Based on the unified solution of shear strength in terms of two stress state variables for unsaturated soils, the unified solution of Coulomb’s active earth pressure for unsaturated soils is proposed which considers the effect of the intermediate principal stress rationally. Moreover, the comparability of this unified solution is analyzed. Influential characteristic of each factor is discussed by comparing with the unified solution of Rankine active earth pressure. The result shows that: the effects of intermediate principal stress and matric suction are significant; the Coulomb’s active earth pressure decreases until zero with the increase of the unified strength theory parameter and matric suction, which indicates that the consideration of the intermediate principal stress effect can make materials give full play to their self-bearing capacities and strength potential, bringing obvious economic benefits to the project; the Coulomb’s active earth pressure increases with the increase of the grading angle of retaining wall and slope angle of backfill, but decreases with the increase of the effective internal friction angle and matric suction angle; meanwhile, the influence of external friction angle is less significant. The proposed solution has very broad applications to the determination of soil pressure. The unified solution of Rankine active earth pressure is its a special case. It provides references for design of slope and foundation pit.

Tensile strength is an important mechanical index of clay. A set of tensile strength apparatus is developed to test the tensile strength of clay and the specimen-preparing methods on tensile strength are studied. Tensile strength tests for two types of swell-shrink soils are respectively designed to discuss the influences of water content, dry density and drying-wetting cycles. Analysis of the data shows that there is an exponential relationship between tensile strength and water content; and the tensile strength reaches the peak value at optimum water content, saturation degree of 66%. And the change rule of tensile strength with water content is impacted by the forms of water in soils. It is also found that tensile strength linearly increases with dry density and the increase rate is significant. As experienced three dry-wetting cycles, the tensile strength sharply decreases to 20% of the initial value and its impact is especially obvious during the first and second cycles. Through theoretic and contrast analysis, it is found that the decrease of tensile strength of swell-shrink soils with drying-wetting cycles is due to the deterioration of microstructures.

Based on virtual soil-pile model, the influence analysis of pile tip soil properties on single pile settlement is conducted in terms of layered soils. Firstly, considering stress dispersion in soil medium, the soil between pile tip and bedrock surface is assumed to be virtual soil-pile. Furthermore, the virtual soil-pile with variable cross-section is divided into finite segments. Then, the surrounding soil layers of pile is modeled as ideal elastoplastic transfer model; and regarding the developmental plastic depth as an independent variable, a recursive method for calculating axial force and side friction of pile in layered soil is derived by load transfer method. On this basis, parameters selection of load transfer model is studied; meanwhile, the factors of virtual soil-pile length and the feasibility of acquiring stress dispersion angle through inversion analysis are analyzed. Finally, by means of the presented method, the difference of load-displacement curves has been examined when pile bottom sediment and soft substratum exist. The results show that: the calculated load-displacement curves of pile tip and pile top agree well with the measured curves considering stress dispersion of pile tip soil; the presented method can take the influences of pile bottom sediment and soft substratum on load-displacement curves into consideration.

Dredger fill changes from fluid to soil and its void reduces markedly during the consolidation of reclaimed ground under vacuum loading. Because both permeability and compressibility are functions of void ratio, the consolidation coefficient of dredger fill changes greatly. The prediction by current design method is much different from engineering practice because the consolidation coefficient in this method is taken as a constant. In order to account for the effect of varying void on consolidation, the relation between consolidation coefficient and effective stress is derived; and a consolidation model for dredger fill is built by applying this relation to Barron’s consolidation theory. The model is proved by comparing the result of this model with that of model test. The analysis result shows that the consolidation coefficient of dredger fill decreases with the void during the consolidation. Therefore, the consolidation of reclaimed ground is rapid in the initial stage and slow later. The quick decrease of the permeability of soil near the drainage board area will hinder the consolidation of soils in the farther area. Therefore, compared with the result of current design model, the consolidation of reclaimed ground is slower in the later stage and the whole period of consolidation is longer.

Sediments and fluids characteristics are the two important factors affected the formation and accumulation of methane hydrate in deposits. In order to study methane hydrate occurrence in sediments, the influences of unsaturated fluids on hydrate-bearing sedimentary strata under high-pressure seepage environment must be discussed. The difference of fugacity has been adopted to be the driving force in this reaction. The hydrate intrinsic kinetic constant has been written in an Arrhenius-type equation. A two-dimensional (horizontal, vertical) model coupling the aspects of unsaturated flow fluids-sedimentary characteristics-hydrate formation kinetics has been established. The evolution of porous hydraulic parameters such as water content, to-saturation coefficient, pore water pressure and hydrate saturation have been discussed. Under the conditions setting in the simulation, the results show that the pore water pressure becomes larger with time during the reaction proceeding. Under the same conditions, the increase of strata temperature and hydrate saturation will increase the pore water pressure. The hydrate saturation causes less influence; while the sediments matric suction, to-saturation coefficient and hydrate intrinsic kinetic constant will vary inversely with the pore water pressure; and the to-saturation coefficient has more strong effects.

In order to reveal the deformation and failure characteristics of salt rock with tilted unsalted interlayer, uniaxial and triaxial compression experiments are carried out. With failure mechanism analysis of stratified rock mass, the variation and impact factors of bonding stresses are investigated; and the failure characteristics and mechanisms of tilted layered salt rock are revealed. The research indicates that tilted interlayer obviously affects the deformation and failure characteristics of rock samples. The failure patterns are related to the position of interlayer. Cracks initiate near interface firstly and then expand to interlayer and salt rock. Trans-granular fracture is observed and its mechanism is revealed. The overall performance of salt rock and characteristics of cracks are influenced by the thickness of interlayer; comparing with thin interlayer salt rock, macro crack tends to appear in the middle thick interlayer more easily. Confining pressure reduces the morphological difference of cracks in different positions of interlayer and the difference of bonding stresses on interface. The analytical result provides an important reference for the study of the cavern construction optimization and stability and tightness research of the salt caverns in the salt rock mineral area with stratum inclination. What’s more, optimizing the geological section to meet the demand of stability and tightness simultaneously is recommended before cavern construction.

