Slices models for slope stability analysis, such as Sweden simplified method, Bishop simplified method, etc, are still the mainstream methods of slope analysis and evaluation. However, due to the rigid body assumption, how to simplify and calculate the forces between rigid slices is the most critical problem for slices model. A reasonable simplified condition will result in a solving approach to different methods, and also is the main and direct reason that affects the difference of calculation results. However, even a simplified condition must be given, slices methods are still important means of the analysis of slope stability at present. Considering the fact existing, this paper attempts to present a more reasonable approach to calculate the force between slices. In this study, the forces between slices will not be simplified by adopting the cellular automata method, and the forces will be obtained by an iterative solution so that the only statically determinate convergent solution is obtained for the statically indeterminate slope slices model. The corresponding program software CASlope is developed. Example results show that this approach is feasible and the results are reasonable.

To investigate the deformation behaviour and the brittleness evolution characteristics of Linghai granite, a series of conventional triaxial compression tests was performed. Complete stress-strain curves under various confining pressures were analyzed in detail. The relation among the specimen surface deformation, damage law and microcrack propagation was discussed on the basis of static strain measurement data. Typical microcracking modes were captured dynamically. Based on the improved post-peak energy method, the brittleness of Linghai granite under triaxial loading conditions was evaluated. The results indicate that there are localised deformations that correspond to the positions of crack initiation and propagation, which are revealed from the multiple long strain gauges showing larger compressive strains in the strain-time curves. Several microcrack propagation modes at the post-peak failure stage, which are related to the stress-controlled loading type to some extent, are observed from dynamic strain monitoring results. Brittle failure of Linghai granite is observed at the confining pressure up to 70 MPa. As the confining pressures increase, the brittleness of the rock decreases initially, and then increases and finally decreases. Experimental results provide an insight into the study of rock failure mechanism and are useful for stability analysis of underground engineering structures.

Due to the large size of rockfill, it must be scaled by a certain percentage so that it can be used to carry out laboratorial triaxial test. However, the particle shape of rockfill may be different before and after scaling. Therefore, it is important to evaluate the effect of particle shape on the deformation behavior of rockfill material. In this study, a high-precision 3D laser scanning technology was introduced to analyze the spatial shape of the rockfill particles with the sizes of 2.5-5 mm, 5-10 mm, 10-20 mm and 20-40 mm of Zipingpu concrete face rockfill dam. Further, triaxial tests were carried out to determine the correlation between particle shape index and particle breakage. The results show that there exists an approximate semi-logarithmic linear relationship between the particle breakage and the particle sphericity for Zipingpu rockfill. With the increase of confining pressure, the effect of particle shape on particle breakage decreases and the effect of particle strength on particle breakage increases. The particle strength of Zipingpu rockfill gradually decreases with the increase of the particle size, but the particle breakage decreases with the increase of the particle size, which may be due to that the particle shape of Zipingpu rockfill become more and more irregular with the decrease of particle size. Therefore, the impact of the particle shape on the particle breakage should be paid enough attention before and after scaling.

When a load is applied on soil around the pile group foundation of the existing bridge pier, it can result in the settlement and lateral deformation of soil and negative skin friction, which would have a great influence on the deformation and bearing capacity of the piles. To investigate the stress difference among piles at different locations under one-side load and surrounding load, model tests of 3×3 pile group foundation were performed to determine the change rules and the differences of axial force, negative skin friction, neutral point position, and safety factors of bearing capacity of piles among different piles. The results show that under the surrounding load, the axial force and skin friction values of the piles at the corner of pile group reach the maximum, those of side piles are the second, and those of the pile in the centre reach the minimum. In addition, the neutral points of the piles in the corner locate at the deepest, those of the side piles locate slightly higher, and those of the pile in the centre are closest to the pile tops. Under one-side load, the axial force and skin friction of the piles in the row near the side load and in the middle row change similarly to those under the surrounding load, and piles in the row far away from the side load are less affected by the load, and there is no negative skin friction acting on the piles. The change rule of the neutral point position in each pile is also similar to that under the surrounding load. Under the side load, the axial force, negative skin friction of the piles at the same location are smaller, but the neutral point position is higher than those under the surrounding load. The safety factor FS of bearing capacity of each pile under the surrounding load condition is less and decreases faster than that under the one-side load condition. This study could provide references for design of pile-group foundation under different loading modes.

