Based on thermodynamics and stress theory of continuum mechanics, a series of stress state variables of unsaturated soils, which covered all stress state variables used in literature, are deduced from the equilibrium equation of soil skeleton. It is shown that there are more than five combinations of stress state variables for unsaturated soils, most of which are related to physical properties such as void ratio and degree of saturation. Stress state variables should have dimensions of stress, while void ratio, degree of saturation and their combination could not be regarded as stress state variables. Both the simplified Bishop’s effective stress equation and the Fredlund’s two stress state variables are simplified forms of stress state variables. The influence of void ratio is not included in the simplified Bishop’s effective stress equation, while the dependence on void ratio and degree of saturation are neglected in the Fredlund’s two stress state variables. When choosing stress state variables for unsaturated soils, rationality in theory and convenience in application as well as correctness in logic should be taken into account

The consolidation characteristics of a saturated silty clay under thermal loading are discussed. Test results show that, for various confining pressures, the evolution patterns of whether pore pressure or volumetric strain is similar. In the process of temperature rising, the pore pressure increases continually with elapsed time, and the pore pressure also reaches to a peak pore pressure (or a steady value) when the specimen is heated to a required higher temperature value. Furthermore, this peak pore pressure increases as the confining pressure increases. Besides, the consolidation volumetric strain (i.e., shrinkage deformation) by heating is greater than the volumetric strain (i.e., expansion deformation) by cooling; in other words, the specimen totally takes on a shrinkage effect after a complete process of heating and cooling.

Water-sensitivity and structural property of loess which have been widely studied are the two marked characteristics compared with the other soils. But most test researches are always conducted by using soil samples taken only in the vertical direction, which means only the vertical mechanical behaviour is taken into account. It is absolutely both reasonable and correct to study the vertical mechanical properties of loess mainly loaded vertically by upper foundations, but for those constructions with complicated stress state at different parts such as loess slopes, tunnels and/or underground structures, the anisotropy of loess might be more significant. The undisturbed Q4 loess in Lanzhou is sampled both from vertical and horizontal directions; then triaxial shear tests are conducted to investigate the difference of shear strength and deformation parameters under different confining pressures. The shear strength parameters obtained from triaxial test are also compared with those from direct shear test. The test results show that the anisotropy in mechanical behaviours of Q4 loess in Lanzhou is remarkable. The anisotropy in shear strength is mainly externalized by cohesion but the friction is nearly the same. Both the vertical and horizontal deformations can be characterized by conversion of brittle failure to plastic failure while the confining pressures are increasing. The difference in deformation modulus of loess between vertical and horizontal directions is quite obvious, i.e. the deformation modulus in vertical directions is averagely twice than that in horizontal directions. The test results are explained by using the structural strength and corresponding forming mechanism of loess. The anisotropy in mechanical behaviours of loess can be considered as another type of structural property of loess, which should be paid more attention in order to get more safe and reasonable parameters.

By the application of computer digital image measuring technique to sample deformation measurement of soil static triaxial test, non-contact measuring of soil specimen deformation was realized. For Nanning saturated fine sand, loose and dense samples were prepared respectively by dry-rapping method and wet-tamping method controlling the different initial moisture content. Combining the digital image measuring technique with traditional triaxial deformation measuring method, consolidation drained shear tests were made under different confining pressures. The influence of sample preparation methods, initial moisture content, degree of compaction and consolidation pressure on relation of stress and strain and different of tangent modulus 5 parameters of Duncan-Chang model by means with the two measurement methods are researched. The test results show that the forms of sample shear failure are decided by the sample preparation method, initial water content, density and confining pressure; and the relations between stress and strain are affected also. There were differences of axial strain and deviator stress by image measuring and traditional triaxial cause because the different shear failure forms and the location of shear band. Duncan-Chang model 5 parameters are affected not only by the sample preparation method, initial moisture content and, density, but also by the methods of measurement.

In order to compare microstructure of lunar soil and lunar soil stimulant, formative processes and five grain type of lunar soil are introduced. We analyze microstructure of lunar soil through lunar soil photos. There is a composition test on pozzolan that is taken as main materials of lunar soil simulant. Microscopic image analysis tests on pozzolan have been done. Test and analysis results show that there are agglutinates in lunar soil. Agglutinate grains take on branching form of organization and closed cell and have metallic iron globules. Main component content of pozzolan is similar to that of lunar soil. Pozzolan crushed show obvious angular, slightly rod-like and little peak value of aspect ratio, which are similar to lunar soil. But complexity factors have little deficiency, which account for pozzolan as simulant materials have not enough coarse and angular.

In order to overcome the defects such as short single cycle footage, low utilization of blast-hole, too many blasting fragment pieces assembled near the roadway heading face, insufficient excavation of dip roadway floor and so on, that often occur in dip roadway excavation, the new cut formation of wedge cut with assisted holes that used to enhance the thrown effect of fragment pieces is designed. The designing of the new cut formation bases on the new idea that using the middle empty holes to assist the broken cut holes in throwing the fragment pieces out of the cut cavity during the blasting. For the sake of proving the advantage of the new cut formation, many in-situ experiments are conducted in the Huaheng coal mine. By means of changing the layout of the cut holes, changing detonating sequences, setting special auxiliary holes to increase the throw distance of the blasting fragment pieces, and using mixed arrangement type of auxiliary holes, three kinds of design schemes of dip roadway blasting excavation are formulated. Through the analysis of single cycle footage, depth of the blast-hole left after blasting, outline of the laneway after blasting, thrown distance of the fragment pieces, lumpiness of the fragment pieces, and height of the fragment pieces in front of the roadway heading face, the perfect blasting design is put forward, of which the overall indicator is the best. Thanks to the technique, the single cycle footage of the coal mine increases by 40% everymonth. The success of the test proves the validity of the technique to overcome the defects in coal mine dip roadway excavation.

