The allowable values of the factor of safety in gravity retaining wall design in various Chinese codes are discussed, and it is concluded that these allowable values are generally smaller than those adopted in many foreign country codes. Theoretically, the factor of safety is defined as the resistance divided by earth pressure, where the earth pressure is the one determined by using the non-corrected strength parameters instead of the reduced strength, and thus the resulting allowable factor of safety should be larger than that adopted in the design of the equivalent retaining structure. To resolve this problem, limit state design with partial factors is applied to retaining wall design problem. For Class 1 structures, its target reliability index is 3.7. Assuming the coefficients of variability of 0.2 and 0.1 for cohesion and friction, respectively, the calibrated partial factors are 1.301, 1.209 and 1.170. The calibrated partial factors are validated by the ‘ratio of safety margin’ theory proposed by the authors’ research team. The discussion of this paper provides some useful guidelines for code makers.

Based on the current research on constitutive model of rock-soil mass, a new constitutive model is developed. It can describe the four different behavior characteristics of rock and soil, including elastic-plastic softening, elastic-plastic hardening, perfect elasto-plastic, elastic fragile. There are four parameters of new constitutive model. The existing constitutive models such as Duncan-Chang model are only a special case of the proposed constitutive model. Its parameter calibration is performed. The basis of constitutive model parameters calibration is that the correlation coefficient between the behavior characteristics described by the new constitutive model and experiment data is located in the optimal state. The behavior characteristics of post failure of rock-soil mass are analyzed, it is found that the behaviours of post failure of rock-soil mass can be described with quasi stress-strain relation. The experiments are used to verify the applicability of the developed model. The results show that the model is suitable to describe the behavior characteristics of rock-soil mass and can provide a reference for research on virtual internal bond model, spring model and discrete element method, etc.

The relationship between volumetric water content of soils and their effective dielectric permittivity is the theoretical basis of time domain reflectometry (TDR) technique. In this study, effective dielectric permittivity of a reconstituted expansive soil sampled from Nanning is measured by time domain reflectometry at different volumetric water contents, and three approaches including empirical equation method, theoretical bound method and theoretical model, are used to simulate the variation of the effective dielectric permittivity. The measured values of dielectric permittivity are found to be located within bounds of the Hashin-Shtrikman and Wiener, but the range of the solutions from the Hashin-Shtrikman bound is narrower than the Wiener bound’s. The effective dielectric permittivity of Nanning expansive soil is overestimated by Topp’s empirical equation, since the bound water in the expansive soil leads to a lower effective dielectric permittivity. The effect of bound water can be considered in Looyenga empirical equation in which three phases are separated and better agreements between predictions and tested results are obtained. Theoretical models have advantage in conceptualizing the effects of internal structure, but for the Maxwell-Garnett model and differential effective medium model, the predicting results of taking liquid phase or gaseous phase as a continuous phase are not satisfactory, since the three- phase configuration is greatly changed during water content increasing. The performance of these two theoretical models can be improved if the saturation are introduced as dependent weight functions to incorporate the calculated results of liquid phase and gaseous phase.

The methods of complex variable function and Green’s function are applied to derive analytical solutions to the scattering of SH-waves induced by a buried elliptic inclusion near a vertical interface in the elastic half space. Firstly, the analytical model to be solved is divided into two rectangular domains along the vertical interface. The conformal mapping method was used to map the outer boundary of the elliptic inclusion into a unit circle in the domain I. Moreover, in these two domains, the method of images is applied to construct scattering wave fields and an applicable Green’s function satisfying boundary conditions of rectangular domains. Secondly, with the aid of interface ‘conjunction’ technique, infinite algebraic equations are established by adding an additional system of forces at the interface and then solved by truncation, which meets the displacement and stress continuity conditions at the interface. Finally, some specific examples are given to estimate surface displacement amplitudes. Numerical results show that incident wave, incident angle, the position of inclusion, material properties and vertical interface to some extent affect the distribution of ground surface displacement.

Based on the formulae of normal displacement and shear displacement of crack faces from linear elastic fracture mechanics, this paper aims to investigate the stored excessive elastic strain energy during the fracturing process of a rock with a single crack by applying external loading. The normal stiffness and shear stiffness of the crack are assumed be constant rather than traction-free. Two deformation models, i.e. the non-uniform deformation mode and the uniform deformation mode, are applied to calculate normal and shear displacements. In addition, based on equivalence of elastic strain energy, the analytical expressions of equivalent Young’s modulus and shear modulus of 2D fractured rock mass containing regularly distributed non-persistent cracks are studied. The comparison between the analytical expression and a closed-form solution derived by Amadei and Goodman (1981) shows that the uniform deformation mode provides the same result. The comparison of normally distributed cracks between the analytical expressions and FEM shows that the non-uniform deformation mode considering crack interactions obviously underestimates the elastic moduli, while the uniform deformation mode predicts directional moduli within 10% deviation of FEM results. The obtained analytical expression can be used as one of effective methods to estimate the equivalent elastic modulus of rock mass under specific conditions.

