Experimental studies are preformed to investigate the mechanism of 3D crack propagation and coalescence in rock under compression, which is significant to understand the damage mechanism and instability of rock mass. Since internal cracks in rock cannot be directly observed and the CT scanning technique is lack of real-time capability, a type of transparent rock-like material is developed to investigate the mechanisms of 3D cracks propagation and coalescence. This transparent material has similar cracking behaviour as rock material, and thus it is convenient to observe the internal crack initiation and propagation. Uniaxial compression tests are conducted on transparent specimens containing build-in single and double pre-exciting cracks using a RMT-150B multifunction automatically rigid servo testing machine. The patterns of 3D cracks propagation and coalescence of specimens with different bridge angles and crack spacings are obtained by detailed observation. Additionally, the influence of different preexisting crack number and crack spacings on peak strength is examined. At last, a theoretical explanation is presented to explain the crack propagation process. The results show that the propagation and coalescence modes of secondary cracks are controlled by different bridge angles and crack spacings. The secondary cracks included wing crack, anti-wing crack (growing in opposite direction with wing crack), and petal-shaped crack (caused by tensile and shear stress). The coalescence of all kinds of cracks leads to the final failure of specimens. The existence of cracks significantly reduces the compressive strength of specimens, and as the number of crack increases, the peak strength exhibits a trend of decline. Furthermore, the crack spacing also affects the peak strength. This research results provide a valuable guide for analyzing the mechanism of rock damage and instability.

The saturated soil is simulated as a two-phase saturated medium in theory so far. Using (u is the displacement amplitude of corresponding wave, p is the porous pressure) solutions of Biot’s dynamic equation decoupled dilational waves, associated with Somigliana’s representation integration, the numerical implementations of dynamic response for saturated soil with drainage are able to complete. These results can also be confirmed in the fields experiments. This paper describes the dynamic property of the saturated soil with drainage in dynamic triaxial test, and makes the comparison between the numerical implementations of flux drainage from equations and measured results in triaxial test. These analysis results show that the performance on computation by Somigliana’s integration with drainage is true. The analysis of dynamics response with drainage for the saturated soil employed in these achievements can be used in analyzing the soil dynamic issues in the future.

Failure of slope is progressive process. The shear strength parameters c、? have different exerting extents during the destroy process, so the strength reserve is different. Thus, the determining of double safety factors based on the strength reserve and the slope stability evaluation based on double safety factors are worthy of investigation. On the basis of softening regularities of geo-material and assumption of linear distribution of strength parameters, the correlation between the reduction factor of cohesion and the reduction factor of inner friction angle is deduced in this article. Then the correlation is introduced into Bishop’s slice method, the non-proportional double safety factors Bishop’s slice method is developed. The slope stability can be evaluated with comprehensive safety factor based on weight of strength parameters contributing to the resistant shear force. This method not only overcomes the blindness of the existing double safety factors method to determine the double safety factors but also provides a new approach to applying the double safety factors method to engineering practice. Finally, the proposed and traditional methods are both applied in an example and the resemblance of their output validates the suggested method.

In the traditional strength reduction method, only the minimum safety factor and the critical slide surface can be acquired as its reduction region covers the whole slope model. However, in the slope engineering practice, we should treat the slope area not only surrounded by the global critical slide surface, but also located in the vicinity of the secondary slide surfaces. Thus, a modified strength reduction method is proposed to overcome the shortcomings of the traditional strength reduction method in searching the secondary slide surfaces. In general, the yield regions of slope are always distributed on both sides of a slide surface, which presents as a shear zone. Thus, a slide surface and its safety factor can be obtained by reducing the strength parameters of local slope area located within the shear zone. Based on the above understanding, it is supposed that the shear zones are distributed along the logarithmic spiral. Shear zones at different locations and ranges can be gained by changing the form of the logarithmic spirals, and then different secondary slide surfaces with their safety factors can be achieved through the strength reduction method. At last, the proposed method is verified through two case studies, and the effect of the shear zone breadth on calculated results is discussed. The results show that the proposed method can get not only the critical slip surface, but also different secondary slide surfaces corresponding to any safety factor.

As one of the major laboratory test methods, the triaxial test are often performed to gain the strength and stress and strain behaviors of soils. There are two great important factors during consolidation process of the triaxial test, i.e. the dissipation of pore pressure and the completion status of the consolidation, which determine the conduct of the following shear process. Based on equal strain assumptions, an analytical solution is presented to analyse the pore pressure and the consolidation degree throughout the drainage consolidation process of the triaxial test. According to the laboratory triaxial consolidated drained (CD) test, the pore pressure and the consolidation degree of the drainage consolidation process are checked by using the analytical solution. The comparative results show that the two curves calculated by the analytical solution, including the curve of consolidation degree and the curve of pore pressure dissipation, are in good agreement with the corresponding curves gained by laboratory triaxial shear tests, which verifies that the analytical solution can be applied to analyze the drainage consolidation process of triaxial shear test.

