Unsaturated soil mechanics has become an active researching area in recent years. However the understanding of some fundamental problems is inconsistent and some basic concepts are confusing. This paper aims at the fundamental problems, e.g. the choices of state variables and effective stress of unsaturated soils, the definition of suction, the limitation of axis-translation technique and the characterization of unsaturated soil structure. In these problems, the choices of state variables and effective stress of unsaturated soils play an important role in the theory of unsaturated soil mechanics and constructing constitutive models. Firstly, this paper briefly reviews expressions of effective stress of unsaturated soils and discusses their advantages and disadvantages. Secondly, this paper points out that there does not exist such matric suction beyond a certain limit in real site because of cavitation of capillary water. However the laboratory axis-translation technique conceals the possibility of cavitation and modifies soil behavior. Thus the validity of theories of shear strength and deformation of unsaturated soils based on it is a question worth study. Thirdly, the paper suggests that the characterization of unsaturated soil structure should include the distributions of pore water and pore air, the interaction among all phases (e.g. capillary and physical-chemical forces), except for the elements of saturated soil structure, e.g. fabric and inter-particle actions. Finally, this paper further combs the confusable concepts. The purposes of these discussions are to attract attention of domestic scholars on these problems, to set up correct understanding about them, and to promote development of unsaturated soil mechanics.

Coulomb’s theory of earth pressure is still playing an important role in calculating soil earth pressure and is well known all the world. By analyzing the mechanical characteristics of the limiting equilibrium soil wedge behind the retaining wall, especially the force between the soil wedge and retaining wall, some modifications to traditional Coulomb’s theory of earth pressure are presented. Considering that it is not necessary for the force (which is defined as earth pressure) between the wedge and the retaining wall surface to be on the point of sliding or to be in a condition of limiting equilibrium; the direction of the earth pressure can not be determined, but it must be limited in an allowed range according to the friction angle. Therefore, the active earth pressure is defined as the maximum value of the force within the allowed range; and the passive earth pressure is the minimum value of the force within the corresponding allowed range. This paper also considers that the soil wedge and the retaining wall are two different objects; and the soil wedge is formed only because of the potential failure surface of the soil itself. In other words, it is not necessary for the force between the wedge and the wall surface to achieve a critical state. As a matter of fact, the force between the wall surface and the soil wedge is equivalent to the force exerted on an object (soil wedge) by another object (retaining wall); and even if the soil wedge slides along a failure surface, the two objects do not need to slide each other along their contact surface. The formulas of the modified active earth pressure are derived and an approximate solution is presented for the passive earth pressure calculation. Examples prove that the results of this study are significantly different from the classic Coulomb’s theory. The works of modifying Coulomb’s theory of earth pressure deserve much more attention.

The processes of dehydration shrinkage and water-swelling of expansive soil correspond to the drying and wetting stages of the soil-water characteristic curve (SWCC). The SWCC plays an important role in defining the hydro-mechanical behavior of unsaturated soils. In order to study the SWCC of Nanyang expansive soil in full suction range, three suction control or measurement methods are used: the pressure plate method for low suction range of 0-1.5 MPa, the filter paper method for suction range of 0-40 MPa, and the vapor equilibrium technique for high suction range of 3-368 MPa. The results show that the SWCC has obvious hysteresis phenomenon during the drying and wetting in the low suction range. When the suction is greater than 300 MPa, the hysteresis effect of SWCC disappears; and the drying and wetting curves are almost overlapped. Test data of soil-water characteristic by the filter paper method are located between the main drying and wetting curves of the SWCC. When the suction of is 367.54 MPa, the soil water content is only 0.325%, and it almost approximates to zero. Void ratio not only changes with suction, but also it is influenced by the suction history and suction path. In the relation between void ratio and suction, the curve of wetting path is above that of the drying path; and they become the same at low suction. There is a clear hysteresis of the relation between void ratio and saturation degree, due to the different suction paths; and they will be consistent when close to saturation. The saturation degree increases with the increasing void ratio when suction changes. The experimental data obtained by the vapor equilibrium method have an obvious linear relationship between void ratio and saturation degree; while the test data obtained by the pressure plate method are no an obvious linear relationship in the low suction range.

The study of the interaction between lunar lander footpad and lunar soil during the moon landing has great importance owing to the strong impact which affects safe landing and normal performance of inner precise instruments. Since horizontal slide is the complicated landing process, model tests on the process of dragging footpad in simulant lunar soil are carried out with the self-developed test device to investigate the changes of horizontal and vertical forces on footpad. The influences of relative densities, sliding velocities, penetrations and turn angles on these forces are further studied. The results show that the horizontal forces appear almost linear increase with the increase of vertical forces. Besides, the forces increase with the growth of relative density of soil, horizontal velocity and penetration. The forces are greatly influenced by the turn angle. When turn angle of footpad is (8°, 26°), the force of lander footpad is the least. When the turn angle is (0°, -26°), which means bulldozing process, the force is the largest. This turn angle (0°, -26°) is appropriate for the buffer structure of lander to dissipate impact energy in slip process.

Gongbei tunnel is an important part of Hong Kong-Zhuhai-Macao Bridge cable; its foundation pit bottom is dealt with cemented soil. The aggressive ions of the sea will easily have an effect on the cemented soil as the underground pressure water is connected to the brine. Through the laboratory experiment and theoretical deduction, the mechanical properties and durability of cemented soil in the corrosion environment are studied as follows. Laboratory cemented soil and corrosion environment are made to test the unconfined compression strength of the samples soaked at 90 d, 180 d and 270 d respectively. Result of this laboratory experiment indicates that: under the influence of erosion environment, the maximum value of uniaxial compressive strength of cement soil will be reached between 90 d and 270 d. After the soak period exceeds 90 d, the solution that contains MCN shows up an entirely negative effect on the compressive strength of cemented soil. Under low concentration (1.5 g/L and 4.5 g/L), the chemical reaction between MSN and cemented soil achieves an optimum value; in comparative conditions the compression strength damage of cement soil is minimal. The effect of magnesium chloride on compressive strength of the cemented soil samples is much larger than that of magnesium sulfate. Based on the principle of chemical kinetics and the chemical equation between magnesium sulfate and the cement hydration products, combined with the assumptions, the relationship between ion concentration and corrosion time is deduced as (C1/C2)β=t2/t1; that is, when the soil samples are eroded to equivalent compression strength, the reaction order of concentration is inversely proportional to corrosion time. According to the formula, the strength of cemented soil soaking in low concentration solution for a long time is predicted by the strength of cemented soil soaking in high concentration solution for a short time. At last, the theoretical results are verified by experimental data.