Mudstone is sensitive to moisture, especially under loading coupled effect, its mechanical properties may decrease dramatically. The compression tests were carried out under loading and drying-wetting cycle joint effect to reveal the evolution process of mudstone performance with environment and loading influence. The compacted samples swelled when immerged into water for the first time. But, they would be compressed during drying-wetting cycle process subsequently. Then, pore size analysis testing was performed on compacted mudstone, the pore size distribution curves had two peaks, which corresponded to diameters 0.3 μm and 10 μm, respectively. Pore volumes reduced for drying-wetting cycle and loading joint effect, which was more obviously for large size pores. And, based on swelling testing, compression test and pore size distribution curves, one model about pore-moisture-loading interaction was established. The whole pore volume includes two parts, inter-volume and extra-volume of agglomerate. On the basis, the crushing-filling procedure of mudstone agglomerate was analyzed under drying-wetting cycle, and the mechanism of interaction between mudstone pore-moisture-load was revealed. This research can provide some theoretical references for mudstone subgrade construction.

According to fracture characteristics of rock caused a pick and its related data, breakage mechanisms and breaking effects of rocks influenced by different cutter spacing are analyzed. Cutting tests with different cutting depths and cutter spacing are performed using a rotary cutting machine with a single pick. The cutter spacing is optimized by three indicators, i.e. specific energy, coarseness index and cutting force, and the optimum ratio of cutter spacing to cutting depth for one type of sandstone is determined. Experimental results show that breaking process of the sandstone is divided into five phases, i.e. the generation of fine chips in a primary crushed zone, the formation of nucleus and its energy accumulation, the extension and connection of cracks, the falling of fracture and the regeneration of fine chips in a secondary crushed zone. The fracture is dominated by tensile failure and meanwhile accompanied by squeezing and shearing failure. Under the optimum cutting conditions, the higher coarseness index occurred when the rate of blocks is increased, and specific energy is lower. Through analyzing rock breaking process and the optimization of cutter spacing, it provides reference for breakage mechanism of rock by the roadheader and its pick arrangement.

The dilatancy and propagation of cracks under compression and shearing are essential to hydromechanical behavior of fractured rock masses. To explore the influence of dilatancy and crack propagation on seepage flow, a new shear constitutive model is developed to describe the features of the pre- and post-shear curves based on residual strength of rocks. Then, by combining the relationship between shear deformation and fracture aperture, a coupled stress-seepage model for fractured rock mass subject to shear dilation process is derived by the minimum potential energy principle and cubic law. In addition, a formulation for determining fracture aperture during crack growth process is obtained by assuming the initiation of mode-I cracks under compression-shear conditions. This model can not only consider shear dilation of rocks, but also imply the effect of crack propagation on the evolution of seepage flow. It is found that the newly developed shear constitutive model is suitable for shear displacement curves with different rough surfaces. A stress dependent seepage model is established by applying the shear constitutive model and is validated by existing coupled shear-flow tests, which is capable of predicting the changes in hydraulic properties of fractured rock masses under compression-shear stress conditions. It is noticed that the stress-seepage model is reliable to reflect hydromechanical behavior of crack propagation by applying the equivalent fracture to the representative of different fractures.

To understand scale effect of coarse-grained materials, the forms of scale effect are systematically clarified. The scale effect is divided into dense degree scale effect and mechanical properties scale effect. We analyze results of previous studies and summarize the performance of the law. The scale effect of dense degree is that the maximum or minimum dry density increases with the growth of the maximum diameter, which is regardless of scale methods. The law can be explained in terms of gradation, particle shape and compaction energy. With regard to the scale effect of mechanical properties, for the specimens with the same dry density, both shear strength and modulus decrease with the increase of the maximum diameter, which is resulted from compaction degree and particle strength. However, under the criterion of the same relative density, the law of scale effect of mechanical properties is unclear, and thus further studies need to be conducted.

The purpose of this study is to present the development of a 1D elasto-viscoplastic (1DEVP) constitutive model to describe the time-dependent behavior of soft clays. Firstly, based on Bjerrum’s concept of time lines, a new concept of viscoplastic strain rate lines is proposed, and after the relationship between viscoplastic strain rate and viscoplastic strain increment is deduced, the 1DEVP model for soft clays is established. Secondly, the relationship between the newly developed model and other three well-known elasto-viscoplastic(EVP) constitutive models are theoretically discussed. It demonstrated that the new model with no essential difference from the other three EVP models is simpler in form and clearer in physical meaning than others. Thirdly, the established 1DEVP is used to develop analytical solutions for consolidation-creep coupling effect, strain rate effect, stress relaxation effect. Lastly, based the one-dimensional consolidation tests on Ningbo soft clays, the model parameters is calibrated. Furthermore, the model is used to simulate 1D consolidation-creep tests on Ningbo soft clays, 1D staged constant rates of strain tests on Wenzhou soft clays and 1D relaxation tests on Hongkong marine clays. The comparisons between experimental results and simulations show that the newly proposed 1DEVP model has a good predictive ability for the one-dimensional time-dependent behavior of soft clays.

