It is of great significance to study the effect of repeated fluctuation of reservoir water level on the stability of slope. Considering the problem of water-rock interaction, uniaxial and triaxial compression tests were conducted on argillaceous sandstone at different dry-wet cycles, respectively. The results show that the deterioration degrees from high to low are as followed: peak strength, cohesion and internal friction angle. Meanwhile, the deterioration degrees of mechanical parameters increase with increasing the times of dry-wet cycles. In combination with Drucker-Prager criterion (D-P criterion) that widely used in geotechnical plastic mechanics, a new D-P yield criterion is put forward for argillaceous sandstone under different confining pressures and dry-wet cycles. By comparing the experimental value with the theoretical peak value calculated by the modified D-P yield criterion, the error is found less than 10 percent. Hence, the D-P yield criterion is feasible for argillaceous sandstone under dry-wet cycles in practical applications. This study provides a theoretical basis for yield strength of rock under the arbitrary confining pressure in practical engineering.

The values of rock dynamic parameters directly influence the results of dynamic calculation in rock structural engineering, which shall be validated with the results obtained from practical engineering projects. As the frequency and amplitude of actual dynamic loading could change within a specific range, a series of cyclic loading and unloading experiments was conducted by considering the variations of these two factors. The effects of these two factors on the dynamic characteristics of sandstone were analysed, such as dynamic stress-strain hysteresis curve, dynamic elastic modulus, damping ratio and damping coefficient. The results showed that the shape of stress-strain hysteresis loop was closely related to the frequency and amplitude of cyclic loading and unloading. It was found that the higher the frequency was, the plumper the hysteresis loop was, and the greater the cyclic loading amplitude was, and the narrower the hysteresis loop was. When the frequency of cyclic loading increased from 0.02 Hz to 1.00 Hz, the damping coefficient of sandstone gradually decreased, whereas the damping ratio and dynamic modulus slowly increased. The varying ranges of damping coefficient and damping ratio were relatively larger, while the growth range of dynamic elastic modulus was smaller. When the stress amplitude of cyclic loading increased from 10 MPa to 35 MPa, the damping ratio of sandstone gradually decreased, whereas the damping coefficient slightly changed and the dynamic elastic modulus slowly increased. However, the effects of stress magnitude on the dynamic elastic modulus were more significant. Therefore, this study can provide useful references for reasonably selecting dynamic parameters of rock mass.

Model tests and numerical simulations were carried out to investigate load transfer characteristics of short stiff piles with big caps. With increasing loads, the lateral shear slip damage occurred on ground soil underlying the pile cap, whereas punching shear was found in the vicinity of the pile tip for soft foundation reinforced by short stiff pile with a large cap. Thus, a novel block displacement method was proposed to analyse the load-displacement relationship of ground soil underlying the pile cap, and the load transfer method was put forward for the lower part of the pile. Based on the displacement compatibility of the upper and lower parts for a single pile, the load-displacement relation (P-s curve) for the short stiff pile with the large cap was obtained by using the iteration method. Meanwhile, the calculated P-s curves were compared with the measured results in model tests, which showed a good agreement between them. Furthermore, the presented calculation model investigated the effects of pile cap-body diameter ratio D/d as well as the ratio between the pile length and pile cap diameter L/D on the bearing capacity of the short stiff pile with large pile cap. The results demonstrated that suggested ranges were D/d>2 and L/D<10 for the application of pile reinforced soft foundation with pile caps.

This study aims to establish a criterion of acoustic emission (AE) and charge signal for different failure forms of coal at various loading rates. Physical experiments were conducted on coal from a work-face of Wulong mine, and the AE and charge signals were monitored during the failure process of coal at different loading rates. Then we thoroughly investigated the relationships among AE-charge signals and mechanical properties, cracking, energy release at various loading rates. The results showed that with higher rates and earlier stress adjustment, uniaxial compressive strength improved, the release of accumulated deformation energy aggravated, failure modes changed from static load-induced instability to similarity-dynamic-load induced instability, more dynamic characteristics were present, and the risk of compressible rock burst of coal increased. At different loading rates, the AE-charge signal variations were consistent with stress mutation, which were further analysed quantitatively. Hence, based on the obtained AE-charge signals, we proposed a criterion for different coal failure modes. In different mines and work-faces, we could forecast or make an early warning to the risk of compressible rock-burst induced by mining, according to the monitored AE-charge signals in combination with variations of the mining pressure field.