In geotechnical engineering, fractured rock mass is often in the post-peak deformational state. It is important for predicting stability of structure to study post-peak stress-strain relationship of fractured rock mass. Based on the evolution of post-peak strength parameters, firstly, a method for solving the relations of post-peak stress-strain of rock and stress-shear displacement of crack is proposed. Then using Mohr-Coulomb strength criterion, regarding maximum principal strain of rock and shear displacement of crack as softening parameters, assuming the strength parameter as a piecewise linear function of softening parameters, in the two cases that the rock mass fails along the fracture and through the rock, the method for solving relation of post-peak stress-strain of rock mass with multiple penetrative crack sets is presented. Finally, in case study, the post-peak stress-stain curves are given in the two cases that the rock mass fails along the fracture and through the rock; and the effects of mean space, normal stiffness and shear stiffness of cracks on post-peak strain are discussed. The results show that the smaller the mean space, normal stiffness and shear stiffness of cracks are, the larger the axial strain of fractured rock mass is.

The seepage failure of rock and soil and the seepage design for underground engineering are closely related to the seepage calculation. A meshless collocation method based on barycentric Lagrange interpolation for solving seepage free surface problems in 2D is presented. Embedded the irregular seepage domain into a regular rectangular domain, the seepage governing equation is extended into the regular rectangular domain. The unknown function of seepage in seepage domain is approximated by barycentric Lagrange interpolation in regular rectangular domain. Then, the barycentric Lagrange interpolation collocation method (BLICM) can be used to solve seepage problems in the regular rectangular domain. The governing equation of seepage problem is discretized by using BLICM into algebraic equations. Using notations of differentiation matrix and Kronecker product of matrices, the system of algebraic equations can be rewritten as matrix form. The boundary conditions are discretized by barycentric Lagrange interpolation and applied by using replacement equation method or/and additional equation method. The final location of free surface is obtained by iteration computation of BLICM in regular rectangular domain. The numerical results compared with others numerical methods indicate the correctness and high accuracy of the proposed method.

Traditional layer-wise summation method cannot take the three-dimensional deformation of foundations into account, which results in very large calculation error on foundation settlement. In order to reduce the range of correction factor of layer-wise summation method, a calculating method of flexible foundation settlement is derived based on e-p curves from confined compression test, which can consider three-dimensional deformation of the foundation. The new calculating method can consider lateral deformation of the foundation soil on the basis of layer-wise summation method when the compression amount of each layer calculated by the traditional calculating method is multiplied by a correction factor. Further, by adjusting the pressure distribution on the foundation base, an approximate calculating method of rigid strip foundation settlement is derived. The numerical example shows that the new calculating method considering three-dimensional deformation can solve the problem that it will get smaller settlement on the small size foundation when using traditional layer-wise summation method. From the calculation results, settlement of the small-size rigid foundation calculated by new calculating method is larger than that calculated by traditional method; while settlement of the large-size rigid foundation calculated by new calculating method is smaller than that calculated by traditional method, which may be one of the reasons that correction factor of layer-wise summation method is in a wide range. The application of this new method can reduce the calculation error of layer-wise summation method.

The dynamic reduction of confining pressure of salt rock gas storage surrounding rock during construction period is simulated by uniaxial compression tests and triaxial unloading confining pressure tests. The effect of temperature on dilatancy characteristics of salt rock are analyzed by the unloading tests. The results show that all the tests are the process of stress increasing. Furthermore, both unloading confining pressure tests and uniaxial compression tests have similar deformation stages of dilatancy curves. The similar deformation stage includes the volume compressed segment, the stable dilatancy segment and the accelerating dilatancy segment. Those two types of tests also have the similar compressed limit. However, due to the different loading modes, the ratio of stable dilatancy in uniaxial tests is larger than that in unloading confining pressure experiments. The dilatancy rate and the value of dilatancy are smaller in unloading confining pressure tests. The unloading confining pressure tests have greater dilatancy under the same deviatoric stress. Higher temperature can promote the dilatancy of salt rock. That means the increasing of temperature can make the accelerating dilatancy point appear forward, the initial stable dilatancy segment shorter, the accelerating dilatancy segment longer, the dilatancy rate and the value of dilatancy higher.

In order to further study the soil arching behind the sheet-pile wall and analyze the earth pressure load regulation and transmission characteristic, the field large-sized test and indoor model test are carried out by monitoring research. In addition, the test results can further reveal the regularities about soil arching and load distribution. Based on the comprehension of the action mechanism about the soil arching structure, the ratio between the earth pressure on the back side of pile and pressure on the breast board is regarded as a standard which is used to assess the arching. A test model box is designed and constructed in order to load behind sheet-pile wall at a rock slope. The earth pressure cells are fixed on the dorsal surfaces of sheet and piles to test the earth pressure during future 21 days. The test results show that the earth pressure ratio increases firstly and then tends to be stable as the development of time; therefore, the time effect is relatively obvious. In addition, multiple test conditions are also designed and done to make further analysis about the relationship between the soil arching and transfer behavior of earth pressure behind sheet-pile wall. The comparative researches are done about different baffle rigidities, pile spacings, filler properties and arrangement effects of sheets, which explains the transfer behavior of load behind sheet-pile wall in the passive soil pressure state.