Through the laboratorial research, the coupling correlation between stress and P wave velocity are studied under uniaxial cyclic loading and unloading ways to create the experimental model based on the characteristic that the velocity gradient is usually high at elastic stage and then begins to level out at plastic stage for burst tendency coal and rock samples from deep coal mine. Also the parameters of the derived model are analyzed. Studies indicate that: ①In the unloading process, the P wave velocity decrease with the drop of stress because of the reopen of closed crack and weakening in the compaction degree between grains; ②The starting P wave velocity of next cyclic is higher than the starting of first cyclic, because that partial crack and pore that closed permanently in the compression phase can not reopen when stress decrease; ③The calculated correlation coefficients to measured value show that the experimental model has the better fitness , and can be well applied to describe the P wave changing sensitivity degree of samples under stress.

Natural soft clays are generally subjected to the effects of soil structure. So evaluating the influence mechanism of soil structure on the compression behavior of soft clays is of utmost importance. One dimensional consolidation tests are performed on two natural soft undisturbed clays to obtain their compression curves, which are compared with the collected consolidation test data for 39 natural soft clay samples collected in literatures. It is observed that the compression curves under logarithmic coordinates show an inverse S-shape and obvious consolidation yield stress. The oedometer test data can be well represented by two straight lines under bi-logarithmic coordinates and the consolidation yield stress can be easily obtained. Relationships between compression index in the post-yield state and water content at yield point for total 41 clay samples can be well quantitatively presented, which is consistent with the quantitative relationships for remolded clay samples between compression index and corresponding water content at yield point. Comparison results show that the compression behavior of natural clays in the post-yield state is mainly controlled by the water content at yield state, which is independent of the resistance of soil structure. Compression curves for natural clay samples lies above the corresponding curves of remolded samples can be explained as being induced by the difference in water contents.

A new reinforced materials as to strengthen the rib of conventional geogrid is proposed on the basis of horizontal-vertical reinforced soil, which has the effect of three-dimensional reinforcement. A series of pull-out tests of geogrid with strengthening ribs are performed to investigate the interface behavior between sand and strengthening rib geogrid inclusions in terms of different forms of reinforcements. 36 groups of pull-out test data are received by a large number of tests. The impacts of tooth height and tooth spacing of reinforcement on ultimate pull-out resistance are discussed. Test results show that the ultimate pull-out force of strengthening rib geogrid significantly larger than ordinary ones in the same normal stress. Ultimate pull-out resistance increased as the increase of the height of strengthening rib; while it decreased when increased the rib spacing under the same degree of rib height. From the test results, the coefficient of apparent pull-out friction is evaluated. The interaction mechanism between soil and geogrid with strengthening ribs is further analyzed. The influencing factors of ultimate pull-out resistance are discussed; and a new theoretical formula is proposed to determine the pull-out resistance. The comparison between theoretical values and experimental results is in good agreement.

Currently, in China, compaction test with five-point method is the most basic earthwork test in the various specifications of soil test. The test has some drawbacks such as needing much more test soil and test time, great effort and meeting trouble in data processing. In order to overcome these shortcomings, different soil samples being prepared is dehydrated under the conditions of the timing constant pressure that its moisture content is higher than its plastic limit water content. There is a law that the moisture content after being dehydrated was stable, it is always lower than the optimum water content under standard compaction work. Based on the characteristic, optimum moisture content of the soil could be calculated by dehydrated moisture content and the correlation coefficient multiplied. Soil optimum moisture content and maximum dry density under standard compaction work could be obtained quickly and accurately using single-compaction method. After improving the method of compaction test, it is simple, fast and reliable comparing with the five-point method.

Rock creep deformation and failure characteristics are discovered and rock creep velocity and propagation process under different stress paths are represented through argillaceous sandstone creep tests. A nonlinear creep model for argillaceous sandstone is set up. Creep characteristics of surrounding rock in underground engineering under the condition of plane strain and support are represented through the application of lateral and anchoring constraints, respectively. The research results show that anchoring constraint can increase the critical stress value that rock produces stable creep and decrease creep velocity evidently. The parameters for creep constitutive equation is determined by the fit of experimental data according the improved visco-elastoplastic creep model. The calculated strains according to the obtained parameters are in good accordance with experiment results, which can be effectively validated that the improved creep model describes the creep characteristics very well. This research is very significant for the study of deformation and stability of surrounding rock mass in deep underground engineering.

For the lime modified expansive soil, two series of tests are carried out. The first ones are water vapor adsorption tests, which are carried out using water vapor adsorption apparatus. The results indicate that with the increasing of lime percentage, the hydroscopicity of expansive soil decreased first then started to increase. And the minimum hydroscopicity is obtained at the point when the content of lime is 6%. In order to study the main reason causing the change of hydroscopicity, the second tests, i.e. the nitrogen adsorption tests, are performed on modified expansive soil. Analysis of the change of porous structure is made. It is found that with the increase of lime percentage, BET microporous specific surface area changes as the fold line; and the minimum value is obtained when the content of lime is 6%. With the increase of lime percentage, accumulating mesoporous volume decreased. With the increase of lime percentage, mesoporous diameter increases first, then decreases, and the maximum value is obtained when the content of lime is 6%. That the hydroscopicity of expansive soils is inversely proportional to the mesoporous diameter is summarized; and the optimal lime percentage is 6%.