Lime soils are widely used in road engineering. With the rebuilding and expanding of road engineering, it may cause many problems, such as how to treat the lime soil and how to reuse it. The compression and strength mechanical properties are discussed by comparing remolded lime soil with lime soil and pure soil. The results show that, compared to lime soil, the compression coefficient (a1-2) of remolded lime soil increases 2-3 times; the unconfined compressive strength decreases 30%-40%; the internal friction angle increases about 1.2 times and the cohesion reduces 40%. Degradation coefficient (DC) is introduced to evaluate the degradation degree of remolded lime soil. The essential degradation cause of the remolded lime soil is analyzed from the standpoints of aggregation, carbonization and hardening. Compared to pure soil, the clay (d <0.002 mm) content of lime soil decreases and coarse grain (d >0.074 mm) content increases. So, the internal friction angle of lime soil is improved. But the cemented structure between lime soil particles is broken by remolding effect. The cohesion of remolded lime soil decreases because of the remolding; and the other mechanical properties of remolded lime soil are affected.

The magnitude and distribution of sidewall friction of open caisson are very important to caisson structure design and the selection of sinking-aid methods. However, the computation of sidewall friction is still quite empirical, which often leads to significant deviations from real situation. In order to further understand and master the distribution rule of the sidewall friction of open caisson during the process of sinking, a new type of micro-friction-sensor is invented to measure the sidewall friction directly; and its accuracy, stability and reliability are verified. Then, it is used in model test along with static strain test system. The result shows that the variation characteristics of sidewall friction can be divided into three phases with the increase of sinking depth. In phase one, the increase of sidewall friction and depth is close to linear relation; in phase two, the sidewall friction keeps increasing with depth, yet the rate of increase drops and the value reaches the peak at a particular depth; in phase three, the sidewall friction drops gradually with the increase of depth. The mechanism of this phenomenon is analyzed with theory of limit analysis. The test result can be treated as a complement of in-situ monitoring, and both provides fundamental data for analytical and numerical computation for sidewall friction of open caisson.

Dilatancy is a special material property of soil; and it is important for sand to describe its dilatancy accurately in the constitutive modeling. Through a large number of triaxial tests, it has been found that dense sand and loose sand show different dilatancies and stress-strain relation characteristics under the same loading condition, which indicate that the dilatancy of sand is not only related with the stress condition , but also with its physical state. The state parameter theory explains the combined effects of stress condition and physical state on the dilatancy of sand successfully. Hollow cylindrical torsional shear apparatus can be used to perform monotonic shear tests in different principal stress directions. Through such experimental tests, the difference in the dilatancy of sand in different principal stress directions is examined even under the same initial stress condition and physical state, which shows that the principal stress direction is also important to the dilatancy of sand. A state parameter including the principal stress direction is proposed by analyzing the effects of principal stress direction on the dilatancy of sand in the tests; and a dilatancy equation considering the principal stress direction is established. Compared with the experimental results, the method is proved to be correct and accurate.

The typical experiment apparatus of rock friction sliding was designed and improved. Under low normal stress, the static friction coefficients for 8 silicate rocks were studied experimentally based on the apparatus. The surface topography of the sliding surface was characterized by the statistic parameters and asperity; and the influence of roughness parameter on static friction coefficient was analyzed. When the sliding surfaces are polished, the static friction coefficients of phyllite, quartzite, lithic sandstone, and quartz sandstone are 0.38-0.47; and the ones of conglomerate, gravelly coarse quartz sandstone, medium grained lithic sandstone, and medium grained quartz sandstone are 0.83-1.07. The two roughness parameters which are the maximum depth of profile valley (Rm) and the maximum height of profile peak (Rp) make the asperity produce different actions and induce to different static friction coefficients. With the increment of the two roughness parameters, the static friction coefficient is increased with exponential law. The amounts of asperity are less on the sliding surfaces of phyllite, quartzite, lithic sandstone, and quartz sandstone. The amounts are more on the ones of conglomerate, gravelly coarse quartz sandstone, medium grained lithic sandstone, and medium grained quartz sandstone. With the increment of asperity amounts, the static friction coefficient is increased during rock frictional sliding.

Based on the theory of comprehensive structure potential, the stress-strain relations of intact loess, remolded loess and saturated loess have been researched by independent developed true triaxial apparatus. It has been reflected that the relationship between stress ratio structure parameter and generalized shear strain along with consolidation pressure and stress path. The mathematical model has been established. Meanwhile, the reliability of true triaxial apparatus and stress ratio structure parameter has been validated. It is shown that the stress-strain relations have distinct difference for different soil states. The stress-strain relations tend to the characteristic of strain softening or hyperbola for the intact loess. The stress-strain relations tend to the characteristic of hyperbola and strain hardening for the remolded or saturated loess. The soil structure reduces gradually with the increasing of shear deformation and water content. The consolidation pressure ? c and intermediate principal stress patameter b are the influential factors for soil structure. The simulation formula can accurately reflect the theory and experimentation. It is convenient for application in projects.