Soil structure is an intrinsic determinant of its strength and deformation. The loess structure is determined by the link structural strength and the friction structural strength. Based on this, some new and reasonable structural parameters of loess are suggested based on soil static strength, such as the link structural static strength potential parameter m? 1, the friction structural static strength potential parameter m? 2 and the structural static strength potential parameter m?. These structural parameters have clear physical meaning, and can be used to reflect the comprehensive structure, the link structural and frictional structure of loess. The stress-strain behavior and the variation of the structural parameters of loess are studied under static triaxial stress conditions. The stress-strain formula of intact loess based on the structural parameters is provided. So we can simulate the stress-strain curve of intact loess by structure parameters and the stress-strain curve of saturated remolded loess which has the weakest structure. Some parameters such as strength index c, the value of the initial tangent modulus Ei, failure ratio Rf, stress level S and tangent modulus of elasticity Et can be worked out from the stress-strain formula of intact loess. These parameters can be used in finite element analysis of loess stress-strain under static loads.

The pile group head of jacket platforms is elastic constrained head. According to the elastic constrained head, a method to determine p-y data of jacket platform pile groups using the nonlinear foundation beam model of pile group is proposed. The lateral displacement of pile head is first determined using the pile group model with the elastic constrained head and p-y data of a single pile under total loads for the method. Additional lateral displacements due to pile group interactions are determined using Poulos’s interaction factors. The superposition result of above two calculations is taken as the initial lateral displacement of pile head of iterative calculations. The modified factor, Ym, of p-y data of the pile group associated with the first calculation is determined by calibrating p-y data of a single pile based on pile head loads and initial lateral displacement. For following calculations, p-y data of pile group associated with each iteration are determined using the updated Ym and then are used to determine the lateral displacement of pile group head and new Ym until the relative difference between Ym(i) and Ym(i-1) associate with the iterations i and i-1 is less than the tolerance. Because the method uses the p-y data considering pile group effects and the pile group model with the elastic constrained head to analyze the pile group interactions, results can objectively reflect interactions of jacket platform pile group and the deformation behavior of pile group with the elastic constrained head.

An effective thermal conductivity model is presented based on the basic solution of ellipsoidal matrix-inclusion problem using homogenization techniques. This model integrates well the effects of the shape, volume fraction, space distribution of inclusion and the interaction between inclusion and matrix on the effective thermal conductivity. The anisotropy of conductivity, which is mainly induced by the aspect ratio of inclusion and the ratio between conductivity of inclusion and matrix, is also included in the model. On this basis, by characterizing the geomaterials as heterogeneous media composed of solid matrix and embedded ellipsoidal voids, the influences of void shape, porosity and saturation degree on the effective thermal conductivity are discussed. Finally, the proposed model is used to predict the effective thermal conductivity of Gaomiaozi bentonite(GMZ01); and the predicted results are compared with the experimental data and the predictions by other models. The results show that the proposed model has a better predictive capability for the effective thermal conductivity of the GMZ01 bentonite. Given the complex pore structure of the bentonite, multiscale homogenization techniques should be used to yield more accurate predictions. The research results may provide a helpful reference for better understanding the coupled thermo-hydro-mechanical behaviors of buffer materials for the high-level radioactive waste disposal.

The contact area will make flow tortuosity more obvious and influence the flow rate significantly. In order to investigate the effect of contact area on fluid flow through joint, a conceptual flow model is established to obtain the contact influential coefficient. And the influential coefficient is introduced into the equation for flow rate calculation considering the effect of tortuosity; a new equation considering the effect of contact area is derived. As the contact area makes the calculation of tortuosity coefficient more complicated, an algorithm for calculating the average tortuosity coefficient is proposed based on the joint aperture distribution. An artificial marble joint is taken as an object and the surface topography data are scanned. In one hand, flow tests are carried out on the joint specimen under five levels of normal stresses. In the other hand, on the basis of the average tortuosity coefficients and contact area ratios of joint specimen under five levels of normal stresses calculated according to the joint surface topography data, the volumetric flow rates through joint specimen are predicted by the new equation considering contact area presented in this paper and Zimmerman's equation, respectively. Then the predictions of the two equations are compared with experimental observations. It is shown from the comparison results that the predictions of the new equation agree well with experimental observations, while Zimmerman's equation overestimates the flow rate through the joint, thus validating the veracity of the new equation.

Based on anisotropy failure criterion with the method of macro-meso incorporation, failure properties of sand considering rotation of principal stress axis are analyzed. The criterion is a function of loading stress, degree of fabric anisotropy and geometric relationship between fabric and stress. So it can describe the effect of mesostructure on failure properties with any rotation angle of principal stress axis. According to the characteristics of hollow cylindrical torsional shear tests, the failure expression considering rotation of principal stress axis is derived in general orthogonal coordinate system. Considering the geometric relationship between loading stress and sand meso-fabric, the established expression can analyze failure properties under the condition of principal stress axis rotation. Rotations of principal stress axis cause the changes of failure properties when anisotropy exists. The greater degree of anisotropic causes the more change of failure. There isn’t any change when the material is isotropic. Therefore, it is indicated that the nature reasons of different failure properties under the condition of principal stress axis rotation is the exist of anisotropy. The verification with hollow cylindrical test results show that the established expression can describe the failure properties with different angles of loading stress. The different failure properties caused by sand anisotropy under the condition are verified preliminarily.

Zhanjiang formation formed in early Pleistocene is widely distributed in Leizhou peninsula in China. The most typical gray clay of Zhanjiang formation is known as its strong structural performance and had produced a series of geotechnical engineering problems in actual engineering. According to the stratigraphic research, this typical gray clay of Zhanjiang formation has different sedimentary environments and compositions in different areas of Leizhou peninsula. Thus, the differences of geologic origin make the engineering properties of gray clay changing with the area. In the actual projects, it is difficult for engineers and researchers to master the engineering mechanics effect of gray clay due to its regional characteristics. Some tests by X-ray diffraction and scanning electron microscope (SEM) are carried out to study the microstructure characteristics of gray clay such as mineral composition and its content, and microstructure in different areas. The results show that there are differences of mineral composition and its content. Different proportions of mineral composition compose different microstructures of gray clay which have obviously regional characteristics. The revelations of micromechanism of gray clay which lead to its strong structural performance and the analysis of its regional characteristics make for cognition and use of this clay.