The generations of blocks and discontinuities network model of rock mass are the basis of various mechanical analyses and field analyses. The cutting process of complex blocks is analyzed, the corresponding algorithms are described, and a C++-based program is developed. The concept of oriented shell is added to the data structure of block in order to describe the multiply-connected characteristics of block. The discontinuities can be simple convex polygon and complex concave polygon. General blocks are cut by finite discontinuities, through calculating the intersecting lines of faces, generating loop vectors by searching intersecting lines, generating face vectors by analyzing the relationship of loop vectors, generating shell vectors by topologically searching face vectors, generating blocks by analyzing the relationship of shell vectors and so on, and then a general method of block cutting is proposed. The validity check of obtained loop vector, face vector, shell vector and blocks is conducted during the cutting process. The final blocks and discontinuities network model are obtained when satisfying with topological validity check. Four examples are selected to validate the method. The results show that the method can effectively cut complex blocks and the discontinuities can be complex polygon including concave surface, and indicate that the method is universal.

Current research on geocell-reinforced embankments is mainly concerned with the static load condition, and little attention is paid to dynamic loading conditions. To investigate the mechanical property of geocell-reinforced embankment under stepped cyclic loading, a series of model tests is performed, using the loading device of USTX-2000, on geocell reinforced embankment with different reinforced layers, different heights of geocell and different welding distances. The vertical and normal deformations at the slope of embankment are monitored under stepped cyclic loading, and compared with those of the embankment under fixed amplitude cyclic loading and static loading, by which the difference of mechanical properties under different loading schemes are analyzed. It is shown that geocell can significantly increase the capacity of the embankment under stepped cyclic loading condition, and reduce the vertical cumulative settlement. At fixed reinforcement spacings, the reinforcement effect of two or more layers is more significant than that of single layer. Both the increase of the height and the decrease of the welding space of geocell can improve the capacity of the embankment under stepped cyclic loading conditions and reduce the vertical cumulative settlement. Reinforcement can reduce the normal deformation of slope of embankment under stepped cyclic loading. Both the increase of the height and the decrease of the weld space of geocell can reduce the normal accumulated deformation of the top and the middle of slope in the same amplitude of stepped cyclic loading. Under the effect of stepped cyclic loading, when the number of vibration is greater than or equal to 8 000 or the amplitude is greater than or equal to 80 kPa, the vertical cumulative settlement of embankment exceeds the fixed amplitude cyclic loading. When the number of vibration is greater than or equal to 9 000 or the amplitude is greater than or equal to 90 kPa, the normal accumulated deformation of the top of slope exceeds the fixed amplitude cyclic loading. Under stepped cyclic loading, both the vertical cumulative settlement of embankment and the normal accumulated deformation of slope are larger than those under static loading, and the reinforcement can effectively reduce the difference between stepped cyclic loading with static loading in the normal accumulated deformation of slope.

Cohesion and friction angle are reduced by the same amount in the traditional strength reduction method, and the reduction parameter at the critical state of the slope is considered as the safety factor of this slope. Thus, the strength reduction method based on single reduction parameter is defined as equivalent proportional reduction method. For a target slope with the certain slope configuration and rock density, the combination of cohesive and friction coefficient, at which the slope is at the critical state, is not unique. The strength reduction by the same amount is simply one of possible choice, but not necessarily the best one. Based on the above understanding, the - critical state curve is first established under different slopes shapes and rock mass densities; then it is proposed that cohesive and friction coefficient are reduced along the short path from the current coordinate of strength parameter to - critical state curve, and a scheme of matching reduction is proposed; finally, the difference between these two reduction methods is analyzed through three examples. The results show the critical states of equivalent proportional reduction method and double reduction method are not the same, the range of the former is larger than the that of the latter

The similarity model test is one of the important means to study the engineering properties of red-bed soft rock; the test result reliability depends on the correct selection and rational proportion of a series of similar materials. In view of the softening characteristics of the saturated soft rock, a series of mixture ratio tests on similar material with orthogonal design method is conducted by adding the quicklime into the traditional similar materials. Four influential factors are considered in all the tests, including the ratio of iron pore powder to barite powder to aggregates, the content and ratio of iron pore powder to iron pore powder plus barite powder, the content of gypsum, and the content of quicklime. Each factor is classified into 5 levels. In total, 610 specimens in 25 groups were prepared. For each group of specimens, the physico-mechanical indices, such as softening coefficient, density, unconfined uniaxial compressive strength, tensile strength, cohesive force, etc. were tested. On the basis of these test specimens, the influential factors of the concerned parameters were analyzed; meanwhile the applicability of the ratio result were examined. At last, the similar materials of three typical soft rock specimens were prepared to test the softening coefficient; and a comparison analysis was also carried out with the real soft rock test. The test results reveal that: 1) the physico-mechanical properties of the similar material fabricated according to the similarity ratio proposed in this paper may vary considerably, and this can meet the requirement of the geomechanical model test of soft rock; 2) the physico-mechanical properties of the similar material demonstrate an evident variation law that facilitates the identification of the optimum mixture ratio for any soft rock material; 3) the softening coefficients of the saturated similar material, in the three comparative tests, are consistent with the real soft rock to a great extent.

