As a review paper, the discourses on the mechanism and the definition of rockburst of authoritative experts were quoted, based on these discourses and according to the different mechanisms of rockburst, rockbursts are divided into the sliding mode rockburst resulting from fault-slip events and strain mode rockburst resulting from the failure of the rock. Combined with the specific accident cases, the mechanism and characteristics of the two types of rockburst are analyzed. Finally, based on the analysis of the mechanism, the quantitative forecast and numerical simulation of rock (coal) pillar strain mode rockburst, enclosing rock strain mode rockburst and sliding mode rockburst are introduced, in which, the author’s latest research of quantitative forecast and numerical simulation for strain mode rockburst after incompatible deformation by using non-Euclidean geometry model is included.

Extensive laboratory tests and field observations show that soft clay exhibits significant rate-dependent behavior. Based on published data from the view of the rate dependencies of preconsolidation pressure and undrained shear strength. The rate- dependencies of soft clays under both 1D and 3D conditions are first studied. The applicability of five rate-dependency equations (two equations in exponential form and three equations in logarithmic form) in correlating the preconsolidation pressure and the undrained shear strength with strain-rate is then discussed. Rate parameters under both 1D and 3D conditions for various clays are estimated by using five rate-dependency equations and correlated with Atterberg limits. Furthermore, the rate-dependencies of soft clays under complex stress conditions including shear vane tests and pressuremeter tests are also analyzed. Then, the uniqueness of rate-dependency between 1D and 3D, between triaxial compression and extension, between different OCR (over consolidation ration) conditions is discussed. Finally, the influence of applied strain-rate on the stress-dilatancy of Hong Kong marine clay for triaxial compression and tension tests under different OCR conditions is investigated with some typical stress-dilatancy equations. The results show that the existing typical stress-dilatancy relationships need further enhancement (e.g., amount, anisotropy) for better describing the stress-dilatancy of soft clays.

Based on the theory of unsaturated porous media proposed by Zienkiewicz, considering the inertia, viscous and mechanical couplings and the compressibility of fluid and solid particles, semi-analytical solutions for one-dimensional transient response of single-layer unsaturated porous media are obtained with the example of a typical boundary condition problem. During the solution procedure, the dimensionless displacement governing equations in matrix form are derived firstly. Then the the nonhomogeneous boundary conditions are transformed into homogeneous boundary conditions and the eigenvalue problems of the governing equations are solved. After that, the variation coefficient method and the orthogonality of the characteristics functions are utilized to obtain a series of ordinary differential equations with their initial conditions. Applying the precise time-integration method, the semi-analytical solutions of the ordinary differential equations are developed. Finally, several numerical examples are provided to prove the correctness of the present solution and investigate the features of one-dimensional transient response of unsaturated single-layer porous media. This method can be extended to any other boundary conditions.

A series of large-scale consolidated drained triaxial tests were conducted to research particle breakage of Zipingpu rockfill material under monotonic and cyclic loading for different initial void ratios. The results show that there exists a hyperbolic relation between the input plastic work and particle breakage, regardless of the initial void ratio or confining pressure under both monotonic and cyclic loading. The dilatancy factor at peak decreases with the increasing of particle breakage and there exists a nearly linear relation between particle breakage at peak and the dilatancy factor at peak on semilog coordinate, irrespective of initial void ratio. Test results in this study also indicate that there exists an approximate linear relation between peak principal stress ratio and the dilatancy factor at peak, regardless of the initial void ratio. This study results can provide useful references for state-dependent elastic-plastic constitutive model considering particle breakage and finite element analysis of Zipingpu concrete face dam during Wenchuan earthquake; and help to further understand the particle breakage of rockfill materials.