To investigate the bearing performance and deformation characteristics of geogrid-reinforced sand soil foundation under the action of dynamic loading, the static and dynamic load failure tests on the pure-sand foundation and geogrid-reinforced foundation were carried by using a self-designed large scale foundation model test device of 300 cm (length)×160 cm (width)×200 cm (height). The variation laws of foundation bearing capacity, foundation settlement, foundation soil pressure, dynamic acceleration response and geogrid strain were analyzed. The bearing capacity and deformation characteristics of reinforced sand foundation under dynamic load were revealed, and the influence of static and dynamic loads on the bearing capacity of reinforced foundation was compared and analyzed. The results showed that the bearing capacity of the monolayer reinforced foundation was improved by 1.12 times as compared with the pure-sand foundation, and the axis settlement of the foundation was reduced by 24%. Moreover, the anti-deformation ability of reinforced-foundation soil was significantly improved compared with pure-sand foundation. The reinforcement changed the failure mode of the foundation, the failure mode of pure-sand foundation was the punching-shear failure, whereas reinforced foundation was whole shear failure under dynamic loading. The existence of the reinforcement had obvious diffusion and attenuation effect on earth pressure and the acceleration peak respectively, which could effectively reduce the transient deformation of the reinforced-soil under dynamic loading.

The nonlinear consolidation settlement of soft clay was mainly estimated by numerical method till now, by which the engineering application of nonlinear consolidation theory was greatly constrained. The classical nonlinear relationships, both e-lg?' and e-lgkv, were adopted in this paper, and the approximate analytical solutions with consideration of constant self-weight stress were obtained by the transformation of variables and iteration methods under case that the ratio of compressibility index Cc to permeability index Ck was not equal to 1. When the ratio of Cc to Ck approaches 1, there is little difference between the results by this solutions and that by analytical solutions with case of Cc /Ck=1. The deviation of approximate solutions from numerical solutions slightly increases with the deviation of Cc/Ck from 1 and increase in external load. When an usual external load in an actual project is considered, if the value of Cc /Ck varies from 0.9 to 1.1, the maximum deviation of this solution from the finite differential solutions varies within 2%. If the value of Cc /Ck varies from 0.75 to 1.25, the maximum deviation varies within 5%. If the value of Cc /Ck varies from 0.5 to 1.5, the maximum deviation varies within 10%. The consolidation rate decreases with an increase in the ratio of Cc /Ck in case of constant load. When Cc /Ck1, however, the consolidation rate decreases with increasing the external load.

In mountain region, landslide has a significant influence on the force and deformation of existing tunnels. The existing researches generally separate the landslide and the tunnel to investigate. However, very little attention is paid to the combination of the two kinds of engineering problems. Moreover, most studies remain on the monitoring analysis of practical engineering. It is rare to analyze the landslide-tunnel interaction by theoretical analysis. In this study, firstly, the landslide thrust is deduced based on the transferring coefficient method and the limit equilibrium method. Secondly, imposing the landslide thrust on the existing tunnel, and establishing elastic foundation beam model for the landslide-tunnel interaction, the internal force and displacement of arbitrary tunnel section are solved by the transfer matrix method. In addition, three-dimensional numerical model is established to simulate the landslide-induced tunnel deformation. The numerical results are compared with the theoretical method, and good consistency is obtained. Finally, the influences of the landslide thrust, the foundation coefficients in the sliding and the sliding bed, tunnel lining stiffness and tunnel length are analyzed. The results can provide certain theoretical support for reinforcement and maintenance of existing tunnels in landslide area.

Based on linear s-lnr relation of land subsidence from a single well pumping, equations of land subsidence from group wells pumping distributed as line and area were derived by applying the method of integration. For the observation point far away from the center of group wells, after simplifying the equations above, the result of formula reveals that the linear s-lnr relation is satisfied too. The result is proved by the monitored data from the third party literatures including classical theoretical solutions, numerical solution and engineering example. The monitored data agrees well with the new formula. The correlation coefficients of fitting curves are 99.91%, 98.67%, 96.60%, 99.63% and 99.43%. So the linear s-lnr relation of land subsidence from group wells pumping distributed as line and area are verified well.

