In the classical crack-tip stress field, the singular stresses (r1/2 term) near the crack tip are characterized by a single parameter, i.e. the stress intensity factor K. Therefore, the O(r1/2) term and the secondary non-singular term (T-stress) are always ignored. However, the stress and strain fields around the crack tip are influenced considerably by the T-stress. In this paper, the conventional fracture growth criteria, i.e. the maximum tangential stress criterion is modified to take into account the effect of T-stress, and is used to study the fracture growth when subjected to modes I, II or mixed mode (I/II) loading. The results show that: (1) In the pure mode I loading condition, if the T-stress is negative (compression), the crack growth is stable. However, if the T-stress is positive (tensile), the crack grows along the crack direction only when , otherwise, the crack growth direction will be changed. (2) In the pure mode II loading condition, the crack growth direction and the loading capacity are influenced not only by the stresses along the fracture direction (T-stress) but also by the normal stress on the fracture. (3) For the mixed loading condition, the T-stress consideration in the fracture criteria leads to better agree with the experiments. The positive T-stress results in an increase in crack growth angle, while the negative T-stress decreases that angle. The results reveal that it is benefit to control the T-stress to change the crack growth direction, in order to avoid the most dangerous crack growth direction or stopping/slowing down the overall structural fracture.

It is known that frost heave pressure goes through the process of initiation, development and dissipation when rock cracks under freeze-thaw cycles. Moreover, both the frost extension of cracking and the degree of frost damage of rock depend on frost heave pressure. Therefore, an analytical model is established for frost heave pressure in an idealized cylindrical closed crack, based on the theory of thermodynamics, seepage mechanics, interfacial mechanics and elastic theory. Besides, the effect of moisture migration on the value of frost heave pressure is investigated. Without considering the moisture migration, the results show that the frost heave pressure increases rapidly with the increase of water saturation in crack and rock elastic modulus Es. Under the condition of and 94%, the frost heave pressure exceeds 15 MPa, which is high enough to drive rocks to initiate cracking. In addition, the rock starts to crack once the water saturation reaches its own critical value, . For rocks with the low permeability below 5×10?14 cm2, a high frost heave pressure can be produced by frozen water in the crack, which easily leads to frost crack propagation. However, for rocks with the permeability above 10?12 cm2, the frost heave water pressure could not cause any damage. The main reason for crack frost propagation in high permeability rock is the disjoining pressure in a microscopic unfrozen water film between ice and rock interface after water frozen.

Seismic response of micropiles in a liquefiable soil is one of the key components that must be addressed in aseismic design. In this paper, the seismic responses of a micropile in liquefiable soils, including lateral displacement and acceleration, moment distribution and pore water pressure, are analyzed based on dynamic centrifuge testing and three-dimensional effective stress analysis. A dynamic centrifuge test is performed on a saturated sandy soil of a 57% relative density, using an input excitation of a sine form with a peak shaking amplitude of 1.516 m/s2 and 1 Hz frequency under a 40g condition. A numerical simulation is conducted to investigate the distributions and variations of lateral deformation and acceleration at the pile cap, the bending moments of the pile at different buried depths, the acceleration and excess pore water pressure in the liquefiable soil. An inversion analysis is performed on the experimental results based on the three-dimensional effective stress method and the concepts of multiple shear mechanism plastic model and state surface of liquefaction front. Comparison between the physical and numerical models indicates that the computed results agree well with the measurements of the centrifuge test, and that the dynamic response of the micropile is controlled by the field condition of the liquefiable site. It is also indicated that the maximum amounts of the lateral deformation and residual deformation at the pile top are 78 mm and 30 mm respectively. The largest bending moment and residual bending moment of the micropile occur at deeper buried depths. It is shown that the seismic response of micropiles in liquefiable soil can be effectively analyzed by combining the centrifuge test and numerical simulation.

In recent years, the demand of exploiting and utilizing new resources is increasing significantly as the consumption of traditional energy resources increases. As a result, the study of thermal properties of geotechnical materials has drawn more attention because geothermal energy is a kind of clean, sustainable and renewable energy resource. Soil thermal property consists of thermal conductivity, thermal diffusivity and heat capacity. Thermal conductivity is the most important soil thermal property, and it not only determines the speed of heat conduction and temperature field in soils, but also is a primary design parameter for various kinds of geothermal heat pumps, geothermal energy piles and other geothermal related structures. The engineering background and research motivation of soil thermal conductivity is firstly introduced, and followed by the definition of soil thermal conductivity and the analysis of its influence factors including moisture content, density, mineral component, temperature, etc. Based on the literature review, the summary of soil thermal conductivity prediction models is also presented. Model assessments including the advantages, disadvantages and applicability of each model are also presented afterward. Finally, some recommendations and suggestions have been provided for future research regarding geothermal applications.