Soil is a granular medium, and its strength and deformation characteristic behave marked grain size effect. Based on the theoretical model of cell element and triaxial shear tests, the analysis of grain size effect is presented. The matrix-reinforcing particles cell element model of soil is proposed according to cohesion and friction effect of interaction between particles of varying size. The cell element is composed of matrix and reinforcing grain. Many tiny grains are grouped into the matrix and reinforcing grain is a sand grain, so the macroscopic soil is simplified as a medium that consists of a number of cell elements. Introduced the generalized spherical strain and equivalent strain, the stress-strain relation and yielding stress formula are derived from strain energy accounting for the grain size effects. A series of triaxial undrained shear tests are performed for test samples with reinforced particles of different sizes and volumes fractions, and test results of stress-strain curves and yielding stresses are presented on the size effect. The experimental results and theoretical analysis show that the size effect of the strength and deformation of soil is increased with volume fraction raising and decreasing of particle size, which suggests strong size effect. The theoretical prediction of size effect is in good agreement with that of the test.

Based on the domestic and foreign representative calculation methods of rockfall impact force, in view of the factors of impact force, a set of impact force testing equipment is designed. Different rockfall qualities, impact speeds, incident angles, properties and thickness of buffer material as influence factors are selected in experimental design to obtain the change rules of impact force. The results show that the maximum impact force gradually reduces with the decrease of rockfall weight or the lower impact velocity, the force in 2 cm thickness of buffer layer is reduced about 90% than without it; the buffering effect is obvious; and the maximum impact force reduces with the incident angle lowering; but the little decrease amplitude of impact force value as the incident angle become smaller. Combining with impact testing results, a calculation method of maximum impact force associated with each influence factor, is obtained. By checking, it is shown that the error is small, so it can provide certain reference basis for treatment design of rockfall hazards.

For the current status of research that it rarely considers the impact of the nature of soil when calculating water and earth pressures, using self-made testing tank as research tool through measurement study by using three types of soil samples with various permeability coefficients, the measured water and earth pressures are compared with the theoretical values obtained by water and earth pressures calculated separately and together method; and the variation laws of pore water pressure and earth pressure are analyzed. The experimental results show that the values obtained by water and earth pressures calculating together method is lower than measured values; but still acceptable under certain conditions. Accompanied by active displacements, it generates negative exceed pore pressure in soil behind wall. Dissipation of negative exceed pore water and transfer of pore water pressure are both affected by permeability of soils which further affect the distribution of water and earth pressures on retaining structures. From the experiment results we can speculate on that the water and earth pressures by calculating together method is reasonable when the permeability of soil reduces to certain level.

The developed flexible uniform pressure loading device and traditional rigid loading device are applied to large-scale geomechanical model test, of which loading effect is comparatively analyzed by using the stress nonuniformity coefficient ?. The test results show that, with the loading stress and the distance from the surface increasing, the stress nonuniformity coefficient ? decreases, and stress distribution tends to uniform. Comparing to the rigid loading method, flexible uniform pressure loading device is more beneficial to uniform stress dispersion and forming uniform stress field. With the flexible uniform pressure loading device being applied to the test, the stress field is nearly uniform where the depth from the surface is 300 mm in the test condition; and the ? is 0.086 when the initial stress has been applied; and the corresponding ? is 0.283 with rigid loading method applied in the test. Even though the distance from the surface is short or the loading stress is low, the flexible uniform pressure loading device also produces a good uniform loading effect, and when the on-load pressure is 0.28 MPa, the ? is only 0.174 where the depth from the surface is 150 mm. The device has the function of reducing range of boundary effect and improving model test accuracy.

Theoretical analysis and field test are carried out on the mechanism of bulge phenomenon under geomembrane in plain reservoir of which the bottom paved with geomembrane. It is pointed out that the uplifting or bursting problem will occur during surrounding embankment construction, reservoir water quick reduction or groundwater level rising. A field test is carried out at Dezhou Datun reservoir, during which different loadings over geomembrane, different amplitudes and rates of groundwater level rising are simulated and controlled. The pore pressure at different depths in soil is measured and the bulge phenomena under geomembrane are observed. The influences of different influence factors on gas field are analyzed. The results show that the groundwater level rapidly rising results in gas pressure increasing under geomembrane and the gas pressure value depends on the rising amplitude of groundwater level; different uplift rates have little influence on gas field under geomembrane. A certain loading over these geomembrane can effectively ensure the safety of geomembrane.