Taking the near field of nuclear waste repositories in fractured rocks as the subject of study, a simplified conceptual model for three-dimensional water flow and heat transfer in single-fracture rock is proposed. The mathematical model, taking into account of distributed heat source and saturated single-fracture rock of infinite extent, is formulated and solved by using a Green function approach, in which a fundamental solution of the governing differential equations after Laplace transform is employed. The fracture surface is discretized by rectangular elements. The singularities in the integral equation are handled through analytical integration in polar coordinates; and a numerical procedure is developed to solve the transient temperature distributions in fracture water and rock matrix. Two numerical examples with special flow field are provided for illustration of the proposed method with comparison of an analytical solution based on 1D rock thermal conduction; and other numerical examples with distributed heat sources are extended for characteristics of flow and heat transfer in single-fracture rock and the sensitivities to flow velocity, rock thermal conductivity and heat source intensity. The calculations show the following observations: Comparing with the direct Gaussian method, the proposed analytical approach to handling the singular integrals is more accurate. The temperature of water in the fracture calculated by using the semi-analytical method is lower in upstream and higher in downstream than the analytical solution, due to the fact that the former method takes into account 3D thermal conduction in the rock matrix, whereas the latter assumes 1D conduction. Without interior heat source, the greater the rock thermal conductivity, the lower the temperature of facture water, due to more heat exchange between fracture water and rock matrix. The larger the fracture water velocity, the more significant of the influence of the inlet fracture water temperature on the temperatures of fracture water and rock matrix. The effects of the distributed heat source on the temperatures of fracture water and rock are more sensitive in downstream as a result of the heat advection of fracture fluid flow.

Analytical formulas of anisotropic equivalent elastic parameters are established by using mechanical method of complex material. An example is made to show the feasibility and efficiency of the analytical formulas. Meanwhile, the coordinate transformation and jointed factor of regular jointed rock mass are considered. The rules of equivalent elastic parameters of regular jointed rock mass are studied. Variation rules of equivalent elastic parameters under different jointed factors are analyzed. Combining with the properties of columnar jointed rock mass in Baihetan dam site, the anisotropic characteristics and values of elastic parameters are analyzed. The results show that the jointed rock mass can degenerate into isotropic medium when the stiffness is equal; and the calculated values by using the method in the paper are well in accord with the test results in the dam site. The cylinder deflection has more effect on the equivalent elastic parameters. The results also show that after deflection, the horizontal elastic modulus and vertical modulus are reduced by 54% and 17% respectively. According to the curves, the higher the angle of cylinder, the greater reduction the elastic modulus would have. This change rule of equivalent elastic parameters revealed obvious anisotropic property of columnar jointed rock mass.

Based on the centrifuge model test, models of subgrade upon slope in permafrost area are completed using the mechanical similarity. Impacts of soil mechanical parameters, subgrade height and ground gradient on the deformation characteristics and stability of subgrade are researched. As well as the deformation failure mechanism and failure modes are researched by comparing rubble stone subgrade and ordinary subgrade. The results show that the soil mechanical parameters, the subgrade height and the ground gradient have obvious effects on the stability of subgrade. The deformation mutation point takes place at the freezing-thawing interface; and the deformation mainly concentrates on the soil layer above the freezing-thawing interface. Under the experimental conditions, the reasonable subgrade height is about 5 m. With the subgrade height of 5 m, the critical value of the ground gradient is about 1:6 and the transverse deformation increases. Both vertical and transverse deformations due to permafrost thawing show a positive nonuniform feature. The essential cause of failure is the deficiency of shear resistance strength of the weak belt. The soil layer above the freezing-thawing interface slips along the freezing-thawing interface. Failure modes of the subgrade are divided into cracking failure in shallow layer, cracking failure in deep layer and failure of integral slippage. Besides, the horizontal displacement and settlement of the rubble stone subgrade are smaller than the ordinary subgrade; and the rubble stone subgrade has better whole stability.

In recent years the frequent geological disasters make deep research on deformation mechanism for rockfill increasingly urgent and important. Rockfill has the characteristics of irregular shape and prominent interlock behavior; and a simple program is used to generate arbitrary shape of particle clusters. Taking tetrahedron as core, arbitrary shape of particle clusters are generated by crystal cell reproduction method. The particle clusters are used as basic unit to generate aggregation of particles so as to simulate the large scale triaxial drained shear test for rockfill and study the mesoscopic mechanism of rockfill deformation by analyzing the change law of mesoscopic parameters in experiment. The analysis shows that the stress-strain curves and volumetric deformations from numerical simulation are in agreement well with the test results; and the arbitrary shaped clusters are much better than circle particles in simulating the rockfill. The breakage rate is the bridge linking the properties of macroscopic parameters and mesoscopic parameters. With the analysis of the change law of mesoparameters among four stages of particle breakage, the mechanism of rockfill deformation is given.

A new approach based on the limit analysis upper bound theorem is proposed to study the seismic stability of earth-rock dams. First, according to the upper bound analysis, a perfectly plastic soil model is assumed with an associated flow rule. Then, the dam slope is divided into horizontal slices with regarding the sliding surface as an arbitrary surface. In order to obtain the maximum anti-seismic capability of dams and the corresponding sliding surface, the multivariate function is established by the energy-work balance equation and optimized by intelligent algorithm. For the study of the influences of shear strength parameters and horizontal slice number on the maximum anti-seismic capability and the sliding surface, the new approach is applied to a typical earth-rock dam with core wall. It is shown by the results that the horizontal slice number has a great influence on the sliding surface but a small influence on the maximum anti-seismic capability. When the horizontal slices achieve a certain number, the maximum anti-seismic capability of dams is ultimately a stable value. Meanwhile, the shear strength parameters of the rockfill materials have a great influence on the maximum anti-seismic capability. Compared with the limit equilibrium method the result illustrates the correctness and feasibility of the proposed method.