The aim of this study is to explore the long-term influence of tunnelling on the adjacent pile foundation under cyclic change of groundwater level. A three-dimensional centrifugal model test is conducted to investigate the long-term effect of tunneling on a pile group while the groundwater level changes cyclically. The long-term surface settlement, additional pile head subsidence, additional bending moment and additional axial force of pile are examined. The results show that the long-term surface subsidence is influenced more significantly by the cyclically varying groundwater level induced, particularly, by precipitation. The long-term surface settlement increases with the number of cycle, showing an attenuation-type deformation; it is not convergent after three times of water level fluctuation cycle. The long-term additional subsidence of pile foundation is remarkable, accounting for more than 50% of the total settlement. Generally, the values of the long-term additional axial forces of front pile and rear pile are positive; if the total axial force increases, it can adversely influence the existing bearing pile. The inflection point and the maximum of the additional axial force are different. The inflection point number of bending moment of the front pile and rear pile decreases after three times of groundwater level cycles. However, the maximum additional bending moments of pile become significantly larger. The plastic hinge of pile body appears when the maximum additional bending moment of pile reaches the ultimate bending moment, which is extremely dangerous to the existing bearing flexible pile that passes through thick layer of weak soils.

Cyclic loadings such as traffic loads, wave loads and earthquake can cause principal stress rotation, which has a significant impact on the long-term settlement and stability of soft ground. In order to study the long-term dynamic characteristics of saturated soft clay under the traffic load, a series of hollow cylinder shearing experiments is performed on the soft soil by adopting three types of principal stress paths including heart-shaped line rotation, circular rotation and directional shear in deviatoric stress space. The undrained cumulative plastic deformation, pore pressure, critical dynamic stress ratio and dynamic strength characteristics are compared among specimens under different experimental conditions. The results show that (1) the critical stress ratio increases, in order, from circular rotation, heart-shaped line rotation and unequal amplitude in tension, to triaxial compression. Accordingly, the cumulative plastic deformation and pore pressure limit decrease in turn for the stable cumulative stage of axial strain. (2) When the vibration times is greater than certain values, the axial cumulative strains of the stable-type specimens have good linear relationships with the logarithm of time under different principal stress rotation paths. Based on this, the cumulative axial plastic deformation equation is developed. (3) Along with the cumulative axial plastic deformation, the pore pressure growth in the soil under different dynamical loadings also shows three kinds of development trend. A three-stage and two-phase mode is found in the pore pressure-time curve and the strain-time curve under the principal stress axis rotation path. (4) When the other conditions are the same, the strength of soil under heart-shaped rotation is higher than that under the circular rotation, and lower than that under the unequal amplitude in tension and compression of triaxial path.

Using the self-developed triaxial servo-controlled seepage equipment for thermo-fluid-solid coupling of coal containing methane, a series of tests is conducted on coal samples during the process of gas loading-unloading at different temperatures to simulate the seepage characteristis of coal under the condition of temperature increase induced by increasing depth of mining. Meanwhile, parallel experiments on helium are also carried out to investigate the slippage effect of low permeability coal seam. The experimental results are shown as follows: (1) During the loading process, the axial strain increases and radial strain decreases linearly with the increase of gas pressure. During the unloading process, the strain of coal shows the same trend with the loading. The slope of strain varies with increasing gas pressure and temperature. (2) During the loading process, the permeability of coal shows a quadratic parabola trend with the increase of gas pressure, and it reaches the minimum at about 3.0 MPa. At the early unloading stage, the permeability of coal decreases slowly and then increases with the decrease of gas pressure. The permeability of coal during the loading process is lager than that during the unloading process. (3) Variation of permeability caused by slippage effect during loading process is greater than the amount of unloading process. What’s more, variation of permeability caused by slippage effect declines in an exponential function with the increase of gas pressure.

In this study, the Lyapunov index method is adopted to study time series under each loading strain level during the experiment on an evolving stress relaxation system of the deep coal seam. Through reconstructing phase space of deep seams, phase locus corresponding to each strain level presents an evolution on encircling its historic stress median no-repeatedly. By analyzing the time series with the fractal geometrics method, self similarities among different evolving stages of the stress relaxation system have been determined quantitatively. In addition, the relationship between classic mechanics and nonlinear science is also reflected. Nonlinear dynamical indexes of the evolving stress relaxation system are obtained to verify that it is a complicated system with typical chaotic properties. Moreover, the chaotic degree and complicated degree might descend with the development of the stress relaxation system of deep coal seams. It is found that the most obvious chaotic degree appears at the critical stage when the competition between softened factors and the enhanced ones for mechanical features of coal seam is the most drastic. Whereas after macroscopic failure the system shows the lowest chaotic and complicated degrees.