Tunnel leakage happens very often through shield tunnel joints and grouting holes. It can induce the pore pressure decrease around the tunnel and then result in tunnel and ground settlements. In tunnel engineering, Grouting is usually used to prevent tunnel leakage. What’s more, the synchronous grouting is considered to be the first barrier of tunnel leakage. To investigate the effect of grouting on tunnel leakage, the permeability of inert and cement grouting is firstly studied using laboratory tests. The experimental results show that the permeability of both inert and cement grouting decreases with time and consolidation pressure at the early ages of grouting. The stable permeability of inert grouting is higher than that of the cement grouting. Thus, the cement grouting rather than the inert grouting can be used to prevent tunnel leakage. Then, the predicting method of tunnel leakage-induced pore pressure decrease is suggested considering the effect of grouting and tunnel lining. The calculations results show that the cement grouting can reduce the tunnel leakage-induced pore pressure decrease. Furthermore, the effect of grouting on tunnel leakage-induced pore pressure change depends on the permeability of tunnel lining. Finally, a dimensionless parameter RP is proposed to describe the relationship of tunnel leakage-induced pore pressure change and the permeability of grouting and tunnel lining. The results show that the grouting can reduce the decrease of leakage-induced water pressure, while, the effect of grouting on tunnel leakage depends on the relative permeability of tunnel lining and surrounding soils.

This study aims to extend the scope of cutting plane algorithm for calculating the constitutive equations in the bounding surface model. Moreover, a new algorithm should inherit the simple and efficient framework in the classical elastoplasticity model. The maximum limit of incremental step was employed to propose a modified cutting plane method by using an adaptive sub-stepping algorithm. This method was also applied to an improved bounding surface model for structured clay. Experiments were conducted on Shanghai clay to validate the bounding surface model using the modified method. The validation of the modified method was proven through obtained results. According to parameters of Shanghai clay, numerical studies were also carried out to simulate undrained triaxial tests, undrained simple shear tests, undrained triaxial cyclic tests and oedometer test, respectively. The maximum limit of the incremental step was further determined. Finally, an undrained single shear test was simulated under the large strain step condition by using the modified method. As a result, the accuracy and stability of the modified cutting plane method were verified.

To study the post-liquefaction stress-strain relationship of saturated Nanjing sand, shear moduli reduction in the liquefaction state, the dilation state and the unloading state were investigated under different loading conditions. As results, the effective confining pressure, the initial shear loading, and the amplitude of cyclic loading have less effect on shear moduli reduction of saturated Nanjing sand in the liquefaction state and the dilation state. However, these factors have great effect on shear moduli reduction of saturated Nanjing sand in the unloading state. As the effective confining pressure and the cyclic loading amplitude increasing, the post-liquefaction shear moduli of saturated Nanjing sand in unloading stage are all stronger and the shear moduli soften faster under the subsequent cyclic loadings. Meanwhile, as the initial shear stress increasing, the post-liquefaction shear moduli of saturated Nanjing sand are also stronger, but it softens slower under the subsequent cyclic loadings. Based on the test results, the decay function of the shear moduli of saturated Nanjing sand in different loading stages are also given.

To improve the waste slurry discarded from a pile foundation project in Shanghai, the curing agent (calcium lime) is used as an additive material. The deformation on the waste slurry surface with various additives is measured during settlement at rest. The vacuum preloading tests are conducted for the slurry flocculation. The variation in water discharge during vacuum preloading and the particle distribution in the slurry flocculation at various locations as the tests completed are then investigated. The changes in the size distributions of the pores in the waste slurry are also explored using MIP tests before and after the vacuum preloading tests. The hydration mechanism of the calcium lime on the improvement of waste slurry is illustrated using the scanning electron microscope image processing. It shows that the settlement rate on the waste slurry surface accelerated after adding calcium lime, and soil particles are joined together by lime hydrate crystals to form the skeletonized soil-flocculation aggregates with larger volume, looser structure, and enhanced permeability. Among the calcium lime contents of 0.2%, 0.5% and 0.8%, the surface settlement of the waste slurry with the content of 0.5% is fastest both during the settlement at rest and in the vacuum preloading test. Most of the pore sizes of waste slurry decrease from between 3,184 to 11,308 nm before improvement to between 681 to 3,244 nm after vacuum preloading with flocculation, indicating that the total pore volume decreased by 60% to 79%.