The coefficient of earth pressure at rest is a critical parameter in the field of geotechnical engineering, which is the foundation of determining the stress state in horizontal site and the calculation of earth pressure at rest. However, at present the study of the coefficient of earth pressure at rest is not enough. The theoretical formula of coefficient of earth pressure at rest of normally consolidated soils is derived from the analysis of the internal stress state of sand prism with natural accumulation stayed in a critical state. Compared with the formula k0 =0.95-sin? (k0 stands for the coefficient of earth pressure at rest, ? stands for the inner friction angel of soil; the same as below) proposed for the coefficient of earth pressure at rest on sandy soil and with the formula k0= 1-sin? proposed for the coefficient of earth pressure at rest on clay; the results show that the formula for the coefficient of earth pressure at rest matches very well with the regular formula for sandy soil and has a little bias with the formula proposed for clay. The formula and its derivation of the coefficient of earth pressure at rest in this paper are scientific and reasonable to express its origin, which makes up the blank in this study area, provides the theoretical support and plays a certain role in the promotion of the derivation of the coefficient of earth pressure at rest of indoor, field test and empirical formula.

Based on the binary SEM image, a method to express the magnitude of particle congeries area accumulation versus the corresponding area was constructed. Afterward, the definitions of the curvature coefficient and the nonuniformity coefficient of the aforementioned curve were established in order to study the features of the curve for different types of soils. For different area ratios, the proposed method were utilized to discuss the distribution characteristic of particle congeries figured in the SEM photos respectively for Guazhou arenaceous silt and Tianjin clay successfully. The study shows that the area accumulation curves, the curvature coefficients and the nonuniformity coefficients of the silt are significantly different from that of the clay. The curves for the arenaceous silt are smoother than that for the clay and have the trait of the grading curves for the poorly graded soils. Whereas, the curves for the clay exhibit obvious multistep, namely have the feature of grading curves for the soils with certain particle lost. Furthermore, the curvature coefficient and the nonuniformity coefficient of the arenaceous silt are larger than that of the clay. The suggested technique is a new means to evaluate microstructure in soil mechanics and can be employed to quantitatively analyze the difference between SEM images for diverse soils.

High-speed railway track and nearby structures are subjected to low-amplitude and high-cycle loading. Additional settlements of track and structures may be caused by irreversible strain accumulation of soils under the low-amplitude and high-cycle loading. At present, the theories described the deformation characteristics of soils have two kinds, stress-strain hysteretic model based on classical plastic theory, e.g. bounding surface model, and strain accumulation model based on empirical law obtained from the cyclic triaxial tests, e.g. Bochum accumulation model. Based on the existing test studies and classical elastoplastic theory, a strain accumulation model is proposed to predict strain accumulation behavior of soils subjected to low-amplitude and high-cycle loading. The model describes the accumulation law of plastic volume strain via a logarithmic law; and it can measure the strain accumulation. The direction of strain accumulation is determined using the flow rule of the modified Cam-clay model. Finally, with the simulation of test data, it is shown that the proposed model can predict strain accumulation behavior of soils subjected to low-amplitude and high-cycle loading. It has widely application prospect.

From the investigation of seismic hazards of subgrade in Wenchuan earthquake, the anti-sliding sheet pile wall has been proven to be an excellent seismic structure in bedding or accumulation slopes. In order to understand the seismic performance better and optimize the seismic design, large shaking table test is performed. In order to ascertain the seismic response characteristics of anti-sliding sheet pile wall, acceleration histories recorded at Wolong station in Wenchuan earthquake are scaled and used to excite the model. The test results reveal the distribution pattern of earth pressure along the pile, seismic response of displacement and accelerations in slopes under ground shaking. The research demonstrates that the distribution of seismic earth pressure is nonlinear along the pile; and vertical seismic load has amplification effect on horizontal acceleration. Therefore, the seismic earth pressure under bidirectional loads is larger than that under unidirectional horizontal load. However, the difference between them is not obvious in seismic zones of basic intensity VII and VIII. Furthermore, landslide thrust, soil resistance of sliding bed, and displacement of pile are proportional to the input seismic peak acceleration. It means that, with the increase of input peak seismic acceleration, the amplification ratio of acceleration increases and the material shear strength of sliding surface reduces. The landslide thrust, soil resistance and pile displacement increase; and the increment rates become larger. In addition, based on the test results, reasonable values of seismic coefficient Cz are proposed for the structure; and the values of Cz are 0.2, 0.35 and 0.4 in the seismic zones of basic intensity VII, VIII and IX respectively. The test results are helpful to reveal the seismic mechanism and can provide reliable basis for seismic design of anti-sliding sheet pile wall.

According to the technical requirements of safe disposal of municipal solid waste incinerator (MSWI) fly ash, the fly ash is solidified by cement. The unconfined compression strength and destruction characteristics of fly ash solidified blocks with different cement contents, curing times and soaking times are studied. Then the erosion mechanism of leachate is analyzed. The test results show that the fly ash solidified blocks are destructed quickly after soaking in leachate with cement content less than 5% and curing time less than 3 days. The erosion of leachate has a great influence on the strength of fly ash solidified blocks. The fly ash solidified blocks present an obvious strain softening characteristic after soaking. The strength of fly ash solidified blocks without soaking in leachate accords with the mode of y=a[1-exp(-bt)]. With the increase of cement content and curing time, the unconfined compression strength increases and the failure strain decreases. But with the increase of soaking time, the strength increases firstly and then decreases; the turning point is about on 5-7 days; and the failure strain increases linearly. The erosion mechanism is that leachate can inhibit hydration reaction and destroy hydration products of fly ash solidified blocks. The results can provide theoretical basis and data support for the safe disposal of MSWI fly ash.