According to the character of grouted bolts foundation in jointed rock masses for transmission lines, the model tests are designed based on the similarity theory. The model tests are carried out with focus on the following five factors: elasticity modulus of rock, elasticity modulus of mortar, trace lengths and joint inclination angle in rock mass, embedding depth of the bolt in the rock mass on four different levels of every factor. The experimental results reveal that the elasticity modulus of rock is the most important and sensitive factor influencing the pullout strength of grouted bolt foundation, and the factors next are embedding depth of the bolt in the rock mass, joint inclination angle in rock mass, elasticity modulus of mortar and trace lengths in rock mass. The destruction forms in the model tests are mainly shear failure and split failure. The force transferring mechanism of the bolt is also analyzed. It is concluded that the bond strength between the mortar and the rock, the mortar and the bolt are the main factors controlling test results and affecting the pullout strength of the bolt and failure status. It should be paid more attention to control the quality in the procedure of construction.

Active cooling roadbed employed block-stone layer in permafrost regions is an important engineering measure. Thereinto, it is one of the important contents of frozen-soils engineering applied research to study the convection heat transfer process of block-stone layer, and the actual determination of micro wind speed in block-stone layer resulting from low pressure or temperature difference has always been being one of the problems to be solved. This study has got air flow characteristics in block-stone layer for the first time by high-precision micro wind-speed detector. The results indicate that the convection process is closely related to the temperature change; the rapid cooling is completed in accelerating and stable stage, thereafter slowly cooling process; when the temperature difference in crushed-rock layer is higher than the start temperature difference, the correlation between natural convective velocity and temperature difference is linear. With the cooling process the convective characteristics is different in the same condition of temperature difference. The influence of surface temperature fluctuations on the bottom temperature along the depth direction weakened; and there is a linear correlation between retardancy of temperature and depth. The asymmetry of temperature change is caused by different heat transfer ways and heat transfer efficiencies. At the same time, there is the scope of heat conduction in block-stone layer, in order to ensuring the cooling effect thickness of block-stone layer must be greater than the scope of heat conduction. The experimental research results provide strong support for theoretical analysis and simulation and have a great significance for further understanding of cooling mechanism and improvement.

A 1:5 scaled model test is conducted in order to find a simple and feasible technical means to improve electro-osmotic efficiency in soft clay foundation. The test program involves two different electrode spacings which are 2m×1m and 1m×0.5m. By monitoring drainage, current intensity, voltage, water content and soil pH values, two tests are compared in terms of treatment effect, energy efficiency and electrodes corrosion. The results show that: halving the electrode spacing and keeping the voltage gradient could improve electro-osmotic treatment efficiency effectively including faster drainage rate, more reduction in water content, less energy consumption and less interface resistivity while it would also lead more changes in pH values of soil and more anodes corrosion; moreover, in the condition of halved spacing, smaller voltage drop took place at the soil-electrode interface; but the proportion of the voltage drop to supply voltage is higher.

In light of the problem that the rudaceous coal-bearing soil which ranges along with Guangzhou-Wuzhou Expressway can’t content for soil subgrade filler claim, two improvement schemes of the soil by respectively mixing quicklime and cement are put forward and improvement tests are accordingly taken for contrasting study according to the results of test in laboratory. The result indicated that the effect by mixing cement is evidently better than that by mixing quicklime of the rudaceous coal-bearing including consolidation properties, California bearing ratio (CBR), water stability, etc.; its shear strength properties, unconfined compressive strength and resistance deformability are all evidently increased by mixing cement; in factors effecting strength of the rudaceous coal-bearing soil, cement mixing ratio is the most salient, followed by age and water content, and a fitted formula which is affected by multi-factors is put forward according to the test result of unconfined compressive strength. The rudaceous coal-soil which is mixed with about 3 percent cement is entirely able to meet the claim for subgrade filler, which offers useful reference to roadbed treatment in coal-bearing soil region.

The failure process of tunnel constructed in homogeneous rock is studied by using model test and discrete element method (DEM); and the laws of the stress of surrounding rock and their displacement during the failure process are analyzed. The results of model test match the results of numerical simulation perfectly. From the results some conclusions can be drawn. The surrounding rock of the arch crown deforms obviously after excavation. The failure of the surrounding rock initiates from the arch crown and develops upward progressively until a stable caving arch is formed. The radial stress and tangential stress in a given range of the surrounding rock decrease after excavation. The range in which tangential stress of the surrounding rock decreases expands outward during the tunnel failure process. The vertical stress of the arch bottom increases after collapse. The displacement increment from excavation to collapse is the biggest and that during the collapse is the smallest.

Based on experimental research on long-term performances of geosynthetics-reinforced earth retailing walls, comprehensive study of strains of geogrid, lateral deformation of panel and earth pressure at the stages of filling, loading and continuing measurement is carried out, and scale effect of model chamber on geogrid’s strain is analyzed. The tests results show that the strains of geogrid increase with time and then have a tendency to become stable after enduring 6 months. As distance from panel increases, the strains of geogrid increase firstly following by decrease slowly; and then geogrid’s strains in cohesive soil develop and increase along the length of geogrid with time changing. Besides, due to the effect of chamber-soil friction and small gap between loading plate and chamber, the measured values of strain for middle rib are larger generally than strain values for lateral rib. Finally, horizontal earth pressure behind retaining wall is twice as large as earth pressure calculated by method of Rankine; and the maximum value of vertical earth pressure only account for 1/3 of the summarization value considered load and dead-weight of soil, which showed the geogrid used for reinforcement will significantly induce the redistribution of stresses in backfill.

The triaxial compression creep test on concrete specimens of tunnel lining structure is performed by the MTS-815 hydraulic servo system. The stage of accelerating creep appears when the ratio of compression creep strength to compressive strength is about 0.85. A combined viscoelastic plasticity model is developed after the characteristics of concrete creep deformation were analyzed, which can be used to describe the creep behavior and whole creep process of the tunnel lining structure. The combined model is characterized by a visco-elastoplastic deviatoric behavior and an elastoplastic volumetric behavior. The viscoelastic constitutive law corresponds to the Burger model, and the plastic constitutive law corresponds to a Mohr-Coulomb model. The characteristics of viscoelastic plasticity can be described by the combined model. The problem which the typical Burgers model only can be used to describe viscoelasticity creep is resolved perfectly. In addition, the methods to solve triaxial compression creep model parameters are given and the parameters of tunnel lining structure are determined.