Based on the cohesive crack model, a virtual node extended finite element method (XFEM) was developed to simulate the fracturing process of quasi-brittle materials, such as rock and concrete. Numerical principles and governing equations of this method were also proposed. Typical numerical examples were presented, including tension fracture of the three-point bending beam, I–II mixed mode fracture of a single edge notched specimen and fracture of multiple cracks in the Nooru-Mohammed experiment. Moreover, these results were compared with existing solutions or experimental results. It is found that this method is suitable to simulate fracturing process of quasi-brittle materials dominated by opening-mode cracks. Pre-assignment of crack-growth paths is not necessary for the proposed method, in comparison with the node-separation finite element method (FEM). Contrary to the plastic-damage FEM, this method also can reliably simulate the propagation of multiple cracks. Meanwhile, it is not necessary to introduce crack-tip elements and calculate stress intensity factors in comparison with the normal XFEM. Lastly, it is particularly applicable to the acquirement of convergent computational results by contrast with the XFEM with higher-order enriched elements. Moreover, the developed method can be easily embedded into conventional finite element software to solve complex problems by applying powerful nonlinear computational functions of the latter for its displacement description based on elements.

Large shaking table model tests were conducted to investigate the seismic response of low-angle soil slope in permafrost regions, main influencing factors and the evolution process of the landslide subjected to vertical and horizontal seismic loadings, respectively. When shaking table tests were conducted on the 8°low-angle slope, the horizontal natural frequency of vibration decreased sharply at the failure state, however there was no apparent change in the vertical direction. The failure of slope model was characterised by completely slipping along the interface between the unfrozen soil and ice layer, while the deformation of soil was not observed inside the sliding body. The amplification coefficients of peak ground acceleration (PGA) increased with the elevation of the slope. Compared with little change of amplification coefficients before sliding, more obvious changes occurred along the surface of the slope after sliding. The peak amplification coefficient subjected to the horizontal seismic loading was greater than that subjected to the vertical one. The peak amplification coefficient in the soil-ice interface was obviously lower than that in the upper soil and lower ice layer. The pore water pressure in the soil-ice interface of the slope increased when the input acceleration reached a specific value. The weak soil-ice interface and the increasing pore water pressure were the main internal factors inducing the permafrost landslide subjected to seismic loading.

In this study, a self-developed ‘THM coupled with triaxial servo-controlled seepage apparatus for gas infiltrated coal’ was employed to study the deformation and permeability laws of sandstone. Seepage experiments were carried out at the same gas pressure and under different unloading confining pressures. Then, a seepage model was established on the basis of strain parameters of sandstone. The results showed that gas infiltrated sandstone exhibited clear brittle feature during the failure process and sudden jumps occurred among the stress, strain and flow immediately after the peak stress point. Both changes of stress-strain and permeability showed obvious periodical characteristics, which means that the gas permeability was zero during pore and micro-fissure compaction, and elastic deformation stages (I and II). During unloading and yield stage (III), new fractures were initiated and the permeability increased slightly. During unloading failure stage (IV), the stress dropped significantly, the strain increased sharply, and the gas flow also increased greatly when fractures connected with each other, which indicated that this period was dominated by the permeability. The strain, stress, and gas flow all tended to be steady after failure (V). At failure state of sandstone, the radial strain and volumetric strain were far larger than the axial strain, due to the effects of gas pressure and the Poisson. The unloading failure mode of bedding sandstone was mainly dominated with extension and shear fracture along beddings. Based on Kozeny-Carman equation and fracture flow theory, the permeability model related to strain was established. Finally, the permeation mechanism of gas infiltrated sandstone at different failure states was revealed.

Loess is a soil with a special structure. Currently the studies of influence factors on loess structure are mostly qualitative analysis using single factor. Structural index reflects the initial structure of loess. To study the influence of the physical parameters on the initial structure, firstly, unconfined compressive strength tests are conducted on the undisturbed and remolded loess samples with different water content taken from Xi’an and Lanzhou. Next, the corresponding structural indexes of loess are obtained. The analysis shows that the value of structural index decreases with the increasing of water content, dry density, minimum density ratio, plasticity index and liquidity index. In addition, it is also related with the geological period and geographical distribution of loess. Secondly, according to the above conclusions, a comprehensive physical index is constructed by the fundamental physical indexes. The relationship between the structure and the comprehensive physical index is further analyzed. The empirical formulas are obtained which can directly calculate structure index quantitatively by the comprehensive physical index considering the distinction between the geographical zonation and geological periods. Finally, by the comparisons of the structure indexes obtained by test for another loess field in Xi’an and the structure indexes calculated by the empirical formulas herein, their rationality and applicability are verified. These formulas provide a simple and easy method for quantitative calculation of the structural index by physical indexes to measure in the future.