Theoretical and experimental analyses are conducted for the pipe jacking force of the great depth pipeline in pipe roof pre-construction method. The pipe jacking force is closely related to the vertical soil pressure. Reference to the calculation method of vertical earth pressure in tunnel excavation, there are two kinds of frequently-used theories to calculate vertical earth pressure in pipe jacking, i.e. M. M. Promojiyfakonov theory and Terzaghi theory. After analyzing the adaptability and defects of the two kinds of theories, new modified theoretical formula based on M. M. Promojiyfakonov theory and Terzaghi theory is got to calculate vertical earth pressure; and MATLAB program is written to calculate the numerical solution of new modified theoretical formula. Both the soil arching effect and the holding force under the arching of soil are considered in modified theory, which is more consistent with the actual soil deformation. M. M. Promojiyfakonov theory, Terzaghi theory and modified theory are used to calculate jacking force of two pipelines at different depths in pipe roof pre-construction method of Xinle Ruins Station of Shenyang Metro; the calculated results are compared to the measured results. It is shown that the results of M. M. Promojiyfakonov theory and Terzaghi theory are much larger than the measured values; the results of modified theory are slightly larger than the measured results. So that the results of modified theory are more suitable for estimating pipe jacking force of great depth pipelines.

Some results of static analysis considering the interaction between superstructure and composite foundation are obtained. However, the research on the response under dynamic loading, especially under earthquake loading, is very inadequate. Via relevant tests, ABAQUS are verified by comparison with EERA. The results show that the elastoplastic constitutive model based on Drucker-Prager yield criterion and the method of finite element and infinite element coupling can adequately simulate the nonlinear dynamic characteristics of soil and the infinite boundary. To the actual problem，the whole finite element model of rigid pile composite foundation-raft-superstructure system is established. And dynamic elastoplastic time-history analysis is made; the differences between rigid pile composite foundation and pile foundation in seismic response are analyzed. The seismic mechanism of rigid pile composite foundation, including the dynamic responses of piles, cushion and raft, is studied under different earthquake intensities; and the earthquake response and seismic performance of superstructure are analyzed deeply. It is shown that cushion has no damping effect almost under minor earthquake; rigid pile composite foundation is superior in seismic performance to pile foundation under major earthquake. The stronger earthquake is, the better damping effect is. But it is limited, the damping coefficient is over 0.8 generally. The results can provide references to engineering practice.

The shallow gas with a certain pressure is found in depths of 15-35 m in many sections of Hangzhou Metro Line No.1 during the detailed exploration. It not only constitutes a threat to the safety of investigation, but also influences the normal construction of the metro structures. It will be a threat to the later operation of metro if unsuitable prevention and control countermeasures are adopted. Shallow gas is one of the challenges met in construction of Hangzhou Metro Line No.1. However, at present, the case study with such large scale shallow gas is few in the history of metro construction at home and abroad; and the methods for preventing shallow gas disaster and the experiences are quite limited. Based on the studies of engineering properties of shallow gas and soil containing gas in Hangzhou area, considering the characteristics of Hangzhou metro engineering in different construction stages, and combining the practical engineering experience and the results of laboratory test and field monitoring, some pertinent disaster countermeasures are proposed. The countermeasures may supply assistance to metro construction and later operation, and also can give a reference to underground constructions in similar areas.

Based on the deformation law of a station yard with subgrade disease, supplementary geological exploration and analysis are performed. A program of karst foundation strengthening is developed as follows. ①Strengthen karst foundation and soil on the top surface of bed rock through grouting. ②Strengthen the original composite foundation through high-pressure jet grouting. ③Strengthen the cushion layer and the soil layer 3-5 m above it by low-pressure grouting. ④ Strengthen karst foundation by the grouting curtain in the side of subgrade. ⑤Build a monitor system to estimate the strengthen effect. The calculation settlement and bearing capacity characteristic value of the strengthened subgrade is 12.4 mm and 455 kPa respectively, which can satisfy the design requirement of the subgrade. And the monitoring results show that the deformation of subgrade trends to be slower after the strengthening of karst foundation and soil on the top surface of bed rock; and the deformation presents wholeness with a sharply reduced deformation rate after the accomplishment of jet grouting piles and the reinforced area by low-pressure grouting. Six months after the reinforcement, the deformation trends to be steady, which certifies the validity and availability of the strengthening program.

Based on the source-sink method, a three-dimensional additional stress formula caused by double-o-tube (DOT) shield tunnel construction soil loss is derived. The additional load caused by additional bulkhead pressure, the friction between soil and shield shell, and soil loss is added to study the total distribution of additional stress on adjacent underground pile foundation. The results show that, the additional squeeze pressure stress is caused in front of the excavation face, while the additional negative tension stress is caused in back of the excavation face. The greater pressure and tensile stresses occur near the depth of DOT shield tunnel axis. The friction between soil and shield shell plays a dominant role in total additional load, especially in the y-direction. The additional load in direction perpendicular to the segment is larger than the additional load in driving direction but has less range of influence. The additional load is small in the vertical direction, which has the opposite direction additional load near the tunnel axis compared to the additional load at each ends and its curve likes bow-shaped distribution. By the comparison with the results of numerical simulation, centrifuge experiments and field measurement, it is proved that the research on influence of DOT shield tunnel excavation on neighboring piles is reliable through analytical analysis.

The search and visualization of dangerous blocks as an important part of safety management of modern underground mine, has not been solved until now because of various reasons. By adopting the research thinking of the subsystem-global mapping and making the two codes of GeneralBlock and 3DMine available with compatible DXF file, the technology of automatic search and systematical visualization of dangerous blocks of underground mine is studied. It takes the developing tunnels at the level of 255 m and 305 m of zinc polymetallic orebody in Tongkeng mine for example. Firstly, the survey data of structural plane are input into GeneralBlock to calculate and analyze the blocks; and the DXF files including dangerous blocks are output. Secondly, those DXF files are input into 3DMine in order to make the dangerous blocks visualization; and their solid models are saved and input into roadway models in 3DMine for measuring the positions of dangerous blocks precisely. At last, the dangerous blocks in the developing tunnels are made systematical visualization by marking them on red alert. The application results show that 27 movable stable blocks and 1 unstable block are found in the developing tunnels by using this integrated technology. Research results testify that the automatic search and systematical visualization technology of dangerous blocks of underground mine can be achieved successfully, which can be regarded as a reference for safety management, disaster warning and fractures consolidation of underground mine.