By using the self-developed true-triaxial rockburst testing machine, the rockburst processes were reproduced in laboratory and the sound signals of rockburst process were monitored. The combination index of Meyer cepstral coefficient, spectral centroid and short-time average zero-crossing rate, which can quantitatively describe the sound characteristics, was used as the feature extraction information of typical destructive phenomenon of rockburst process. Then, Gaussian process, a machine learning method for solving small sample, nonlinear classification problems, was used to construct an intelligent identification model. Thus, the intelligent identification of typical failure phenomena in a rockburst process was realized. In addition, in order to overcome the shortage of traditional rock burst prediction methods, which emphasize on trend prediction but can not distinguish the development stage of rock burst process, a multilevel, progressive and dynamic prediction method of laboratory rockburst was developed based on the strategy of intelligent recognition + trend prediction. The variation laws of acoustic characteristic indexes such as quiet period, harmonic mean value and chromatographic vector mean value before rockburst were taken as the precursor information of rock burst. The prediction results of different laboratory rockbursts indicate that the method is feasible and lays the testing foundation of the sound-based method for in situ rockburst prediction in the further.

The coarse-grained materials are normally in three-dimensional (3D) stress state which changes with time and location. To investigate the strength characteristic of coarse-grained materials at different coefficients of intermediate principal stress, large-scale true triaxial tests were performed under isotropic consolidation and proportional loading conditions. The results show that the intermediate principal stress has an important effect on the strength of coarse-grained materials. The strength of coarse-grained materials in 3D stress state increases greatly compared to the strength in conventional triaxial stress state. The relationship of the major principal strain and the difference between the maximum and the minimum principal stresses in 3D stress state is steeper than that in conventional triaxial stress state. The difference between the maximum and the minimum principal stresses increases from 39% to 50% when the coefficient of intermediate principal stress increases from 0 to 0.25. Both the interlocking force c and internal friction angle ? increase with the increase of the coefficient of intermediate principal stress b. Especially, the interlocking force c increases significantly when b increases from 0 to 0.25. With the same minimum principal stress, the ratio of deviatoric stress to spheric stress at the failure state is the smallest with b=0, whereas the failure stress ratio reaches the maximum with b=0.25. However, the failure stress ratio decreases with the increase of b from 0.25 to 0.75. The stress ratio at failure state decreases with the increase of the minimum principal stress at the same coefficient of intermediate principal stress.

Supercritical CO2 is a kind of fluid, which is in a special state between gas and liquid, with unique properties such as low viscosity, high diffusivity, and zero surface tension. Thus, supercritical CO2 can be used as fracturing liquid to assist crack initiation and propagation without leading to the damage of reservoir zones. In this study, the mechanisms of supercritical CO2 fracturing were obtained by studying and analysing the characteristics of rock breaking and fracturing with supercritical CO2. The results indicate that, owing to its lower viscosity, supercritical CO2 can easily penetrate into micro-pores and micro-cracks and build fluid pressure systems with varied magnitudes in rocks, which results in tensile and shear failure. In conventional hydraulic fracturing, the initiation pressure is high, and the fractures are single or multiple straight cracks. Most fractures penetrate through mineral grains with high strength along the same direction, and the fracture sections are smooth and flat. However, the initiation pressure of supercritical CO2 fracturing is lower than that of conventional fluid fracturing methods; the fracture network is complex, and fractures are connected with each other. Generally, the induced fracture majorly initiates along the lower-strength grain boundaries, but seldom penetrates the mineral grains. Moreover, the fracture planes are rough. This study provides theoretical support for implementation of supercritical CO2 fracturing technology.

Double row long-short piles are new type of pile foundations, which can make up for the deficiency of common double-row piles and shows a good potential of engineering application. It is important to study the calculation method of landslide thrust. Based on the model of Winkler elastic foundation beam, considering the deformation coordination between the loading segment and anchoring segment and limit equilibrium condition, calculation method of the landslide thrust of the flexible pile is derived. Distribution form of landslide thrust behind the pile is trapezoidal. Comparing the theoretical calculation results and the monitoring data of model test, it is found that the error of the landslide thrust calculation value of rear-row pile is 7% to the experimental results, and the error of fore-row pile is 14.4%. Therefore, it is proved that the calculation method of landslide thrust of the double row long-short piles are feasible.