Slope instability is a dynamic progressive failure process from qualitative to quantitative, which is one of the key points and difficulties in slope investigations. A theoretical framework of the slope progressive failure is constructed by considering strain-softening behavior of geotechnical materials and the calculation of kinetic. The simulation of slope progressive failure is implemented into the dynamic explicit solution module of ABAQUS software. The extension process of shear band is revealed according to the developments of plastic strain. The partition evolution law of material parameters at the slip surface is determined by the softening constitutive model. The slope slip surface is obtained by the equivalent plastic strain. Slope can be divided into landslide body, sliding zone and slide bed by the location of the sliding plane. The development of kinematic variables of internal characteristic points in the partition zones is studied to reveal the progressive failure process of the slope. Based on temporal-spatial distribution of material parameters along the sliding plane, the safety factors of slopes are obtained at different evolutionary stages using the vector sum method. By comparing the vector sum method with the Bishop method, it is found that the safety factors corresponding to peak and residual strength parameters are relatively close using these two methods. The sliding plane determined by vector sum method is also located among sliding planes by Bishop method, which to a great extent demonstrates the procedure is rational and reliable. Finally, the influence of material softening characteristics on slope stability is analyzed. It is noticed that when the residual cohesion is increased but other parameters are constant, the sliding plane is deepened, the initial value of safety factor is reduced, the appearance time of rapid decreasing phase of the safety factor is delayed, the duration of rapid decreasing phase is extended, and the safety factor after slope stability is increased. When the threshold of the corresponding equivalent plastic strain of the residual cohesion is increased but other parameters are constant, the sliding plane is deepened, the initial value of safety factor is reduced; however the appearance time of rapid decreasing phase of the safety factor is substantially constant, the stable time is delayed, and the stability safety factor is slightly increased .

To investigate the influence of wetting-drying cycles in seasonal variation on the long-term performance of red clay, the traditional sample preparation method is improved, and the moisture content variation of compacted soil during wetting-drying cycles is controlled in the regime of saturation from the optimum water content to full saturation. Considering the dripping humidification, the 2% and 4% water content samples experienced by the wetting-drying cycles were prepared. The consolidated drained triaxial shear test and compression test were conducted to analyze the effects of wetting-drying cycles and moisture change amplitude on the engineering properties of compacted red clay. The results show that cohesion c and internal friction angle decrease with increasing cycle number N, and the decrease is significant in the first wetting-drying cycle and then becomes stable. However, the rate of decrease of is less than that of c, and the decrease of shear strength under low confining pressure is more drastic than that under high confining pressure during the cyclic wetting-drying process. Compression modulus E1-2 also decreased with increasing N and the decrease rate is less than that of c but larger than that of . Similarly, c and E1-2 decrease slightly with increasing moisture change amplitude of the cyclic wetting-drying compacted red clay samples. At the moisture content in range of 4%, the compacted red clay shows relatively stable mechanical behavior. It is suggested that the red clay should not be directly used as the backfill material, but can be used as the backfill material of the embankment core, if the drainage and slope protection measures are taken effectively.

Unsaturated hydraulic properties of rock fractures have been widely studied in geotechnical, energy and environmental engineering. In this study, the three-dimensional laser scanning testing is used to obtain the morphology properties of a granite fracture; and then the influences of morphology properties on the unsaturated hydraulic properties are analyzed. By analyzing the variation of aperture distribution under various stresses including tensile stress, compressive stress, shear stress, a model is established to describe the relationships between capillary pressure and saturation of rock fractures under complex loading. A relative permeability model for unsaturated fractures is developed based on the analysis of the surface microcosmic features. The model is verified by experimental data subsequently. Finally, a variational relative permeability model for unsaturated fractures is established. This work is helpful for the research on hydro-mechanical coupling mechanism of unsaturated fractured rock masses.

To investigate tunnel vertical earth pressure caused by surface surcharge, a finite element model is established based on the measured results of tunnel structure deformation and earth pressure of indoor model test. The developed model is used to analyze the influence of the compressibility of soil layers acrossed by tunnel and soil layers overlaying on tunnel on tunnel structure deformation and surrounding soil pressure under surface surcharge. The results show, the compressibility of soil acrossed by tunnel is smaller, the vertical earth pressure on the tunnel caused by surface surcharge is greater, the corresponding tunnel lateral earth pressure is smaller, which is more likely to cause lateral elliptical deformation in the tunnel structure. The lower the compressibility of tunnel overlaying soil is, the lower the surface surcharge-induced vertical earth pressure on the tunnel is, the smaller the deformation of tunnel structure is. The vertical earth pressure of the tunnel caused by surface pile soil(surcharge) is greater than that calculated by the theory of soil column. Thesis results can provide some references for the design and operational phase control of subway shield tunnel in soft clay.