Static load test is a most direct way to determine bearing capacity of large diameter pedestal pile. Because of the difficulties in loading process and high test cost, load test is not feasible in practical engineering. In general, bearing capacity can be obtained by means of the reduction of pile tip and side resistance based on the empirical parameter method of Technical Code for Building Pile Foundation in large diameter pedestal pile design. In this way, the bearing capacity of large diameter pedestal pile can not be well used. Based on the modifying of technical specification for large-diameter pedestal cast-in-place pile foundation, the test data of piles are statistically analyzed. The characteristic value of tip resistance and shaft resistance are presented using for reference from Technical Code for Building Pile Foundation and empirical values given by local codes of Beijing, Tianjin, Zhejiang, Fujian and Guangdong. The characteristic values of tip resistance and shaft resistance given in the present paper are compared with the measured values of test pile and calculated values obtained by Technical Code for Building Pile Foundation. It is found that the characteristic values proposed in the present paper are reliable and applicable in practical engineering.

It is difficult to search the dangerous block combination and acquire the safety factor precisely because of the numerous planes of weakness, the complicated block system and the diverse slip modes in rock slopes. Based on block element method, the search of dangerous block combination and the calculation of stability safety factor are conducted. At first, a new method for searching dangerous block combination is proposed, according to the results of block element method through strength reduction calculation. Then, the method for computing safety factors is studied; and the double line intersection method is suggested to calculate the safety factor when the abrupt theory of displacement is adopted. At last, several typical examples are calculated with the proposed methods; and the results are compared with those obtained by using the rigid body limit equilibrium method. The results show that it is more accurate by adopting the block element method, because the force and bending moment equilibrium and the nonuniform distribution of stress on the planes of weakness are considered. The proposed method is expected to be widely used to solve stability problems in complicated rock slopes.

In order to analyze reliability of loess high slopes, 266 natural ultimate state slopes are measured in Gansu area. According to the characteristics of topography and lithology, the studied region is divided into 4 subareas. The correlations of ultimate state slope height and width are given by means of double log-linear regression in different confidences for the 4 subareas respectively. Meanwhile, 1 024 groups of physico-mechanical indices obtained from reports are statistically analyzed. Based on the above, the ultimate state equation is built by means of Bishop's method. Reliability analysis for slopes under different confidences in 4 subareas is carried out with the Monte Carlo method. The effects of slope height, slope ratio and variability of parameters on the probability of failure are systematically analyzed. The results show that the slope ratio in Lintao-Yongjing subarea is smaller than 31.5? if the probability of failure(Pf) is kept within 10%. In the condition of the average coefficient of variation in Longdong subarea, Qin’an Lanzhou-Dingxi-Huining subarea and Tianshui-Qin’an-Tongwei subarea, the maximum of Pf of loess slopes is less than 10% with safety factor(Fs) of 1.3, as well as Pf less than 20% with Fs =1.2.

Water inrush is one of the typical geological hazards during tunnel construction in karst area; an attribute recognition model of water inrush risk evaluation is established based on attribute mathematic theory in order to control the water inrush risk and guarantee tunnel construction’s safety. Firstly, formation lithology, unfavorable geology, groundwater level, topography and geomorphology, contact zone of dissolvable and insoluble rocks, layer and interlayer fissures are selected as first grade indices of attribute recognition by considering of karst hydrologic and engineering geological conditions; and the unfavorable geology is divided into water-bearing structure, karst water system and fault fracture zone as second grade indices. The weights of first grade indices are determined by frequency statistic method through collecting and collating information of typical water inrush examples of karst tunnels; and the weights of second grade indices are determined by judgment matrix established by analytic hierarchy process (AHP). Then attribute measurement functions are rigorously constructed to compute attribute measurement of single index and synthetic attribute measurement. Lately, the identification and classification of risk assessment of water inrush in tunnels are recognized by the confidence criterion. The water inrush risk of Jigongling tunnel on the line of Fanba expressway is evaluated based on the established attribute recognition model; and the evaluation result agrees well with construction situation, so as to provide an effective method for water inrush risk evaluation of karst tunnels.

Reasonable method for determining supporting load of deep tunnel is a critical scientific problem during design and construction at present. Taihang mountain tunnel goes through gypsum breccia region of about 4 410 meters on Shijiazhang-Taiyuan high-speed passenger railway. In situ experiments on supporting load were carried out to investigate statistical surrounding rock pressure of primary support and contact pressure of lining. According to Terzaghi theory, Bierbaumer theory, Xie Jia-xiu theory, Code of Design on Tunnel of Railway and M. M. Promojiyfakonov theory, surrounding rock pressure is calculated; and comparison between the theoretical values and the monitoring data is made. The results show that surrounding rock pressure is dominated by vertical mainly. Calculation result by Terzaghi theory is mostly close to monitoring value. Obviously, vertical and horizontal distribution patterns are different from traditional calculation methods. Vertical pressure is recommended as double peak type or uniform distribution. Horizontal pattern is recommended as folded line type. Furthermore, primary support and lining bear load compatibly. The ratio of lining is about 50.1%. The results can provide references for construction of deep tunnel on high-speed railway in gypsum breccia stratum and similar geological tunnel.