The soil around the pile is regarded as saturated porous medium; and the horizontal dynamic impedance of saturated soil is obtained by using continuum mechanics; and the pile-saturated soil-pile interaction is modeled with dynamic Winkler spring and wave-interference by regarding the superstructure as a beam model; through the force equilibrium at the pile cap level, the pile groups and the structure were coupled; and dynamic interaction of saturated soil-pile-structure under harmonic SH wave is also investigated. The simplest case of a 2×2 pile group is formulated; and the effect of the structural parameters on the dynamic behaviors is analyzed; especially studied the seismic properties of the saturated soil-pile structure system. Numerical calculations show that the pile spacing, length-diameter ratio, ratio of modulus between pile and soil had great effect on the dynamic behaviors of coupled system of saturated soil-pile. The influence of pile spacing on earthquake amplification coefficient is concerned with external excitation frequency; the aseismic effect is better when the ratio of modulus between pile and soil is smaller and the damping of structure and pile is larger; and the greater the length-diameter ratio, the greater the structural deformation produced by seismic action.

Because of existence worldwide of interlayer staggered zone, there have been many engineering rock mass stability problems and geological disasters occurred over the world. In this paper, some cases of distortion failure or destruction of engineering rock mass structure induced by interlayer staggered zone are summarized. Rock mass structure controlling theory is adopted to categorize the failure modes of rock mass with staggered zone into tension crack failure, block-fall and interfacial shear slip failure. After that, it is shown that shear strength model is the main point when modeling for the mechanical behavior of interlayer staggered zone. A deep review of the existing test methods and the modeling theory for its shear behavior shows that the elasto-viscoplastic model is the fundamental constitutive model for describing the mechanical behavior of interlayer staggered zone.

Soil nailing wall is a more economical way of retaining and protection and has been widely used in engineering practice; but the design theory is still relatively deficient. This paper analyzes the main problems in the design of soil nailing retaining wall, namely, soil nail forces and displacement calculation; based on practical engineering experiments and considering the affect of the construction process with the incremental method, supposing that the active earth pressure acted on the retaining wall are equal to the sum of the soil nail forces, a simple method for calculating soil nail forces is proposed. Basing on the elastic theory, the formula for calculating the displacement of the soil nailing wall under excavation is put forward. The rationality of the method and formula is checked by engineering case studies.

In the beginning of 2008, South China encountered the rare extreme snow climate; and in this period many geological hazards happened. In order to study the correlation between the extreme snow climate and geological hazard, this paper explored the evolution characteristics of homogeneous loose debris slope in the condition of extreme snow hazard. The landslide triggered by snowmelt is compared with rainfall to obtain the similarities and differences and then the general law of snowmelt triggering landslide is analyzed through a specific example. Two physical quantities named degree-day factor and snow rainfall equivalent are used to establish the simplified effective snowmelt model based on the characteristics of extreme snow climate in Southern China. The evolution characteristics of the debris slope are simulated from seepage, stability and deformation under the different snow rainfall equivalent proposed above. The results show that a certain snowmelt rate and quantity have no influence on the seepage of the slope. With the snowmelt continuing, the metric suction in the unsaturated zone decreases gradually which causes the whole stability decreasing. The horizontal and vertical displacements of the slope surface increase gradually and the deformation of the front is larger than the back in the horizontal direction as well as the down dislocation in the back and protrusion in the front. After a time of snowmelt the slope fails and slides.

The three dimensional effect in low strain integrity testing of cast-in-place large-diameter pipe pile(PCC pile) is mainly presented by the difference of velocity responses at the points with different radius angles between exciting and receiving points. Based on the analytical formula founded by the authors, the dynamic response characteristics of every vibration modes are studied. The variations of total velocity responses on pile top along circumferential and radial directions are analyzed. The research indicates that: the total velocity response at 90º point mainly derives from the axisymmetric mode. The reflected wave crests at 45º, 135º and 180º point are close to the axisymmetric mode; but the incident wave crests have a greater difference. The total responses on pile top mainly derive from the first order vibration mode along radial direction, while the other modes along radial direction have little contribution. The velocity responses are the superposition of low order modes; but the circumferential modes higher than tenth order have little contribution. The incident wave crests derive from many modes, while the reflected wave crests mainly derive from the axisymmetric mode. The high-frequency interference waves mainly derive from the first order non-axisymmetric mode, the peak value and phase of which are different from various points. The high-frequency interference at 90º point is the least; and the phase from 0º to 90º point is contrary to that from 90º to 180º point. The velocity responses have little difference among different points along radial direction but have great difference along circumferential direction.

The stability discriminant of slopes and the position determination of potential sliding surfaces are always the research hot spots in geotechnical engineering. According to the upper bound method, and assuming the slope material obeys Mohr-Coulomn criterion, the internal and external works are calculated respectively by dividing slope body into blocks with regarding the potential sliding surface as a group of several sectional planes. Then the multivariate function to calculation critical height of slope is established and optimized by mathematical method, which is determined to the critical height and sliding surface shape. Through the example analysis and comparison with the previous research, the results show that the method in this paper is reasonable and effective, and can be used to determine the potential sliding surface and discriminate slope stability.