The permeability of rock is apparently dependent on the direction or anisotropic, which is resulted from the rock-forming geological processes. At present, theoretical and experimental studies on directional permeability properties of sandstone remain insufficient, and there is lack of quantitative assessment on the anisotropy of permeability. The transient tests are performed on sandstone from CCUS by orthogonally sampling cores. Experiential results indicate that there exists a power function relation between the permeability and the confining pressure. However, as the confining pressure is increased up to 20 MPa, the difference of permeability inorthogonal direction declines gradually, which shows a convert of permeability from anisotropy to isotropy. The pressure sensitivity of the permeability in the vertical coring samples is generally lower than that in the horizontal ones. Furthermore, a theoretical model is deduced in the Cartesian coordinate for the anisotropic permeability, which can be used to explain the vector properties of rock permeability. The parameter of anisotropy ?k is also proposed on the basis of the standard deviation and then is employed to normalize the heterogeneity of permeability in different reservoir sandstones. It shows that the proposed method is quantitative to describe variations of anisotropy in the seepage field system.

Geofabriform method is a new method for constructing fine-grained tailings dam construction. To study the basic mechanical performance of tailings dam geofabriform, the frictional and compressive properties of small-scale tailings dam geofabriform are investigated by means of slope sliding test, direct shear test and unconfined compression test. The tailings dam geofabriform is made of geotextile and tailings sand, which both come from a construction site in Yunnan province. The slope sliding test of geofabriform shows that the coefficient of friction between the geofabriforms is larger when the moisture content of the geofabriform is lower. The internal friction angle and cohesion of the geofabriform consolidated for 8 days are around 28º and 40kPa respectively, based on the direct shear test. Under the unconfined compression condition, with the compression of the geofabriform consolidated for 1 day and 3 days, the tangent elastic modulus increases first, then decrease and finally increase again. Due to the combined effect of geotextile and tailings sand, the ultimate compressive strength of the geofabriform is much larger than that of the pure tailings sand. When compressive stress reaches the ultimate compressive strength of the tailings dam geofabriform, geotextile tearing often occurs at the bottom and close to the edge of the geofabriform, The failure mode is analyzed using the active earth pressure theory.

Model tests are carried out on the large-diameter single pile and the conventional long pile under lateral cyclic loads in clay, which aims to investigate the behavior of pile foundations of offshore wind turbines under the action of cyclic loads from wind and wave. Based on the backbone by API and Masing rules, a cyclic p-y curve is formed, where p is load and y is displacement. The method is presented to calculate the cyclic degradation of a pile subjected to lateral cyclic loading, using a typical clay model in the literature, which modifies the undrained shear strength by accumulated absolute plastic shear strain. By relating cycle numbers and cyclic amplitudes of lateral load to the cyclic degradation of soil around pile, the method can consider various types of lateral cyclic load in offshore projects. The results by model tests and theoretical calculations demonstrate that, under the same cyclic load, the cyclic resistance capacity of a large-diameter single pile with larger stiffness is better than that of a conventional long pile. For the design of a large-diameter single pile foundation of the offshore wind turbine, it is more rational to consider the post cyclic lateral capacity of the foundation with the residual strength of clay.

Rock creep is an important factor affecting the stability of rock engineering. The long-term strength of rock, which shows an obvious size effect due to rock heterogeneity, is an important mechanical index for evaluating the long-term stability of rock engineering. The aim of this study is to investigate the size effect of long-term strength of rock. Firstly, on the basis of classical power law creep model and damage mechanics principle, a nonlinear creep damage model is established for describing the whole process of rock creep. Then, by comparing the numerical results of uniaxial compression creep test based on the proposed model with the results of indoor uniaxial compression creep test, the applicability of the model is verified. At last, numerical simulation tests are performed on seven rock specimens with different sizes under uniaxial compression creep. The size effect for long-term strength of rock is analyzed. The numerical simulations indicate that, the long-term strength of rock decreases with the increasing of size; but it will be constant when the size of the rock exceeded a specific size.

Based on the current solution of the squeezing displacement due to cavity expansion in a semi-infinite soil layer, the horizontal surface and inclined slope displacement boundaries have been revised. The current analysis method is improved via the methods of coordinate transformation and superposition. Hereby a relatively simple solution to the soil compaction displacement field due to spherical cavity expansion is derived under nonaxisymmetric displacement boundaries, and the effects of the slope angle and the distance between the spherical cavity and the free boundary are analyzed. The results indicate that the inclination boundary condition has a significant influence on the squeezing displacement of both sides of the spherical cavity. With the increase of inclination, the influence of boundary inclination angle on the soil compaction displacement become more and more significant, and the larger the distance from spherical cavity position to the free boundary, the less significant effect on the soil squeezing displacement. This research may provide practical guidance for the design and construction of static piling on the adjacent slope and the issues related to cavity expansion.