The large-scale landslides, widely distributed in southwest mountains and canyons, are prone to produce instability due to rainfall. In Gushui hydropower station，not only is the extremely large volume of Zhenggang large landslide accumulation body up to more than 4 750×104 m3, but also the thicknesses of many ultra-deep landslide are over 50 m. Under the influences of the heavy rainfall in 2008, the landslide started to deform again and even more seriously. According to the results of field engineering geological investigation and analysis of geologic structure characteristics and hydrogeology conditions, the mechanism of rainfall infiltration and stability in the current situation was qualitatively analyzed first. Then the whole landslide was transfixed slipping surface in the state of creep deformation; so it should be treated such as excavating and reinforcing. Combined with the analysis result, quantitative study of the seepage properties of landslide showed the dynamic sliding mechanism that water infiltration caused deformation, evaluated the stability and treatment, thus proposed effective drainage measures before and after the treatment under the rainfall. The results also showed that the saturation line and backwater was formed due to saturated soil by infiltration. Before the treatment, the backwater was seriously persisted in local landslide and slip surface; and the worst stability came out 4 days later after the rain stopped; and the maximum thickness of backwater increased up to 6 m. However, with backwater reduced obviously, the stability was significantly increased by the treatment. The results truly reflected the situations and laws of the stability of landslide, which were consistent with the results of field exploration. The results were significantly improved by the treatment. Drainage measures proposed would keep the landslide’s stability efficiently. The research results will be useful and meaningful in the same projects as reference.

According to the deformation characteristic of surrounding rock in deep thick top coal roadway, guided by the support idea of controlling at first, pressure yielding behind and then resistance, four schemes of pressure relief anchor cable box beam support system are designed, of which beam arrangement mode are lateral beam, longitudinal single beam, longitudinal double beams and longitudinal-lateral combined beam respectively. The results of numerical test show that, among the four schemes, longitudinal-lateral combined beam scheme is best in surrounding rock controlling; lateral beam scheme is relatively worst. The four schemes are applied to field test. The monitoring results show that the surrounding rock deformation of test section with longitudinal-lateral combined beam scheme is least; the sides deformation is 120 mm and the roof deformation is 90 mm. Meanwhile, bolts and anchor cables have the largest supporting reaction in this scheme. The force measuring anchor cable box beams are designed and made. The monitoring results show that box beam stress conditions are reasonable and coordinate, material is used fully. In the four schemes, supporting and protecting components reaction corresponding with the effect of the roadway control, and the effect of anchor cable yielding ring is obvious. Mechanisms of support systems are analyzed. Under the geological conditions of deep thick top coal roadway, longitudinal-lateral combined beam and longitudinal single beam schemes of pressure relief anchor cable box beam support system are better for controlling surrouding rock deformation in a cost-effective way; and the supporting demand is satisfied.

A simplified analytical approach is proposed to study the behavior of piled raft foundation with rigid raft subjected to coupled loads in layered soils. Based on the shear displacement method, using the transfer matrix, the pile-pile interaction under vertical load is analyzed. Adopting the modified subgrade modulus, the pile head-pile head interaction under horizontal load is analyzed through the finite difference method. Based on solutions for stress and displacement in layered elastic half space, the interactions between pile head and soil surface, soil surface and pile head, soil surface and soil surface subjected to coupled loads are taken into account to determine the flexibility matrix of the pile group-soil system. Then the load-settlement relationship of rigid piled raft foundations subjected to coupled loads, load and settlement of the piles along the depth are obtained. Comparison between the finite element results and the calculated results is carried out; and the feasibility of the present method for analyzing piled raft foundation and the superiority of the modified subgrade modulus are proved. The piled raft foundation with rigid raft subjected to coupled loads is analyzed. The calculating results indicate that the horizontal load will cause the pile raft foundation oblique.

Based on many dynamic test results about rockfill materials of domestic earth-rock dams, the dynamic residual deformation characteristics of the rockfill materials are studied. During the process of collecting test data, the primary residual deformation model presented by Shen Zhu-jiang is modified along with considering the affect of consolidation ratio and fitting the test data well. Through dynamic residual deformation analysis and calculation of ideal concrete faced rockfill dam (CFRD), the similarities and differences between the primary residual deformation models presented by Shen Zhu-jiang, Kong Xian-jing, Ling Hua and this paper are given. Then, the rationality of this improved model is elaborated. In addition, based on the statistic analysis of test results, the average, upper and lower envelopes of the curves of - and - are proposed to calculate the seismic residual deformation of the rockfill materials. With the application of the above, the deformation behaviors of ideal CFRD are numerically analyzed. The results show that the parameters of c4 and c5 have great impact on the residual deformation of the dam; but the parameters of c1 and c2 have almost no effect. It’s significant to apply the mean curve parameters in seismic design of medium and small earth-rock dams and the result has certain reference value.

The cone penetration tests under seven different gravity fields were studied by a commercial distinct element method (DEM) code PFC2D (particle flow code) in order to investigate the relationship between cone penetration tests under high and low gravity fields. The numerical results show that cone penetration has an apparent effect on the upper zone of the ground under low gravity fields, while it has a significant effect on the lower zone of the ground under high gravity fields. The maximum (steady) values of the normalized cone tip resistance and the maximum values of normalized mean stress and deviatoric stress increase linearly with the reciprocal of gravity accelerations. The soil around the penetrometer undergoes apparently loading and unloading process during the penetration process, with the smaller gravity field corresponding to the more apparent loading and unloading phenomenon, while the upper zone of the ground undergoes the loading and unloading process more apparently than the lower zone under the same gravity field.