Using the digital-laser dynamic caustic system, the impact tests are utilized to investigate the interaction mechanism between a circular hole defect and mode I moving crack. The result shows that, under impact loading, the fracture face is relatively smooth and the growth path is straight towards the hole before the coalescence of moving crack and the round hole. After the moving crack initiates at the top of the round hole, the fracture face is relatively bumpy and the growth path is relatively bent. When the mode I moving crack propagates towards the adjacent round hole, the round hole plays a suppressive effect on crack growth velocity and dynamic stress intensity factor of moving crack. The suppressive effect becomes apparent with the increase of round hole diameter. After the crack-hole coalescence, the moving crack is arrested abruptly and the crack tip becomes blunt. The initiation toughness of the blunt crack increases about 9.58%-13.87% than that of the tip fracture. The crack growth velocity and dynamic stress intensity factor of the blunt crack show obvious jump when the crack initiates again at the hole, indicating that the more energy consumption is needed for blunt crack initiation.

Brazilian tests are conducted on shales with different bedding angles, and a high speed camera and an acoustic emission (AE) system are used to investigate the effect of stratification on shale mechanical properties, crack propagation and AE characteristics. The results indicate that: (1) Brazilian split destruction process can be divided into three stages: compaction stage, elastic stage and destruction stage. (2) The anisotropic characteristics of tensile strength, splitting modulus and deformation at the stress peak point are highly distinct. The most obvious effect of stratification is found at the angle of 30°, whereas the smallest effect is found at the angle of 90°. (3) For samples with angles of 0° and 15°, the failure surfaces are normally along the direction of bedding plane. At the angle of 30°, the failure surface intersects bedding planes and coincides with the loading baseline surface. In contrast, fracture surfaces deviate from the loading baseline surface with varying degrees and turn out to be curved surface for samples with angles of 45°, 60°, 75° and 90°. (4) 30°and 45° are transformation angles of failure mechanisms. At 30°, tension failure occurs along bedding converts to cross-cut bedding plane failure along the loading baseline, while cross-cut bedding plane failure happens along the loading baseline converts to cross-cut bedding planes failure with varying degrees of shear slip at 45°. (5) AE activities and energy releases demonstrate obviously with increasing bedding angles caused by the bedding plane, which results in the anisotropy of failure mechanism. Moreover, there exists a good linear relationship between the peak value of AE energy rate and tensile strength, which indicates that the peak value of AE energy rate can reflect the variation of tensile strength well.

Three-dimensional problems of consolidation with vertical drains are sometimes simplified as plane-strain ones to facilitate the operation of numerical analyses. It requires establishing an equivalent relationship between the three-dimensional consolidation with a vertical drain and the two-dimensional consolidation with a sand-wall drain. Analytical solutions for two-dimensional consolidation with a sand-wall drain are, therefore, the important basis for determining the equivalent consolidation parameters. To analyze the applicability of the existing approximate solutions of equal-strain consolidation with a sand-wall drain, an analytical solution for free-strain consolidation is derived based on the rigorous governing equations for soil element volume. An important effect of drain resistance on consolidation is considered. A general drainage boundary condition is taken into consideration, including fully drained condition and impeded drainage condition. Comparisons are made between the existing solutions and the proposed solution. In addition, the effects of Poisson’s ratio, vertical and horizontal drainage on consolidation are investigated. It is shown that the existing solutions for consolidation with a sand-wall drain are approximate but sufficiently accurate. The consolidation is mainly governed by horizontal consolidation. Vertical consolidation is however negligible. Consolidation rate increases with the increase of Poisson’s ratio. This factor should be considered when a sand-wall drain is equivalent to a vertical drain.

When initial density and confining pressure of non-cohesive soil are considered in Pastor-Zienkiewicz (PZ) constitutive model, multiple sets of parameters are required to simulate the same type of material. Thus this model is pretty difficult to be applied in practice, which to a great extent limits its development. To solve the existing deficiency of the PZ model, some recently proposed state parameters are introduced into the dilatancy formula, plastic potential direction vector, loading (unloading) direction vectors and plastic modulus equation of the original PZ model, separately. It is found that the predicting capability of the modified model has been improved in terms of mechanical behaviour of the non-cohesive soil, particularly by considering the initial density and confining pressure. Then the calibrating method for each parameter in the modified model is discussed in detail and the physical meaning of each parameter is further clarified, according to a practical example. Finally, the predictive ability of the modified model is verified by the experimental data. It is shown that predicted results are in good agreement with experimental data, which indicates the modified model is reasonable and feasible.

To analyze the influence of an adjacent foundation on ground stiffness, a new concept of ground stiffness influence coefficient is introduced, which is defined as the ratio of the ground stiffness with the adjacent foundation to that without. Based on the Mindlin solution, an analytical expression of the influence coefficient is derived from a system of two rigid, surface, rectangular foundations under three-dimensional elastic half space. Through calculating the influence coefficient, we analyze the influences of clear distance, side-length ratio of foundations and Poisson’s ratio of the ground soil on ground stiffness when the adjacent foundation exists. It is shown that: (1) The ground stiffness can be improved when the effect of the adjacent foundation is taken into account. (2) The influence of the adjacent foundation on ground stiffness shows a positive correlation with the side-length ratio and the Poisson’s ratio, and a negative correlation with the clear distance. (3) The influence is more significant in clay ground than in other kinds of ground, such as silt ground. (4) It is not necessary to consider the influence of adjacent foundation when the clear distance is larger than 15 times the side length of the foundation or the side-length ratio is less than 0.1. In other cases, however, we should decide whether or not it is necessary to consider the influence of both the clear distance and the side-length ratio.