Based on a large number of test results and accumulated test experiences, some specific principles of scale of test gradation are summarized. The influence of gradation situations in coarse aggregate on its filling relation is investigated based on the variation of the content of , where is the aggregate with diameter ≤5 mm. It is shown that the more excellent the gradation is, the better the filling relation of the coarse aggregate is; and the coarse aggregate can reach a larger dry density; otherwise, the worse filling relation of the coarse aggregate makes it difficult obtain a larger dry density. Under low confining pressure, the influence of gradation situations on dynamic properties of rockfill is not obvious; and with the increase of the confining pressure, the dynamic properties of rockfill with good gradation obviously become satisfactory. Owing to its good roundness of particles, the influence of gradation situations of sandy gravel on its dynamic properties gradually decreases. According to the above rules, for high earth-rock dams, it is necessary to adopt coarse aggregate with good gradation as dam materials; and the advantage of mechanical properties of the coarse aggregate under high stress may be employed to control the integrated deformation of dams and to increase their aseismic performance.

Traditional studies on soil strength are mainly concerned with the soils subjected to the compressive shear stress state, and only a few addressed the strength characteristics of soils subjected to uniaxial tension, triaxial tension or combinations of tenso-shear. Based on the analysis of the existing results, a tenso-shear coupling strength model is developed with considering various combinations of tenso-shear in saturated clay. The failure envelope and the peak strength line as well as the strength under multidirectional tension are analyzed, and the corresponding strength formulations with specific derivation process are obtained. Different types of saturated clay are selected for the uniaxial tensile test, direct shear test and triaxial shear test. It is shown that the proposed model performs very well. Due to the complete theoretical basis and simpler strength formulation, the proposed strength model can be conveniently applied in practice, yielding better results in understanding and analyzing the behavior of saturated soils subjected to combinations of tenso-shear, such as slope instability caused by rainfall infiltration or earthquake loading.

A basic solution for ground settlement caused by horizontal movement of retaining wall is derived. And then a simplified method is proposed for computing the ground settlement caused by arbitrary mode of movement of wall based on the basic solution. The FEM numerical solution is then used to verify proposed simplified method by comparison of their solutions to ground settlement, which is caused by three modes of horizontal rigid movement for translation, rotation about bottom and rotation about top, respectively. Finally, the simplified solution is applied to prediction of measured data in the field. The predicted settlement curve of ground is similar to the measured one; and the location of the maximum settlement is close to the measured one as well, showing that the simplified analytical method is practicable.

The critical fracture characteristics of rock, such as the length of process zone and crack mouth opening displacement(COD), are the fundamentals to solve the fracture problems of rock by using fracture mechanics. Traditional measure methods, such as strain gauge and LVDT, etc., cannot obtain the full-field displacement of rock specimen. In that case, it's difficult to achieve the above mentioned critical fracture characteristic of rock. Digital image correlation (DIC) method is a optical measure method which can obtain the accurate full-field displacement of the specimen by comparing the original image and deformed image. A series of three-point bending tests is performed on the semi-circular rock specimens(relatively hard marble and relatively soft yellow sandstone), and the DIC method is used to measure the full-field displacements of the specimens. The obtained displacements are further analyzed, and the critical fracture characteristics of the specimens including the length of process zone and the crack mouth opening displacement are revealed. It is shown that the length of process zone in marble is obviously shorter than that in yellow sandstone, while the values of COD of yellow sandstone are higher than the ones of the marble with same crack length, which demonstrates that the soft yellow sandstone is prone to be affected by boundary effect. Those researches can help us to further understand the fracture process of rock ,and to solve rock engineering problem using fracture mechanics.

Drying-wetting cycle usually occurs in compacted embankment fill due to periodic variation in precipitation and evaporation, which affects the long-term stability of embankment and subgrade. A series of compression tests are performed to measure the variation in vertical compression strain of compacted loess subjected to cyclic drying-wetting by using pneumatic oedometer apparatus so as to study the effect of drying-wetting cycle on deformation behaviors. The relationships between vertical compression strain and vertical pressure ( - ) at different initial compaction degrees are fitted with the existing models; and the relationships between secant modulus and as well as number of drying-wetting cycle are established. The results show that of compacted loess tend to increase with numbers of drying-wetting cycles and then keep fairly stable after 5-7 cycles; Moreover, the increment of remarkably affect by . The - curve of compacted loess can be depicted by the Gunary model exactly under uniaxial compression condition, and its behaviors are unchanged due to repeated drying-wetting cycle; The relationship between and as well as also can be formulated in exponential form without considering the effect of water content. It can be concluded that the change in deformation behaviors is irreversible after the compacted loess is subjected to drying-wetting cycle.

To investigate the fracture mechanical behaviors and acoustic emission characteristics of of Beishan granites with different particle sizes, we used the MTS815 Flex Test GT rock mechanics testing system and acoustic emission three-dimensional positioning monitoring system to conduct three-point bending tests on the four different particle sizes of Beishan granites. The experimetal results show that under the condition of three-point bending test, there is still comparative large bearing area in the section of specimen after the peak. The bigger is the particle size , the lower is the peak load is, the lower is the yield load’s percentage of peak load, and the smaller is the fracture toughness value. But the bigger particle size allows larger total deformation, and the residual strength is also higher. For the fine-grained or medium-fine grained specimens with a relatively low strength, the acoustic emission events firstly appear near the tip of notch, where tensile stress is larger. In the coarse-grained or medium-coarse grained specimens, acoustic emission events initially emerge in the upper part in which point load is applied. Regardless of the fine-grained or coarse-grained granite, the acoustic emission events concentrated in the stable stage of crack extension which is the main stage of rock fracture. The statistical results indicate that the bigger the particle size of granite, the more the acoustic emission events is, and the broader the distribution of acoustic emission events is.