The basal stability of soil nailed excavation systems has been investigated using Terzaghi’s ultimate bearing capacity theory; and a simplified procedure is proposed for estimating basal heave stability. It is assumed that the basal heave failure is caused by bearing capacity failure; but only part of the failure surface presented in Terzaghi’s bearing capacity is mobilized to evoke basal heave failure. The important factors affecting basal stability have been identified; and correlations between them and factor of safety are discussed from the results of parametric studies. Comparison with other methods has shown that the method proposed is more reliable to estimate basal heave stability.

This paper aims to propose a method for probabilistic analysis of load-displacement hyperbolic curves of single pile using Copula. Firstly, the uncertainties of probabilistic load-displacement hyperbolic model are transformed to the uncertainties of two curve-fitting parameters. Then the Copula theory is adopted to model the joint probability distribution function of the two curve-fitting parameters. The direct integration method is used to carry out the reliability computations for serviceability limit state of single pile. Finally, an example of load-test dataset for Augered Cast-in-Place (ACIP) pile is illustrated to demonstrate the validity and capability of the proposed method. The results indicate that the proposed Copula-based method can provide a more general and flexible way of modeling and simulating the bivariate probability distribution of the two curve-fitting parameters in isolation from their marginal probability distributions, which facilitates the generation of normalized load-displacement hyperbolic curves and thus results in more accurate reliability results for serviceability limit state of single pile. A strong negative correlation between the two parameters underlying the normalized load-displacement hyperbolic model can be observed. Neglecting such negative correlation will overestimate the probability of failure. In addition, the Gaussian Copula, often adopted out of expedience without proper validation, is not the best Copula for modeling the dependence structure underlying the two curve-fitting parameters. If the Gaussian Copula is adopted, the probability of failure of ACIP pile will be underestimated significantly.

The existing studies of ultimate pullout capacity of plate anchor which employed linear and nonlinear Mohr-Coulomb failure criterion have been achieved in sand. However, the Mohr-Coulomb failure criterion is unsuitable for calculating the ultimate pullout capacity of plate anchor in rock foundation. Based on the Hoek-Brown failure criterion, a curved failure mechanism is constructed and the expression of pullout capacity of plate anchor is derived in the framework of upper bound theorem of limit analysis. The upper bound solutions of ultimate pullout capacity and failure surface for plate anchor are obtained with the help of variational calculation. To evaluate the validity of this method, the ultimate pullout capacity of curved failure mechanism is calculated by using equivalent parameters of Mohr-Coulomb criterion when parameter B=1. The solutions are compared with the results presented in the existing literatures. The good agreement between the results of curved failure mechanism and the solutions computed by linear multiple blocks failure mechanism shows the upper bound solutions calculated by curved failure mechanism is effective. According to the results of parameters analysis, both the ultimate pullout capacity and failure surface decrease with the increase of rock parameter B when other parameters are fixed.

Explosion effect on a structure can be considered as a corporate results from the force shock and the heat shock. Based on the Biot’s theory of thermodynamics, the dynamic response of tunnel lining embedded in saturated thermoelastic medium, due to an internal excitation (heat/mechanical shock), is investigated theoretically by using the presented coupling the thermo-hydro-elastodynamic model. Using Flügge theory of thin shell and the general solution of thermo-hydro-elastodynamic model, the motion equation of tunnel lining is obtained; and the solutions of stress, displacement and pore pressure are derived by using the consistency condition at the interface of lining and soil in the domain of Laplace transformation. Numerical results are obtained using inverse Laplace transform and are used to analyze the effect of the internal excitation on stress, displacement and pore pressure for different rigidities and thicknesses. The results show that, under shock of blast, the lining acted as a good barrier for the surrounding soil, and bigger rigidity have better protection effect; the stresses on the interface between lining and soil decreased with lining rigidity increase, and bigger rigidity lead to faster stresses decrease; the rigidity and thickness of lining influence the displacement response evidently, but influence temperature response lightly.

For the sutra cases of Shijing Mountain based on the investigation in site, the lithology, terrain and the geologic structure are analyzed. The potential slopes with rockfall are confirmed initially. Four typical profiles are selected on the slopes and surveyed by BJSD-2B laser profiler. With the results of the numerical simulation, the locations A~F on the profiles are confirmed as the potential rockfall sources with the bigger possibility to fall. The volumes of blocks situated at A~F are gained in site. On the basis above, the kinetic characters(run-out distances, bounce height, kinetic energy and the translational velocities )of the six blocks are predicted and analyzed by the help of the rocfall software. The simulation results show that the rockfall of the block D threatened the life safety of the tourists less than other blocks, but it could destroy the sutra cave No.7 greatly. Oppositely, the rockfall of the other five blocks would threaten the life safety of the tourists greatly, but they wouldn’t destroy the relevant caves. The suggestion that the key locations and blocks should be reinforced as quickly as possible is given.

Considering the multiple failure modes of anchor system such as pulling out of grouting mass from surrounding rock, the tensile failure of the pull rod and the shearing failure of interface between pull rod and grouting mass, etc., applying systematical reliability principle and limit equilibrium method, a direct solution of failure probability of anchor system of rock slope with double slide blocks is proposed based on Monte Carlo sample principle. Meanwhile, taking the time variation of shear strength indices of soft sliding surface c and ? and the erosion of anchor bars into account, a time-variation reliability model of anchor system of rock slope with double slide blocks in consideration of multiple failure modes of anchor is established. The calculation results of examples show that the time-variations of shear strength indices of soft sliding surface c, ? and the resistance between grouting body and surrounding rock have great impact on failure probability of anchor system, while erosion of anchor bars has less impact on it.