Tetrapod bucket foundations are formed by four suction buckets into a square shape and connected by superstructure into a whole to provide better resistances against severe marine environmental loads than single bucket foundation. Tetrapod bucket foundations can serve as foundations for offshore platforms or wind turbines, and show a broad application prospect. To investigate the bearing capacities and failure mode for single and tetrapod bucket foundations, and optimizing the design of tetrapod bucket foundations, this paper established a large number of 3D finite element models of tetrapod bucket foundations in soft soil and systematically investigated the effects of bucket separation distance on uniaxial capacities and corresponding failure modes with different foundation embedment depths and soil strength heterogeneities. Results indicate that vertical capacity factors are almost independent on the bucket separation distance while horizontal capacity factors are evidently dependent on separation distance, and moment capacity factors are especially affected by separation distance. All uniaxial capacity factors of tetrapod bucket foundations increase as embedment depth increases, but decrease as soil heterogeneity index increases under the same bucket separation distance ratio. By analyzing uniaxial capacities, corresponding failure mechanisms and group effect factors of tetrapod bucket foundations, optimal bucket separation distance of tetrapod bucket foundations was defined, which can provide reference for optimizing the design of bucket separation distance of tetrapod bucket foundations.

To investigate damage constitutive model of rock mass with intermittent joints in engineering practice with the assumption that the rock microscopic element strength obeys the Weibull random distribution function, the Mohr-Coulomb failure criterion is used to describe the strength of the microscopic element and the microscopic damage variable is deduced. A formula for macroscopic damage variable is deduced using energy and fracture mechanics theory and considering geometrical features and mechanical properties of the joint. Finally, based on the Lemaitre strain equivalence hypothesis, considering the macro-meso defect coupling effect, the composite damage variable is derived. A damage constitutive model of rock mass with intermittent joints under the action of macro-meso defect coupling is established based on the Mohr-Coulomb criterion. The results show that: (1) The damage model established by the Mohr-Coulomb criterion as the statistical distribution variable describing the strength of the micro-element can well reflect the characteristics of rock deformation and distribution of internal defects, which reflects that the micro-element strength of rock is influenced by the stress state. (2) The theoretical curve established by this model is in good agreement with the test curve of intermittent jointed rock mass. (3) The change of the macro damage variable and peak strength of jointed rock mass with the joint tilt angle and the change of damage variable considering the macro-meso coupling action and the peak strength of jointed rock mass with the joint tilt angle are generally consistent. (4) The relationship between the equivalent elastic moduli of macro-meso coupling and the crack connectivity rate exhibits nonlinear negative correlation. When the joint angle is constant, the relationship between the damage variable and joint length exhibits nonlinear positive correlation. When the connectivity rate is small, the relationship between the macroscopic damage variable and the internal friction angle of jointed rock mass shows a linear negative correlation, and the linear relations become nonlinear relations when the connectivity rate reaches a certain extent.

Equivalent linearization in frequency domain, represented by SHAKE2000, is the mainstream approach for seismic response analysis of soil layers. Due to the severe unreasonable results in soft soil sites, its improvement becomes a research hotspot, which mainly adopts frequency-dependent method (FDM), but there has been no substantial improvement. This paper assumes that dynamic shear strain and vibration velocity present a constant proportional relationship in FDM. Based on wave equation, exact solutions for the relationship of the two variables were derived on three typical site models, i.e., the homogeneous half space with one-way traveling wave, the single uniform soil layer, and the layered soil site. The reasonability and deviation degree of the assumption were studied through numerical experiments. Results show that the constant proportional hypothesis only applies in the case of unbounded homogeneous medium with one-way traveling wave. For actual layered soil sites, the relationship between dynamic shear strain and vibration velocity strongly depends on the frequency of the wave and the location of the observation point. If we ignore the reflection wave in ground seismic response analysis, the use of constant proportion assumption under one-way traveling wave will make significant deviation to the results. The deviation caused by the assumption may span four orders of magnitude. Even for a single uniform deposit, the deviation is also very significant. For practical ground motion calculation, this assumption clearly exists qualitative error in theory, and the quantitative deviation can not be acceptable, so it must be abandoned and need to find another more reasonable approach.

