With the development of society, huge muddy production with silt and sludge as representatives gets rather close to the municipal refuse. Moreover, muddy materials have very complex and diverse traits and pollutants inside, resulting in a critical issue that we should dispose and utilize them economically and safely. Although the compositions of the various muddy materials are different, they share many similar physical and chemical properties. The ways to treat these materials are also consistent, including quantity reduction such as dehydration, harmless treatment like pollutant stabilization, and reclamation such as reusing after pretreatment. Furthermore, some special properties of muddy materials can also be utilized to solve certain problems. For a long time, there had scientific research and processing technology with various kinds of muddy materials in all walks of life. However, problems arose such as disunion of concept and index, isolation between technology and methods that all affected the development of basic theory and techniques. To solve these problems, this paper proposes the establishment of disciplines direction — the muddy material science and technology. We review main types and problems, hot issues encountered during the treatment process as well as several typical researches in muddy materials. The problems of the treatment with mud are analyzed and summarized.

A three-dimensional centrifuge model test was carried out to investigate tunnelling effects on pile group in expansive clay. A volume loss of 2% was controlled. During the test, ground settlement trough, additional settlement, additional bending moment and axial force of pile due to tunnelling were monitored. The results show that ground settlement troughs are affected obviously by the distance between the centre axis of pile group and tunnel face. The additional settlement of pile shows linear growth with the tunnel excavation from -0.75D to 1.25D; and it continues to settle even when the tunnel excavation reaches 1.25D away. The settlements of front piles and behind piles are different; and the pile cap tilts. Positive additional bending moment presents at the top of front pile while negative bending moment presents at the bottom. However, only positive additional bending moment occurs at the bottom for behind pile. Moreover, the maximum additional bending moment of the front pile occurs near the axis of tunnel; and it is much bigger than that of the behind pile. The additional axial force of the front pile increases with the tunnel construction and reaches maximum near the axis of tunnel. The additional axial force of the behind pile increases too; but it reaches maximum near the top of pile.

At 15:00 PM on October 3, 2009, the 260×104 m3 of the pile-stabilizing fill slope slid suddenly in Panzhihua airport in Southwest China, so as to reactivate the Yijiaping landslide located at the downside of the fill slope and cause 2 years interruption of Panzhihua airport. Sliding mechanism of the pile-stabilizing fill slope under rainfall conditions is studied using a centrifuge modeling technique. A series of centrifuge tests on the reinforced model are conducted respectively during different accelerations under natural and rainfall conditions, reproducing the slide process of the fill slope. The analysis of the centrifuge model test results is performed to investigate the sliding mechanism of the landslide. Under natural condition, deformation of the slope includes crest settlement and creep of soft layer mainly; and the slope appears creep state. Under rainfall and groundwater conditions, tensile fracture of rear and compressive failure of front are obvious; and the slope shows progressive break characteristic and slides finally. The piles in the slope stressed larger, and the closer pile to the rear end of the slope, the larger the pile’s pushing force, which causes the progressive failure of piles from rear to front. The irrational pile spacing among three rows of piles may account for the incapability of piles in retaining the slope. It can be concluded that the occurrence of slope failure is contributed to the soft layer at the soil-rock interface and the high pore pressure in rainfall condition, which is found to be 3.7 times the pore pressure under natural condition. Comparing the model test with the prototype test, the sliding mechanism of the fill slope can be recognized as shear creeping-progressive breaking-sliding mode.

According to the results of limestone samples under uniaxial compression test, conventional triaxial test and unloading test, energy evolution characteristics of limestone during the whole-process deformation are analyzed. Results show that rock failure laws of energy evolution are different between loading and unloading. The confining pressure restrains the internal crack growth. The absorbed total energy U and elastic energy U e in conventional triaxial compression test are larger than that of uniaxial compression test. The elastic energy U e continues increasing before peak strength. However, it remains the same during the unloading stage. The elastic energy U e with release availability mainly accumulates in the loading process prior to the unloading process. The geostress state before engineering excavation determines the releasing amount of elastic energy. The dissipated energy U d increases quickly near the peak strength. But the increasing velocity of dissipated energy under unloading stress path is faster than that under loading stress path. The rapid increase of dissipated energy indicates the occurrence of rock failure in loading and unloading stress paths. The absorbed total energy U and elastic energy U e increase with the confining pressure increasing. The absorbed total energy U, elastic energy U e and dissipated energy U d decrease with the unloading velocity increasing.

In light of traditional instrument which can’t consider the factors of stress and temperature in the geotechnical engineering, a set of multifunctional soil-water characteristic curve test instrument is developed. Taking silt as study object, the soil-water characteristic curve tests under the effect of different consolidation stresses are carried out. The volume shrinkage deformation is also measured at the same time so as to amend miscount caused by volume change. It is found that the shrinkage is obvious in the drying process; and the smaller consolidation stress is, the more shrinkage is. But the soil volume remains unchanged in the wetting process. The consolidation stress has a great influence on the air-entry value and wetting (drying) rates. The air-entry value and wetting (drying) rates are greater when the consolidation stress increases. Finally, the mesoscopic tests under the action of different consolidation stresses are carried out for revealing its influence mechanism. The test results show that the large pore between the soil aggregate is compressed mainly by consolidation stress; but the pore between soil particles is less effect by it. The soil-water properties are related to the pore size and its distribution mode closely. The large pore has influence on the air-entry value and the distribution mode has influence on the wetting (drying) rates.