In the current large-strain consolidation theory of double-layer soft soil foundation, it is assumed that the seepage in the soil follows Darcy’s law. However, it has been recognized that the seepage in soft clay may deviate from Darcy’s law under low hydraulic gradient. The effects of non-Darcy flow, non-linear characteristics, and time-dependent loading are discussed comprehensively, and a model for the finite-strain consolidation is developed in Lagrangian coordinate by employing the excess pore water pressure as variable. Meanwhile, the corresponding finite difference solutions for this model are provided. On this basis, the numerical solutions for the proposed consolidation model are verified by comparing with the numerical solutions for the large-strain consolidation of a single soil layer foundation with non-Darcy’s law. The influences of upper layer parameters m1、i11 and sublayer parameters m2, i12 on consolidation behaviors and the differences in the dissipation of excess pore water pressure and in the consolidation settlement are investigated. The results show that the influences of m1, i11 on consolidation behavior are more evident than m2, i12. The dissipation rate of excess pore water pressure in upper layer increases with the increase in the value of m2 or i12, whereas the consolidation rate of double layer foundation decreases with the increase in the value of m2 or i12. The consolidation rate of double layer foundation under large-strain assumption is larger than that at small-strain, whereas the final settlements under both large-strain assumption and small-strain assumption are the same.

The physical model test is an effective method for landslide and its anti-slide structure research. The application of infrared thermal imager to landslide physical model test is based on infrared detection and thermal imaging principle. With the infrared thermal imager, the surface temperature of the landslide model can be effectively, accurately and real-timely detected and recorded, which helps to analyze the temperature field variation of the anti-slide pile area and study of the effect of anti-slide pile soil aching from the angle of energy. The experimental results reflect the distribution of temperature field in the anti-slide pile area, prove the existence of the soil arch effect from the temperature field, and reveal the process and mechanism of soil arch formation; Anti-slide pile effectively prevent the landslide thrust expand to the front of the pile, and play the reinforcement effect. The stress field, displacement field and energy of rock and soil masses are interrelated in the process of landslide developing. The application of infrared thermal imaging technology presents a new research method to study the soil arch effect of anti slide piles, which makes it possible to study landslide control structures such as anti-slide piles from the aspects of energy; and its feasibility and reliability has been proved in the physical model test.

Rock consists of mineral grains at the mesoscale. The grain boundaries form an interface network and grain sizes are also uniformly distributed. Besides mechanical properties of mesostructure, the geometrical characters (i.e., topology of grain interface network and grain size distribution) also contribute to fracture energy. The aim of this paper is to explore the effects of geometrical characters on the fracture energy and get insight into the mesomechanism of macro fracture energy of rock. Three-point-bending tests are carried out on marble specimens to obtain the fracture energy of marble. Then a rock slice is taken from fracture surface to obtain the image of mesostructure of rock using the scanning electron microscopy (SEM). The technique of digital image processing is used to obtain the grain interface network. Through analyzing of the interface network, it is found that both the topology of interface network and the grain size distribution present a strong fractal feature. There exists a close correlation between macro fracture energy and the fractal dimensions, in which the fracture energy increases with the increase of the fractal dimension. The relationship between fracture energy and fractal dimension of interface network can be well characterized by the quadratic function, while the relationship between fracture energy and fractal dimension of grain size distribution can be characterized by the linear function. The research results reveal a mesomechanism of macro fracture energy.

A series of shaking table tests on the saturated sands are conducted. The embedded aluminous tube in the sands is pulled laterally to simulate relative motion between liquefied sands and tube. The theories and methods of fluid mechanics are introduced to analyze the mechanical effects induced by the flows of liquefied sands. The real-time apparent viscosity of liquefied soils and the viscous shear force acting on the tube wall are calculated according to the measured pulling force. Fluid characteristics and flow effects of the liquefied soils are analyzed. For the soils during shaking and after shaking, their flow effects are rate-dependent and dependent on the excess pore water pressure. The influences of initial relative density on the flow effects are also discussed. The experimental results show that apparent viscosity of the liquefied sands is positively correlated with their initial relative density. The apparent viscosity and the viscous shear forces acting on the tube wall are negative correlated with the excess pore water pressure in the soils. Furthermore, the viscous shear forces acting on the tube wall are strongly rate-dependent. Therefore, the rate-dependent soil-structure interaction is suggested to be considered for the seismic analysis of structures in the liquefiable site.