The swelling property of bentonite-sand mixtures is important to evaluate the long-term performance of deep geological repository of nuclear waste. Through the comparison of swelling test results of different kinds of bentonite-sand mixtures, we can conclude that, for pure bentonite and bentonite-sand mixtures with low sand mixing ratio, montmorillonite void ratio em and vertical stress ?v relation is a unique line in their bilogarithmic coordinates, and we can predict the volumetric strain after water uptake under a given vertical stress and the swelling pressure at constant volume for mixtures with different sand mixing ratios; while, for the mixtures with high sand mixing ratio, the em-?v linear relation is satisfied only before sand skeleton is formed. When the stress is larger than deviation starting stress, the em-?v relation deviates from the line. Moreover, the larger the sand mixing ratio, the smaller is the deviation starting stress. Before sand skeleton comes into being, sand particles are surrounded by montmorillonite, hence the vertical stress is mainly burdened by montmorillonite, and the swelling quantity at full saturation including swelling deformation and swelling pressure is determined by the montmorillonite content per unit volume; after the sand skeleton is formed, the vertical stress is burdened by montmorillonite and sand skeleton. The deviation starting stress when sand skeleton is formed can be determined for mixtures with different sand mixing ratios according to the concept of sand skeleton void ratio, and the range of sand mixing ratio in which the sand skeleton may be formed can also be obtained.

Based on the upper plastic limit analysis theory, the kinematically admissible velocity field of a capped rigid pile is obtained through both numerical and model tests. According to the energy balance principle and unified strength theory, an equation is developed for the ultimate bearing capacity limit solution of the pile, in which the effect of the intermediate principal stress is considered. In addition, the variations of the pile ultimate bearing capacity with the cap size, soil cohesion and internal friction angle are analyzed through parametric analysis. It is found that the upper limit solution increases more significantly for the pile with considering the intermediate principal stress than those without. A reasonable range of the unified strength parameter is from 0.1 to 0.2 in calculating the rigid piles bearing capacity. With the increase of the cap size, the pile bearing capacity increases. When the ratio of cap diameter D to pile diameter d is greater than 2, the bearing capacity increases more significantly. With the increase of soil cohesion and internal friction angle, the bearing capacity increases drastically. By comparison with experimental results, it is found that the proposed method yields better results than the existing theoretical methods.

Drucker-Prager yield criterion is widely used in rock engineering with the benefits of its simple form and clear physical meaning. However, its drawbacks are gradually revealed after years of research and application, such as lacking the effect of stress angle and the oversized tensile-shear region. Therefore, we propose a modified Drucker-Prager yield criterion in this study. In compressive-shear region, the same form of Drucker-Prager yield criterion is reserved, but in tensile-shear region, the original conical surface is replaced by the combination of a spherical surface and a cross section. This method can not only consider the real tensile strength of rock, but also satisfy the condition that the yield criterion passes the point of uniaxial tensile strength. A corner model is introduced to describe different strength characteristics of rock under triaxial compression and triaxial tension. To satisfy the condition that the yield criterion goes through the point of uniaxial compressive strength, a new method for calculating the value of k is suggested. Then the calculated results of the yield criterion are further fitted with that of true triaxial tests to verify the applicability and accuracy of the modified yield criterion. By comparing the results of the modified criterion with that of Mohr-Coulomb criterion, it is shown that the modified yield criterion can describe experimental data well, which is more realistic than the later.

To reveal the stabilization mechanism of lignin based industrial by-product treated soil, the lignin is utilized to improve the silt of subgrade. The unconfined compressive strength, pH value, microstructure, chemical elements, mineral composition and functional groups of lignin and lignin treated silt are tested to evaluate the mechanical properties and pH values of silt after treatment. The difference of microstructure between silt and lignin treated silt is compared. The interactions between lignin and soil are also discussed based on the results of chemical analysis. A possible stabilization mechanism of lignin treated silt is proposed. The experimental results show that the unconfined compressive strength of lignin treated silt increases with the increase of lignin content, while the lignin content exceeds a certain range, the strength of soil exhibits a decreasing trend. The optimum percentage of lignin for silt is approximately 12% and curing time has an important influence on the strength of treated silt. The relationship between unconfined compressive strength and pH value of treated silt is approximately linear. Compared with natural silt, the microstructure of treated silt is more compact and stabler. Soil particles are bonded with cementing materials and no new clay minerals are formed after treatment. The engineering properties of silt are improved by the addition of lignin through hydrolysis, cation exchange, protonation and electrostatic reaction, which leads to the reduction of thickness for double-layer and filling the pore. The proposed model of microcosmic mechanism can provide a theoretical reference for engineering application of lignin based by-products.