The geo-encased stone column is a new type of column which is wrapped with geosynthetics. Because the stone colume is encased by geosynthetic materials, its stiffness and shear strength is effectively improved. Thus far, however, limited research efforts have been devoted to the geo-encased stone column, especially in terms of aseismic behavior. In this study, shaking table tests on the composite foundations of geo-encased stone columns and stone columns were carried out to investigate the dynamic responses and aseismic behaviors of the two types of foundations. The acceleration, column-soil stress ratio, displacement of two composite foundations are monitored, and the failure phenomena were observed under the action of seismic waves of different types and amplitudes during the test. It is shown that the acceleration amplification factors in the horizontal direction at the top of the geo-encased stone columns and the soils between these columns are about two times that of the stone columns, and for the same saismic wave, the peak column-soil stress ratio of geo-encased columns is about three times that of stone columns. Compared with the stone column composite foundation, the geo-encased stone columns foundation forms narrower cracks in a larger area with the excitation of a 0.9g artificial seismic wave. The geogrid decreases the final settlement of the stone columns by 51%. It can be concluded that the geo-encased stone columns performs better than the stone columns during shaking.

To obtain the static undrained shear strength of the saturated marine soft clay in the Pearl River Estuary in Guangdong, a series of tests are conducted, i.e. the unconsolidated undrained (UU) shear tests, the consolidated undrained (CU) shear tests, the unconfined compression tests, the static T-bar penetration tests in the centrifuge model foundation of normally consolidated soft clay, the field vane shear tests, and the cone penetration tests (CPTs); empirical formulations of soil mechanics are also employed. Based on these tests and calculations by the formulations, the distribution patterns of the undrained shear strength of the marine soft along with the depth clay are determined and the validity of different methods mentioned before are evaluated comprehensively. In order to acquire the cyclic degradation behaviors of the marine soft clay in this area, cyclic triaxial tests, cyclic T-bar penetration tests in the centrifuge model foundation of normally consolidated soft clay, and cyclically horizontal loading tests on a single pile in the field are carried out respectively. It is found that there is a double logarithm linear relationship between the cyclic secant modulus ratios of the soft clay and the number of cycles, from which three types of degradation factors for three testing condtions are achieved. The results can provide a guide for the offshore structure design in this area, and can also be used for reference in other engineering projects.

Using the self-developed digital image correlation method based on the particle swarm optimization algorithm, the spatiotemporal distributions of rotation angles of principal strain axes for wet sandy soil specimens are investigated under uniaxial compression. Using the bicubic spline interpolation, the rotation angles of principal strain axes at any positions are obtained to study the evolution of the rotation angles of principal strain axes with the longitudinal strains before the occurrence of a shear band, at its center, in the vicinity and at both sides of the band. It is found that with the increase of longitudinal strain, the rotation angles of principal strain axes change from a scattered distribution to a stable one; and at last, its range in most regions is -10°-10°. After the occurrence of an apparent shear band at the late stage of strain-hardening, at the center of the band, the rotation angles of principal strain axes tend to be stable or descend slightly, within a range of -5°-5°, while those at both sides of the band still increase. At tip of the band, the evolution of the rotation angles of principal strain axes is complex.

To investigate the deformation and failure characteristics of the anchor and soft surrounding rock in the soaked tunnel, two scale model tests are carried out based on the building Yangtze River bridge project at Jijiang. These two model tests are soaked state tunnel anchor model M2 and natural state tunnel anchor model M3. The reduced scale ratio is 1:30. By analysing the data from key measuring points, the deformation characteristics and rules of the tunnel anchor and its soft surrounding rock at different states are achieved. The results show that either on the surface or in the interior of the tunnel anchor, the measuring displacement of M2 is greater than that of M3, but the failure load value of M2 is lower than that of M3. There are various failure modes caused by different moisture contents. For model M2 with a moisture content of 7.39%, the cracks appear firstly at the surface of tunnel anchor hole and the west tunnel anchor, following at the ground surface and the east tunnel anchor. For model M3 with a moisture content of 5.36%, the cracks are initially observed at the ground surface of the upper part of the tunnel anchor, then at the side face of the tunnel anchorage hole and finally in other parts. Furthermore, the terrain also results in the deformation difference of tunnel anchorage with the same water content.