Through the dynamic tests on the CAS-1 lunar soil simulant (national standard sample) which is developed by the Institute of Geochemistry and the National Astronomical Observatories, Chinese Academy of Sciences with the GDS resonant column test instrument, the effect of the different void ratios and the different confining pressures on the dynamic shear modulus G and the damping ratio ? is analyzed. The test results show that, the dynamic shear modulus G is smaller as well as the damping ratio ? is larger, and dynamic shear modulus decreases rapidly as well as damping ratio increases sharply with the increase of shear strain in the real lunar surface environmental conditions with low stress level and large void ratio. By using the Hardin-Drnevich model, the average fitting curve of G/Gmax and ? /? max with the change of the dynamic shear strain ? under different void ratios and different confining pressures are obtained; and the correlation formulas between dynamic parameters of the maximum dynamic shear modulus Gmax, the maximum damping ratio ? max, the reference shear strain ? r and the stress level ? are discussed. Accordingly, the range intervals of Gmax, ? max, ? r within the drill sampling depth (0-3 m) of the lunar soil as well as the recommended values of G/Gmax and ? corresponding to each of the shear strain are presented.

Gravel cushion has been widely applied to various fields of civil engineering; but its strength and deformation characteristics is still not very clear; so it is needed to further study. Through the laboratory model test and finite element analysis of the gravel cushions with 5 different thicknesses in high stress, the strength and deformation characteristics of gravel cushions with different thicknesses are revealed. Comparing the finite element analysis and test result, the variation characteristics are obtained that the load and deformation, the vertical stress and lateral pressure of gravel cushions with different thicknesses are approximately linear. The thickness of gravel cushion has some effect on the settlement. All else being equal, the settlement increases with the thickness of gravel cushion increasing. Under certain loads, the compressive deformation of gravel cushion approximately trends to be stable within a short period of time; and then the variation of compressive deformation is less as time increases. A test determination method is proposed to measure the lateral pressure coefficient, Poisson’s ratio and deformation modulus of gravel soil, so as to provide references for other soils.

In a dynamic replacement (DR) process, the length and the subsidence of the replacement pier need to meet the design specifications. The times of tampings and the fillings to fulfill a DR process is generally determined by field test. Understanding the forming mechanism of the replacement pier, and quantifying the factors such as energy releasing time, motion characteristics of fillers, stress of pounder bottom, etc, play critical roles for optimizing the DR process. In this paper, a model test method to study the interactions of the fillers and soil under the multiple impacts of the pounder is established. Experimental rusults show that each acceleration-time curve in DR can be divided into four stages, representing different phases of the soil-fillers motions under each pounding. Through analyzing the force-displacement (F-s) curves of the pounder in this process, the characteristics of each F-s curve are distinguished and the intrinsic mechanism is discussed. It was shown that with the increasing of filling times, the energy releasing time during each pounding decreases and the maximal force increases, signifying the gradual forming of the replacement pier in the soft soil foundation.

The CO2 breakthrough pressure (threshold pressure) is one of key indicators for the caprock sealing ability and safety of CO2 geological storage. Different breakthrough pressures experiments have been carried out for different caprock examples around world, and set up a fundamental database for caprock sealing ability evaluation; however, very few researches has been carried out on CO2 breakthrough pressure experiment under different pressures and temperatures. Based on the characteristics and range of breakthrough pressure equipment, the CO2 breakthrough pressure measurement apparatus is developed. The apparatus integrates several methods, such as, step method, continuous method, pulse decay method, and dynamic method. It covers the ordinary range of breakthrough pressure in cap-rocks. Experimental studies of CO2 breakthrough pressure in water saturated argillaceous siltstone sample from Xujiahe formation are carried out; for the Xujiahe formation is an important target formation and seal formation for CO2 aquifer storage in Sichuan Basin. The CO2 breakthrough pressures of core samples under different temperatures and pressures are studied in the experiment. The experiment results show that the breakthrough pressure increases with temperature increasing and CO2 pressure decreasing within the studied pressure and temperature range. The experiment results have the same tendency as the theoretical results deduced by theoretical method; however, there is a discrepancy between experiment results and theoretical results. The discrepancy might be caused by the variation of contact angle, pore-geometry heterogeneity, and experiment method; and this need further study.