Natural montmorillonite was studied by adapting ionic soil stabilizer (ISS) with different concentrations. The water vapor isothermal adsorption experiment was conducted on both raw and modified samples in the range of relative humidity (P/P0) from 0 to 0.95. By analyzing water adsorption velocity curves derived from adsorption results and d001 variation with P/P0, montmorillonite hydration mechanism was defined and association analysis of hydration-pore was implemented through void ratio variation with P/P0 that display the boundary relative humidity for pores in different scales adsorbing water. In addition, associated evolutionary mechanism of hydration-pore for montmorillonites conducted by ISS was indicated by different pore size analysis methods such as X ray diffraction (XRD), nitrogen adsorption and mercury injection test. The results show that, for calcium montmorillonite, the cations interact with water molecules firstly at the range of 00.8-0.9, weak adsorbed water gets into macropores sostenuto. Ionic soil stabilizer regulates the characteristics of adsorbed water by changing the physico-chemical parameters (cations and basal surface of crystal layer) to influence the hydration processes of pores.

Permeability evolution of the joint is a key scientific problem for the hydraulic characteristic of rock mass. To further understand the permeability variation of coupling granite joint during shearing under high normal stress, the permeability of coupling granite joint was measured during shearing. To solve the sealing problem of high-pressure pore fluid during shear and the measurement of joint permeability, we designed a shear flow testing system for high-pressure porous fluid. Shear flow tests were conducted to obtain the variation of permeability with shear displacement, and the relationship between shear stress/normal deformation and shear displacement. Experimental results show that the joint permeability tends to decrease after shear stress peak under high-stress conditions, though dilatancy occurs during shearing. During the shearing process, the damage mechanism of dilation, compression and asperities contribute to the change of permeability. Under high-stress conditions, the dilatancy characteristics of joints are limited, but the shearing mechanism of asperities is enhanced. The joint permeability under joint shear is thus reduced.

Triaxial compression tests were conducted on marble specimens to obtain a single fully-penetrated cracked surface under different confining pressures, and fractured specimens were reinforced by "cement grouting" and "bolting-grouting" with a self-made fixture. Triaxial compression tests were carried out again on the reinforced specimens under the original loading failure confining pressure. By combining with the laser scanning technique and scanning electron microscopy, we analysed the reinforcement effect and mechanism of grouting and anchoring. Experimental results show that the bonding effect of grouting reinforcement causes the cohesion of the fracture surface of the fractured marble, and the shear and tensile effect of the anchor can further improve the strength of the fractured surface. The shear displacement corresponding to the peak shear stress of marble cracked surface strengthened by grouting is almost equal to that of marble strengthened by bolting-grouting, which indicates that the shear and normal loading effects of anchors can be effective when there is a certain shear deformation. Finally, based on the Mohr-Coulomb strength criterion, a shear strength formula of cracked marble bolting-grouting surface with low surface roughness and filling degree greater than 1 is proposed, which is proved to be reliable. To some extent, the experimental results could provide reference for support optimisation of deep hard rock engineering.

The properties of the coarse-grained soil are significantly affected by its gradation. The gradation of the soil is usually descripted by the particle size distribution curve (PSD), and the gradation equation is a quantitative description for the PSD. A continuous gradation equation was developed in this paper, and its applicability was verified by many PSDs of the rockfill material, transition material and filter material used in earth-rock dams. Furthermore, the range of gradation parameters of the coarse-grained soil was summarized, and the applicability of the gradation equation was analyzed. The main conclusions are as follows: The commonly used gradation parameters of coarse-grained soils are in the range of -2 Related Articles | Metrics

Trident cracks are the ubiquitous form of rock mass defect in nature, it has a significant influence on mechanical properties of rock mass. Uniaxial compression tests were conducted on cement mortar samples containing prefabricated trident cracks. The crack initiation, propagation and coalescence processes were recorded by a camera at the same time. The strain fields of the specimens were obtained by the digital image correlation (DIC) method. Combined with the PFC2D software, we studied the strength characteristics, crack modes and crack propagation rules of specimens under different angles of ? and ?. The results show that trident cracks have an obvious weakening effect on uniaxial compressive strength of specimens. When ? is constant at 120°, the sample reaches the maximum compressive strength when ? is 30°. When ? is constant at 90°, the compressive strength of the sample decreases first and then increases with increasing ?, and the maximum compressive strength is achieved when ? is 45°. There are three types of cracks generated in the specimens, including tensile crack (mode I), shear crack (mode II) and mixed crack (mode III). These three types of cracks usually initiate from the tip of crack. The mode I crack propagates along the loading direction normally and usually does not extend to the sample boundary, but the mode II and mode III cracks normally propagate to the sample boundary at a certain angle to the loading direction. By analysing the geometry of cracks and the composition of microcracks that further leads to macrocracks, it is known that the failure of the specimen containing trident cracks under uniaxial compression is tensile failure. The strain fields obtained by the DIC show that when the load reaches a certain stage, stress concentration occurs at the tip of a crack, then microcracks begin to develop and form micro-cracked areas, and finally, macro-cracks are generated. Through the analysis of the principal strain and shear strain fields, it is found that the failure of specimens is tensile failure, and the shear strain has little influence on the crack propagation process.