Heterogeneous discontinuities with different properties are widely distributed in rock masses. The unconformity is usually beyond consideration; however the deformation and strength characteristics of heterogeneous discontinuities yield the basis for the stability evaluation of rock mass. In this study, samples from the typical sandstone-mudstone interbeds in the Jurassic in Three Gorges Reservoir Region are tested. The joint roughness coefficient (JRC) values of flat discontinuity and four different irregularly shaped discontinuities are calculated in a quantitative approach using the fractal geometry theory. Then the numerical shear tests on these five kinds of samples are carried out by the PFC2D software. The shear stress-displacement curves are acquired for heterogeneous discontinuities with different roughnesses under various normal stresses. According to numerical results, the strength characteristics of heterogeneous discontinuities are analyzed using Barton’s JRC-JCS model and compared with the homogeneous discontinuities. At last, on the basis of shear failure mechanisms of heterogeneous discontinuities, two types of modified Barton criteria (i.e., type I and type II) are proposed by introducing a concept of strength factor, which are applicable to strength evaluation of heterogeneous discontinuities. The results indicate that shear strength of heterogeneous discontinuity falls between those of two homogeneous rock discontinuities with the same joint roughness, and is close to strength of soft rock discontinuity at low normal stress, which follows the modified type I Barton criterion. At high normal stress, the strength of heterogeneous discontinuity is close to that of the homogeneous discontinuities of hard rock, which follows the modified type II Barton criterion. In practical project, the evaluation of rock mass stability can employ the modified criteria in combination with the stress state of heterogeneous discontinuities, which improves on some of the shortcomings of previous studies.

To investigate the relationship between the permeability and temperature of tectonic coal containing gas, a series of tests on tectonic coal samples containing gas is carried out at different stress conditions using a self-developed triaxial hot-fluid-solid coupling seepage testing device. Experimental results show that the permeability of tectonic coal samples decreases with the increase of temperature, and there is a negative exponential function relation between permeability and temperature. Moreover, the relationship between the permeability loss rate and the effective stress follows the Boltzmann distribution with the variation of experimental temperature. There exists an effective stress threshold of approximately 4.515 MPa for the permeability loss rate. There are two stages identified in permeability change of tectonic coal, i.e. the accelerating stage and the smooth stage, and the transition temperature from the accelerating phase to smooth phase is about 45 ℃. Within the range of 21-80 ℃, the sensitivity order of the permeability coefficient is 10-2, which means the effect of temperature on permeability of tectonic coal sample is not obvious. At the same time, the increase of effective stress results in the decrease of the temperature sensitivity coefficient.

Along with the acceleration of urbanization process, a large number of engineering landslides are induced by artificial slope cutting in loess region. Feifengshan landslide has been selected as a typical case in order to study the mechanism of typical engineering landslide. Detailed field investigation has been done, meanwhile, in-situ water penetration test and soil water characteristic curve (SWCC) test on undisturbed loess and paleosol have been conducted in this study. The experimental results show that: The permeability coefficient of loess can reach medium grade about 10-4 cm/s, if loess has structural problems, such as small holes, cracks, and so on. The air entry value of paleosol is 11.5 kPa obtained by SWCC tests , which is higher than that of loess (9.0 kPa); at the same time, curve also indicates that the evolutions of strength and deformation of paleosol and loess will differentiate dry-wet cycle. With the increase of suction, the growth rate of shear strength of paleosol is lower than that of loess during drying phase, and the elastic volume shrinkage rate of paleosol, caused by compressing, is also lower than that of loess in the same process. Paleosol will have bigger expansion deformation than loess during wetting phase, because the clay content of paleosol is higher. On the basis of experimental results analysis and field investigation, it is founded that artificial slope cutting is the most important factor inducing Feifengshan landslide, which triggers a series of physical and chemical chain reactions, especially reinforces coupling effect between soil and water and accelerates the occurrence of landslide. Finally, we put forward the failure mode of Feifengshan landslide characterized by two-way development of sliding surface and typical progressive destruction, simultaneously having part of geological features of slumping.

In applying Asaoka’s method to predict the post-construction settlement of a road foundation in Shenzhen, it was found that the predicted results can be different by three times, depending upon the chosen initial time and time spacing. In this paper, the mathematical formulation of Asaoka’s method is analyzed and the effect of the error of fitting parameters on the predicted results is discussed, from which the concept of the error amplification coefficient of fitting parameters is defined. The reliability of predicted results by Asaoka’s method is evaluated. The results show that one of the major reasons that affect the predicted results of Asaoka’s method is that the expression of settlement prediction is very sensitive to the fitting parameter ?1, and the sensitivity increases with the increase of ?1. Hence, without scarifying the accuracy of the linear fitting method, the time spacing is increased, and ?1 can be reduced, so that the influence of the fitting parameter error on the predicted results can be effectively reduced. In predicting the foundation settlement based on Asaoka’s method, the reliability of prediction is not only dependent upon the accuracy of linear fitting , but also on the error amplification coefficient of fitting parameters.

According to various excavation depth and the formation conditions, a database of 80 engineering case histories on the urban rail transit foundation pit in Beijing is analyzed, and the ground deformation due to the foundation pit excavation in Beijing sand & gravel zone and clay soil zone is studied. The results show that 1) The distribution of the maximum ground displacement approximately follows a normal distribution or half normal distribution form. 2) The average value of maximum ground settlement is 0.11%H in sand & gravel zone and 0.20%H in clay soil zone, where H is the excavation depth. 3) The ground deformation is related to the pile horizontal displacement on or out of the pit. 4) When the insertion ratio increases, the ground settlement decreases. The system stiffness exerts significant influence on the ground settlement in clay soil zone. 5) The main influence zone around foundation pit is 0.6H or 0.7H. The results can be used to calculate the lateral deformation of similar projects in Beijing and other areas, helping guide the related engineering design and construction.