Prediction of ground deformation caused by pipe jacking is very important for safe construction. During the construction of pipe jacking, the ground deformation will be influenced by: ①the attitude of pipe jacking machine and the supporting pressure on excavation face; ②the jacking process and the course of changing pipes; ③the grouting process, etc.. To analyze the ground movement during the construction of pipe jacking, a numerical approach is proposed. The Mindlin solution to vertical and horizontal force is used to calculate the ground deformation induced by the face support pressure and the lateral friction resistance, respectively. The Sagaseta method is adopted to consider the ground loss induced by over excavation. A complex variable method proposed by Verruijt is used to simulate the ground movement caused by grouting. Then a 3-D analytical solutions for ground movements induced by pipe jacking is obtained and applied to a case study. The results indicate that considering the grouting pressure can simulate the ground deformation induced by pipe jacking reasonably.

Considering the damage characteristic of the brittle rock, the brittle damage model is introduced; and the expression of damage variable under 3-dimension isotropic damage condition is derived. The analytical solutions of elastoplastic stress and plastic radius of the circular hydraulic tunnel are deduced, considering the seepage influence, damage influence and principal stress corresponding relations under different engineering conditions. The research about brittle characteristic shows that the plastic radius of surrounding rock will increase after considering brittle damage. The more distinct the brittle characteristic is, the more larger plastic radius will be. Whether the first principal stress of rock nearby the tunnel is radial stress or tangential stress, the consideration of seepage will expand the plastic zone of the surrounding rock. In the process of internal water pressure increase from zero, the plastic radius will gradually decrease accompany with the increase of internal water pressure, until the plastic zone disappears in the end; however, when the internal water pressure exceeds the critical pressure; the plastic zone appears again; the plastic zone will gradually increase.

Due to nonlinearity of ground material’s behavior, the error will be induced using traditional superposition method to calculate bearing capacity. The main cause of superposition error is that the bearing capacity factors based on failure mechanism of weightless soil is not applicable for weight soil. Based on slip line method, variation trends of error induced by superposition method is analyzed; and bearing capacity factors which could reduce the error effectively or eliminate the error completely are calculated. The investigation results show that, bearing capacity calculated by traditional superposition method is conservative; the error increases with increasing of internal friction angle and decreases with increasing of cohesion. Error induced by superposition method will be reduced effectively or eliminated completely if bearing capacity factors considering soil weight.

As a technique for protecting adjacent buildings, partition wall has got an application to deep pit engineering. However the design of its parameters still lacks theoretical guidance. Based on a large-scale deep pit case, a 3-D finite element model is presented using the software ABAQUS. The behavior of soil is assumed to conform to modified Cam-clay model; and the actual construction sequence is simulated. Parameters relevant to the partition wall are analyzed in detail. The calculated and measured settlements of the adjacent building demonstrate the rationality of the optimization analysis of these parameters. It is shown that the ground surface displacements, the transverse angular distortions of the building and the maximum horizontal displacements of the retaining wall decrease with the increase of the partition wall depth. Nevertheless, the rate of the decrease of transverse angular distortions tends to be small if the partition wall depth is greater than a certain value. Thus there exists a reasonable depth for the partition wall. As the partition wall moves toward the adjacent building, the transverse and lengthways ground settlements, the differential settlements as well as the transverse angular distortions of the building decrease. Theoretically, the closer the partition wall is to the adjacent building, the better its effects are. The installation plane range of the partition wall also has significant influence on the adjacent buildings. Usually the range of the protected building can be the appropriate range of the partition wall. It is also shown that the stiffness of the partition wall does not affect the surface settlements and the angular distortions of adjacent building significantly, a medium-stiffness partition wall is sufficient to meet the engineering demands.

In order to ensure the construction safety of karst tunnel, the unfavorable geological body can be accurately predicted using integrated prediction method so as to significantly reduce the potential for disasters. Setting the Zijingshan Tunnel of the Da-Guang’nan Expressway as the project background, firstly, the developing characteristics of the karst of Zijingshan Tunnel is studied through geological analysis. And then, long-distance detection of unfavorable geology is done with the TSP (tunnel seismic prediction) method; so an accurate prediction of the ZK221+960 underground river’s location is obtained; and a short detection of the underground river's direction is done using geological penetrating radar. The detection result is consistent with actual disclosure direction. For the actual situation of Zijingshan Underground River, drainage tunnel program is eventually made after many analysis of various treatment program. Under the premise of ensuring the safety and duration of tunnel construction, the hole drainage program is optimized. The optimized processing program achieves good results in the Zijingshan Underground River’s governance, which has some reference significance for the similar engineering.

To provide the theorical basis for electromagnetic emission (EME) forecasting dynamic disasters of the coal or rock mass, the relations between EME energy and loading mechanical energy are analyzed preliminarily by the theory of electromagnetic wave and damage mechanics. The coupling relations between EME and energy accumulation during the deformation and failure of coal under uniaxial compression are investigated by laboratory tests and the energy theory. The results show that: the relations between loading mechanical energy and EME energy during the deformation and failure of the coal sample is positive linearity, and the ratio is 0.1074 times with elastic modulus of coal samples; the EME energy is not only proportional to peak elastic strain energy of coal, but also proportional to the impact energy index of coal. The EME signal reflected the difficulty level of the coal dynamic damage, Therefore, the EME can forecast dynamic disasters of coal or rock mass.

Based on the tested ground temperature, salt dilating and frost-heave deformation etc. in the subgrade filled by the sulphate saline soil in site in corps reclamation area of Xinjiang Autonomous Region, using the multiple linear regression theory in SPSS software, the paper discusses the rule of the salt dilating and frost-heave deformation, the factors affect the salt dilating and frost-heave deformation, and the calculation method of the salt dilating and frost-heave deformation occur in the sulphate saline soil subgrade. The main conclusions are as follows. ①In one freezing and melting cycle, the tested ground temperature—salt dilating deformation curve can be separated into three phases, and the property of each phase is different. ②When the ground temperate is going up, the factors affecting the salt dilating and frost-heave deformation and the calculation method of the deformation are different from it when the ground temperate is lowered down. ③In one freezing and melting cycle, the maximum salt dilating deformation in the subgrade occurred when the ground temperature rises; thus the formula based on the data tested during the temperature rises is recommended to calculate the maximum salt dilating deformation.