Multi-footing system composed of multiple separate footings is a common type of offshore foundation. Based on failure envelope theory, the failure mode of multi-footing system on sand is analyzed in this study. Additionally, the corresponding analytical method of the bearing capacity is established and its feasibility is verified. The comparison of the calculated load safety factors between the single footing and multi-footing system is conducted in different loading paths. The analytical method of foundation bearing capacity is also conducted based on the combination of the failure envelope theory and partial factor method. From the analysis of failure mode, head sea side footings with four-legged platform firstly reach the failure envelope as result of increasing horizontal load, and their failure modes further become sliding. However, the sliding of head sea side footings would not occur due to the constraints among other footings. Since the head sea side footings bear less horizontal and vertical loads with increasing the horizontal load, the back wave side footings have to undertake larger load. At last, back wave side footings also reach the failure envelope, which results in the failure of the multi-footing system. It is shown that load paths have critical influence on load safety factors, and thus they should be specified when calculating load safety factors. Since the size and shape of failure envelope depend on a variety of factors, the calculation of bearing capacity needs to consider the failure envelope under particular conditions during design of foundations.

The settlement of Shanghai World Financial Center (SWFC), which is piled-raft foundation, is analyzed in this study. Then the corresponding foundation pit resilience, settlement developing with time and structure stiffness are investigated, and the variation of superstructure stiffness is predicted according to the settlement as well. It is found that the settlement progress of deep piled-raft foundation in soft soil is depending on a various of factors such as loading rate, pit resilience, additional stress transferring in the scopes of shaft and its embedded soils, dewatering and so on. The settlement-time curve of foundation is not an ideal hyperbola, even if the loading rate of construction is relatively stable. At the initial stage of building construction, the low settlement rate is mainly related to the rebound of soil deep foundation pit. While the high settlement rate is found when the building reaches completion, which is probably caused by the additional stress that has transferred to the soil layer under pile toe. It is found that 4.5 m thick raft in settlement development is still considered as an elastic body, and the contribution of superstructure stiffness to the raft is limited. Moreover, the superstructure stiffness is gradually formed with the progress of construction. Three methods are employed to conduct the back analysis of settlement of the SWFC, and calculated results are compared with the settlement data. It is shown that in-situ data is fitted well with the results obtained by the Mindlin solution, referring to “Shanghai code”, in which the proper experience coefficient is considered as 0.40-0.55 and the water buoyancy is set as 70%. By analyzing interaction of deep piled-raft foundation in soft soil, the superstructure stiffness should not be overvalued.

The immigrants from Houziyan hydropower station are resettled on the accumulation fan of Jiangkou gully debris flow. Based on field investigation on the formation conditions and motion characteristics of the debris flow, the hazard scope of debris flow occurred in 1958 with a frequency of 50-year return period is obtained, and the motion parameters of debris flow are determined with the rain-flood method. The fluid level distribution and velocity field are simulated, and the hazard scope and the velocity field of Jiangkou gully debris flow are obtained by using CFX software, based on Bingham rheological model. The simulation results show that the average velocity of a 50-year return period debris flow passed through the accumulation fan is 5.76 m/s, while the maximum velocity is 13.59 m/s. The hazard zone obtained by the numerical simulation is larger than that occurred in 1958, since the deposit of early debris flow filled up the original ground, especially the ground near the channel, making the debris flow flood and expand more easily toward both the flanks of the channel. This research provides a new approach to designing prevention project and delineation of hazard zone of debris flow.

In the construction of the high rockfill dams, it is crucial to accurately determine the mechanical parameters of prototype rockfill materials. Parameter inversion has been shown to be a potential method to solve this problem. In the traditional inversion methods, however, a large amount of finite element analysis is required so that the inversion method is time-consuming and of low efficiency. The response surface method can effectively overcome the above problems. The existing response surface methods are originally designed for the instantaneous parameters of rockfill materials, without including the rheological parameters. Considering both instantaneous and rheological parameters, a more reasonable response surface function is developed, and an inversion scheme is proposed for determining both the instantaneous and rheological parameters of high rockfill dam, which can greatly improves the efficiency and precision of the parameter inversion. The proposed method is adopted to evaluate the instantaneous and rheological parameters of Shuibuya concrete face rockfill dam. It is shown that the calculated settlements of the inversion parameters agree generally well with the measurements with regard to both magnitude and trend, showing that the proposed method is applicable.