The behaviors of strength and deformation of natural soils differ from that of remolded and unstructured soils due to structural property. The effective consolidation stress method for the undrained shear strength evaluation of remolded and unstructured soils is reviewed and the corresponding equations are derived. The equations of the method for the undrained shear strength evaluation of structured clays are hereafter established in terms of the strength envelopes simulated by two segment lines with different slopes (tangents of internal friction angles) and intercepts (cohesions). The component of effective stress in the equation of effective consolidation stress method should be multiplied by a revised coefficient of 0.8 for the effective stress range less than the yielding stress. The undrained shear strengths of structured soils can be then determined if the consolidation stresses prior to shearing would be known. The applicability of the effective consolidation stress method is validated using the isotropically consolidated triaxial test results of soft structured Lianyungang clay. The undrained shear strengths estimated by the effective consolidation stress method are in good agreement with the laboratory results. Furthermore, the structure-induced yielding stress predicted by the effective consolidation stress method has a certain but negligible deviation from that fitted by the laboratory data.

By lots of creep test researches, artificial frozen clay shows an obvious creep behavior under high deviatoric stress. When artificial frozen clay in low stress level, a viscoelastic creep deformation presents. While in high stress level, viscoplastic creep deformation presents. On the basis of Nishihara model, parabolic viscoelasto-plastic creep constitutive model is constructed by adding a viscoelastic element and parabolic yield function instead of plastic yield. The modified Nishihara model can describe the shear creep characteristics of artificial frozen soil well. Based on niche technology, genetic algorithm and viscoelasto-plastic theory, the model of viscoelasto-plastic displacement back analysis for artificial frozen soil has been set up by finite element program generator (FEPG). With the measured data, displacement back analysis is conducted during the excavation of deep freezing shaft. And the obtained parameters of mechanical model for frozen soil agree with the test results. The deviation between numerical calculation results and measured data is less than 6%; so the inversion results verify the feasibility of the model. The constitutive model parameters, which obtained by displacement back analysis for artificial frozen soil, can reflect the average of frozen wall construction process and effective freezing range. So it has wide representation and important engineering significance for frozen wall stress field, displacement field and stability prediction.

Cyclic impact tests on sandstone under different axial and confining pressures are conducted with a modified three-dimensional split Hopkinson pressure bar (SHPB). The main research contents are change property of energy absorption per unit volume, relationship between energy absorption per unit volume and average strain rate, effects of axial compression and confining pressure on energy dissipation of sandstone under cyclic impact loads. Confining pressures are set as 4, 8, 10 and 12 MPa; and four levels of axial stresses are 49, 84, 105 and 125 MPa respectively. Incident waves in input bar are approximately equal and incident energy is equal to 230 J. The results show that energy absorption per unit volume increases with the increasing number of cyclic impacts, when the cyclic impact loadings are a constant. Energy absorption per unit volume has a good linear increasing relationship with average strain rate. The linear-fitted slope K of average strain rate and energy absorption per unit volume indicates a trend of “increase, constant, then decrease” with the increasing axial stress. When the axial stress is smaller, K increases and then becomes lower with the increasing of confining pressure. The smaller axial stress is, the bigger the confining pressure corresponding to slope’s turning point is. After the axial stress reaches to 125 MPa, K continually reduces with the increase of confining pressure. The conclusions may provide the theoretical basis for blasting design of rock mass engineering under different in-situ stress conditions.

Natural soils are structured and anisotropic in the process of sedimentation. Within the theoretical framework of breakage mechanics for geomaterials, a new binary medium model (BMM) for structured soils with initial stress-induced anisotropy is formulated. Breakage mechanics for geomaterials idealizes the structured geomaterials as binary medium consisting of bonding blocks (the bonding element) with strong bonding and weakened bands (the frictional element) without bonding; and the bonding blocks will break and transfer to weakened bands gradually during the loading process. The bonding blocks are assumed as elastic-brittle constitutive model with transverse-isotropy; and the weakened bands transferred from the bonding blocks are assumed as nonlinear elastic Duncan-Chang model. By introducing the breakage ratios and local strain coefficients reflecting the influence of soil structure and anisotropy, the binary medium model for structured soils with initial stress-induced anisotropy is established; and the determination of model parameters in detail is given. Finally the capability of the new model to model the stress-strain properties of anisotropic structured soils is verified and compared with the triaxial tests of artificial structured soils with initial stress-induced anisotropy. The computational results demonstrate that the proposed constitutive model can simulate the mechanical properties, including stress-strain and volumetric deformation, of structured soils with initial stress-induced anisotropy.

Based on orthogonal test, the softening mudstone interlayer effected by three factors, i.e. temperature, brine density and immersion time, is studied. The extremum difference analysis and variance analysis are respectively applied to analyzing the results of the test. Any two factors are the binary liner regression; the relative importance ratio according to the standardized regression coefficients for the uniaxial compressive strength and elastic modulus of mudstone interlayer is used to build judgment matrix; analytic hierarchy process is used to calculate weights of various factors. The results show that the temperature and the immersion time play a decisive role on the softening process of mudstone interlayer; and the weights of analytic hierarchy process are more than 98%. The greatest influencing factor is immersion time whose weight is about 80%. The change of brine density has hardly any effect; therefore it can be neglected in the practical project. With the immersion time extending and the temperature increasing, the cracks in mudstone interlayer are gradually intruded into solution; meanwhile, the uniaxial compressive strength and elastic modulus show a decreasing trend. In view of the insignificant role the brine density plays in the whole experiment, only temperature and immersion time are taken into consideration to establish the equations.

Based on some simplification and assumption of one-dimensional finite-strain consolidation theory and Barron axisymmetric consolidation theory, a finite-strain consolidation governing equation for vertical drains ground is established. Based on the finite-strain consolidation equation for double-layered vertical drains in soft soil and the computing program in some literatures, the power function relationships of permeability coefficient-void ratio k=ced and effective stress-void ratio e=a( )b for soft soil are introduced; and some examples calculation of consolidation behaviors of double-layered vertical drains ground are carried out under instantaneous loading. The obtained results show that: (1) Some important influence on the consolidation behaviors of double-layered vertical drains ground of various parameters in power function permeability and compressibility relationships for double-layered soft clay soil are as follows. With the increase of parameters c1, c2 of power function permeability relationship (the increase of permeability), or parameters a1, a2 of power function compressibility relationship for soft soil, the void ratios in the radial and vertical directions decrease faster; and the dissipation velocity of excess pore water pressure and the settlement rate are also accelerated; the settlement development rate is faster than the dissipation velocity of excess pore water pressure. (2) The void ratio and average excess pore water pressure at the interface of double-layered soft soil present significant mutation, of which the distribution curve along the depth is divided into two sections with different consolidation behaviors.