Artificial ground freezing is widely used in the construction of underground engineering. The calculation of freezing temperature field is not only the basement of theory research on artificial ground freezing, but also an important basis of the freezing construction design. Single-circle-pipe freezing method is often applied to artificial ground freezing projects, while the analytical solution of this situation has never been published. Thus it’s necessary to get the solution to benefit our engineering practice. Based on the theory of analogy between thermal and hydraulic problems, using conformal mapping, mirror reflection of sources and sinks, and the superposition of potential function, this paper gives an analytical solution to the steady-state temperature field of single-circle-pipe freezing and demonstrates its correctness by numerical thermal analysis. Comparison of the analytical solution with the numerical thermal analysis shows that the analytical solution is precise enough and will afford reasonable guidance in single-circle-pipe freezing projects. A simplified formula is also given and is proven to be precise enough. Based on the simplified formula, the thickness formula of the outer part of frozen wall is derived. The analytical solution, simplified formula and thickness formula can provide guidance and theoretical basement for construction and design of single-circle-pipe freezing projects.

In western region of China, there are numerous alpine valleys. The special topography develops a large numbers of mountain disasters of rock-fall. It badly threatens the safety of people. Generally, the shed is the most effective engineering method. However, the impact behavior of rock shed is a complicated dynamic process, especially in the aspect of contact response between slab and rock-fall. The application of rock-fall shed is seriously restricted because of the current research status. In view of the present disadvantages of Herzt contact law, a theoretical method of dynamic response of reinforced concrete (RC) slab is proposed based on the numerical indentation test and the Olsson’s model. Compared to the presented methods, the results obtained in this paper are much closer to the results of finite element method. Meanwhile, the impact forces obtained by Herzt model are larger because of without considering of plastic deformation. As a much reasonable method, it can be used to analyze the dynamic response of RC slabs subjected to impact load of rock-fall. At the same time, it also provides an effective numerical analysis method for the application of indentation test to civil engineering.

Using DDARF (discontinuous deformation analysis for rock failure) method, the uniaxial and biaxial compression experiments of double-fractured rock mass are numerically investigated. Influences of fracture angle and lateral stress on behaviour of rock mass are studied and the corresponding stress-strain relationship curves are analyzed. Moreover, crack initiation stress and the peak strength of specimens are obtained. In the biaxial compression experiments of double-fractured rock mass, strength envelopes of 45° fractured specimen under different lateral pressures are obtained. Using DDARF mothod, the axial load-displacement curves and crack propagation for four kinds of specimens with different anchor angles are obtained, which can simulate the anchorage effect on fractured rock. The simulation results and the previous splitting tests under similar conditions agree well with each other. The parameters obtained from biaxial compression experiment are used in a practical engineering example; and the differences of damage state between intact rock mass and fracture rock mass are analyzed after excavation of underground cavern by means of the equivalent mechanical character. At the last, the DDARF method is adopted to analyze the stability of an underground cavern with four groups of random joints; and then crack propagation process of jointed rock mass around the cavern are obtained. Besides, the effect of anchor on fractured rock mass are studied through monitoring the displacements of the key points.

Based on the multiphase pore elasticity theory, the propagation equations for the various elastic waves in unsaturated soils are derived. According to the boundary conditions at the interface, the relationships between different wave amplitudes of various potential functions are established; and then the reflection and transmission problems of incident shear wave at the interface between unsaturated soil layers with different saturation degrees are discussed. Three compressional waves and one shear wave exist in the infinite unsaturated soils, therefore four reflected waves and four transmitted waves are generated when the shear wave propagates at the interface between different unsaturated soil layers. The theoretical expressions for the amplitude ratios and energy ratios for the various waves are derived; and the numerical analysis is then performed. In the numerical examples, the influences of incident frequency, incident angle and saturation degrees of the upper and lower soil layers on the proportionality coefficients of the reflected wave energy and transmitted wave energy (namely energy reflection ratios and energy transmission ratios) are discussed respectively. It is shown that the energy reflection ratios and energy transmission ratios are affected by both the incident angle and the incident frequency. Additionally, the influence of the saturation degrees of the upper and lower soil layers on the reflected and transmitted waves should not be ignored.

Based on the elastic wave theory, the stress wave field will separate when it is oblique incidence across a linear elastic joint. The propagation of stress waves in rock mass including oblique joints is analyzed; and the computation of transmission and reflection coefficients is conducted with a universal distinct element code (UDEC), according to different approaching times of incident, transmitted and reflected waveforms. When stress waves is oblique incidence across a single joint, the simulated relationships between transmission coefficients and joint stiffness, incident angle agree well with the classic theoretical solutions, as well as the reflection coefficients. Furthermore, the situation of stress waves propagate through a set of parallel joints is analyzed. The results show that the transmission coefficients Tpp, Tss of the same kind wave firstly increase to maximum values and then decrease to stable values with the increasing of joint space; and the stable value becomes smaller when the number of joints is larger. Additionally, the critical values of joint space when transmission coefficients approach maximum values are approximately identical for different numbers of joints; but the inflection values of joint space for the situation of stable values gradually become larger with the increasing of joints number.

A new method for determining nonlinear fracture toughness of frozen soil is presented based on the method of J integral as well considering nonlinear behavior characteristics of frozen soil. It is well known that J integral can be expressed as J =Je +Jp, in which Je represents elastic component and Jp represents plastic component. The calculation of Je is relatively simple, but the calculation of Jp is quite difficult. Because in the process of calculating Jp, one parameter Up, which represents the plastic work of sample, is very difficult to determine. To simplify the calculation, rewriting Jp as an expression of Je, Jp =qJe, then J integral expression becomes as J=(1+q)Je. That means when Je and factor q are determined for one kind of sample, J integral can be gotten easily. This method is named as correction-factor method. The correction factor q is obtained by increments of load and displacement according to curve of load vs. displacement (P-Δ curve), in which includes two cycles of loading-unloading. Using the suggested method, nonlinear fracture toughness of frozen soil in-situ is determined; and the results in comparison with the data of related literature show its reasonableness and availability.