Gypsum, lime and sodium carbonate are selected as additives to eliminate the disbenefits of organic matter on cement stabilization. Response surface methodology (RSM) based on central composite rotatable design (CCRD) is employed for experiment arrangement and optimizing the ratios of additives, in which response values are the unconfined compressive strength values of 7 days and 90 days while the contents of additives are independent variables. Meanwhile, one-factor and mutual-influence analysis are used to investigate the effect of each factor on response values and evaluate mutual-influence among the three independent variables. The results show that strength as response value becomes more sensitive to the content of gypsum and lime as the age increases. However, the situation is reverse for sodium carbonate. With the age of 7 days and 90 days, the mutual-influence for lime and sodium carbonate, gypsum and sodium carbonate are significant respectively. On the premise of humic acid content is 6%, cement content is 15%, the optimal ratios for additives with the age being 7 days and 90 days are presented and once the optimal ratios for additives are applied, the actual strength of stabilized muddy soil can reach 623 kPa(7 d) and 1 213 kPa(90 d).

A model test performs to study the failure progress of different slopes morphology soil slope by a self-made one-way vibration test-bed. The test includes eight test models and three patterns of slope (concave slope、convex slope and combination of concave and convex slope), The failure of slope is induced by applied initial displacement with the vibration test-bed and the failure progress is recorded with camera, slope models deformation and failure characteristics are summarized and analyzed. The result shows that the concave slope failure reason is slope surface formation of transverse crack drive the whole slope destroyed, convex slope failure is due to the slope top formation of transverse crack drive the whole slope destroyed, combination of concave and convex slope failure by the raised portion of slope surface control. Further studies proves that the more intense the degree of the concave and convex parts is, the more strong the slope damage is. The study results are useful to understand better the progresses and forms of soil slope failures with different surface morphologies and provide a scientific basis for treatment of engineering slope.

Linear equation between peak stress and Young's modulus with confining pressure is obtained from triaxial compression for limestone under different confining pressures. The relationships between strain energy and strain under three confining pressures are fitted by using stress-strain curve, which can obtain the energy value of limestone in anyone axial strain. Several energy changing equations of several stages in the compression process are analyzed, so as to explain the more several rock failures under higher confining pressure in consideration of the rock sample doing work to hydraulic oil. Based on crack extension degree affects the value of the Young's modulus, the concept of crack development coefficient F is proposed; and the empirical formula to strike F using ? under the three confining pressures is given; and the real rock mechanical parameters of various stages during the limestone compression are got; the plastic platform appearing near the peak intensity under high confining pressure with changing tendency of F is explained.

Dynamic triaxial test is performed to study dynamic strain amplitude of Qinghai-Tibet frozen clay and Lanzhou loess under staged dynamic loading with different loading frequencies, confining pressures and negative temperature conditions. The results show that the dynamic strain amplitude is basically unchanged with increasing vibration cycles under the same load level; and average values are used to characterize dynamic strain amplitude. The law of that dynamic strain amplitude changes with dynamic stress amplitude are the same; namely dynamic strain amplitude increases with increasing of dynamic stress amplitude with different loading frequencies, confining pressure and negative temperature conditions. Dynamic strain amplitude decreases with increasing of loading frequencies, but the decreased rate is becoming more and more small that it tends to be a stable value, which increases with the increasing of loading series. With increasing of confining pressure, dynamic strain amplitude for Qinghai-Tibet clay is basically unchanged and for Lanzhou loess decreases gradually. Dynamic strain amplitude decreases with decreasing temperature. The most important effect on dynamic strain amplitude is dynamic stress amplitude, vibration frequencies of dynamic load secondly, temperature thirdly, and confining pressure with minimal impact.

Expansive soil as a kind of special soil, its shear strength value is different from the common cohesive soil. In practice, it is proved that in the design of expansive soil slope, not only the soil block strength, but also the crack surface strength must be considered. It is impacted by the crack surface scale and occurrence, in particular, to the strong expansive soil which grows extremely crack surface. In order to simulate more realistic shear failure of strong expansive soil under the control of the crack surface, the in-situ direct shear experiment of strong expansive soil for South-to-North Water Transfer Project in Nanyang section is made; through matching the data of strong expansive soil, and shear strength influencing factors are analyzed. The results show that all of the crack surface development degree, the slope angle and the crack surface fluctuation degree have great influences on the strong expansive soil shear strength.