The soil strength will decrease to a certain extent due to the unloading effect induced by deep foundation pit excavation. The decrease of soil strength will then change the lateral earth pressure exerted on the supporting structure system. So the actual lateral pressure will underestimated using soil strength parameters prior to unloading, which will result in a difference between the calculated deformation and the measured results of the supporting system. It is not sufficient to just monitor the deformation of support system during deep foundation pit excavation. In order to precisely predict the results, it is critical to understand the pattern of support system deformation under unloading effect. To investigate the influence of unloading effect on the soil strength and soil deformation on the bottom and sides of the excavated area, a case study is conducted on an excavation site of Xiangluowan Project of Fuli in Tianjin by monitoring the deformation of bottom rebound, supporting system and surrounding soil deformation during the entire excavation progress. Reduced soil strength of Xiangluowan project was calculated from measured results, which is used as the basic perimeters for the numerical simulation in later phase of research. ABAQUS, a commercial FEM software, is used to established a 3D model of the foundation pit excavation. By the model, the deformations of bottom rebound, supporting system and surrounding soil are simulated using in-situ soil strength and reduced soil strength as perimeters respectively. The deformation patterns of bottom rebound, supporting system and surrounding soil under two different conditions are studied; then the results from the finite element model analysis is compared with the actual in-situ monitored results. It is found that the disturbed zone imposed by unloading effect is located with 3-4 meter below the bottom of foundation, the percentage of strength reduction of the disturbed zone induced by unloading is up to 20%-35%, the percentage of strength reduction of the surrounding soil is 10%-15%, the results derived from FEM analysis is proved to be consistent with the field test results. It shows, through this paper, that it is practical and reasonable to consider the unloading effect during deep foundation pit excavation, thus the same method and philosophy is suggested to be applied in future deep foundation pit design and engineering.

During the installation of horizontal jet grout columns, the high pressurized fluids injected into the soil can induce the expansion of the foundation soils, causing significant movements of the soils around jet grouting locations. The soil deformation caused by installation of horizontal jet grout columns can be considered as the combined effects of the expansions induced by the high pressure and the volume of fluid. Based on a case history of the construction of horizontal jet grout column, a finite element model is developed for analysing the soil deformation caused by installation of horizontal jet grout columns. In this model, the soil deformation caused by installation of horizontal jet grout columns is simplified as the expansion of a round hole, the jetting pressure and the volume of injected fluids can be considered as a whole. A method is proposed to estimate the influencing radius of the jetting pressure, and the equations for determining the ratio of volume expansion are provided. Based on these, the soil deformation can be obtained by the finite element analysis. Comparisons of the measured values and numerical results indicate that the numerical results are in good agreement with the measured data when the influencing radius is 6 times that of the column, demonstrating the applicability of the developed finite element model.

In this study, the movement and failure modes of overlying strata at Haishiwan mine are analyzed for different mining modes at mountainous field by using a numerical approach named continuum-based discrete element method (CDEM). In combination with the results of physical modeling experiment and the surface displacement data from field monitoring, the movement and failure modes of overlying strata caused by the ‘push mode’ along slope and the ‘traction’ mode along inverse slope are obtained. Besides, the upheaval phenomenon of slope basis induced by cantilever structure effect is confirmed through numerical analysis and in-situ observation. The study shows that under the same geological conditions, tension and shear failures are prone to appear along the slope crest, but shear creep deformation is prone to arise at the slope foot. As a result, mining along a slope has a greater impact on the slope stability than that along inverse slope. Therefore, an industrial ground is recommended to be built on the low inverse slope with some degrees of intact strata.

The stability analysis of slope has been a classic research field in geomechanics and geotechnical engineering. Numerous research results and analytical methods have been proposed in this regard. To overcome the defects of imbalance thrust force method adopted in the practical specification, both two-dimensional and three-dimensional formulations of strength margin and overload margin downslide thrust are developed, based on the principle of the vector sum analytical method of slope stability. Two 2D numerical examples and a 3D numerical example are investigated. The results of 2D numerical examples indicate that the variation of obliquity of sliding surface has less significant influence on the direction of vector downslide thrust than on the direction of residual downslide thrust. The vector downslide thrust curve is smoother than the residual downslide thrust curve, which helps avoid the problem that the precision of the residual downslide thrust is significantly influenced by the variation of obliquity of the polyline-shaped sliding surface. The results of vector downslide thrust and residual downslide thrust are the same when the sliding surface is a single polyline, and the result of 3D numerical example by the proposed method is consistent with the common understanding of the slope stability. The proposed procedures are used to calculate the downslide thrusts of a landslide in the western China under three working conditions, including the gravity condition, rainfall condition and seismic load condition. The results show that the vector downslide thrust result has been better improved compared to the residual downslide thrust. Finally, the potential improvements in the vector downslide thrust method are discussed.