In order to study the strength and deformation characteristics of Q3 sand loess from Heifangtai, Yongjing, Gansu province, using the true triaxial apparatus developed by Xi'an University of Technology, true triaxial tests are carried out under conditions of different confining pressures (50, 100, 200 kPa), ratios of intermediate principal stress (b =0, 0.25 and 0.5) and water contents (5%, 10% and 15%). The test results illustrate that: (1) The shear failure modes of Q3 sand loess are displayed by lateral bulge failure and monolete shear failure under the true triaxial conditions; and there are other modes for few soil samples including taper failure, double-slit shear failure and T-shape shear failure. (2) The deviatoric stress difference of Q3 sandy loess increases with the ratio of intermediate principal stress when the confining pressure and moisture content are the same, while the ratio of intermediate principal stress (b) is different. Meanwhile, the stress-strain relationship curve shows hardening and obvious shear dilatancy phenomenon. However, the ratio of intermediate principal stress effects the relationship between generalized shear stress and generalized shear strain; the relation curve increases accordingly with the increase of b. With the increase of b, the strength decreases; and the relation curve between the generalized shear stress and the average spheric stress ratio (p/q) becomes flat gradually, and decreases accordingly. (3) With the increasing of water content, the Q3 sandy loess transfers to semisolid state from initial solid state. The cohesions with different ratios of intermediate principal stress all declines remarkably; but the internal friction angle increases a little.

Cementation plays an important role in the stress-strain and strength behavior of cemented sand. The calcium oxide cemented sand samples prepares artificially in twice mixing water method. In order to speed up the carbonation reaction, samples are then placed in a humid room filled full of dry ice. Until samples curing are finished, consolidated drained triaxial tests and CaCO3 quantitative chemical tests are conducted on artificially cemented sand which have different cement ratios. The influence of cement ratio on the physico-mechanical characteristics of cemented sand is analyzed. Defined the product of CaCO3 content and Ca(OH)2 content as the chemical index to estimate the degree of cementation, an modification is given on Mohor-Coulomb strength theory using the chemical index . The test results show that twice mixing water method can effectively control the initial moisture content of artificially cemented sand; the formation of carbonate cements have a great influence on physico-mechanical properties of artificially cemented sand. At different confining pressures, samples show stress softening. With the increasing of cement ratio, stress softening are enhanced, cohesion and increment of frictional angle increase. Cemented sand has property of pressure sensitivity. Confining pressure and cement ratio can effectively control dilatancy. With the increasing of , cohesion and increment of frictional angle increase. The quantitative relations of cohesion, increment of frictional angle vs. are obtained. Combined with the test results, the relationship of shear strength parameter and is established to modify Mohr-Coulomb strength theory. This modified model can reflect the influence of carbonate cements on strength of cemented sand.

As one of the basic issues, the evolutions of humidity field must be studied to avoid roadbed water damage. To study changes and distributions of the subgrade humidity field affected by compaction methods, 4 kinds of scale-down laboratory subgrade physical models with different compaction methods were tested. Through analyzing tow-dimensional infiltration rainfall test results，the influences of different compaction methods on the extended distances in roadbed humidity disturbed zone, humidity distributions in front disturbed zone, humidity gradients and humidity values of embankment were summarized. Their mechanisms were explained and relevant engineering suggestions were proposed. The results show that extended distances in roadbed humidity disturbed zone decrease linearly with the increase of compaction degree. The whole humidity disturbed zone can be divided into front and back zones. The front zone humidity shows a quick linear decrease distribution and humidity gradient increases linearly with the increase of roadbed compaction degree. Embankment humidity is affected significantly by the compaction degree of its upper roadbed. The combined effect reveals that embankment humidity decreases linearly with the increase of roadbed compaction degree. It should place a strict control on the upper subgrade impaction to avoid deeper subgrade damages due to water in engineering construction.

Railway track and nearby structures in the vicinity of high-speed railway are exposed to low-amplitude small-strain vibrations as high-speed railway traffic construction activities arise. In the finite element (FE) method, how to determine the stiffness tensor E is important to build a rational predicting model to analyze the settlement of track and nearby structures in the vicinity of high-speed railway. Shear modulus G and damping ratio D which will vary with shear strain amplitude are two important stiffness and damping parameters. It is found through resonant column test study that the shear modulus G of soil due to vibrations must be determined by the strain amplitude and the confining pressure, even though under small-strain (shear strain amplitude 10-6<? <10-4) vibrations, will still show a remarkable nonlinear characteristic. The Davidenkov model based on the classical Hardin-Drnevich model is proposed to fit the experimental tests of the curves of shear modulus G/Gmax and damping ratio D vs. ? powerfully. However, the shortcoming of the Davidenkov model is that the fitting parameters are too much, and the key parameter ?0 can’t be determined through a definite physical method. This study proposes an simplified two-parameter Davidenkov model through using a reference shear strain ?ref which will be determined through the ratio of the shear strength τf of soil in static triaxial tests versus the maximum shear modulus Gmax at a given stress state instead of the fitting parameter ?0 on the basis of Mohr-Coulomb failure criterion, and verifies the feasibility of the proposed method through tests finally.