In order to study the salt rock damage evolution characteristics in a certain ground temperature, brine and ground stress of the geological environment during the construction process of energy underground storage, the acoustic emission (AE) method is used to analyze the uniaxial compression damage process of salt rock with the combined effect of temperature and saturated brine. It is found that the uniaxial compressive strength and elastic modulus of salt rock samples are slightly lower after soaked in brine for 30 days. The changed peak strength of salt rock caused by saturated brine is not obvious. It is means that brine mainly works on dissolution effect on salt rock while with little soak weakening effect. The stress-strain curves and the acoustic emission-strain curves of salt rock have good consistency in the process under uniaxial compression tests; and the cumulative AE number increases with the increasing brine temperature. The salt rock samples soaked in the brine are usually with lower AE rate and AE number after the temperature and brine actions. In the construction period of salt cavern, salt rocks may lose slightly the strength but gain damage recovery due to brine. This feature of salt rock is conducive to security and stability during salt cavern construction.

Considering the viscous damping of the soil and the soil-pile torsional coupled vibration, regarding the pile as a one-dimensional rod and the soil as three dimensional axisymmetric viscoelastic medium, the frequency-domain torsional vibration response of cast-in-place concrete large-diameter pipe pile (PCC pile)in viscoelastic soil is theoretically investigated. The analytical solution in frequency domain is obtained by Laplace transformation method and the method of separation of variables. The solution is compared with the solution of a solid pile and plane strain solution extended to pipe pile to verify its rationality. Moreover, by analyzing the effects of pile length and viscous damping coefficient of soil on the velocity admittance and complex dynamic stiffness, the regularity of variations of the torsional vibration characteristics with parameters is obtained. Analysis shows that the complex dynamic stiffness at the top of pile increases with the increase of viscous damping coefficient of outer soil and the oscillation amplitudes of the velocity admittance decrease; while the effect of viscous damping of inner soil on the complex dynamic stiffness and velocity admittance is little. The complex dynamic stiffness increases and the oscillation amplitudes of the velocity admittance decrease with the increase of the pile length. However, when the pile length increases to a critical length, the effect trends to vanish.

In order to study the aseismic behavior of the compound supporting structure which is composed of frame structure of anchor cable in the upper part and sheet pile wall in the lower part, the large-scale shaking table model test is carried out. By imputting Darui wave excitations of different intensities, acceleration response and dynamic earth pressure response of compound supporting structure are investigated. Test results of dynamic earth pressure are compared with that calculated by code method and Mononobe-Okabe method (M-O method). The results show that: (1) The horizontal and vertical accelerations are magnified differently along the height of compound supporting structure under the seismic excitations with different intensities; and acceleration amplification effect increases with the increase of peak excitation acceleration. (2) The dynamic earth pressure increases with the increase of peak excitation acceleration at the back of sheet pile wall which is the lower part of the compound supporting structure; and the distribution of dynamic earth pressure along the wall height is small in the upper part and large in the lower part. (3) The current methods including code method and M-O method for calculation of dynamic earth pressure are reasonable in low intensity seismic area, but they need to be modified in high intensity seismic area.

When batter pile group are under lateral load, the foundation piles in pile group are subjected to radial load, axial load and bending moment. In order to study the influence of axial load on lateral bearing capacity of batter piles, large-scale model tests of three single piles and a 1×2 batter pile group have been conducted in a larger soil tank. It is found that the upward axial force on the pile top decreases the lateral stiffness and the lateral ultimate bearing capacity of batter pile; while the downward axial force enhances the horizontal stiffness and the lateral ultimate bearing capacity of batter pile. Based on the assumption of wedge failure of the shallow soil around piles, considering the influence of axial load on piles, a formula for calculating the lateral ultimate resistance force of soil is deduced. Then, a p-y curve method for resistance force of soil around piles is proposed. Finally, the formula and the proposed p-y curve method for batter pile are verified by model tests and field tests.

In order to study the degradation mechanism of the remolded saturated loess under long-term seepage, dilute acetic acid of pH=4 was used as percolation fluid so as to accelerate the deterioration. After a series of long-term conventional percolation tests and triaxial percolation tests under different confining pressures were conducted using Yangling Q3 loess as samples, the permeability coefficient was measured; and the relation between the permeability coefficient and time was analyzed; then the degradation mechanism of the remolded saturated loess under long-term seepage was studied. The results show that the permeability coefficient decreases with time; and the relationship between permeability coefficient and time meets power function. The permeability coefficient is the most sensitive parameter reflecting the degradation degree of the loess, which is determined by void ratio in the minimum void ratio surface. Loess seepage rate of deterioration is defined by the ratio of void ratio in the minimum void ratio surface and initial void ratio, to describe the degradation degree of the remolded saturated loess in long-term seepage. At last, the relationship between loess seepage rate of deterioration and permeability coefficient is proposed based on the nonlinear permeability model of loess, as well as the relationship between loess seepage rate of deterioration and the percolation time is derived.

Aiming at effectively improving the electroosmotic efficiency and effects, the influence of three different arrangements of the electrodes, namely rectangular array, hexagonal array, and parallel and transposition array, on the effect of electroosmotic drainage through laboratory model tests is investigated. Aluminum tube electrodes are adopted; and the treated area, applied voltage and treated time are designed to assure the consistency among the three electrode arrangements. Drainage, the rate of current decrease, cracks on the soil and coefficient of energy dissipation are analyzed and compared after the tests. It is revealed that the cracks of rectangular array are regular and its drainage effect is ordinary. Parallel and transposition array has the maximum drainage, the largest rate of current decrease as well as the best performance based on the coefficient of energy dissipation. The cracks of hexagonal array are the most complex, which run through the lengthwise section and have a strong impact on the electroosmotic efficiency at later stages. Before the extensively developing of cracks, the rate of hexagonal array’s current decrease is very small. According to the results of laboratory tests, it is recommended to employ the parallel and transposition array primarily for electroosmotic treatment.