The permeability of fractured rock mass is essentially governed by its geometrical characteristics, connectivity and filling materials of fractures, which commonly displays a significant correlation with the magnitude of the geostress field. Due to the limitations of existing models for the prediction of permeability, a novel empirical model (called ZRF model) is established to estimate the permeability coefficient of fractured rock mass by considering various factors. Particularly, the proposed model represents the permeability coefficient as a function of three geological indexes, i.e. buried depth (Z), rock quality designation (RQD) and filling substance designation (FSD). These three parameters are easily calibrated with packer hydraulic test data and borehole acoustic television images. Then the proposed model has been successfully applied to predict the permeability coefficient of fractured rock mass at the site of Yagen-II Hydropower Station and two other sites at which the borehole data are available in the literature. Advantages of the ZRF model include more clear physical significance and easy parameterization in engineering practices over the other existing permeability estimation models. The results indicate that the proposed ZRF model provides a feasible tool for the estimation of permeability of fractured rock mass at various field conditions.

During the long-term service period of the municipal solid waste (MSW) landfill, the strength parameters of the waste body change as time elapses. In analyzing the stability of the MSW fill, we decompose the waste body into layers according to the age, and propose two failure mechanisms based on the upper bound theorem of limit analysis, including the rotational and translational mechanisms. To confirm the above two mechanisms, a comparative analysis is performed on a slope composed of horizontally soil layers, and then the case of the landfill slope layered by aging is analyzed by assuming the two mechanisms above. It is shown that the factor of safety and failure mechanism by the upper-limit solution of limit analysis are very close to those deduced from the reduction finite element analysis, and the critical failure mode obtained by the multi-block mechanism is more reasonable. In addition, the stability of the Coll Cardús MSW landfill in Spain with complex layers by aging is analyzed, and the results are consistent with those of the reduction finite element analysis. Our results indicate that, as the fill age increases, the landfill slope will become more and more stable in the absent of pore pressures. The fill age can be divided into low, middle and high levels. The safety factor of the landfill slope layered by aging is between those of the homogeneous slopes with middle to high aging levels.

To improve the coal recovery rate, it is crucial to properly determine the coal pillar size, supporting patterns and parameters of the roadway, since this can ensure the stable of butt entry with fully-mechanized mining. The evolution of principal stresses, deformation and failure are studied using FLAC3D with different pillar widths in haulage roadway 20103 of Wangjialing coal mine. It is shown that the most reasonable and economic pillar widths range from 6 m to 10 m. The final reasonable pillar width is determined to be 8 m based on the geological production conditions. Theoretical analysis and borehole observation show that the main roof breaks at 7 m next to the gob. The property and structure of the surrounding rock and the stress distribution are asymmetric along the axis of roadway, resulting in asymmetric roof strata behaviors including the subsidence of roof near the pillar and even the dislocation, embedding, step convergence of rock on the abutment of coal mine. The roof near the pillar and the abutment of coal mine are the key positions of deformation failure. A comprehensive supporting technology is proposed including high strength anchor-beam-net, asymmetric cable-beam and cable truss. The mechanism of supporting measures for surrounding rock is analyzed. It is shown that the pillar width with 8 m is reasonable and the proposed measure can control the asymmetric deformation failure. The proposed supporting measure has been successfully applied to Wangjialing mine area.

The seismic response of rock slope is one of the most significant issues in the field of geotechnical engineering. In this study, the intake slope of Zipingpu project is investigated to study the influence of Wenchuan earthquake based on the monitoring data. From the perspectives of time and space, the characteristics of supporting load and slope displacement have been analyzed before and after Wenchuan earthquake. Furthermore, the synergistic effect of slope shape, rock mass structure and supporting load under seismic force is discussed. The results show that the sudden increases both in deformation and supporting load of slope are principally caused by the seismic force, and thus earthquake is regarded as the main influencing factor. It is found that the change of supporting load caused by the earthquake is below 100 kN, which is less than 5% of the design load. Permanent displacement of the slope in horizontal direction is within the range of 12 mm and the deformation distribution in depth is controlled by the structural plane of rock mass. Particularly, the maximum displacement concentrates in the middle part of the slope. Free surface amplification effect has been found along the horizontal direction, while there is no obvious vertical amplification effect along the elevation direction. Therefore, this research can not only underpin the understanding of the dynamic response mechanism of anchored rock mass, but also provide references for the aseismic design of slope engineering controlled by the displacement .

Fuzzy evaluation systems have been widely applied to comprehensively evaluate the slope stability. When material parameters of rock masses are used as evaluation factors, due to their discreteness in practical engineering, these parameters are required to be evaluated in advance to obtain characterization parameters. Then the characterization parameters are employed in comprehensive evaluation. From the viewpoint of rock masses resisting on external interactions, the characteristics of rock masses of Dagangshan abutment slope are studied in depth by considering the range and sequence of deformation and the safety factor before and after slope failure. Specifically, these characterization parameters are obtained by analyzing layered sensitivity and evaluating each parameter and the membership degree is derived by the area distribution method. Besides, the comprehensive sensitivity of mechanical parameters of rock masses is investigated by referring safety factor and the displacement. The weight of each factor is further determined by its sensitivity, and then applied in the comprehensive evaluation system. It is found that the deformation of rock masses mainly depends on cohesion when the displacement is regarded as reference. The frictional angle has more effect on slope failure than the cohesion when the safety factor is regarded as reference. The results also indicate that the sensitivity characteristics of displacement are better than those of safety factor. The results of comprehensive estimation reveal that rock masses have a group of intensive mechanical parameters. It is also noticed that evaluation results are better when the safety factor is used as reference, which is in accordance with the results of sensitivity analysis and practical circumstance, and thus the reasonability of the proposed method is verified.