According to the Biot wave theory and the complex function combined with multi coordinate method, the wave scattering dynamical equation of liquid- filled pipe in saturated soil impacted by P waves is solved. The effects of fluid property in the pipe, incident wave angle and buried depth on the distributions of dynamic stress concentration coefficient and pore pressure concentration coefficient are analyzed. The results show that, in low frequency elastic wave incidence, dynamic stress concentration coefficient and pore pressures concentration coefficient are relatively uniform distributions around the periphery of the pipe; when the incident frequency increases, the distributions become complex, but the peak values decrease. In low or middle frequency elastic wave incidence, the dynamic stress concentration coefficient and pore pressure concentration coefficient of the pipeline filled with water or oil are relatively smaller than air medium. But in high frequency elastic wave incidence the opposite is the case. When the fluid medium in pipeline is water, the influences of the incident angle and depth of pipeline on the dynamic stress concentration coefficient and pore pressure concentration coefficient are analyzed. With the increasing of incident angle, the distribution also occurs at certain angle deflection. When the incident wave is perpendicular to the pipe, the peak value of pipe dynamic stress concentration coefficient around the periphery of the pipe is relatively large. With the increasing of depth, dynamic stress concentration coefficient and pore pressures concentration coefficient show shock decrease.

According to the special environment in deep seasonal permafrost regions, the repeated frost heave and thaw settlement properties of silty sand under different initial moisture contents, dry densities, overburden pressures, and freeze-thaw cycles were extensively investigated by means of laboratory experiments. Testing results show that: the freezing temperature is -1.03 °C. The deformation of frost heave and thaw settlement presents waves of ups and downs and then levels off in freeze-thaw cycles. Samples with higher dry density present expansion, while samples with lower dry density present compaction after repeated freeze-thaw cycles. The overburden pressure restrains the amount of frost heave, while increasing the amount of thaw settlement. However, the overburden pressure reduces the frost heave ratio and thaw settlement coefficient in every freeze-thaw cycle. At the optimal value of initial water content, the height of sample does not change after repeated freeze-thaw cycles. Due to water supplying to the samples, the water contents of samples frozen and thawed in the open system are greater than the initial water content. The increases of dry density and overburden pressure effectively restrain the outside water supply. The frost heave ratio and thaw settlement coefficient in every freeze-thaw cycle levels off after 4 freeze-thaw cycles.

As anisotropic granular materials, soil’s dynamic characteristics depend on the stress path. Foundation soil is in heart-shaped line rotation stress path of the maximum shear stress space under train load. Yet it is not reported that the heart-shaped line stress path is imitated in laboratory test. The simplified form of the soil’s dynamic response is proposed, which is justified by analyzing the stress path of foundation soil caused by train load. For imitating the stress characteristics produced by train load, the loading methods of three types of hollow cylindrical apparatus (HCA) are derived: (1) The two-directional vibration HCA can achieve the heart-shaped line stress path by determining the forms of axial force and torque. The variation of spherical stress p is similar to the load waveform of axial force. When the internal pressure is equal to external pressure, the variation of coefficient of intermediate principal stress b will like cosine curve. (2) When the three-directional vibration HCA imitates stress characteristics produced by train load, p can remain unchanged or b can be kept at the constant of 0.5. (3) With the control of the axial force, torque as well as the internal and external pressures, the four-directional vibration HCA can imitate stress characteristics of the foundation soil under train load. By setting a constant b, the linear relationship between p and q can be realized. Besides the constraint conditions on loading parameters are derived, under which the normal stress in soil is compressive stress and the direction of intermediate principal stress is kept at the diametrical direction all along.

With the great development of agricultural hydraulic engineering in China, irrigation induces serious landslide disasters at the edge of loess plateaus. As a typical region of loess landslides induced by agricultural irrigation, the Heifangtai plateau in Gansu province has been paid closed attentions in recent years. To investigate the deforming and failure mechanism of loess slope due to infiltration, undisturbed specimens are obtained from the backwall of a typical landslide in Heifangtai and are used to conduct three series of stress-path triaxial tests to simulate the process of slope failures. Through conducting anisotropically consolidated undrained (ACU) tests and saturated dead-load (SDL) tests with increased pore-water pressure, the initiation and failure processes of saturated loess are investigated. Through conducting unsaturated dead-load wetting (UDL) tests, the initiation and failure processes of unsaturated loess are investigated. Based on above results, the mechanism of infiltration-induced loess landslide is elaborated from the aspects of stress paths and microscopic structure of loess. Besides, the typical physico-mechanical properties of loess are analyzed. The results show that the saturated loess has a unique critical state line which reflects the major stress level corresponding to the thickness of the loess; the state boundary line of the loess can be approximated by its critical state line; the soil-water characteristic curves (SWCCs) obtained from the UDL tests are affected by the stress state, showing lower water content at higher stress level.

It has important value of theory and practical significance to understand the mechanical characteristics of rock structural planes by taking direct shear tests for shear strength of simulation rock structural planes with different sizes (especially large sizes). Based on MSJ-DST, the direct shear tests of simulation rock structural planes in sizes 20 cm×20 cm, 40 cm×40 cm, 60 cm×60 cm, 80 cm×80 cm and 100 cm×100 cm are taken under the normal stresses from 200 kPa to 1 000 kPa. Then, the characteristics of shear strength of rock structural planes with different sizes are studied. The results indicate that the stresses and deformation characteristics of simulation rock structural planes under different normal stresses are similar; and the peak shear displacements are generally fluctuated in the vicinity of a certain value. Under the same normal stress, the peak shear strength values of simulation rock structural planes with different sizes are fluctuated in the vicinity of a certain value; and the residual shear strength values rise slightly as the size increases. Under five grade normal stresses, the variations of peak shear strength and residual shear strength with different sizes are approximately same; and the ratio of residual shear strength value and peak shear strength value increases and tends to be stable as normal stress increases.