A suitable yield criterion is the key to elastoplastic solution of axisymmetric circular tunnel. Many scholars observed corresponding solutions based on criteria of Mohr-Coulomb、Drucker-Prager and Hoek-Brown. The paper establishes power- hardening model, power hardening-perfectly plastic model based on Drucker-Prager failure criterion of axisymmetric circular tunnel; and acquires the elastoplastic solutions, taking strain-hardening effect into account during the calculation to meet the needs of practical engineering. For a project example, the paper compares the results of power hardening-perfectly plastic model based on Mohr-Coulomb failure criterion and Drucker-Prager failure criterion, and power-hardening model respectively. The influence of power-hardening index n on elastoplastic solutions of surrounding rock is analyzed. The study results show that the effect of strain-hardening is of great impact on the stability of surrounding rock, and it is of more practical value to adopt power-hardening model for the rock of great strain-hardening effect.

In order to improve the description method of variation of structural parameters, according to the existed problems of the stress ratio structural parameters, this article proposed the equation description of initial structural parameters that may take the initial water content and the confining pressure of structural loess into account, and also established a more reasonable description of equation between the structural parameters and the generalized shear strain, which is introduced into the strength theory of structural loess. Then the distribution of structural parameter and the potential sliding surface of structural loess slope are evaluated through the numerical computation. The results show that: the newly established initial structural parameters can reflect the distribution of initial structural parameters in structural loess slope; the upper limit of structural parameters is the initial structural parameters, while the lower limit of structural parameters is the structural parameters that the structural strength is completely lost; and its value is equal to 1.0. In the stability analysis, the potential sliding surface can be determined through the distribution of structural parameters to some extent; but it just can be effective under the circumstance that the plastic zone in slope shows the obvious banded concentration distribution.

To avoid pavement structure failure induced by subgrade settlement, index and criterion for settlement control applicable to cut-and-fill subgrade for mountain-highways are studied. Firstly, the Prony series expressions for three-dimensional Merchant model are deduced based on constant Poisson’s ratio hypothesis; and the settlement calculation models of cut-and-fill subgrade were established with ABAQUS. The calculations indicate that post-construction settlement curves of cut-and-fill subgrade are spoon-shaped or S-shaped. The difference in curve shape reveals the time-space coupling effect in subgrade settlement. Secondly, the following hypothesis is proposed based on post-construction settlement mechanisms: the post-construction settlement curves of cut-and-fill subgrade are central symmetric S-shaped curves composed of two parabolic segments. The mechanical response of asphalt pavement under parabolic subgrade settlement indicates that the additional flexural-tensile stress is linear with quadratic term coefficient of parabolic equation which, consequently, can be adopted as settlement control index applicable to cut-and-fill subgrade. Finally, the method of defining settlement control criterion applicable to cut-and-fill subgrade is presented based on the failure modes of asphalt pavement under traffic load and subgrade settlement. The research findings provide a technical foundation for post-construction settlement control of cut-and-fill subgrade.

Mud screen is formed to reduce the jacking friction in long-distance and large-diameter pipe jacking; the form and wholeness of mud screen determine the lateral friction resistance for pipe-jacking; the existing formulae of lateral friction resistance do not take account of the effect of different forms of mud screen; there are large differences between the calculation results and the actual friction resistance. The possible forms of mud screen are classified and analysed based on different construction conditions; the method for determining the forms of mud screen is presented. The influence of the injecting pressure on the thickness of mud screen is discussed; the lateral friction resistance caused by mud slurry is calculated using nonlinear fluid mechanics. Finally, a formula of lateral friction resistance for pipe jacking with different forms of mud screen is put forward using the assumption of stable mining face. The new calculation method is used in an engineering case; the calculation results are compared with the measured data; it is shown that the existing formulae are conservative and the formula of this article is more reasonable.