Numerical simulation is the most efficient way to investigate the mechanism of grout diffusion in jointed rocks, but one of the most critical challenges is to accurately track the grout flow in numerical simulation. This study was focused on the diffusion process of self-expanding grout in the flat fracture surface and considering characteristics of two-phase flow in the grout diffusing. Moreover, the VOF function was employed to describe the distribution of two-phase medium, and then was solved by Youngs method based on the principle of geometry. Thus, the tracking of grout moving interface was realised. Finally, the proposed method was examined by an example of the self-expanding grout with a known density freely diffusing with time in a plat fracture surface. The results showed that the analytical solutions were in good agreements with obtained locations and contours of the grout moving interface at different times in the example. Hence, the results verified that the proposed method had a high calculation precision and exhibited a sound effect of interface tracking. This study lays a solid foundation for developing the numerical method for simulating the diffusion processes of self-expanding grout flow.

Slurry liquidity has an important influence for grout diffusion and migration. By conducting orthogonal liquidity test on viscosity-varying grouting materials with different pumping time, the influence of water cement ratio and the additive agent on the properties of slurry flow are analyzed. Flow degree of time variation characteristics of the viscosity time-varying slurry are investigated. The changes of the flow of the slurry machine rational are discussed considering the effects of the additive agent on slurry hydration and hardening. Slurry liquidity in the application of grouting in cataclastic rock is studied. The results show that while increasing the amount of water cement ratio, additive #1, additive #2, additive #3, the flow of the slurry augmented. Water cement ratio is the most significant influence factor of fluidity, followed by additive #2, additive #3, additive #1. For the main factors of the influence on slurry viscosity followed by additive #3, additive #1, water cement ratio and additive #2. Slurry fluidity change presents three-stage development: slow increase, fast increase, and stable state. When nearing slurry pump time, slurry flow has significant change respectively. Grouting application in cataclastic rock mass shows that the use of grouting slurry considering time-varying viscosity can reduce refilling time for the ordinary cement slurry, shorten the grouting time, save the grouting material.

Diverse risk events and risk possibilities are involved in the design, construction and maintenance of highway cut slopes because slope risk and elements at risk are varying during these period. It is therefore necessary to conduct differentiated risk assessment and life cycle risk control. In this paper, a technical framework is proposed for life-cycle risk analysis, assessment and management of highway cut slopes based on a modified Hall three-dimensional morphology of systems engineering. Critical points of risk zonation, data collection and information cataloging during risk analysis in a typical design, construction and maintenance life cycle are clarified. Distinct definitions and analysis methods of failure probability in the process of hazard analysis are subsequently elucidated. Meanwhile, the definitions and measures to recognize hazard-affected targets and to evaluate their damage and space-time probability as well as vulnerability and other essential indices are determined in the process of hazard consequence analysis. The applicable methods for qualitative and quantitative risk estimation in different stages of slope engineering are also revealed. Technical procedures and key implementation points in each aspect such as slope risk assessment and benefits evaluation, comparison-and- selection and planning of solutions and countermeasures for risk regulation, monitoring, verification and modification are further discussed, respectively. Principles of determination of risk tolerance standard, ideas to enhance the accuracy and reliability of risk assessment are finally proposed. Research findings of this paper preliminarily standardized the procedures and technical program for risk assessment and management of highway cut slopes during their life cycles. It is hoped that this paper will bring forth new ideas and gradually develop and perfect the framework in practice.

null

To precisely describe the strength of rock materials under a variety of complex stress conditions, we developed a new nonlinear tri-parameter strength criterion based on the theory of twin T2 strength. Compared with experimental results of five types of rock, the rationality and applicability of the proposed criterion were discussed separately under five stress states, i.e., ?1>?2 =?3, ?1>?2 =?3, ?3=0 and ?1>0, ?2>0, ?3>0 and . The results shows that the parameters, A, B and C, can be determined by simple tests, though their physical meanings are reasonable. The different ultimate stress ratios and correspond to different strength theories. The limit line in the plane of the proposed criterion is among the single shear strength theory, Mises criterion, and the twin shear strength theory. Furthermore, the criterion conform with the triaxial symmetry and the convexity of yield surface of the isotropic material. The theoretical prediction curve can reflect the interval characteristics of the principal stress and the hydrostatic pressure effect in rock materials. Moreover, it is in good agreement with the strength trends of some materials at complex stress states.

Isotropic compression and conventional triaxial compression experiments are performed on the T2b marble samples taken from Jinping underground laboratory, China, and some issues related to the isotropic compression deformation characteristics, the effect of confining pressure on the post-peak behavior and the pre-peak progressive failure process have been analyzed and discussed. The test results show that the volumetric strain under isotropic compression presents a linear-nonlinear-linear change when subjected to a confining pressure increase. An apparent volume change inflection point (sensitive stress) is observed, which can be related to stress history. There is a brittle-semi brittle-ductile transition trend with increasing confining pressure in the results of the conventional triaxial tests, and the change of the strength is consistent with the change of the isotropic compression curves which is in isotropic pressure vs. volumetric strain plot. Furthermore, the increased confining pressure has an effect on each stage of the progressive failure process, and other stress thresholds of the marble increase approximately linearly with the confining pressure except the crack closure deviatoric stress threshold. The initial crack closure process of the marble occurs twice, one under isotropic compression and the other under the deviatoric stress loading condition. The above results of Jinping T2b marble is valuable for achieving a better understanding the rock failure mechanism and preventing disasters in deep underground engineering.