In this study, hydraulic fracturing experiments were carried out on the large raw coal to investigate the influence of bedding planes on fracture propagation by using a true triaxial hydraulic fracturing experiment system. The extension rule of hydraulic fractures in coal was analysed to reveal the formation mechanism of net-cracking based on the spatial distribution. The results show that the hydraulic fracture tends to bifurcate and swerve at the weak bedding. Then the developed bedding and fracture system provide the premise conditions to establish a fracture network. From the pump pressure curve, its frequent fluctuation indicates a remarkable feature of the net-cracking in coal rock. There are four basic modes on the initiation and extension of hydraulic fractures, and the fracture network is the combination of the four modes. In-situ stress difference coefficient and injection displacement both have great influences on the hydraulic fracture geometry. The small in-situ stress difference coefficient is more likely to form the fracture network. However, a relatively simple fracture geometry is easily formed when the fracturing fluid displacement becomes high, which conversely results in a poor fracturing effect. The experimental methods and corresponding results can provide reference and basis for the design and optimisation of hydraulic fracturing parameters.

The hydraulically connected wrinkle length is a crucial parameter for evaluating leakage through composite liner. Rare objective and efficient hydraulic connectivity analysis method for wrinkle network exists. Based on the simplification of head distribution around a hole, search length is proposed as the criterion to determine the connectivity between wrinkles. A digitized hydraulic connectivity analysis method for wrinkle network is proposed by simplifying wrinkles into line segments. The proposed method has been programed and applied to both wrinkle networks reported in the literature and a GM/GCL liner in a good condition. The results indicate that the definition of search length gives an objective criterion for determine the hydraulic connectivity between wrinkles and can reflect the influence of different parameters of the liner (e.g., hydraulic head, interface transmissivity, hydraulic conductivity, etc.) on the hydraulic connectivity between wrinkles. The location and dimension of a wrinkle are easier described by simplifying wrinkles into line segments. The application of the proposed method is simple and efficient. The results of leakage evaluation based on hydraulic connectivity analysis are conservative. In the process of leakage calculation, the determination of hydraulically connected wrinkle length is dependent on the hole frequency when a unified hydraulically connected wrinkle is used to describe the wrinkle network. The proposed method has a promising application future of acting as a key link of completely digitized leakage evaluation.

To study the effect of particle shape on shear modulus of sand, four sand samples with different particle shapes and grain sizes are prepared, including Fujian standard sand and artificial quartz sand. Image Pro Plus and Matlab program can extract the geometric parameters of particles from images of four samples obtained by scanning electron microscope. Two geometric parameters, radius angularity and form index, are improved and studied. Through statistical data of sand particle, the results show that radius angularity and form index describe different aspects of particle shape. As such, shape factor of sand is proposed based on radius angularity and form index to generalize particle shape. Then, a series of model tests in K0 condition is conducted to obtain the shear modulus of samples under different vertical stresses by bender element deployed in the sand. The tests show that shape factor is closely related with shear modulus of sand. Finally, based on Hertz-Mindlin contact model, shape factor is introduced to derive the formula of sand’s shear modulus, which agrees well with the measured one.

Geotechnical engineering in China including urban rail transit, highway, high-speed rail construction, hydraulic engineering is featured by large scale, long construction period, complex operation and frustrating situations regarding project safety. Various accidents have been reported from time to time, resulting in serious social impact and huge economic loss. This paper presents the main progress in the safety risk management of geotechnical engineering in China from 2010 to 2017, including (1) national major scientific and technological projects, achievement awards, specifications for safety risk management; (2) safety risk analysis and evaluation method of large geotechnical engineering; (3) safety risk monitoring and early warning of large-scale geotechnical engineering construction; (4) information technology and platform of construction safety risk management of large geotechnical engineering; (5) the intelligent analysis and control of construction safety of large geotechnical engineering. Based on the analysis of the above research progress in China, new challenges in the safety risk management for geotechnical engineering are identified as follows: (1) application and promotion of quantitative risk assessment theory and technology; (2) risk decision theory; (3) the implementation of the whole-process safety risk management plan; (4) shared platform for the construction of safety accidents in geotechnical engineering; (5) analysis and control of geotechnical engineering parameters based on big data mining technology; (6) research and construction of intelligent geotechnical engineering.