Because of the low resolution of CT images, it is difficult to quantize the loess microstructure accurately. Therefore, in this paper, SEM images are associated with triaxial CT images for the investigating of the microstructure of the collapsible loess. Firstly, the reasonable training samples are extracted from a large number of SEM images with different magnifications. The supervised classification of the CT images is carried out based on these training samples. For a comparison, supervised classification under gray classification of CT images is also made. The results show that: supervised classification of CT images based on SEM images exhibits better performance on the quantization monitoring of the change rule for loess microstructure. It is shown the high quality for monitoring larger pores, aggregates, clay aggregate particles and closed mineral substance of the loess sample during the triaxial shear test. It is believed that this supervised classification based on SEM images could provide a new sight for researching loess microstructure.

Composite linings are often adopted in highway tunnels. The new Austrian tunneling method is used in tunnel’s designing and construction. For weak wall rock section, steel arch(or grid arch) + grouting + shortcrete network of united initial support form are adopted. The key to achieve dynamic design is to find a good method to evaluate initial support safety evaluation about shape steel reinforced shotcrete. This paper gives a systemic study on initial support safety evaluation and established a numerical calculation method. Firstly, according to the monitoring data, this paper analyzes the final settlement value and horizontal constringency of wall rock and makes it as a basis for parameter inversion. Secondly, adopting the mechanical parameters from inversion, the initial support internal force about shape steel reinforced shotcrete is calculated by stratum-structure numerical method. Finally, initial support safety is calculated and evaluated by safety factor calculation method of shape steel reinforced shotcrete. From the study of this paper, we find that the initial support of shape steel reinforced shotcrete can exert the strong support effect of shape steel. And that, when the shotcrete is upgraded, concrete plays a major supporting role. The space between shape steels is not prominence in increasing the initial support safety factor.

On the foundation of non-seismic active earth pressure theory, the seismic active earth pressure formulas, in the conditions of filling surface incline, wall back incline, filling soil having the characteristic of c and ?, having the c and ? action between the wall back and filling soil, the filling soil having horizontal and vertical acceleration, were obtained with differential layer method, these formulas included their distribution and total force. At the present time, the every active earth pressure formula in earthquake and non-earthquake condition is the special example of this paper’s construction. The total earth pressure calculated with differential layer method is consistent with other method such as Coulomb’s earth wedge equilibrium method; but its acting position is raised up; therefore, the design of retaining wall oughted to pay enough attention to this aspect.

The authors think that there exist two kinds of size effects for the stability of surrounding rock mass of an underground opening; namely they are the constitutive size effect and the structural size effect. As a tentative inquiry, the authors try to investigate the actions of the two kinds of size effects in the finite element analysis (FEA) for an underground opening model. So two cases are assumed: the first one, the mechanical properties at any point in surrounding rock mass are the negative power functions of distance between the point and the opening center; and the opening span changes from small to large, that is, the constitutive size effect and the structural size effects are considered simultaneously; the second one, the mechanical properties of surrounding rock mass are constants, but the opening span changes from small to large; that is, only the structural size effect is considered. The elastoplastic calculations under a plane-strain condition are carried out for the two cases; and the stresses, displacements and plastic zones in the surrounding rock mass are analyzed. The results show that: the displacements and plastic zones in the the surrounding rock mass are smaller; and the stresses are comparatively large, considering the constitutive size effect of the mechanical properties; but the situation of without considering the constitutive size effect is the other way round.

The Niucheding tunnel in Guangzhou-Wuzhou Expressway is taken as the engineering background, the analytical and numerical calculation method for the second lining time of tunnel under the conditions of rock mass rheological characteristic are analyzed. The disciplines of the arch settlement obtained from the proposed method are compared with that from the monitoring data; then the numerical calculation model is founded to analyze the influence of stress releasing coefficient on the stability of surrounding rock mass of tunnel. The analysis results show that, the disciplines of the arch settlement obtained from the proposed method are the same with that from the monitoring data, which validates the correctness of the proposed method. The stress of surrounding rock mass is released gradually with the tunnel face moves on; and the releasing rate is slow in the III degree of rock mass, which shows obvious spatial effect. According to the numerical calculation results, if the stress releasing coefficient is larger or equal to 60%, the whole area of the disturbed zone is larger than the area the bolt can reinforce. Finally, the appropriate second lining time is obtained so as to give some guidance for the real practice.

It is difficult to simulate the whole process of shield tunneling and divide the overlapped shield tunnel. Combined with overlapped tunnel in Shenzhen Metro Hongling Zhonglu-Laojie-Shaibu interval to establish a more comprehensively 3D numerical simulation method, reflecting the characteristic of the whole tunneling process and divide shield tunnel from its horizontal direction on the basis of the index of Mohr-coulomb, longitudinal partition according to displacement rate of change. The results show that: the horizontal adjacent partition is : no-effect area when F≤F? , weak-effect area when F?＜F＜0, intensive-effect area when F≥0; longitudinal partition is：Less than 0.3 mm/d for the non-effect zone，0.3-1.5 mm/d for the low impact zone, greater than 1.5 mm/d for the high-impact zone. The results applied to the Shenzhen Metro Hongling Zhonglu-Laojie-Shaibu intervals overlap tunnel, and achieved good results, and provide theoretical supports for the numerical simulation of the whole tunneling process and the design and construction of shield overlapped tunnel, and it also provides a reference for similar projects.