Targeted research is performed on several problems about the application of the soil nailing wall combined with row piles in tunnel deep foundation pit support. Based on the theoretical and numerical analysis of engineering examples, the selection principle are proposed for the slope ratio i of soil nailing wall, the diameter-separation distance of the row pile as well as the relative stiffness between row pile and inner support while the inner support is applied. It is shown that 1) the smaller the value of i , the more stable the soil nailing wall, but the larger the moment and the displacement on the row pile are. The value of i should take a small value when the sliding stability of the soil nailing wall is fulfilled. 2) The larger the diameter and space between the piles, the smaller the horizontal displacement of the row pile, so that less concrete and steel rebar is required, resulting in a more cost-effective design, though the protection for the soil between piles will be more difficult. Hence, the diameter and the space between the pile should assume larger values when the engineering geological conditions is good and the water level is low, and vice versa. 3) Supporting effect of weak pile with strong support is better, but the construction interference for the surroundings is larger, while the cost of the strong pile (non-cantilever) with weak support is less. To achieve the purpose of safety and reliability and economical efficiency, it is necessary to consider engineering geological conditions, surrounding environment requirements, construction period, cost, and so on, in selecting the relative stiffness of the pile need.

Seismic dynamic stability analysis of arch dam abutment has been a hot topic in geotechnical engineering. However, studies are not yet mature on the evaluation methods and criteria of dam abutment dynamic stability. The advantage of the finite element method and limit equilibrium method are combined to improve the abutment dynamic stability analysis method. Then the seismic dynamic overload method and limit cumulative displacement criterion are proposed. Firstly, a real-time dynamic analysis is conducted based on the finite element method. Secondly, multi-grid method is adopted to calculate the resistance force and sliding force of abutment potential slip blocks through stress interpolation, it can determine the main sliding direction and calculate the dynamic safety factor using the limit equilibrium method for each slider. Meanwhile, there is an acceleration integral to get the cumulative displacement of slip blocks for the period within which the safety factor is less then 1.0. According to the slider size and the sliding surface strength parameters, the limit cumulative displacement formulation is proposed and acted as the instability criterion of slip blocks; by comparing the cumulative displacement and the limit cumulative displacement, the stability of the slider is determined. An abutment dynamic stability analysis system SAFEDAM is established from the above ideas. To verify the rationality of the limit cumulative displacement criterion, Shapai arch dam is taken as example which has underwent the Wenchuan 5.12 earthquake in Sichuan province. The results show that for the left bank abutment of Shapai arch dam, the overload stability safety factor Kp0 is 3.2, while for the right bank the Kp0 is 4.2. The example study confirms that the proposed method provides a new path for the dynamic stability of arch dam abutment and evaluation. For the reliability and flexibility to assess the dynamic stability analysis, the method shows its wide applicability in engineering practice.

Based on the results of the centrifuge modeling, the possible effect of spudcan penetration on an adjacent pile is analyzed based on the coupled Eulerian- Lagrangian (CEL) FEM method and the nonlinear foundation beam FEM method. In the CEL FEM method, the strata with large deformation are set as the Eulerian domain, and the deformation responses of the domain are calculated by an Eulerian finite element algorithm. Eulerian meshes are fixed in the space and materials can move through meshes during calculations. Other strata are set as the Lagrangian domain and deformation responses of the domain are calculated by the Lagrangian finite element algorithm during calculations. Material movements are consistent with deformations of meshes for the Lagrangian finite element algorithm. Eulerian and Lagrangian domains are coupled by the penalty function method. Lateral displacements along the adjacent pile due to the spudcan penetration are determined by CEL finite element analyses. Bending moments along the pile are further determined by nonlinear foundations beam finite element calculations. It is shown that the variations of lateral displacements and bending moments along the pile are consistent with the results of centrifuge model tests, showing that effects of the spudcan penetration on an adjacent pile can be analyzed by using the CEL finite element method and the nonlinear foundation beam finite element method. It is also shown that the Eulerian stratum depth has a significant effect on the CEL calculation results. If the spudcan penetration depth is less than 0.75 times the diameter of the spudcan, the Eulerian stratum depth is set as 1 to 1.25 times the diameter of the spudcan. If spudcan penetration depth is larger than 0.75 times the diameter of the spudcan, the Eulerian stratum depth should be set as 0.5 times the diameter of the spudcan below the spudcan penetration depth.