Based on the interpretive structural model and cause-sequence mapping approach, twelve representative factors, either qualitative or quantitative, of seismic liquefaction are selected to construct a Bayesian network (BN) model of seismic-induced liquefaction under the condition of a large number of incomplete data. Based on a set of incomplete data of the 2011 Pacific Coast liquefaction induced by Tohoku Earthquake, the performances of proposed model are assessed comprehensively with regard to the following five indexes: the overall accuracy, the area under the ROC curve, precision, the recall rate and F1 score, and then compared with a radial basis function (RBF) neural network model. It is shown that both the back evaluation and forward prediction of the BN model are better than those of the RBF neural network model, and the BN model also performs well for the case of incomplete data. In addition, the BN model is also suitable for predicting the liquefaction of different soils. Classification imbalance and sampling bias can influence the performances of the models significantly. Hence it is suggested that the five indexes mentioned above can be used to evaluate the performances of evaluation models.

The bulk waves in double-porosity media and the reflection of P1 wave at the free permeable and impermeable surfaces are theoretically derived and solved. The variations of the amplitude reflection coefficients of the reflecting waves with incident angle and frequency under different surface boundaries, as well as the dispersion and attenuation characteristics of the four kinds of bulk waves, are numerically studied. The results show that: (1) P1, P2, P3 and S waves all show dispersion and attenuation characteristics. Therein, P1 wave propagates fastest with lowest attenuation, and P3 wave attenuates fastest with lowest propagation velocity. (2) As the incident angle of P1 wave increases, the amplitude reflection coefficients of the reflecting P2, P3 and S waves increase at first then decrease, but the case of the reflecting P1 wave is just opposite. Moreover, the permeability of the surface has distinctive effect on the amplitude reflection coefficients of P2 and P3 waves and has slight effect on P1 and S waves. Nevertheless, whether the surface is permeable or impermeable, the changing tendency of the amplitude reflection coefficients is consistent. (3) The amplitude reflection coefficients of the reflecting P2 and P3 waves grow as frequency increases. In contrast with the reflecting P3 wave, for the reflecting P2 wave, the amplitude reflection coefficient in the case of free impermeable surface is larger than that in the case of free permeable surface. (4) For the reflecting P1 wave, with the increase of frequency, the amplitude reflection coefficient in the case of free permeable boundary increases firstly and then decreases, but fluctuates (decrease-increase-decrease) in the case of free impermeable surface. The amplitude reflection coefficients of the reflecting S wave just vary oppositely with the reflecting P1 wave.

Due to the continuing consumption of fossil fuels, a large amount of CO2 is emitted into the atmosphere, leading to global warming, sea level rising and other global climate problems. CO2 geological storage is one of the most promising techniques for reducing CO2 emission. However, this approach may induce a series of geomechanical issues, such as ground surface deformation, damage of the cap-rock integrity and existing fault activation. In order to reduce the safety risk from the CO2 geological storage, theoretical analysis, numerical simulation and response surface method are all applied to solve these geomechanical issues. At present, numerical simulation is the most widely used option on account of its exceptional performances in solving multi-field coupling problems associated with large-scale and complex geometry model. The aim of this paper is to provide a comprehensive review of numerical analysis approaches for analyzing the geomechanical issues induced by CO2 geological storage. First, a brief introduction is given about the thermo-hydro-mechano-chemical (THMC) coupling theory of porous media, and the classifications of the numerical simulation methods for settling the multi-field coupling problems are discussed; then, a review of research status about settling geomechanical issues based on numerical modeling is presented in detail; finally, the difficulties of the numerical simulation in solving these geomechanical issues are discussed, from which several suggestions for improvement are offered.

Proper estimation of input parameters plays a crucial role in the numerical simulations of geotechnical engineering. In conventional numerical procedures, all the input parameters are assumed constant and unique, and their variability and cross-correlation are usually neglected, so that significant error can be induced in the simulated results. To resolve the problem, an inversion method must be introduced to obtain mechanical parameters, which are spatially correlated. The random field of a single parameter has a characteristic of auto-correlation, whereas the random field of multi-parameter is not only auto-correlated, but also cross-correlated. Based on ‘two sides of one’ (auto-correlation ??cross-correlation ??remaining auto-correlation), a mathematical inversion model of multi-parameter random field is built. Correlation statistics results of random fields generated by ‘two sides of one’ method show that auto-correlation and cross-correlation are well satisfied in the isotropic correlated condition.