null

The stability and safety of the soil slope will be threatened when the thrust at front end of slope change due to rainfall infiltration. Rainfall simulation system and monitoring system are used to make a comprehensive and real-time monitoring of rainfall-induced slope failure process of unsaturated residual soil. The real-time response of thrust at front end of slope with different locations, slope angles, densities and rainfall conditions are analyzed. Results show that the changing of water content, matric suction, body strain and thrust at front end of slope in the upper part of slope is with great range and speed, while the bottom part of slope with less range and speed. The change of thrust at front end of slope can be divided into slow increase, rise accelerate and attenuation three stages, compared with less angle and lower density slope, the changing of residual thrust in larger angle and higher density slope is more strong and with a higher peal value. The new slope is easier to fail than the old one under rainfall since a greater growth force in a new slope. The internal high stresses of larger angle and higher density slope inside can be fully released after repeated rain, i.e., a high level of residual slide force will be happened, and the effect of soil softening will be even more significant. The results are useful and meaningful for further revealing the influence mechanism of unsaturated soil slope under rainfall infiltration in Southeast of Fujian, to further understand the weakening effect of unsaturated soil strength and stability.

Generally, the test data for a specific site are very limited. However, the Bayes method can overcome the limitation of small sample size of test data. This paper aims to apply the Bayesian updating with structural reliability methods (BUS) to update the reliability of soil slopes and estimate statistics of soil properties with limited site-investigation data. The effectiveness of the proposed approach for slope reliability is demonstrated by an infinite soil slope with direct shear test. The effects of prior information of soil properties including sample size of test data, marginal probability distribution and likelihood function on the slope reliability updating are systematically investigated. The results indicate that the BUS method can accurately estimate the statistics of soil properties and update the slope reliability incorporating limited site information and the effects of probability distribution and likelihood function. The probability distribution of soil properties has a significant effect on the updated slope reliability results (e.g., posterior mean and standard deviation, updated probability of failure). It will lead to conservative estimates of the slope reliability results when the normal or lognormal distributions are used. In contrast, the likelihood function has a slight influence on the updated slope reliability. In addition, the uncertainties of soil properties and updated probability of failure decrease as the sample size of test data increases, but they will change slightly once the sample size reaches a certain value.

The physico-mechanical properties of soil vary in different regions. There are great differences on engineering properties for soils formed by different geological depositions. Based on the investigation data of Dalian offshore artificial island, the soils could be subdivided into three independent statistical unit according to the geological origin, i.e., the top marine deposit , middle marine-continental deposit, and deep continental deposit. The physico-mechanical indices of deposits are analyzed by the SPSS software incorporated with mathematical statistics methods. The Gaussian distribution behaviors of deposit soils are hypothesis testing by the skewness and kurtosis method. The shear strength indices are optimized by the Bayesian approach. The results show that there are significant differences in the engineering properties of the deposits and their distribution. The standard deviation and variable coefficient of shear strength indices decrease after Bayesian optimized, and the time-varying predictions of parameters are realized.

Cracks significantly influence the stability of rock engineering, and two parallel cracks are the severe distribution forms of cracks. A series of triaxial compression tests were conducted on sandstone specimens with two parallel cracks to investigate the effect of fracture morphology on its mechanical properties and failure modes. The results showed that strength, elastic modulus and Poisson’s ratio of fractured rock were influenced by both the trace length and the trace length ratio. It was also found that these parameters decreased first, then increased immediately with dip increments, and less varied with spacing between cracks. When the dip angle was less than 30°, the failure plane initiated from crack tips and propagated through the rock bridge along the direction of principal stress. Shear failures localised within one crack, when the angle was between 30°and 60°. For the rock specimen with small trace length and trace length ratio (continuity 75%), but the trace length ratio remained relatively small (trace length ratio <1). For the rock with moderate cracks (continuity falls between 25%~75%) and the trace length ratio of approximately 1, failure modes were controlled by two cracks.