Setting earth berm before retaining structure can reduce effectively its internal force, deformation and embedded depth, and achieve good economic and environmental benefits. Based on various technical codes for excavation engineering can’t provide the design and calculation of the earth berm, and very few research this problem and the point of view is not uniform in geotechnical engineering field, the past calculation methods of earth berm is collated and analyzed firstly, and four categories are induced: (1) adding load before the retaining structure; (2) subsidiary stress of Boussinesq approximation; (3) improved elastic foundation beam method; (4) whole finite element collaborative computing, then the doubts, key points and applicable conditions in these methods are pointed out; next the mechanism of earth berm is summarized. At last, based on elastic foundation beam theory, a new simplified method to analyze earth berm is proposed, which used pressure springs and overload to simulate the back pressure effect, and by comparative calculating of setting earth berm, not setting earth berm, reinforcing earth berm in an excavation engineering, the rationality of this method is verified; and it can provide reference for design and construction of retaining engineering.

It is the essential prerequisite for mechanical analysis of the surrounding rock around underground cavern with the complex variable theory that the mechanical boundary must be transformed from the actual excavation cross-section to the unit circle. According to the Riemann’s existence theorem, the mapping function is established with the finite Laurent series, by which the exterior of unit circle can be conformally mapped to the exterior of carven with the arbitrary excavation cross-section. Based on the boundary correspondence principle, the conformal mapping of source and target domains is transformed into that of the boundary lines of unit circle and cavern. Research on solving the mapping function is carried out with the triangle interpolation theory, by which even and odd interpolation points are repeatedly iterated each other. In order to accelerate the iterative calculation, the calculating points are adjusted by the normal approximation method. Precision of the conformal mapping is ensured when the absolute error between the mapping boundary and the actrual excavation boundary is set as the convergence condition. Moreover, examples of the calculating mapping functions are provided for caverns with three excavation cross-sections. Compared to other methods in the literatures, the conformal mapping of the exteriors of unit circle and cavern with the arbitrary excavation cross-section is achieved more simply, accurately and rapidly by the method in this study.

In order to control the deformation of subgrade under long-term traffic load, the dynamic deformation control method is employed to study the relationship between the critical height and the humidity of subgrade from the viewpoint of mechanics. Firstly, a quantitative formula of resilient modulus, humidity and compactness is obtained according to the result of laboratory resilient modulus test. Then, combining the solutions of dynamic response of the multilayered subgrade structures and the principal of compatibility of deformation between pavement and subgrade, the allowable value of dynamic deformation of subgrade top is derived. Based on the quantitative formula and the control standard of dynamic deformation of subgrade top, the relationships among the critical height, humidity and the compactness are built up. Finally, numerical results are obtained by taking two typical asphalt pavement structures for example. It is shown that, the critical height of subgrade increases with the increasing humidity. Both of the subgrade humidity and the compactness have little effect on the critical height of subgrade when the subgrade humidity is less than the optimum water content. However, when the subgrade humidity is more than the optimum water content, the critical height of subgrade increases sharply with the increasing humidity; and the effect of the compactness on the critical height becomes significant. The results provide a new idea for reasonable value of the subgrade height and humidity.

In order to understand the mechanism of expansive soil canal slope slide, a study is conducted based on the statistical results of 19 landslides at the Nanyang Section of Middle Route of South-to-North Water Transfer Project. The right bank landslide at TS105+400 is selected as typical landslide. Excavating trench exposes internal structure of landslide. A series of geological survey has been done to study the characteristics and evolution mechanism of expansive soil landslide in the area by excavating trench slides. The results show that, most landslides in the area occurred in the strata. Slope stability is jointly controlled by vertical joints in the upper layer soil and the gently dipping big fissures at the toe of the slope, which are filled with highly expansive clay. Sliding surface is composed of the steep tilting cracks at the trailing edge of slope and big gently dipping fractures at the toe of slope. Excavation unloading effect causes the vertical joints’ opening. These joints can extend more than 3 meters into the slope, which destroys the integrity of the slope soil and serves as the main channel for water to go in and out of the slope; there is a high humidity area 4 to 8 meters below the surface of slope. Soil within the area is weak and the sliding surface develops in it. Wetting and drying cycles caused by climate and excavation unloading lead to the vertical crack to open and propagation deep into slope. This effect causes a dramatic attenuation of soil strength. The vertical cracks connected with the big gently dipping fissures at the toe of the slope eventually. The slope slides after these fissure and crack are filled with water after the occurrence of strong lasting precipitation.