The outward dislocation sliding of roadway sides with mudstone interlayer in a large height mining face is a new research topic in roadway support. Convergence measurement and borehole observation are conducted on coal seam 5-2 with soft interlayer in Sandaogou Coal mine. It is found that there are vertical cracks in the roadway roof, and side failure zones occur near the mudstone interlayer but failure zones are more obvious in the pillar side. The roadway deformation is mainly caused by adjacent mining goaf. Both numerical simulation and physical simulation are used to reveal the mechanisms of side outward dislocation sliding. The lateral abutment pressure of adjacent mining goaf induces vertical cracks at the roadway roof, which leads to the outward sliding of roof with the upper side along the soft interlayer. Thus, we propose a self-stable equilibrium arching support model of the roadway with a soft interlayer, and develop support principles to strengthen upper sides and control outward sliding failure, and design a support scheme for sides reinforced by long bolts and roof supported by cables. The designed scheme is verified in engineering practices.

Based on the role of prestressed anchor and the mechanical transmission mechanism of anchorage interface, an anchorage interface stress model is developed. Then an analytical method is proposed to calculate the anchorage stress by considering the shear slip of anchorage interface. On the basis of the prestress loss mechanism, various factors causing prestress loss are analyzed, and then a method for calculating prestress loss is developed under the influence of various factors. According to fundamental theory of the solution of stress distribution, a method is presented to estimate the initial tensile tonnage of the anchor cable. During the excavation of underground caverns, the stress increment is considered as safety margin of initial tension of the anchor cable, and the prestress loss can also be taken into account. From the above, a finite element calculation program is compiled, and a case study of an underground caverns shows that the proposed method is reasonable.

The least absolute deviation (LAD) method has some favorable properties, including intuitiveness, robustness, zero-error, predictability and generalization, compared to the least square (LS) method. The LAD approximation is a "mini-mini approximation", which is minimization of the minimum absolute value of error. Because the best result of the mini-mini approximation must be zero, the zero-error principle is indeed the basic principle of LAD method. The realization of the LAD solution is achieved by a "representative" data processing mode. Because the disturbance induced by the data with big error is removed, the LAD method has a good robustness. Since the representative data are selected according to different applications, the LAD method is widely applicable. When the LAD method is applied to prediction, the endpoint data can be set to be zero-error data, yielding the unequal weights of the data which are closer to the end point. As a result, the endpoint data have the greatest weight, and have a basis without errors, making the proposed procedure more reasonable and more accurate. Based on three engineering cases, the LAD method is applied to predicting the settlement process of rock and soil foundation. By comparing with LS method, it is concluded that: 1) The stability of data processing of LDA method is better, the fluctuation range is smaller, and the forecasted result is more accurate; 2) Ccntradictory and unreasonable phenomenon no longer appear as in the data processing of LS method; 3) when t→∞, although the limit settlement values cannot be verified, the prediction of LAD method is more reasonable than those of LS method, since the LAD method is built on the basis without errors.

In view of the anisotropy of deformation and strength of surrounding rock at different locations(vault and hance) in loess tunnel, the anisotropy of intact loess with respect to structural strength is empatically discussed. The intact loess is sampled from Wangjiagou tunnel along Baoji-Lanzhou passenger dedicated line. A series of tests including shear test, unconfined compression test, tensile strength test are conducted on the intact loess with sampling directions of vertical and horizontal directions. Considering the influence of structure of loess on its anisotropy, the scanning electron microscope (SEM) test is also performed to analyze the main reasons leading to the anisotropy of intact loess. The results indicate that the strain ratio of horizontal direction to vertical direction decreases with the increasing of confined pressure, but firstly increases and then decreases with the increasing of normalized deviatoric stress; the peak (singularity) increases with normalized deviatoric stresses in the manner of power function. A geometric model of anisotropy of structural strength of intact loess is presented; and related strength parameters are also proposed.

In hydropower engineering, engineering slopes on both river sides have characteristics of large scale, high and steep geometry, long service life and high stabilization requirement. Due to real-time changes of geological parameters, slope structure and excavation support, real-time evaluation of slope safety is a key technical issue in hydropower projects. A real-time simulation method based on numerical method is developed by considering the change of excavation progress during slope construction process, the new geological information disclosure, the modification of supporting plans and the correction of numerical parameters. This procedure focuses on real-time analysis and control of the safety of high-steep slopes, which is seldom reported in the previous studies. Based on database technology and secondary development of ABAQUS software, the information of dynamic mapping the construction schedule, geology and support is implemented into the numerical model. Finally, a real-time analysis and control system for slope safety is developed to realize the simulation and forecast of the real-time slope safety status.

Blocks are one of potential failure modes appeared mainly in structurally-controlled rock mass. Factors of safety (FoS) for blocks identified in underground caverns of a huge hydropower station are obtained using the limit equilibrium method and strength reduction method. A simple graphical method for stability analysis based on block parameters is proposed by the calculated FoS of the block, and the reliability of results is validated by in-situ monitoring data. It is found that the FoS values obtained by these two methods are inconsistent even for the same block, which thus further influence the stability judgment. Results obtained by the strength reduction method are conservative since it is affected by software algorithm realization and mesh sizes. FoS of simple blocks (i.e. blocks consist of 3 joints and excavation boundary) can be easily obtained by the limit equilibrium method, while FoS of complex blocks (i.e. blocks consist of more than 3 joints and excavation boundary) are more conveniently calculated by the strength reduction method. Four block parameters, namely vertical in-situ stress, block volume, the maximum vertex depth, and equivalent joint shear strength, in combination with the stability partition chart, are applied to assist in the stability judgment. Once blocks are determined as unstable, the proper support and reinforcement measures should be adopted promptly. These findings can be used to other cases with similar unstable problems of blocks.