The mathematical model for structured soil is a key problem of soil mechanics in twenty-first century. Soil microstructure is the main reason which causes the differences between natural and remolded soils. Stress structural parameter and strain structural parameter were introduced to describe the change of microstructure. At the same time strain structural parameter were introduced into the isotropic consolidation process to describe the effect of spherical stress on structural soil in the process of isotropic consolidation and shearing, which realized the description for the whole process of triaxial shear tests. Modified Cambridge model can make an accurate description for normally consolidated remolded clay in the triaxial compression tests, but failed to make the same precision for overconsolidated clays and intact soil, which had the structural characteristics. Then the strain and stress structural macro parameters were introduced into the modified Cambridge model to achieve structurization. The parameters determination methods for structured modified Cambridge model were similar with the conventional modified Cambridge model parameters determination, but more than the relationships between spherical stress and strain, deviatoric stress and soil structural generalized shear strain. Through numerical simulation, the structural modified Cambridge model well reflected the structural evolution of the intact soil, and described the whole failure process of structure on a variety of stress paths for intact clay in Haikou.

The conventional earth pressure analysis method only considers the shear strength of saturated soil, which ignores the influence of matrix suction as well as its change. Based on the principle of effective stress of unsaturated soil and Rankine's earth pressure formula of saturated soil, the Rankine's earth pressure formula of unsaturated soil is deduced. The earth pressure formula of unsaturated soil under the condition of rainfall is established by the combination of the formula derived above and Iverson’s rainfall infiltration analytical solution. In the formula, earth pressure of unsaturated soil under rainfall infiltration is expressed as a function of time and depth. The study result shows that the earth pressure value calculated by the improved method is relatively larger than the conventional calculation and its action point is higher. With the occurrence, infiltration and stop of rainfall, the value of active earth pressure is at a state of “decreasing, increasing, decreasing and lastly stable”; but the value of passive earth pressure presents a tendency of “increasing, decreasing, increasing and stable”. The phenomenon is due to the change of matrix suction during rainfall. The distribution and change rule of earth pressure calculated by the proposed formula can be applied to design of retaining engineering structure.

The elastoplastic seismic response analysis of earth dam on deep sandy alluviums is an unresolved issue in the seismic study of earth dam due to the complexity of modelling the stress-strain characteristics of sandy soils and the instability of numerical computation. Elastoplastic seismic response analysis of an earth dam on deep sandy alluviums during M6.7 earthquake is carried out by using the u-p fully coupled finite element method for the saturated porous media and the multi-mechanism elastoplastic constitutive model for sandy soils. The dynamic behaviors of dam and foundation and the variations of the excess pore water pressure in generation, diffusion and dissipation are investigated. The results show that there are some differences between the calculated and measured values in accelerations and permanent deformations; but the calculated ones basically reflect their actual distributions; and then the certain accuracy of the constitutive model and numerical method is validated. Based on the numerical analysis results, it can be concluded that the reinforcement measure for the dam and its underlying foundation is not necessary, because the excess pore water pressures in these zones and the permanent deformation of the dam are small. However, an anti-liquefaction reinforcement measure should be taken since the excess pore water pressure is great and the liquefaction may occur in the shallow layer of sandy alluviums near the dam toe.

Multistage and combination retaining structures are employed widely in high slopes; but the seismic earth pressure calculation method of this kind of retaining structure is seldom studied. According to the pseudo-static method and the plastic limit analysis theory, the upper bound solutions are derived to calculate the seismic active earth pressure and its coefficient of three-stage retaining structures including gravity retaining wall and two-stage anchored retaining structures based on strength reduction technique. The following factors including horizontal and vertical seismic coefficients, the angle of the retaining wall back, slope surface shape and its multistage retaining structure, soil cohesion, adhesive strength between soil and retaining wall back are taken into account. The case study of two-step anchored retaining structures indicates that static active earth pressure agrees well with the available research; and the effects of the factors such as strength reduction coefficient and axial force of upper anchored retaining structure are significant on seismic active earth pressure of the lower. Sensibility analysis of seismic active earth pressure of gravity retaining wall combined with anchored retaining structure indicates that the sensibilities of horizontal seismic coefficient, height and angle of gravity retaining wall are significant; and the sensibilities of axial force and angle of upper anchored retaining structure are relative small.

The near-fault ground motion has long been ignored in the seismic design phase of the underground caverns. Hence this issue needs to be seriously addressed. The differences between near-fault and far-field ground motions are firstly discussed, along with the impact of the respective ground motion on underground caverns. Then a modified simulation method of near-fault pulse-type ground motion is proposed. And with the simulated ground motion, the seismic stability of underground caverns for a hydropower station is studied. The results indicate that compared with the conventional far-field ground motion, the near-fault ground motion is characterized by greater Vmax /Amax and Dmax /Amax values (where Vmax, Dmax, Amax are peak values of velocity, displacement and acceleration). And energy of near-fault ground motion is concentrated in 0~1 Hz band. The damage caused by near-fault is far more serious than that of by far-field ground motion, even with same amplitude and spectrum. The proposed simulation method is well logical for considering the information carried in the [1/Tp, 1] Hz. With current calculation conditions, the underground powerhouse is in a potential danger of failure under the near-fault ground motion. Thus further specific risk assessment for near-fault ground motions is expected to check if specialized seismic reinforcement measure is required.

Because of better performances both in pullout capacity and deepwater installation, drag anchors are increasingly applied and play a key role in deepwater mooring systems. During the anchor installation, complicated interaction happens between the anchor and installation line. The penetration and motion of drag anchors are directly affected by the reverse catenary shape of the embedded installation line. Investigating the reverse catenary shape is important for improving the drag embedment performance, precisely predicting the anchor trajectory, and solving the positioning problem of the anchor. In the present work, the reverse catenary equations which can be used to simulate the reverse catenary shape of the embedded installation line both for cohesive and noncohesive soils are derived based on the mechanical model for the embedded line. A technique for real-time measuring the reverse catenary shape based on the tilt transducers and the circular recursive algorithm is also developed. Three typical line profiles are selected to verify the rationality of the experimental technique in the air. The results show high accuracy between the real line profiles and the measured curves by using the technique. The accuracy of the developed equations for simulating the reverse catenary shape of the embedded line is verified by model experiments using the real-time measurement technique. Besides, the model experiments help us to understand the variation of the reverse catenary shape of the embedded anchor line during installation.