Soft rock slope engineering catastrophic mechanism is a hot difficult subject in slope engineering. According to soft rock slopes unsaturated seepage-stress calculation principles, indirect coupling method is used to analyze soft rock slope stability under rainfall infiltration. A method for analyzing seepage effect has been established. The water-softening characteristics of the shear strength of soft rock are given to the corresponding soft rock in transient saturated zone dynamically, simulating the softening effect of soft rock slope in rainfall seepage. Then, combining seepage and softening, a numerical analysis method for soft rock slope rainfall-seepage-softening-cataclysm analysis has been established based on transient saturation region, which is used to research the disaster mechanism of the slope engineering. The results show that, the depth of the potential sliding surface shallow gradually over time; the shape of stability factor-time curve is reverse recovery curve; and the inflection point appears on the curve when the slope begins to soften. The slope stability calculation results are consistent with the actual situation, so as to show that the proposed thought and method have some validity and can provide a theoretical basis for soft rock slope management program design.

Approximate analytical solution and finite element solution for the water flow of single fracture and variable water temperature considering the impact of water temperature on rock fracture seepage have been deduced. Based on the analytical formulas, the seepage numerical equations for two-dimensional fracture network under variable water temperature are formed, which apply to the approximate analytical methods and sub-structure method to variable water temperature analysis of fracture network seepage respectively. Some results can be found as follows. (1) The water head and the hydraulic gradient in single-fissure are nonlinear dependence. Seepage curve is convex curve when the current flow from the higher temperature area to the lower temperature area. In the condition, the water head slants low under liner seepage simplification. On the contrary, seepage concave is convex curve when the current flow from the lower temperature area to the higher temperature area, and the water head slants high. (2) In a single fracture, the hydraulic gradient is smaller in higher temperature area and is larger in lower temperature area. Higher the average water temperature of fracture is, and faster the flow velocity is. There is deflection effect caused by temperature in the fracture network. In the crossing nodes fluid tends to flow to the higher temperature area. Consequently, the effect of temperature change to seepage should be taken into account in the region of high temperature and high temperature gradient.

According to a large number of findings on seismic hazards, the damages of tunnels during earthquakes depend on tunnel depth to a great extent. In order to study the effect of tunnel depth on lining, the earthquake responses of mountain tunnel in eight different depths are calculated by using the finite element method; and the seismic force loading method is verified, which confirms the rationality of the seismic load input method. In order to analyze the change of peak acceleration/displacement with increasing tunnel depth, the results of peak acceleration/displacement in both horizontal and vertical directions on key nodes of lining are extracted; and the influence of different tunnel depths on structural dynamic response is discussed. Some rules that the dynamic parameter peaks of lining decreased rapidly when the tunnel depth increases from 5 m to 50 m are found when calculating the ratio of peak acceleration/displacement on the section of different tunnel depths at vault. In addition, the influence of tunnel depth on peak internal force at tunnel vault is analyzed. And the comparative analysis between the results of the finite element calculation and the model test is made. Finally, the conclusion can be obtained that tunnel depth could not be less than 50 m when constructed in high-intensity earthquake zone, so as to provide reference for related projects.

The frame prestressed anchor cable computational model for system reliability is put forward basing on a series system, which is constituted with main failure modes of the frame beam destruction in tension stage or in working stage, destruction of stringers and beams, destruction of the beam in cross-section and oblique section, and anchor damage. During tensioning stage, the anchor cable force is assigned on the beams in accordance with the principles of displacement coordination and static balance; internal forces of the monolithic beam is calculated respectively by Winkler elastic foundation beam model. In the working stage, to calculate their internal forces, the lateral earth pressure coefficient is regarded as random variables, frame beams as simply supported beam under the earth pressure. Basing on the correlation coefficient matrix of the failure mode function, prestressed anchor computational model for reliability of the system and computer program are then derived. The results of an engineering example show that the probability of failure of the prestressed anchor frame is determined by three factors: cross-section failure probability of in working stage, cross-section failure probability of tensioned stage and anchor bolt failure probability. Assuming the error of three failure modes interdependent system failure probability and system failure probability considering correlation is 8.1%.