Based on the fictitious loading upper bound limit analysis approach i.e. total-displacement-loading approach based on extend mobilizable strength design method (T-EMSD), by which the traditional upper bound limit analysis is equivalently converted into iteration of elastic solution, the elastic finite element based T-EMSD method is realized in commercial FEM software ABAQUS. The numerical T-EMSD method is applied to analyzing the shaft of laterally loaded pile in undrained clay. The load-displacement curves obtained by the proposed method are generally consistent with those obtained by elasto-plastic FEM, and the corresponding ultimate bearing capacities are close to plastic solutions. Under large displacements, the mechanisms underlining the elastic finite element based T-EMSD analysis transform from elastic into plastic mechanisms through iteration process, which finally become similar to the plastic collapse mechanisms. Compared with other mobilizable strength design (MSD) methods, the proposed approach are more accurate in the analysis of laterally loaded pile. Moreover, since corresponding mechanisms can be simultaneously obtained in the process of the numerical T-EMSD analysis, it can be utilized in studying complicated problems whose plastic deformation mechanisms are not acquired yet, and in verifying the results of other numerical methods.

The accuracy requirements of numerical manifold method (NMM) can be reached by changing the density of mathematical covers and the order of cover functions. In this study, NMM coupled with nonuniform rational B-splines (NURBS) and T-splines in the context of isogeometric analysis is proposed. Computational formula for a 9-node NMM based on quadratic B-splines is derived. For the case of crack propagation problems where singular fields around crack tips exist, local refinement technique by the application of T-spline discretizations is incorporated into NMM, which facilitates a truly local refinement without extending the entire row of control points. A local refinement strategy for the 4-node mathematical cover mesh based on T-splines and Lagrange interpolation polynomial is proposed. The results from numerical examples show that the 9-node NMM based on NURBS has higher accuracies. The local mesh refinement using T-splines reduces the number of degrees of freedom while maintaining calculation accuracy at the same time, and T-splines refinement method is a natural truly local refinement algorithm.

The displacement method of geotubes is widely applied to improving soft soil with a high water content and low bearing capacity in reclamation dam and road constructions due to its easy construction technology and significant treatment effect. When the weight of the geotubes is equal to the bearing capacity of foundation soil, the soil attains limiting equilibrium condition. It is necessary to estimate the relationship between the penetration depth and height of the geotubes. In this paper, model tests of three widths of geotubes are carried out, and the relationships between penetration depth and height, the displacement and upheavals of soil are revealed. In order to simulate the process of penetration of three geotubes, the Coupled Eulerian-Lagrangian approach(CEL) in ABAQUS software is used to simulates the process of penetration of geotubes. In this simulation, the geotube and silt are described as Lagrangian body and Eulerian body, respectively. The relationships of penetration depth with height of the geotube as well as displacement of the silt are obtained. The accuracy of the simulation approach is verified by model test results and theoretical result, showing that the coupled Eulerian-Lagrangian approach can be applied to predicting the penetration of geotubes very well.

Lateral spreading due to liquefaction of sandy soils is a significant cause of damage to the important engineering facilities and buildings in earthquakes. FLIP ROSE is a two-dimensional finite element analysis program for seismic liquefaction problems. A strain space multiple mechanism model (cocktail glass model) is implemented in the program for analysis of the granular material. A two dimensional finite element model is proposed to predict and analyze liquefaction-induced lateral spreading in gently sloping ground under the seismic loading. This numerical simulation model is verified using the excess pore water pressure, the lateral acceleration, the arias intensity and the lateral displacement time histories recorded in three sets of free-field lateral spreading centrifuge tests. The computed response shows good agreement with the centrifuge test measurements. The final lateral displacements in the gently sloping ground with the different gradients are predicted by this simulation model after sandy liquefaction. It shows that the gradient of gently sloping ground plays a more and more important role in the liquefaction-induced lateral spreading as the depth increases.

The failure modes of slopes include not only the shear failure but also the tensile failure, especially in the back of frontier of slopes. The traditional yield criterion of Mohr-Coulomb is difficult to deal with the shear failure with the tensile crack modes in the lower bound limit analysis. The paper reformulates the yield criterion of Mohr-Coulomb with respect to the linearized spatial discretization in the lower bound limit analysis based on finite element method. Meanwhile, a pseudo cohesion is introduced to assure that neither the Mohr-Coulomb yield criterion nor the tensional failure criterion is violated, and a lower bound limit finite element method is developed for analyzing the tenso-shear failure of slopes. The numerical examples illustrate the validity of the proposed method, while showing that if only the shear failure is considered, the factor of safety would be overestimated in the slope stability analysis.