Based on the parallel computing cluster platform of ABAQUS, according to the typical stratigraphic profile of downtown area in Suzhou city, a two-dimensional nonlinear analytical model of large-scale deep soft site was established. Comparison of nonlinear seismic effects under artificial seismic wave and far-field ground motion of large earthquake were conducted. The results indicate that: (1) The amplification effect of peak ground acceleration (PGA) under far-field ground motion of large earthquake is especially significant compared with that under artificial seismic wave. There is obvious variability among the PGA of ground surface in different locations due to the horizontal heterogeneity and nonlinearity of ground soils. (2) The deep soft site has a significant filtering effect on the high-frequency ground motion wave whose period is less than 0.3 s. When it is under far-field ground motion, there exists an obvious amplification effect on the long-period seismic wave whose period ranging from 0.85 s to 1.65 s. However, there is also apparent filtering effect on the long-period seismic wave whose period ranging from 2.5 s to 7.0 s. (3) It presents fluctuation phenomenon of PGA distribution with depth and along the lateral direction. There are prominent local focusing effects and filtering reduction phenomenon near the contact surface between different soil layers and in the horizontal heterogeneous soil area; and the PGA amplification effect of soils above 20 m depth is relatively larger than that of other soil layers. (4) Both the spectral characteristics of seismic wave and horizontal heterogeneity of ground soils have obvious influence on the shape of ? spectrum. Besides, three-dimensional curves of spectrum shape those describe the acceleration response spectrum changing with depth are given, which can directly show the soil depth in deep soft site where long and thin underground constructions may occur resonance phenomenon.

Based on the measured rainfall data in the Three Gorges reservoir area, this paper investigates the effect of different initial conditions on different types of soil slope stability. A method to reflect the real initial conditions of the water content in the slope is recommended. Unsaturated seepage analysis is adopted to study the effect of antecedent rainfall on different types of soil slope stability. The typical sand and clay slopes are taken as examples to investigate the effect of antecedent rainfall on slope stability. The results indicate that the effect of the initial conditions on different types of soil slope stability can differ considerably. It is recommended that the steady-state seepage field under average annual rainfall can be used as the initial conditions for transient seepage analysis. The lower soil permeability coefficient, the greater impact of antecedent rainfall can on slope stability, and the longer time affected by antecedent rainfall. It is recommended that more than 15 days antecedent rainfall should be considered for sand and clay slopes stability analysis. Furthermore, more computational effort for sand slopes will depend on the 5 days rainfall before these 15 days rainfall. The short term and high density antecedent rainfall has a significant effect on sand slope stability, whereas the long term and low density antecedent rainfall has a significant effect on clay slope. The cumulative antecedent rainfall can be taken as a criterion to determine the minimum factor of safety. For the sand slopes, the time corresponding to the minimum factor of safety matches well with the time corresponding to the maximum 10-day cumulative antecedent rainfall, whereas, for the clay slopes, it matches well with the time corresponding to the maximum 15-day cumulative antecedent rainfall.

Taking the physical clogging of water source heat pump (WSHP) as research background, based on mass balance equation, a migration equation of particles in porous media and a mathematical model of porosity damage due to particle deposition are established. The proposed model considers the velocity correction parameter, the porosity correction parameter and the probability of capture of particles. The porosity change is dynamic in the process of material migration and plugging by considering the change of porosity. Compared with the traditional filtering model, the proposed model is more systematic and comprehensive with the consideration of the speed reduction factor and flow reduction factor. It provides theoretical basis for solving the physical clogging of reinjection well in laboratory and field tests.

The Jinchuan No. 3 mine in Gansu province of China, is an ultrabasic lean ore and locates at east of fault F17 of Jinchuan mine area. With the improvement of mineral processing and smelting, exploiting of No. 3 mine is on the way. Understanding of in-situ stress state in deep mine can provide the scientific basis to extract and support roadway. In-situ stress measurement was conducted at 1 200 m level to achieve this goal. A improved hollow inclusion stress relief method was employed to obtain three-dimensional stress data of seven measurement sites, which were used to draw section stress-ellipses at different normal directions of each site. The measurement results indicate that at 520 m below the surface, the values of maximum principal stress are about 17-21 MPa, a medium stress level. The directions of the maximum principal stress are within NNW-NNE and NE-WE, and agree with regional stress field basically, and also show a new trend direction. It is probably two faults in No.3 mine area that pose an influence on the direction of the principal stress. However, some dips of maximum principal stress exceed 20° and the largest one is -49°, which reveals a change of horizontal-stress-orient stress field in Jinchuan mine. The result is of great significance in mine design and construction.

A high steep rock hill with two-side slopes near National Road 213 is used as a prototype. Combining with the results of shaking table test, the landslide mechanisms of the high steep rock hill with two-side slopes are researched by a new continuum-discrete element method (CDEM). The research results show: The stress concentration phenomenon appears at the top of the sliding mass firstly; and then some tension failure points and some shear failure points appear there, which expands toward the toe of the sliding mass from the top of that along the structural plane. At the same time, the number of tension failure points gradually increases. At last, the toe of the sliding mass breaks; and then the sliding mass shears out from the toe. The non-consistent movement, the reduce of strength of the structural plane and the difference of distribution and dissipation of the earthquake energy in the sliding body and the sliding bed are three of the major influence factors which induce the landslide. When the accelerations are small, the instantaneous frequency of the accelerations between sliding bed and sliding body are generally consistent; the energy transmittance coefficients of the sliding structural plane and the regulatory frequency band all range in a limitation. With the increase of the seismic intensity, the instantaneous frequency and the energy transmittance coefficients gradually decrease; and then they are steady within the lower limitation. At the same time, the regulatory frequency band also gradually shifts from high frequency band to the lower one.