Underground petroleum storage caverns with the containment of groundwater is one of the methods for storage of petroleum. In the construction process of large water sealing petroleum storage caverns, the appearance of engineering excavation changes the surcharge, runoff and drainage conditions of the local groundwater and disturbs the equilibrium of groundwater seepage field. To ensure the water sealing effects, it is necessary to research the fractured rock mass permeability characteristics and space-time evolution behaviors of groundwater seepage field around underground storage caverns. In the engineering background of the first under-construction large water sealing petroleum storage caverns in China, the three-dimensional groundwater numerical simulation is performed using the method of equivalent continuum and permeability tensor theories combined with analysis of in-situ test data. Under anisotropic conditions, groundwater level change is predicted during different construction progresses. The analysis and prediction results show that groundwater level drops and some zero-water head regions appear when excavation is performed without water curtain; the water sealing effects cannot be guaranteed. During operation period, groundwater table rises to 35 m and tends to be stable. The water sealing system is verified. The studied results have theoretical value and important guidance for water sealing evaluation of underground petroleum storage caverns.

During the construction of permafrost tunnels, some extent of surrounding rock will thaw under the effect of construction behavior; it will affect the pressure on support and the convergence of tunnel. An elastoplastic model of surrounding rock of permafrost tunnel under thawing effect is built; and the elastoplastic solution of the model is deduced. In the model, the surrounding rock is divided into frozen zone and thawed zone, considering the frozen zone rock mass as elastic material and the thawed zone rock mass as elastoplastic material. Interaction between the surrounding rock and the support is analyzed under different kinds of surrounding rock and support conditions. The results indicate that the model can represent the characteristics of the interaction between permafrost rock mass and support under thawing effect; and the strength of shotcrete plays an important role in maintaining the stability of permafrost rock mass and controlling the deformation of permafrost tunnel under thawing effect. This model can be used to identify the maximum allowable thawing depth of surrounding rock during construction of permafrost tunnels. Some measures should be taken to control the thawed zone during construction of permafrost tunnels.

Geotechnical parameters are closely related to the stress environment, sedimentary condition, weathering degree, buried condition and other factors. The statistical data of geotechnical parameters have high space variation and relativity. Sometimes, it is uneconomical to use the noncorrelative statistical parameters in geotechnical engineering analysis and design. The variation law and relevant key factors of geotechnical parameters are researched. The depth or distance from the test point to the bottom or surface of the layer could be correlational with geotechnical parameters. Besides, the criterion and formulas of correlational geotechnical parameters are provided; and the graphical method is used to study the reliability of different calculation methods. Finally, in the two cases of foundation design of conventional island in a certain nuclear power station and the slope desigh of a certain transformer station, the criterion and formulas of correlational geotechnical parameters are applied for optimization design. The correlation analysis shows that the geotechnical parameters varied regularly in the layers should be divided into the correlational type. The depth or distance from the test point to the bottom or surface of the layer could be the correlational parameters; the correlational standard values of geotechnical parameters are more reasonable and economic in geotechnical engineering analysis and design.

Choosing 20 meters long tunnel with typical wall rock condition as the subject of study, which is a part of the Dujiashan tunnel belongs to the Guang-Gan Expressway and located in the earthquake fault zone of the 5•12 meizoseismal area, and comparing the follow items when there exists systematic bolts or not, including displacement of tunnel inner perimeter, axial force of the bolts, internal force of the steel arch truss, pressure between wall rock and preliminary support, and the internal force of second lining are analyzed. The essay also discusses why the bolts support effect is not evident and proposes to adopt the preliminary support method which is jet steel in broken phyllite tunnel in meizoseismal area. The research shows: for tunnels with bolts, axial force of the bolt is smaller; the final displacement of tunnel inner perimeter and the pressure between the wall rock and preliminary support are large relatively; the safety stock of the steal and second lining is not enough. The process of applying bolts will disturb the deep wall rock to some degree and miss the best time to apply the shotcrete and steel arch truss to support the weak and broken phyllite tunnel. What’s more, the damage to the deeper wall rock and the penetrability of the broken rock mass increase, which result in that wall rock relaxation region exceeds bolt setting range and the cohesive force between bolts and wall rock decreases.

The Lüliang airport landslide is a large loess landslide which was found in early construction of the airport. Due to airfield runway next to the northeastern of the landslide, the landslide makes a large influence on the engineering construction and safe operation. In order to assess landslide risk and prevent its damage, the geological features and formation mechanism of the landslide are analyzed based on the detailed engineering geological investigation and testing work. The landslide is considered as a multilayer landslide developed in loess layer. There are three steps on the landslide; and the main sliding surface locates in the paleosol layer of Lishi loess. Three-dimensional geological model is established using GIS; and the stability of the landslide is evaluated by limit equilibrium method and shear strength reduction method. Results from three-dimensional limit equilibrium analyses show that most parts of the landslide are stable or basic stable in natural state. From strength reduction analysis, the safety factor of the whole slope is 1.26. It is shown that the most dangerous areas appear in the posterior margin of landslide II and landslide I-3. In case of loading or rainfall condition, it is likely to slide again. In conclusion, three-dimensional stability analysis can more truly reflect the actual state of landslides, especially for complex landslide.