To study the strength and deformation characteristics of silty clay during wetting-drying cycles, the undisturbed silty clay is taken from a slope located in Three Gorges Reservoir Region, China. A series of triaxial shear tests are conducted under different load conditions after wetting-drying cycles. The test results indicate that the peak strength, effective cohesion and secant modulus all decrease slowly with the increasing number of wetting-drying cycles. While the effective internal friction angle stays at a steady state. The reason for the deterioration of strength and deformation of the silty clay in wetting-drying cycles is the crack development, which undermines the soil structure. However, external loading can positively restrict the crack development caused by wetting-drying cycles. By fitting the test data, the expression of Mohr-Coulomb yielding criterion for the undisturbed silty clay under wetting-drying cycles is modified.

The stiffness of soils at small strain level is different from that at large strain level obtained by normal triaxial tests. Small-strain characteristics of soil play an important role in the analysis of deformation in deep excavations, tunnels and other underground engineering. A series of static triaxial tests equipped with Hall Effect local strain transducers and dynamic tests using resonant column with bender element are carried out to study the small-strain characteristics of typical Shanghai natural clays (i.e. Layer2~Layer6). The results of this study show that the shear modulus increases with the increase of depth of soil layer. Shear modulus of the over-consolidated Layer 6 decays more rapidly than that of the other normally consolidated layers (i.e. Layer 2~ Layer 5). The initial shear modulus and the normalized shear modulus degradation are obtained by the resonant column. Both the triaxial tests and the dynamic tests can reveal the non-linear decay trend of shear modulus with the increasing of shear strain.

Nonwoven geotextiles are typically subjected to unequal biaxial tension in filtration applications, which changes the pore sizes, and can easily cause the failure of material. This work investigates the analytical solutions for the equivalent pore size and pore size distributions (PSDbi) subjected to unequal biaxial tensile strains. The experimental data collected from two nonwoven geotextiles under equal biaxial tensile strains are compared with the analytical solutions. The analytical variations of the pore size parameters under unequal biaxial tensile strain are calculated. It is concluded that the changes in analytical PSDbi depends on the tensile strains and the directional parameter Ka. The experimental PSD curves under equal biaxial tensile strains correspond with the analytical predictions. For changes in O95, the solution predicts more accurately than the PSDbi solution. The solution predicts more accurately for the thicker nonwoven geotextiles. The same conclusion can be drawn from the results under uniaxial tensile strains in the literatures.

Although Van Genuchten model is widely used for fitting soil water characteristic curves (SWCCs), the fitted parameters vary while considering the different forms of parameter and n and whether the largest matric suction point (106 kPa) is involved in the data. Therefore, three forms of parameter ( refers to a) without constraints; b) ; c) , respectively), as well as largest matric suction point were studied in the paper. The relationship between different parameters were discussed and the saturation degrees at matric suction of 106 kPa were calculated using the different parameters under different conditions, which were applied to analyze the relationship of saturation degrees at matric suction of 106 kPa among different conditions. The result shows that in the condition of constrained , discrete degree and interval range of 、 、 decrease when the largest matric suction point is added into the data. Moreover, discrete degree and interval range of decrease and discrete degree and interval range of 、 increase when adding the largest matric suction point under the condition of is not constrained. The results also indicate that the parameter under various conditions of and largest matric suction point shows a linear correlation of parameter without any constraints; but for parameter , the rule does not exist. The saturation degree at the largest matric suction point under various conditions of and largest matric suction point shows a linear correlation of saturation degree at the largest matric suction point without any constraints.

One-dimensional compression and swelling tests are carried out to investigate the responses of pore water pressure under unloading. Effects of various magnitudes and rates of unloading on changes of pore water pressure are analyzed. Based on Terzaghi’s one-dimensional consolidation theory, a calculation method is proposed for predicting the pore water pressure responses under linearly loading and unloading conditions. Coefficients of consolidation under compression and swelling are measured by consolidation and permeability combined testing. The applicability of the proposed calculation method to the pore water pressure responses under unloading conditions is evaluated. The results show that negative excess pore water pressure increases with the increases of magnitude and rate of unloading. At the beginning of unloading, excess pore water pressure reduces at a relatively high rate. This is followed by a reduction of excess pore water pressure at an approximately constant rate. However, this type of pore water pressure response behavior cannot be satisfactorily modelled by the calculation method established based on Terzaghi’s one-dimensional consolidation theory.