The permanent deformation and dynamic stability of high tailings dam are analyzed by using time-history analysis method, global deformation analysis method (equivalent nodal force method and modulus soften method) and limit equilibrium method with Xiaodae tailings pond as an example. This paper analyzes the influence of length of dry beach, height of tailings dam and seismic acceleration on the safety factor and permanent deformation of tailings dams, and analyzes the time-history change laws of safety factors of tailings dams. It is found that the permanent deformation of the tailings dam is different from that of the general earth-rock dam. The permanent deformation in the horizontal direction of the tailings embankment is greater than that in the vertical direction, and the permanent deformation is not monotonous with the dam height. The minimum safety factor of the tailings dam in the earthquake is linearly related to the influencing factors, and the seismic subsidence at the crest has a nonlinear relationship with the influencing factors. The calculation methods of the minimum average safety factor for seismic safety are improved based on the characteristics of the fluctuation and reduction of the transient safety factors in the earthquake. The results show that the seismic performance of the tailings dam of Xiaodae tailings pond can meet the requirements of the corresponding seismic fortification.

Blockiness is a new index to evaluate the integrity of rock mass, and it can represent the degree of rock mass fracture in three dimensions. However, current blockiness evaluation methods have not considered the cutting degree of rock mass, the size of the block, unrestricted basic application conditions and unreasonable ratings of blockiness levels. In this study, the causes of these limitations were analysed in-depth. Moreover, the concept of block volume percentage was defined referring to the calculation principles of the rock mass cutting degree, three-dimensional (3D) block modulus and volume RQD. A comprehensive percentage calculation method of block volume was proposed, in which the scale effect of rock blocks was considered. The ratings of blockiness levels and the values of the classification index were determined, and the improved blockiness evaluation method was developed. By comparing the rating results of the modified blockiness evaluation method and the Chinese codes for geotechnical engineering, the rationality of the modified blockiness evaluation method was analysed. Based on the Tongkeng Zn polymetallic orebody and Wudongde Hydropower Project, the case studies of the modified blockiness evaluation method were conducted. The results showed that the block volume percentages of No. 4 testing area at the 255 m level in the Tongkeng Zn polymetallic orebody, the PD49-1 adit and PD4 branch adit in Wudongde Hydropower Station were 11.18%, 12.847% and 10.168% by using the modified block evaluation method, respectively. In addition, they all belonged to slight-blockiness categories, and the rock mass was moderately integrated. Compared with the conventional blockiness evaluation method, the block volume percentage calculated by the modified blockiness evaluation method can more accurately represent the integrity of the rock mass from a 3D perspective. The findings in this study are significant for accurately characterising the 3D integrity of rock mass.

To investigate the propagation of subway vibration in the saturated poroelastic medium, a dynamic analysis model of the tunnel, lining and the saturated medium was established by using the moving axial excitation technology. The analytical solutions of dynamic responses in the frequency domain were derived by the wave function expansion method and the Fourier transform method. Moreover, an empirical formula for the critical velocity of the saturated poroelastic medium was given. The time-space domain solutions of dynamic responses were obtained by the discrete inverse fast Fourier transform. The results show that, for the tunnel without lining, the critical velocity of the saturated poroelastic medium is only related to the shear modulus and density of the medium, and its value is close to 1.1 times shear wave velocity of the medium. For the lining tunnel, the critical velocity of the medium increases with the increase of the shear modulus of the lining but decreases with the increase of the lining density. The lining has a certain weakening effect on the propagation of vibration. The weakening effect is more obvious when the difference of shear modulus between the lining and the medium is large. When the dynamic response frequency is close to the excitation frequency, the amplitude of the dynamic response becomes high but the corresponding critical velocity becomes low.