The geological sequestration of CO2 in deep saline aquifer is an effective countermeasure for reducing global warming and greenhouse effect. Based on the Shenhua Ordos CO2 capture and storage (CCS) pilot project, the behavior of CO2 in deep saline aquifers is investigated. The transport process of CO2 fluid, the pressure buildup of system and the reserves potential of sequestration are analyzed. This model can provide technological support and save human and financial resources for Shenhua CCS engineering project. First, the model is calibrated by comparing simulated results and measured pressure values. The suitable pressure curve is obtained and the main hydrological parameters are determined at this stage. Then an assumption of CO2 continuing injection for 3 years is simulated based on the former model. The CO2 diffusion, solution behavior, pressure variation and total reserves of strata are analyzed. The conclusions are drawn as follows: the largest distance of CO2 migration is about 350 m; hydraulic fracturing can improve CO2 injectivity obviously; cap rock can effectively prevent the escape of CO2. Simulation results demonstrate that even though the deep saline aquifers of Ordos basin has low penetrability, it is also suitable for CO2 sequestration.

In order to research the effect of the canyon terrain on the stress and displacement of cutoff wall in dam foundation with deep overburden, taking an asphalt concrete core rockfill dam project as a background, a narrow canyon and a broad canyon were simulated and corresponding finite element models were also established respectively. The filling materials of dam and overburden were simulated by the Duncan-Chang E-B model. The interface without thickness was used to simulate the contact relation between the cutoff wall, overburden and bedrock. A 3-D nonlinear finite element calculation was carried out to analyze the stress and displacement of cutoff wall in different canyon terrains. Results indicate that vertical deformation and horizontal displacement of cutoff wall, as well as unequal settlement between cutoff wall and overburden in narrow canyon are less than that in a broad one; the maximum unequal settlement reduced 24.8%. The cutoff wall is less restrained by the broad canyon than that in a narrow one; thus the vertical stress in the broad canyon is greater and increases by 40.3% utmost. The position of the maximum vertical stress of cutoff wall is affected both by the position of neutral point and canyon terrain. The maximum vertical stress is approximately 30% by the weight of embankment and 70% by negative skin friction. The studied results can provide a reference for the design of cutoff wall in dam foundation in different canyon terrains.

Based on the previous definitions of arching effect coefficient, the formula of arching effect coefficient is improved. It is suggested that the arching effect coefficient should be the ratio of measured soil pressure and the sum of overlaying soil pressure and pore water pressure. Combining with the monitoring data of Pubugou dam, the arching effect of core wall is calculated by the improved formula of arch effect coefficient; and dynamical analysis of arching effect is made. Construction technology is the main factor affecting the arching effect coefficient of core wall during the construction. The arching effect coefficient is larger when the filled soil is 0-20 m above the instruments; and decreases with the rising of filled soil. The stress distribution of dam during construction plays an important role in initial water storage. During the impoundment, the arching effect coefficient of upstream side of the core wall shows inverse correlation with water level variation. The arching effect coefficient of upstream side is larger than that of downstream side at the same elevation and 0+001 m. The arching effect coefficient at 0+001 m shows positive correlation with water level variation, where gets the smallest arching effect coefficient. The arching effect coefficient of downstream side of core wall shows positive correlation with water level variation. The arching effect coefficient is larger than 100% on the contact surface between core wall and bed rock. There are three reasons for the above laws, such as wetting, seepage and hydraulic fracture; and the combined effect of the three results in stress variation in core wall.

Compared with traditional offshore foundation structure, bucket foundations which are large cylindrical structures open at the base and close at the top may realize economical advantages over traditional driven piles due to the speed of installation, elimination of the pile driving process, and reduction in quantities and material costs. Though bucket foundation is a new foundation type for offshore wind turbine, there have no standards for bucket foundation designing up to now. In designing the bucket foundation, the simplified limit equilibrium method is usually adopted, in which the soil inside the bucket and the skirt of the bucket are assumed to form a whole foundation. But this method cannot reflect the working mechanism of the internal soil. Based on a foundation of offshore wind turbine in Fujian, adopting the finite element software ABAQUS, the three-dimensional numerical analysis is performed to simulate the working mechanism and evaluate the bearing capacity behavior of bucket foundation. According to the bearing capacity of offshore wind turbine, the simplified model and models with or without compartments are established with the considering of the compartments effect. According to calculation results, the bearing characteristic of the bucket foundation is different from the simplified limit equilibrium method that the internal soil of the foundation yields first in the loading process. The difference leads the simplified limit equilibrium method will over estimate the bearing capacity of bucket foundation. In addition, the error of the bearing capacity calculated by the simplified limit equilibrium method is estimated. The suggestion about the bearing capacity of bucket foundation has reference value to project designer.

The degradation evolution of the bonded geomaterials under different loading conditions is investigated by using distinct element method (DEM). First, considering the bonding behavior at contacts between bonded geomaterials, an equation of degradation parameter, which is used to describe the degradation process in bonded geomaterials, is established based on the micromechanics theory. The equation has microscopic physical significance, which cannot be directly used to establish macroscopic constitutive model. Second, a series of isotropic, constant stress ratio and biaxial compression tests are performed on the numerical specimens by using a two-dimensional DEM code, NS2D; and the evolution of the degradation variable varying with macro variables, i.e. volumetric and shear strains, is investigated under the three loading conditions. Finally, a simple evolution equation for the degradation parameter is proposed; and it is a function of the major principal strain. It is shown that three variables, i.e. bond strength, constant stress ratio, and confining pressure, can influence the degradation evolution of the numerical specimens to a certain extent. The evolution of the parameter varying with the volumetric or deviatoric strains can be easily described by functions in the isotropic and constant stress ratio compression tests. The evolution of the parameter varying with the volumetric strain cannot be easily described due to the dilatancy in the specimens under biaxial compression tests. The equation proposed for the degradation parameter can well describe the degradation evolution observed in the three loading conditions.