The novel amplitude parameters are proposed based on mecroscopic theory for orthotropic sand fabric, the comparative analysis is presented between the magnitude and mecroscopic test results of Toyoura sand. The random arrangement of non-spherical particles naturally forms orthotropic fabric of sand, and the three-dimensional component of its referred orientation are described with two independent angles. Based on the fabric theory of non-spherical particles randomly arranged of sand, the novel amplitude parameters of anisotropic sand on three orthogonal planes are defined. The expressions of the novel amplitude parameters are defined with the mathematical probability statistical methods, the fabric characteristics of discrete particles become the measured parameters in the plane. With the changes of the physical characteristics and geometry of microscopic particles, the amplitude parameters can be used to describe the isotropy, transversely isotropy and orthogonal anisotropy of material. Verification analysis with mecroscopic tests for Toyoura sand shows that the three amplitude parameters can describe the properties in orthogonal plane, the extensive quantitative methods of mecroscopic fabric which own a clear physical meaning are provided.

null

It is significant to investigate the evolution laws of pore structure during the process of backfill body damage in the surface subsidence zone. In this study, the medical X-Ray CT and small loading devices were employed to conduct real-time uniaxial compression scanning test on the backfill body with a mixture of waste rock and tailings. The multi-component structure models were reconstructed for the backfill body under different stress states. The results show that the backfilling porosity increased, as the stress value grew. When the stress value exceeded 80% of peak stress, the porosity increased sharply. Furthermore, pores were mainly distributed along the X-type conjugate surface. The backfilling porosity increased with the percentage of rock. The backfilling meso-structure evolution can be divided into six stages, including continuous compaction of pores, pore expansion after compaction, continuous expansion of pores, integration and expansion of pore groups, coalescence of adjacent micro-cracks, and pore expansion in heterogeneous areas. When the stress of the backfilling was less than 80% of the peak stress, both the CT number and porosity varied within small ranges. Although no significant penetrating cracks were observed in the backfilling, there were two main states, including the compression and expansion of the primary pores and the initiation and propagation of secondary pores. As the stress exceeded 80% of the peak stress, the CT number reduced drastically, while the porosity increased sharply. Meanwhile, the primary and secondary pore structures gradually connected in the backfilling and then formed penetrating cracks, resulting in the damage of the backfilling body.

Using CPT data from the Chi-Chi earthquake, CPT-based liquefaction discrimination methods proposed by Robertson and by Olsen are inspected. The prediction success ratios of the two methods are 82.61% and 80.43% for the liquefied sites, but 31.82% and 44.32% for the non-liquefied sites, respectively. CPT-based methods are reliable for liquefied soils, but not effective for non-liquefied soils. For comparison, the SPT-based liquefaction discrimination method recommended by National Center for Earthquake Engineering Research of USA is tested using data from Chi-Chi earthquake, the prediction success ratio of the procedure is 92.41% and 94.35% for liquefied and non-liquefied sites, respectively. The SPT-based method demonstrates higher prediction success ratio than that CPT-based methods. However, CPT-based soil type chart can reflect soil strength and soil type simultaneously, and can differentiate liquefied soils from non-liquefied soils in Chi-Chi earthquake. Moreover, for the preliminary discrimination of clayey soils liquefaction, CPT-based soil type chart is also better than the clay content which has been always used.

As for the limitation of instability criterion of displacement predictive parameter in the rainfall-induced landslide and also according to the basic principle of the holographic information theory, this paper proposes a new appraisal parameter of vertical displacement direction ratio for slope stability as an effective displacement dynamic parameter in the analysis and evaluation of slope stability. It systematically analyzes the variation and relationship of displacement and groundwater in rainfall-induced landslide, and also determines the formation mechanism and variation of the surface vertical displacement direction ratio at the different stages of stability. Meanwhile, the quantitative relation of the surface vertical displacement direction ratio with the stability of landslide in the compression stage and the plastic deformation stage is explored. Based on the analysis results above, it establishes the overall instability monitoring and early warning criteria of the vertical displacement direction ratio of the landslide by using the basic principles of mathematical statistics. Finally, the posterior analysis and evaluation of the stability evolution process of the vertical displacement directional ratio are carried out by using actual data of A3 and B3 monitoring points of the Xintan landslide. It is found that the results of the analysis and forecast are consistent with the actual deformation instability law of the slope. It reveals that it is effective and practical for the landslide forecast and early warning.

Many landslides in Three Gorges Reservoir area are featured by stepwise increasing surface displacement along the time. Responding composition model is one of the main methods for displacement prediction of landslides. Currently, high-frequency and low-frequency components of inducing factors are usually ignored. A method based on reconstruction of time series by ensemble empirical mode decomposition (EEMD) and particle swarm optimization based support vector machine regression (PSO-SVR) prediction for displacement is proposed. The typical Baishuihe landslide in Three Gorges Reservoir is taken as an example. Firstly, the monitored surface displacement time series from July 2003 to March 2013 is decomposed into trend and fluctuant components by EEMD. The trend component can be predicted using quadratic polynomial equation fitted by the least square method. With EEMD and t-test methods, rainfalls and reservoir levels time series are reconstructed into high-frequency rainfalls, low-frequency rainfalls, high-frequency reservoir levels and low-frequency reservoir levels, respectively. Combined with other common factors, high-frequency rainfalls and monthly variations of reservoir levels are selected as predominant factors for fluctuant displacement components by method of gray relational analysis (GRA). Finally, PSO-SVR is utilized for prediction purpose. Application results show that average relative error is 0.009 8, variance ratio is 0.023 9 and small error probability is 1, which demonstrate preferable effect of the proposed method. For verification and testing, 5 other typical landslides in Three Gorges Reservoir area are presented to test the effectiveness of our method, which can provide references for similar landslides.

