be achieved using the regular statistics of the size effect of the AVC3D. This method reveals the mechanism that the maximum effective dip angle and the roughness coefficient C, which affect the directional change of the structural surface features, tend to be more stable with the increase of sampling size. It shows that once the rock joint reaches the threshold of the anisotropic size effect, the stable anisotropy will appear.

There are strict requirements on surface displacement caused by frost heaving and thawing settlement for the shallow tunnels. To improve the frozen wall design of the contact channel in Zhuhai Urban-Airport Interity Railway transit project, the finite difference numerical calculation method is used to simulate the whole process of the artificial ground freezing method based on the thermal-mechanical coupling theory. The optimal design of the thickness of the frozen wall is achieved by comparing the surface displacement of frost heaving, thaw settlement and the deformation of the tunnel segment in models with different frozen wall thicknesses. Results show that: 1) the finite difference numerical model can effectively predict the development of the surface displacement caused by frost heaving and thaw settlement, and an actual deformation prediction value of high accuracy can be obtained with the known error. 2) The frost heaving and thaw settlement of different models almost have the same feature, but the deformation amount and the influence range decrease with the decrease of the thickness of the frozen wall. When the thickness of the frozen wall is less than 2.5 m, the deformation meets the requirements. 3) The frost heaving and thaw settlement of the soil are not simple reciprocal processes. The thaw settlement is usually larger than the heaving deformation, and the averaged excess amount is around 40%. 4) The greater the thickness of the frozen wall is, the greater the frost heave force is. By optimizing the thickness of the frozen wall, the additional stress and deformation of the tunnel segment can be effectively controlled to protect the structure of the existing tunnel. 5) The thickness of 2.5 m is chosen for the improved frozen wall. The research results are directly applied to the construction of the No. 4 communication channel using the freezing method. Combined with the on-site monitoring test, the deformation of each characteristic point is within a reasonable range, indicating that the optimization scheme is practical and feasible, and can be successfully applied in the design of frozen wall thickness in similar projects.

Stiffened deep mixed (SDM) column is a new ground improvement method for soft soil foundations, and has been successfully applied in ground improvement projects in roadways and railways recently. However, due to inadequate understanding of the instability mechanism of the embankment supported by SDM column, there is no mature theory to guide the design. In this paper, strain-softening model, which can reflect the post-failure behavior of column materials, was used to model the scale-down 1g model test of stability of SDM column-supported embankment. The sequence and modes of SDM column failure were investigated through examining the developments of plastic zone and force changes in SDM columns in the failure process. The results showed that the columns did not fail simultaneously in the process of embankment failure. The columns near the center of embankment failed by compression at first, followed by the sequence bending failures of the columns under embankment slope from the toe to the shoulder. The slip surface did not completely pass through the failure positons of the columns due to the existence of columns. Accordingly, the failure mechanism of SDM column-supported embankment was interpreted. The semi-rigid column method based on residual strength in the currently-used code yielded close factor of safety to the model test, but its feasibility still needs a further study.

To investigate the influence of degree of saturation on small-strain stiffness of soil under evaporation conditions, an experimental device for measuring small-strain stiffness of unsaturated soils was developed. The device consists of three components: relative humidity controlling system, deformation monitoring system, and wave velocity measuring system. The relative humidity generator is used to modulate the relative humidity inside the environmental chamber. The weight change of soil sample is recorded through the weighing module. A digital camera is used to take photos of the upper surface and the side of the soil sample at the same time. The sensors measuring compression and shear wave velocities are designed and fabricated by using piezoelectric ceramic elements. The small-strain stiffness of soils is then determined by wave velocities. The experimental results show that the system can stably control the relative humidity of the environmental chamber, indirectly measure the volume change of soil sample, and monitor the variation of wave velocities of soil in real time. On this basis, the dependencies of wave velocity and small-strain stiffness of soils on degree of saturation can be observed. Shear wave velocity and small-strain shear modulus of soil samples gradually increase with decreasing the degree of saturation, while compression wave velocity and small-strain volume modulus first decrease and then increase with decreasing the degree of saturation. The test system is simple in structure and easy to operate, providing a powerful tool to investigate the effect of degree of saturation on the mechanical properties of soil.

In order to measure the hidden danger of groundwater leakage under natural and artificial flow fields accurately, a new method of hydrogeological logging—sonar seepage vector method is proposed from the point of view of energy measurement. The basic principle, prototype instrument and measurement procedure are introduced systematically. Based on the principle of time difference measurement, the sonar seepage measurement instrument and display and control software have been developed independently, which combine many new technologies such as vector sonar technology, aeronautical orientation technology, pressure conduction technology and GPS locating technology. The sonar seepage measuring instrument can accurately determine the leakage location, leakage passage, flow velocity, flow direction, flow rate and other quantitative indicators. The foundation pit of diaphragm wall in water-rich stratum is studied as an engineering case. The three-dimensional visualized imaging system of seepage field is established to reveal the distribution of three-dimensional seepage field so as to dynamically guide seepage control design of foundation pit, leakage treatment. The effect of leakage treatment is verified by repetition test. The results show that sonar seepage vector method can locate leakage defects in underground engineering accurately, and the test results meet the needs of seepage control in engineering fully in the absence of strong noise interference. The early warning and grading criteria of leakage risk for foundation pit are put forward, which are mainly based on seepage velocity and supplemented by single hole seepage flow rate and permeability coefficient. At the same time, treatment measures must be taken in advance for avoiding leakage risk effectively by combining geological survey, monitoring, in-site inspection, emergency disposal with other means and mutually validation.