Based on the constitutive equations given by Yang and Ahmed et al.[1-5] to predict the liquefied large deformation of sand, the hardening rule of model and the calculation method of elastoplastic modulus are modified. Then, it is extended to 3-D equations and is implemented by using Fortran subroutine program and is implanted into ABAQUS finite-element software. By using ABAQUS software with the advanced model, how the main model parameters to affect the dynamic pore pressures arising and the stress-strain curves of sand, are analyzed in detail, respectively. At the same time, the sensitivities of model parameters are also analyzed. It is proved that the subroutine program can work perfectly to model the liquefied large deformation of sand; and the combination of different model parameters’ values can predict perfectly the different deformation stage of sand. These conclusions verified and given here can be used expediently to fixed the model parameters by using dynamic triaxial tests. Accordingly, a new effectively numerical method used to calculate the large liquefaction is advanced by using the advanced finite-element ABAQUS software with special constitutive model.

In order to study coupled groundwater flow and heat transfer in the near-field of high-level nuclear waste repositories, a meter-scale physical model with regular fractures has been constructed, using rock blocks taken from the preliminary potential Beishan site in China’s Gansu province, to conduct experiments of fracture water flow and heat transfer. As a preliminary theoretical study of the physical modeling, this paper presents a numerical modeling analysis of the fractured rock physical model using an equivalent continuum model. Under the prescribed conditions, the calculations reveal: heat conduction and fracture water flow turn quickly from uncorrelated in the early phase of heating to fully coupled at later times; while the standstill fracture water essentially behaves through heat storage and conduction, the fracture water flow can induce heat convection between water and the rock matrix as well as heat advection; if the fracture water flux is kept constant, then variation of the fracture aperture would exert little influence on the system; if the fracture water velocity is kept constant, then the effect of the variation of the fracture aperture would be significant; higher heat intensity would cause more convection heat transfer from fracture water flow and larger water pressure increase, and the latter effect would be much more significant when the temperature exceeds 100℃; the model system approaches to steady state after about 7 days of heating , with the amount of heat flushed out due to fracture flow equaling the effective heat input.

In analyzing the seepage processes in unsaturated porous media, numerical instability may occur due to the difference between the adopted physical and numerical models under nearly saturated conditions. To resolve the issue and improve the numerical results, three methods are proposed and then implemented into the finite element code U-DYSAC2. Numerical simulations are compared with the experimental data available in the literature, showing that the problems related to numerical stability, convergence and accuracy can be induced by the high nonlinearity of soil water characteristics and hydraulic conductivity functions under nearly saturated conditions; and significant numerical discrepancies can appear in predicting the water content and matric suction of the porous media. Among the three proposed methods, modified Van Genuchten model (MVGM) method yields the best results in calculating the moisture content; while the Line method achieves best prediction for matric suction. Consequently, in analyzing the seepage processes in unsaturated porous media, it is crucial to introduce a proper analysis procedure such that reliable numerical results can be achieved

Accompanying the development of ocean engineering, research on the bearing capacity of pile foundation under the action of wave cyclic loading has been given special attention for the design of offshore structures. Based on the concept of cyclic-softening effect, cyclic strength of the soft soil and Mohr-Coulomb yield criterion are combined together to numerically evaluate cyclic bearing capacity of inclined uplift pile under wave cyclic loading. A perfect elastoplastic finite element method for numerical assessment of bearing capacity of inclined uplift pile is developed. Dynamic ultimate bearing capacity of inclined uplift pile under cyclic loading is studied. The influences of load cycle times, static stress ratio, pile length, pile diameter and pile modulus on the cyclic ultimate bearing capacity of inclined uplift pile are analyzed. It can be concluded that the bearing capacity of the foundation under cyclic loading is less than that of under static load. The effects of wave cyclic loading lead to nonuniform distribution of cyclic strength of soil beside the inclined uplift pile and reduce the cyclic bearing capacity of pile foundation. The results also show that the ultimate bearing capacities of inclined uplift pile improve gradually with the increase of the load inclination, pile length, pile diameter, pile modulus, soil cohesion and soil modulus.

An anisotropic bounding surface model for cohesive soils is developed based on Wheeler’s rotational hardening law (S-CLAY1 Model)，which is integrated in return mapping algorithm. The UMAT subroutine of the developed model is programmed in ABAQUS software. With the developed model, the numerical simulation of undrained triaxial compression test for normal consolidated kaolin clay (OCR =1) is implemented, and the numerical results of the developed model compared with the Modified Cam-clay model in ABAQUS. The results show that the developed anisotropic bounding surface model could efficiently depict the stress-strain behaviors and change of excess pore pressure of normally consolidated clay. Meanwhile the developed model could reproduce anisotropic behaviors of soft clay under loading conditions. Then the similar numerical simulation for over consolidated kaolin clay (OCR =4) is implemented. The results show that the developed model could also efficiently depict nonlinear stress-strain behaviors of over consolidated clay under small deformation conditions.

In the implementation of strength reduction finite element method (SRFEM), the input file will be modified by hand based on various reduction factors to do trial calculation, the process is trivial. In order to simplify the computational process and improve efficiency, on the basis of SRFEM, combined with field variable provided by ABAQUS, a new technique for strength reduction calculation is presented-field variable based strength reduction finite element method. Cohesions and internal friction angles of rock and soil mass are assigned as functions of field variable in ABAQUS; field variable is assigned as a function of step time t, the cohesions and internal friction angles can be reduced with the increase of t, the safety factor of slope can be obtained in a single simulation process, which avoid modifying of input file and without repeating the trial calculation. Combined with some literatures, the validation of this technique for slope stability analysis has been done; the computational efficiency is compared with bisection method in the determination of safety factor. The results show that: the safety factor obtained by this method is effective and reliable; this technique is feasible for slope stability analysis; this technique simplifies the process of calculation; the efficiency of calculation is improved.