In contrast with the level ground, which has a single symmetrical failure mode, foundation soil near the slope presents a variety of failure modes under different working conditions. Based on the upper bound finite element method, a nonlinear upper programming model is developed, in which a feasible arc interior point algorithm is used for solution. The calculation results are converted into the corresponding velocity field and the energy dissipation, and compared with the numerical results in the literature. It is shown that the upper bound finite element method can be used to effectively analyze the slope failure mode and obtain a reasonable solution. By converting the bearing capacity of slope into Terzaghi partial coefficient, and based on the variations of Nc、N? 、Nq, it is found that the variations of Nc and Nq are consistent with the existing monotonous change rule. As for N? , because the volume of the failure body changes due to the change of failure mode, a non-monotonous change rule of N? is observed. By defining the safe distance as the distance a0 /B completely unaffected by the slope, part of the calculation chart is developed, providing a useful reference for slope foundation design.

The saturated hydraulic conductivity is inherently variable. An extension infiltration model considering the variability of saturated hydraulic conductivity is established using the classic Green-Ampt model; and then the corresponding depth of the wetting front and the distribution of water content are determined. Meanwhile, a closed form of the limit state function is presented, based on the combination of the extension of the Green-Ampt model and the infinite slope stability model. Random number sequences of saturated hydraulic conductivity are generated following a lognormal distribution using the Monte Carlo simulation method. For a hypothetical slope that is subjected to steady-state rainfall infiltration, a series of parameter analyses is conducted. The results show that the cases with a smaller coefficient of variability have a smaller failure probability in the initial stage of rainfall infiltration. However, as rainfall progresses, the cases with a smaller coefficient of variability have a larger probability of failure instead. The most likely failure time of the slope is not affected by the coefficient of variability, but depends on the intensity of rainfall. The corresponding probability of the most likely failure time decreases with the increase of the variability.

The upper bound limit analysis method has a more rigorous theoretical basis and clearer physics meaning compared to the limit equilibrium method. Based on the series expansion, the velocity field of a triangle element is expanded through center point velocity and its velocity gradients. Hence the upper bound FEM method based on the center point velocity and its velocity gradients are introduced as unknowns. The method not only enriches the upper bound limit analysis method, but also has a simpler flow equation. The proposed method can be considered as the multidimensional extension of rigid FE upper bound limit method, and has a more rigorous theoretical basis than rigid FE upper bound limit method. The requirement that all the point must strictly meet the limit properties is relaxed and the equal-area polygon is adopted to approximate the yield circle. Two case studies show that this method can steadily converge into the true solution, and has the same convergence as the traditional method by Sloan; when equal-area polygon is taken, the method yields a good result by less number of polygon edges, with significantly improved convergence.

This paper presents results of three-dimensional simulations of the hollow cylinder test using the discrete element method (DEM). To verify the capability of numerical model, the hollow cylinder specimens are sheared at different principal stress directions. The localization in the specimen are examined in terms of the distributions of stresses and strains. The force chain formation and collapse during the loading are visualized. Meanwhile, the formation of the shear bands in the different shearing directions is characterized by porosity and shear strain rate distributions in the samples. The results show that the shear strain rate contour is a better indicator for shear band development than the porosity contours. To explore the micro behavior of the specimen, the evolutions of coordination number and particle rotation at different locations in the sample are also monitored. It is proved that 3D DEM technique can capture the macro-micro behavior of specimen in the complex stress path, which facilitate the interpretation of the stress-strain behavior from physical lab tests.

Simulating large-scale engineering problems with discontinuous deformation analysis (DDA) is extremely time-consuming. The solving process of linear equations normally costs more than 70% of the total computing time, and thus the computing efficiency of algorithms for linear equations is a significant research topic. Firstly, two contents of non-zero storage in the DDA have been described. One is the block compressed sparse row method, and the other is the iterative scheme of non-zero position recording based on the trial-error approach. Secondly, in view of the sub-matrix technology, the block Jacobi (BJ) iteration method and pre-processing block conjugate gradient (PCG, including Jacobi and symmetric successive over relaxation(SSOR)pre-processing) iteration method have been introduced into DDA, and then the key operations of solving linear equations have been analysed. Last, the calculation efficiency of various algorithms for solving linear equations are investigated through two examples of tunnelling excavation. The results show that the direct solution cannot meet the requirements of large-scale engineering computing compared with the iterative method. Although there are few differences of computing efficiency between BJ and BSOR iteration methods, both of them are obviously not as well as the PCG method. Therefore, the PCG method, in particular SSOR-PCG method is highly recommended. Jacobi-PCG is the best method to perform parallel computing, however BJ iteration is also an acceptable choice when there is an apparent inertial advantage of the stiffness matrix.

Rainfall is one of the most critical factors that results in the landslide instability. However, the influence of runoff from bedrock slope on infiltration has not been considered by the existing numerical methods so that unfavorable infiltration is underestimated and the stability of landslide, especially those deep slides, is overestimated. In this paper, based on Richards’s equation and FEM, a 2D simplified numerical method is proposed, which can consider the influence of runoff supply on the infiltration of sliding mass. In the proposed method, calculation domain only contains sliding mass in order to avoid the numerical problems induced by the large difference between the permeability coefficients of landslide body and landslide bed (or slide zone), and rainfall boundaries are corrected according to the horizontal length of bedrock slope and saturation of slope surface. The seepage field and stability of an existing landslide in the Three Gorges Reservoir Area are simulated under the water level fluctuation of reservoir and the rainfall from Oct., 2006 to Dec., 2009. It is shown that the saturated zone in the trailing edge of landslide becomes significantly enlarged when the runoff supply is considered, and thus the stability decreases more significantly, which is consistent with the data from the displacement monitoring. In addition, rainfall trivially influences the stability of landslide if the runoff supply is not taken into account.