A new methodology for slope reliability analysis using support vector machine (SVM) is proposed. The presented method fits the actual performance function of slope via SVM, by performing deterministic computations at some sampling points designed with uniform design method for training SVM. Then, the reliability index and the design point are obtained using first-order reliability method (FORM) and iterative algorithm. Based on SVM model, the failure probability of slope is calculated using second-order reliability method (SORM) and Monte Carlo simulation (MCS). The accuracy and efficiency of the method are demonstrated by comparing with other methods for two illustrative examples. The results show that sampling and constructing SVM in U-space and evaluating performance function in X-space make the procedure easy to perform reliability analysis involving correlated abnormal distribution variables and ready to do SORM. Comparisons among different methods for two example slopes show that the proposed method is more accurate than FORM and has higher efficiency than MCS. In the proposed algorithm, computations of factor of safety and reliability analysis are separate, which makes the method adaptive for both simple problems having explicit performance function and complicated applications requiring commercial software to calculate the factor of safety.

To analyze the influence of bending stiffness of reinforcement on the bearing capacity of reinforced soil structures, the beam element form is applied to modeling for reinforcement. Based on the dynamic relaxation (DR) method, the finite element model of beam element simulating reinforcement is presented. Firstly, the beam element stiffness matrix is constructed and the solution to internal force vector is obtained. Then, the fictitious mass density is defined just for numerical purposes and a mass matrix is established. The beam model is embedded in the existing dynamic relaxation calculation program. To validate the performance and accuracy of the beam element model , the loading simulation of simple beam is conducted. Subsequently model, the numerical study for the reinforced sandy ground physical model test is carried out by combining the beam element model and the DR calculation program (including the sand constitutive and the interface models). The comparison of simulation results between the beam element model and bar element model are conducted. The influences of tensile stiffness and bending stiffness on the bearing capacities of reinforced sand foundation are discussed respectively. The simulation results show that the tensile stiffness has little effect on the bearing capacity of the reinforced sand foundation. The influence of bending stiffness on bearing capacity is related to the reinforcement layout; especially when the shear band occurs in the reinforced sand structures, the impact will increase. Thus, to analyze the reinforced mechanism of reinforced sand structures, the beam element is recommended (with bending stiffness) in the simulation procedure.

Meshes are generated without taking account of existence of cracks in the extended finite element method. However, the enriched nodes and their enrichment types are determined by the interaction between the computational mesh and the locations of cracks. In general, the enriched nodes and their enrichment types are determined according to the enriched element types. To determine the enriched element types, the locations of cracks and the shapes of elements must be considered. The limitations judging the enriched element types with the nodal level set values only considering the locations of cracks are pointed out; and the improved method is proposed. For complex cracks, it is inconvenient to determine the enriched element types with the level set method. An effective method is proposed to determine the enriched element types, enriched nodes and their enrichment types, i.e. according to the intersection number between the cracks and element boundaries and the locations of crack inflection points, the enriched element types are determined; then the enriched nodes and their enrichment types are determined by the enriched element types. The detailed procedure of the method is presented. The case studies of numerical simulation show the correctness of the proposed method.

The contour method of tracking the trajectory of rock crack propagation was embedded into extended finite elements (EFE) platform of ABAQUS to simulate the process of initial crack extension under uniaxial pressure. The result testifies the effective of the simulation method. Then, the arc-circular fracture stress strength factor was embedded into the EFE platform. Using the maximum circumferential tensile stress criterion as cracking criterion, when the stress is greater than that rock fracture toughness, the initial crack within surrounding rock begins to extend. Based on this, a numerical simulation of the simplified round tunnel is carried out with the model test which the zonal disintegration phenomenon is observed. During the simulation process, the subdomain precise integration is adopted to improve the precision. The simulation results show that there appears 3 to 4 layer zonal disintegrations surrounding the deep tunnel. The numerical simulation indicates that there exists zonal disintegration phenomenon in deep surrounding rock mass. Comparing the numerical simulation with the ruin pattern of tunnel model after the model test, it is found that the distribution character is in coincidence with each other. The numerical simulation results show that the EFE method is effective in dealing with the problems of complex fracture in rock mass.

In order to research microscopic changes of loess during triaxial shear process, the loess triaxial test model is established with PFC3D (particle flow code in 3 dimensions) software. And the triaxial shear tests are simulated under different confining pressures of 0 kPa, 50 kPa, 150 kPa and 300 kPa. Compared with the results of laboratory triaxial shear tests, the microscopic changes of particles from the beginning then to the peak and at last to the residual deformation of the stress-strain curve are simulated by PFC3D. And the variations of displacement field and contact stress field are more consistent with the macroscopic phenomena of laboratory triaxial tests. The researches indicate that some closely links are existed between the macroscopic parameters (elastic modulus, Poisson's ratio, peak strength) of laboratory model and microscopic parameters (normal stiffness, shear stiffness, friction coefficient) of numerical model. For example, the elastic modulus is controlled by the normal stiffness; the Poisson’s ratio is controlled by the ratio of normal stiffness and shear stiffness; the friction coefficient impacts on the peak strength of stress-strain curve. With the increase of confining pressure, strain energy increases and volume changes in expansion-to-reduction trend. The research result can provide some references for further study of stress-strain behavior and shear strength characteristics of loess.