For the existence of stope structure body having a great influence on the safety mining of ore block, it is very important to carry out the method for calculating the structure body and its stability. By the application of general block theory, based on the data of structure plane in the developing roadways surrounding the stope, structure bodies in stope were preliminarily worked out by using GeneralBlock code. The authenticity of structure body was analyzed; and its grading table was established. Combined with 3DMine software, the structure body positions were accurately marked by defining the package cuboids. A material mechanics model was established; the minimum area of the fixed structure plane in the fixed structure body was gotten. For the calculated structure body, according to the block mobility theorem and stability theory, using methods of material mechanics and sets, the mobility of structure body during the ore stoping process was analyzed; and the coefficient of structure body stability was obtained. The results for calculating the structure body and its stability provide a good guide for strengthening and optimizing the mining technology of the fractured rock mass.

An induced fracture grouting method has been developed to solve the randomness of the splitting direction of fracture grouting in soil in tunnel. The objective of the method is to control and change the splitting direction of the soil. The results can be used to guide the design and construction of induced fracture grouting. Based on the elastic theory and its basic assumptions, the mechanism of fracture grouting in a single layered soil under the condition that the vertical in-situ stress is larger than horizontal in-situ stress is analyzed. And then the necessity of changing the splitting direction into horizontal as well as the induced method is discussed. The mechanism of induced fracture grouting is analyzed according to the stress concentration theory. By using the finite element method, a research about the stress distribution around the grouting hole is performed, in order to find out how it changes with the lateral pressure coefficient and distance between the grouting hole and the induced hole. The requirements about the lateral pressure coefficient and distance between the above two holes to achieve induced fracture grouting are studied. The formula for calculating critical value of distance between the above two holes is fitted. It is concluded that it appears possible to change the stress distribution around the grouting hole by excavating symmetrically the induced holes at its top and bottom positions; and the distance needs less than 4 times the diameter of the grouting hole. While the distance between the grouting hole and induced hole is less than the critical distance, the stresses of the control points around the grouting hole transform, as a result the slurry would split horizontally. In addition, the critical distance between the induced hole and grouting hole increases if the lateral pressure coefficient increases.

Based on the dynamic consolidation theory of saturated soil and the dynamic constitutive model Pastor-Zienkiewicz III, the wave-induced dynamic response of the silty seabed in the Yellow River delta is examined. Liquefaction potential is estimated using a total excess pore pressure criterion. The liquefaction depth calculated is compared with the site observed data at last. Results indicate that the wave-induced excess pore pressure contains two parts: the transient and the residual one. A quicker reduction with depth of the transient pore pressure and a more significant accumulation of the residual pore pressure in the surface layer tend to happen in the seabed that with a surface hard layer rather than a homogeneity one. Under different wave conditions, the value and the variation trend of the transient pore pressure are similar, while distinct differences are found in the residual pore pressure. Under the annual mean wave condition, liquefaction is impossible to happen. While liquefaction would occur under the 5 and 50 years once wave conditions, a greater liquefaction possibility and a bigger liquefaction depth tend to occur while the three-dimensional effect and a surface hard layer of the seabed are considered, the biggest liquefaction depth is within 2-3 meters under the seabed surface. The calculated liquefaction depth is in good agreement with the depth of the submarine disastrous landforms in the Yellow River delta. It is indicated that the method is effective and reasonable.

In situ core-drilling tests and static loading tests are carried out on rock-socketed pile. The test results show that when using the impact pore-forming construction and the upper part is a weakly cemented soil, pile bottom sediment thickness is generally larger, which is much thicker than the specified value by the foundation code. When the load reached a certain value, sudden and large subsidence is observed, because the end resistance does not take effect due to the existence of a thick layer of sediments. The test results indicate that the current design method of rock-socketed pile which considering both of the end and side resistance is not appropriate when impact pore-forming construction is used and there is weakly cemented soil around the pile. The bearing capacity of single rock-socketed pile should be determined by side resistance in soil and the socketed segment in rock. Numerical analysis is carried out on the distribution of side resistance of the rock-socketed segment and deformation and load carrying capacity of the pile. Under a given applied load, smaller settlement is observed when socketed thickness and rock strength is greater. With the increase of the applied load on pile top, settlement difference caused by the variation of socketed thickness and rock strength increases. The distribution of side resistance of the socketed segment along the vertical direction shows a double-peak saddle shape. When the socketed thickness and rock strength are smaller, the lower peak is higher than the upper peak. When the socketed thickness and rock strength are larger, the lower peak is lower than the upper peak. Under a given applied load on the pile top, smaller socketed thickness would result in larger settlement and side resistance. Due to the thick layer of sediment, the axial force of the rock-socketed segment of the pile gradually decreases from the applied load on pile top to zero along vertical direction, the smaller the socketed thichness, the faster the decrease rate.

Coarse grained material is the aggregation of rock particles with certain gradation, which has special physical and mechanical properties. Based on the results of consolidated-drained triaxial tests and three-dimensional particle flow code theory, the numerical model of triaxial test for coarse grained materials is obtained by programming and redevelopment of PFC3D (particle flow code of three-dimension) from the view of mesoscopic scale. The influence of particle shape on strength and deformation of coarse grained materials is considered by introducing clump particles. The relations among dilation, particle breakage and rearrangement are analyzed. The numerical results show that particle shape is the main factor affecting the strength and deformation of coarse grained materials. Changing the particle shape can significantly affect the mechanical behavior of coarse grained materials in the case of other mesoscopic parameters unchanged. The stress-strain relationship of bonded-particle model (BPM) is agreed with experimental result only under low confining pressure; the deviation is growing larger with the increase of confining pressure. The stress-strain characteristics of coarse grained materials in consolidated-drained triaxial tests are simulated by PFC3D model using clump particles accurately. The BPM and clump particles are both rigid particles. The shear dilatancy is too large because the particle deformation and breakage are not considered.