Unbounded domain problems are frequently encountered in geotechnical engineering and infinite elements are often used effectively for simulation. The weak form quadrature element method is an effective numerical tool in which the computational accuracy is often improved through increasing the order of integration. An infinite weak form quadrature element is developed and applied to the analysis of unbounded domain problems in geotechnical engineering. Based on coordinate transformation, an unbounded domain is mapped onto a standard region where numerical integration and numerical differentiation are conducted and the conventional numerical integration points and weights in the weak form quadrature element method are retained. Numerical examples in the areas of transient seepage, consolidation and elastostatic analysis are given and the results are compared with analytical solutions or those of other numerical methods. It is shown that the infinite weak form quadrature element is simple and applicable to solution of various unbounded domain problems, while conventional elements are used during discretization of the domain of interest. In addition, the dependence on the pole of coordinate transformation can be alleviated considerably with the increase of the integration order of the element. Consequently, computational resources are reduced significantly and accuracy of results is enhanced.

An indirect boundary integration equation method (IBIEM) of high precision is used to analyze the vibration isolation effect of the two-dimensional discontinuous pile-group barriers on plane P and SV waves within a broad band of frequency. This method is based on the single layer potential theory, and the scattered field in vivo and vitro of the scatterer is constructed using the fictitious wave sources close to the surfaces of discontinuous barriers. The magnitude of the fictitious wave sources are determined by boundary conditions, while the total response can be obtained by the superposition of free field and scattered field. It is shown that the proposed method has excellent numerical accuracy in solving broadband isolation vibration problems related to arbitrary arranged piles. For the elastic cylindrical solid pile, the frequency spectrum is quantitatively analyzed, showing that different vibration isolation effects of elastic rowed piles are associated with different ranges of frequency with the incidence of P and SV waves. The vibration isolation effect for different wave velocity ratios, different distances of piles and different rows of piles are discussed, and it is shown that: 1) there exist optimal dimensionless frequencies for prominent vibration isolation effect; and thus it is necessary to carry out optimization design of pile diameter and pile spacing; 2) for softer soil, the isolation effect becomes better, and compared to P wave, SV waves is more sensitive to the pile spacing; 3) Multiple rows of piles should be adopted for low frequency waves, but for the high frequency wave, more than three rows of piles will not lead to significant improvement on the vibration isolation effect.

Data preprocessing completed in a data management system plays an important role in structural health monitoring. The Romanowski guideline (t-test) is selected to eliminate the gross error through comparison. The wavelet denoising method can effectively remove the noise (or error) caused by environmental factors, and yield more reliable monitoring data. However, a consensus has not been achieved on how to select the threshold function, the threshold and the number of decomposition layers. Using the 39 groups of monitoring data obtained in the health monitoring of the Yangtze River tunnel, we analyze the effects of various combinations of thresholds and threshold functions on the results of the wavelet denoising method. It is shown that the denoising effect via 4-5 decomposition layers is the best using the combination of Rigrsure threshold and hard threshold function. A satisfactory result is achieved when the above research are applied to denoise the corresponding period monitoring data. Based on these results, it is shown that the denoised data can effectively prevent the generation of false alarm in the early warning compared to the original data. Finally, a method based on the combination of the wavelet denoising and least squares method is proposed to predict the monitoring data.

A series of uniaxial compression tests is carried out on the rock-like brittle specimens with pre-existing three-dimensional (3D) cracks. Acoustic emission (AE) monitoring technique and computerized tomography (CT) scan are used to investigate characteristics of crack propagation. It is found that the uniaxial compressive strength, crack initiation stress and displacement decrease with the increase of the crack depth in the specimens. The propagation of shear cracks and secondary cracks is the key factor to control the failure of specimens with a deep crack. The main failure mode is shear failure accompanied with transverse fractures. The initiation of Mode-I can leads to occurrence of turning points on stress-strain curves while the AE energy reaches a peak value. After crack initiation, the AE energy drops to original level, and remains constant. The AE energy of Mode-II crack initiation is 8-10 times higher than that of Mode-I. The wing cracks in the loading process forms an anti-symmetrical shape with the crack center line being the symmetry axis. The initiation angle of the wing crack increases gradually with the crack spreading into the specimen. The propagation shape of wing cracks is leaf-like with an anti-symmetrical structure. A large amount of shear cracks and secondary cracks are observed in rock-like specimens. Shear crack plane is initiated towards the extension direction of pre-existing cracks until the pre-existing crack is terminated.