To compare CLoE and Gudehus-Bauer hypoplastic models, a series of finite element analyses were carried out on modeling the strain localization for Hostun sand under plane strain condition. The strain localization was analyzed mainly in terms of different confining pressures and same initial imperfection. The results are drawn as follows. (1) Compressive hardening can be observed by two hypoplastic models; the stiffness of Hostun sand rises with the increasing of confining pressures. (2) The pattern of strain localization from the results of CLoE hypoplastic model is more compatible with the results of experiments than Gudehus-Bauer model. (3) CLoE hypoplastic model can simulate volume expansion before shrink with the increase of load of Hostun sand. (4) The softening behavior under high confining pressures of the elements in strain localization is more obvious in the strain-stress curve based on CLoE hypoplastic model. (5) CLoE hypoplastic model is better than Gudehus-Bauer model in simulating the nonuniform strain localization which is caused by initial imperfection. All in all, numerical results suggest that CLoE model performs better than Gudehus-Bauer model in modeling strain localization. What’s more, influences of confining pressure and initial imperfection can be modeled by CLoE hypoplastic model. However, CLoE model should be improved because of void ratio and intrinsic length scales unconsidered.

Strength reduction technique by finite element method is an effective method for slope stability evaluation; and it has been used more and more widely. There are many factors influence the result of this method; the order of finite element is an important one of them. The effect of element order on the strength reduction method is analyzed through three classical examples, i.e. 2D foundation problem, 2D slope stability problem and 3D slope stability problem. The results show that both linear element and quadratic element will approach to the exact solution from the upper side as the number of elements increases; but for linear element is too ‘rigid’; and it will overestimate the safety factor, so it is dangerous for engineering practice. The error with linear element is about eight times larger than that of quadratic element under the same meshes. The shortage of strength reduction method with linear element can be made up by replacing the linear element with quadratic element based on the convergence of the maximum displacement in the system as the stability criteria. Quadratic element is more suitable than linear element for the shear strength reduction finite element method.

To analyze the impact of geometrical particle parameters on macromechanical properties of the soil in particle flow method, the conversion formula of the granule number and modified porosity of numerical sample group in the 2D condition was proposed; and the numerical simulation and the laboratory model tests were carried out. It conducted 6 groups of parameter validation tests and 24 groups of variables cross-comparison biaxial compression tests, and included 4 types of gradations, 3 types of porosities and 2 kinds of contact models. The results show that the effect of porosity is larger than that of gradation on soil mechanical properties. With the increase of porosity, the peak value of soil strength gradually decreases; and the axial strains of corresponding points increase accordingly. Compared with the results of the Hertz-Mindlin (H-M) contact model which is focused on the nonlinear relation between the contact force and relative displacement; the results of the linear contact model are more reliable. Meanwhile the behavior of volumetric strain transited from dilatancy to negative dilatancy. And it is suggested when the porosity increases the H-M contact model reflected the subsequent dilatancy effect of soil after peak strength better.

In order to study the wearing pattern of the ore-pass wall, and ensure the continuously safe and stable production of the ore-pass, it is necessary to monitor the ore-pass wall. The diameter of the ore-pass is large. It is dusty, damp and lightless in the ore-pass, and there are unstable stones on the ore-pass wall. To safely monitor the ore-pass wall, a vertical ore-pass panoramic scanning imaging device is developed. The device realizes full-face high-resolution and seamless scanning of ore-pass wall with 4 cameras for target tracking, Gigabit Ethernet for data transmitting, high illumination of 4 LEDs for lighting, rotary encoder for depth locating, and combination cable and self-lock capstan for lifting. Screening and stitching the scanned images, the full-face images of the ore-pass come out. The device is applied to a monitoring experiment of the ore-pass wall at a mine. The image data of the experiment visually and clearly reflect the damage degree of the ore-pass wall. That is the basis to study the wearing pattern, and to design and fulfill the scheme of maintaining and repairing ore-pass. Experimental result shows that the device is suit for monitoring complex vertical ore-passes wall.

Efforts have been made to estimate permeability coefficient of soil using cone penetration test with pore pressure measurement (CPTU) in recent years. However, there are large deviations for the current approaches. Moreover, the mechanism of permeability coefficient determined by CPTU can not be explained explicitly. In order to directly determine the permeability coefficient of soil by CPTU in the field, two assumptions are proposed: i) the flow surface area for pore water is assumed to be a spherical crown covered the tip of the cone; ii) negative exponent distribution of initial excess pore water pressure during CPTU is assumed. Based on these two assumptions and the previous approaches for evaluating the permeability coefficient of soil using CPTU, the equation to calculate the permeability coefficient of soil is derived. According to the measured data in the field, the comparison is made between the proposed approach and the existing one. The results show that the proposed modified approach extends the range of the application of existing approaches. The permeability coefficient of soil is influenced by the degree of the tip of the cone greatly, which increases with the increasing of the degree of the tip. For a cone penetrometer with 60 degrees accepted as the reference and specified in the international reference test procedure, the permeability coefficient determined by the proposed approach is larger than that of previous approaches and more close to that of lab and field tests.

A dual-mode fiber Bragg grating soil pressure sensor is invented on the basis of dual membrane oil cavity structure. Biasing pressure experiments are done with this kind of soil pressure sensor. The results show that the sensor can effectively distribute the soil pressure between the sensor and its contacting segment evenly. Under soil pressure, the first diaphragm deflection passes the pressure on to the second diaphragm via transferring pressure oil. The transfer lever and the equal strength cantilever are rigidly connected. The central deflection of the second diaphragm pushes the transfer lever timely, which exerts a equal size defection on the free end of the equal strength cantilever. Then soil pressure could be obtained by the wavelength shift of fiber Bragg grating, which is pasted on the surface central axis of equal strength cantilever. The loading experiment indicates that the sensitivity of dual-mode optical fiber Bragg grating soil pressure sensor is 412.7 pm/MPa; the nonlinearity error is 1.97% FS; and the repeatability error is 3.68% FS.