The traditional strength reduction finite element method (SRFEM) is widely used in numerical analysis of slope stability. However, for the lack of considering geotechnical uncertainties, it still cannot be directly applied to the reliability analysis of slope stability, especially the dynamic stability problems. Thus, based on the SRFEM, the reliability analysis method of slope dynamic stability under seismic loading is proposed. In this method, the finite element limit analysis, dynamic analysis and reliability analysis are included to analyze the reliability problems of slope under seismic loading; and above process is realized in the numerical calculation program. For the traditional method, the acquirement of slope safety factor should be based on continuous manual pilot calculations, which makes batch processing impossible. But in the proposed method, above problem is solved through the improved slope dynamic failure criterion and the corresponding calculation program. The case study shows that the proposed method is significantly different from the general methods; it can relatively comprehensively reflect the dynamic characteristics of rock mass, the variability and correlation of material parameters. The results are closer to the reality. The proposed method not only widens the application scope of SRFEM, but also provides a new effective way for the reliability analysis of slope dynamic stability.

Based on the upper bound limit analysis, the functional program about the slope stability problem was developed by MATLAB. The stability factor (Ncr) and graph of critical slip surface can be visually shown in the program window. The solutions of stability factor (Ncr) were obtained through the calculation results of nonlinear and sequential quadratic programs (SQP), combined with the method of exhaustion to avoid sticking at the local optimization. The results demonstrated the effectiveness of the program. The influence on the stability of slope was evaluated by analyzing the factors in terms of the internal frictional angle ?, slope angles ? and ?. The calculation results show that the stability factor (Ncr) is more sensitive to the frictional angles ? and ? than ? . In addition, if the frictional angles ? and ? are close to each other and the values of ? and ? fluctuate in the narrow range, the slope stability is sensitive to the stability factor (Ncr). In other words, when the low values of ? and ? are inputted in the program, the slip surface would cross below the slope toe, while the slip surface would intersect at the slope toe for the case of steep slope.

In order to study the rock localization phenomenon, based on the examples of the stability of axial pressure bar, applied the mathematical and mechanical knowledge, according to Drucker-Prager criterion, a universal formula can be derived based on the bifurcation theory. On this basis, combined with the finite difference theory, the judging criterion of bifurcation in rock mass in FLAC3D calculation can be worked out. Then under the assistant of own FISH compile language, using the Drucker-Prager criterion, a numerical analysis model is established in FLAC3D to study the rock localization phenomenon and its development process. The results show that in FLAC3D the imbalance force ratio model system came from the 20 000 steps interval gradually tends to be stable, the unbalanced force ratio gradually reduces; the localization phenomenon gradually expands the scale to its stability; and the clarity increases gradually, which confirms that the bifurcation theory can effectively analyze the rock localization phenomenon. The rock bifurcation criteria of rock localization phenomenon and numerical verification given by this paper more scientifically and reasonably expound the origin of the universal bifurcation condition, so as to provide the theoretical support for the explanation of the mechanism and process of the rock localization phenomenon to play a promoting role.

Blast-induced destruction in rock and soil is a research focus in geotechnical engineering. Based on a coupled smooth particle hydrodynamics-finite element method (SPH-FEM) algorithm in the frame of LS-DYNA, a numerical model of analyzing explosion in sand foundation is established to study the explosion effects of small equivalent group charges at different depths of burial(DOB). The crater depth reaching to maximum is prior to the crater diameter reaching its maximum value. The shearing flow of surface layer sand plays significant effect on the evolution of crater diameter. With the increase of DOB, a critical crater diameter can be obtained. Since it can be considered the crater diameter is only related to explosive equivalent and DOB, therefore a basically linear relationship is proposed between crater diameter relative to DOB and scaled distance. Compression waves appearing the similar features of strong discontinuous shock, dominate in the close-in region of explosion. Large deformation of sand is mainly induced by compression. At far field, the blast waves change to the continuous shear waves with reduced amplitude. Attenuation law of peak pressure has a good coincidence with the empirical predictions. The analysis results can provide reference for design and reinforcing of foundation structures to resist blast load.

Based on the U-W formula of governing equations of Biot’s theory for the saturated porous medium, the corresponding 2.5D finite element method (FEM) equations and the viscoelastic absorbing boundary are deduced by introducing Galerkin method and Fourier transform. The solution in wave-number domain can be obtained firstly from the built model for open trenches in the saturated ground; then the fast Fourier transform (FFT) is applied to gain the results in 3D space domain. For the moving loads, the vibration-isolation effects of open trenches in three saturated grounds (homogeneous saturated ground, layered saturated ground and saturated ground with upper layer being single-phase elastic medium) are investigated. It’s found that not only is the vibration-isolation efficiency of open trenches related to the trench depth, but also it’s affected by the interface properties and parameters varying of the layered soil media. The transmission and reflection of waves in different soil interfaces have great influences on the vibration-isolation effects of open trenches. Besides, the thickness of the upper single-phase elastic layer can greatly affect the vibration-isolation efficiency of open trenches in the saturated ground; as the thickness of the upper single-phase elastic layer becomes larger, the vibration-isolation efficiency of open trenches in the saturated ground gets improved.

New bored grouting PHC nodular pile is a new kind of composite pile. And it has a lot of advantages, such as higher bearing capacity, more technologies, and higher security. For the sake of getting to know the compressive mechanical characters of new bored grouting PHC nodular pile, ABAQUS is used to simulate the vertical compressive load test. Comparing the simulated result with the field test result, the reliability of ABAQUS numerical model is acceptable. And it is found that: the radius of nodular stress influence is 2-3 times pile diameter; there is no friction at about 1.6-1.8 times pile diameter distance around of the nodule; ultimate bearing capacity of the nodules increases linearly from top to bottom except the first and the last ones. It is also shown that nodular space, sickness of soil-cement and the elastic moduli of soils of pile side and pile tip influence ultimate bearing capacity of the composite pile; and the elastic modulus of soil-cement can be ignored. Existing formulas of ultimate bearing capacity can’t work for mechanical characters of the new composite pile well. So, by combining theoretical analysis and compressive mechanism, a new formula of ultimate bearing capacity considering nodular influence coefficient is proposed; and its validity is verified by field tests.