Based on the monitoring curves of landslide deformation, the early warning criteria are established separately by previous researchers according to the parameters such as the displacement rate, displacement tangent angle, and improved tangle angle. However, the uniformly fixed criterion is not sufficient for the early warning of the landslide, due to the significant differences in the characteristics of landslides. The applicable range of Time-time curve conversion was obtained by considering the improved tangential angle and the characteristic of creep strength. This study also analysed the developments of creep parameters during the constant velocity deformation stage, until reaching the constant or steady values by using the Nishihara model. The constant deformation rate can be regarded as a comprehensive external performance of the parameters in the creep process of the landslide. Moreover, it was inversely related to critical sliding tangential angle. The constant deformation rate and critical sliding tangential angle were obtained by dividing the stages and conversing Time-time curve of 16 typical graded landslides. There was a high correlation between them. By combing with the maximum tangential angle of the undisturbed landslide deformation process, the lower limit was used as the criterion of landslide warning to ensure the reliability of forecasting.

The Kelvin model is modified by the spring-pot element based on the Caputo fractional derivative to describe the one-dimensional rheological constitutive relation of saturated clay. The rheological consolidation equation is derived with the assumptions proposed in Terzaghi's one-dimensional consolidation theory of saturated soil. The numerical analysis is performed by using the Laplace transform and its numerical inversion based on Fourier series expansions. To verify this method, the numerical solutions by the present method for the cases of the integer order derivative are compared with their analytical solutions. The applicability of the modified Kelvin model is verified by the simulation results of one-dimensional rheological consolidation test in the literature. The influences of the fractional order and the coefficient of viscosity of spring-pot component on the rheological consolidation process of soil are investigated. The calculated results show that the dissipation rate of pore water pressure is greater than that based on Terzaghi's one-dimensional consolidation theory in a long period of consolidation and then less than the latter in the final stage of consolidation. Furthermore, the settlement rate of ground is always less than that based on Terzaghi's theory in the process. Overall, the settlement of ground lags the dissipation of pore water pressure. This phenomenon is obvious and it takes more time to achieve the final settlement with the decrease of the fractional order or the increase of the coefficient viscosity.

The effect of cracks on the stability of surrounding rock of the straight wall arched tunnel was simulated by using different types of finite element analysis software (RFPA2D and ABAQUS). Cracks were normally distributed in the arch bottom, sidewall, arch shoulder, roof and other parts of the tunnel. This study investigated the mechanisms of crack initiation and propagation and the damage-failure patterns of surrounding rock under the confining pressure. The detailed comparisons and discussions of these two finite element methods (FEM) were presented. The model tests using cement mortar were conducted to validate the results obtained by numerical simulations. There was a good agreement between experimental results and numerical simulations. The results showed that cracks at the corner were the weak position of surrounding rock in the tunnel. When the crack at the arc of the tunnel roof formed 45°angle with the arc centre of the tunnel roof, the compressive strength of the tunnel was the lowest and the stability was the worst. Under the confining pressure, tensile failure mainly occurred at the arch bottom and the roof, and shear failure was observed at the sidewall and the crack tip.

The identification of blocks cut by finite discontinues could be realised by which polyhedrons are cut by discontinues and then merged. Although the object of cutting arithmetic is a convex polyhedron, rock mass model in practical engineering is not limited to the convex polyhedron. To solve this issue, the method of convex decomposition was proposed to decompose concave polyhedron model into convex sub-regions. Firstly, rock mass model was established by the surface element method and then a cuboid was set to wrap the rock mass model. Secondly, the cuboid was cut into some sub-regions by all surface element planes based on the algorithm of a plane cutting convex polyhedron. Finally, the convex decomposition of rock mass model was obtained by judging space position relations between sub-regions and the model. The explanation of key algorithms was given as well. In the cutting algorithms, the rough and accurate judgment method was successively presented to determine space position relations between discontinues and blocks. The efficiency of the computation complexity was greatly improved. Two typical examples were conducted to achieve the process and result of this identification method. The results show that the form and number of identified blocks are not limited by this method, which proves its applicability and efficiency of the concave polyhedron model.