This paper introduces the model of pressure solution for granular aggregates established by Taron et al. into the finite element code for analysis of thermo-hydro-mechanical (T-H-M) coupling in porous media developed by the first author. Using the Mohr-Coulomb yield criterion, a hypothetical disposal model for nuclear waste located in a saturated quartz aggregate rock mass with a laboratory scale is simulated. Two computation cases, the elastic analysis and the elastoplastic analysis are designed. Then the corresponding numerical simulation for a disposal period of 4 years is carried out. The states of temperatures, solute concentrations in the intergranular fluid film and at the pore space, migration and precipitation masses, porosities and permeabilities, pore pressures, flow velocities and stresses and plastic zones in the rock mass are investigated. It is shown that, because of the stress adjustment and the increased molecular diffusivity in the elastoplastic analysis, there are obvious changes of solution, migration and precipitation of aggregate medium in the plastic zones, and the seepage field (including pressures and flow velocities of pore water) and the stress field are markedly influenced. But the differences of solution, migration and precipitation of aggregate medium in the elastic zones for two cases are not significant.

In slope stability analysis, one needs to calculate the factor of safety and determine the slip surface. The strength reduction method can yield reasonable value of the safety factor, but it cannot depict correctly the critical slip surface. This paper proposes an approach to determining the critical slip surface based on the displacement filed analysis during the sliding movement. It is shown that the displacement contours near the slip surface are the densest at the critical state of slope. The points along the potential slip surface represent those points at which the change rate of displacement along the direction perpendicular to the critical slip surface reaches the maximum in the vertical direction. By comparing the proposed method with the Spencer method and physical model test of rock slopes, it is shown that the proposed method yields good results in searching the slip surfaces. In addition, the effects of some other factors such as geometrical shape, degree of density and discrete point spacing on the slip surface are discussed.

This paper aims at establishing a set of scaling laws according to three similarity criteria of geometric similarity, mechanical similarity, dynamic similarity, under which a scaled discrete element model can exactly reproduce the prototypical problem. The method is based on the exact scaling laws of discrete element method proposed by Prof. Y. T. Feng. The scaling laws are then extended to the study of rheological behavior of granular materials. A detailed theoretical derivation is given based on the Burgers creep model. The rheological parameters are introduced to the model, and then we can gain the scaling laws in both two-dimensional and three-dimensional cases. Secondly, numerical simulation is conducted on the basis of the theoretical derivation. The results show that some parameters must be scaled to ensure the consistence of simulated results. The scaled discrete element model can exactly represent the original physical problem within the relative error of 3%. This paper also discusses the influence of the time step, viscosity coefficient, particle numbers and scale number on numerical simulation, which provides a good reference for parameters selecting in numerical simulations. Besides, because the scaled model has the same particle number, particle shape, particle compactness and scale number as the physical model, it can reveal the effect of proportional scaling on rheological behaviors.

In the Savage-Hutter (S-H) granular flow model, the earth pressure coefficient that is only related to the material property is chosen to describe the internal stresses of the granular flow. However, the constitutive relation of the granular flow is not given in this model. In this paper, an avalanche dynamic model, which can well reveal the constitutive behavior of the granular flow, is developed by introducing the relationship between the stress and the velocity gradient. In order to relax the assumption for lateral stress, the Von Mises criterion, Drucker-Prager criterion, Mohr-Coulomb criterion and Matsuoka-Nakai criterion are combined to yield four expressions of the generalized friction coefficient for describing the moving properties of the granular material. As one of the advantages of the proposed model, the internal mechanism of granular flow is clearly revealed by introducing a simplified constitutive relationship, and the strength parameters used in the proposed model are readily available. The strength parameters derived by Drucker-Prager criterion and Mohr-Coulomb criterion are also analyzed, and the relationship between the generalized friction coefficient and the Lode angle is deduced. It is shown that the numerical simulations of the proposed model agree well with the experimental results.

In this study, a three-dimensional (3D) dynamical analysis is carried out for foundation piles under the action of a concentrated pulse load at the top centre by both in-situ test and numerical simulation. It is found that the amplitude of the envelope curve of velocity signal varies along the radial direction from point to point at the top plane of a pile. There exists a point where the amplitude of the envelope curve keeps the lowest, and this point is called 3D interference minimum point. The envelope amplitude decreases firstly from pile centre to this point and then increases outwards. The signals of 3D interference minimum points are compared with others in the literature. It is shown that the influence on the point location from pile-soil parameters such as pile diameter, pile length, load pulse width, elastic modulus of pile concrete, shear wave velocity and Poisson's ratio of surrounding soil, can be ignored. However, the point location is highly depending on the Poisson's ratio of pile concrete. The location of interference minimum point is more closed to the pile center when the value of Poisson's ratio of pile concrete is higher. A mathematical expression between the point location and the Poisson's ratio of pile concrete is then deduced, showing that the proposed method can be easily applied to practical engineering.