Under the condition of excavation and unloading, the induced relaxation and cracks of columnar jointed basalts (CJB) were the main challenges for the stability of the dam foundation of Baihetan hydropower station. To analyse the loosening characteristics of CJB, the temporal-spatial evolution of the loosening depths and degrees of class Ⅲ1 and class Ⅲ2 of CJB were studied by the single-hole acoustic wave test and borehole camera inspection which were conducted in the CJB test section of the left bank dam foundation of Baihetan hydropower station. The reason of loosening characteristics was then analysed in detail. The results showed that the loosening depth was mainly caused by excavation. The loosening degree of CJB adjacent excavation surface grew progressively with time by the negative exponent type, with a semi-annual growth period. The loosening characteristics were closely associated with the rock mass quality. The loosening depth, the time-dependent characteristics and distribution with borehole depths of loosening degree were largely different from class Ⅲ1 to class Ⅲ2. The time-dependent loosening was caused by the ageing degradation of strength parameters of rock mass due to the opening of internal joints. This study is useful to understand the unloading mechanical behaviour of CJB and to optimise the excavation and support of CJB in the dam foundation.

Base on the foundation treatment of south anchorage caisson of Wenzhou Oujiang Estuary Bridge, nine groups of field static loading tests on composite foundation with thick cushion and sand pile group of different cushion materials, cushion thicknesses and pile spacings were performed to study the application of this composite foundation and influencing factors of its bearing capacity. The interaction test on sand piles installation was conducted to investigate the impact of sand pile installation on completed sand piles. The results show that thick cushion and sand pile group, which can strengthen the soft foundation very well, is an ideal way of foundation treatment of super-large caissons. Water content of cushion, material of cushion and thickness of cushion have greater influence on bearing capacity than pile spacing. Because of the differences between the impact on sand pile and the impact on soil which are caused by installation of sand pile, it is inaccurate to use compaction effect of static-press pile to analyze the impact on existing sand pile. The impact on existing sand pile is very large and the maximum impact concentrates on the 1/3 pile length below earth surface. At the same time, the better of soil property, the smaller of this impact. A big pile spacing and blocking effect of existing sand pile can effectively reduce the impact.

Two boreholes of G2 and G3 were drilled at Tanggu and Dagang in Tianjin, and undisturbed samples were extracted at various depths. Through laboratory test of all soil samples, this study measured the physical and mechanical properties and conducted consolidation test and repeated loading and unloading tests for clayey soils. By analyzing the test results, this study delineated the characteristics and influence factors of elastic-plastic deformation. Study results indicate that clayey soils in depth less than 100 m are under-consolidated soft soil. Clayey soils in depth from 100 m to 400 m are in the states of over-consolidated or slightly over-consolidated, which causes by long-term over-exploitation of groundwater in the past. Deep clayey soils in depth more than 400 m show the state of normal consolidated. Based on the repeated loading and unloading tests of clayey soils in boreholes G2 and G3, it is found that plastic deformation decreases gradually and elastic deformation remains unchanged, independent on repeating times. The test results also suggest that clayey soils will gradually become an elastic body in the process of groundwater fluctuation, and deformation will rebound when groundwater recovers. By analyzing correlation between elastic-plastic deformation and depth, natural moisture content and clay content of clayey soil, it is found that elastic deformation shows positive correlation with depth, natural moisture content and clay content, and plastic deformation shows negative correlation with natural moisture content and clay content and no obvious correlation with depth.

Compression index Cc and swelling index Cs are two important deformation parameters in soil mechanics which can be obtained by oedometer tests. However, due to time-consuming of the test, several empirical correlations between compression index and some basic physical parameters (especially liquid limit wL, natural water content wn, initial void ratio e0 and plastic index Ip) have been proposed by many researchers. The empirical relationships of compression index all over the world have been compared and it is found that relationships between compression index and natural water content and void ratio are less discrete. Consequently, the unified correlation equations are given respectively. The oedometer tests have been conducted on specimens from 2nd to 6th soil layers in Shanghai. Samples were retrieved by the thin wall sampler. Consolidation test data available in literatures and geotechnical investigation data from 69 sites in Shanghai were collected and analyzed. Empirical relationships between compression index Cc and wL, wn, e0, Ip of Shanghai clay were given. Meanwhile, it was noted that there were highly linear relationships between Cc /n0 and Cc, Cs /n0 and Cs (n0 is initial porosity). Therefore, empirical formulas for predicting compression index Cc and swelling index Cs of Shanghai clay based on n0 were given. The relationship between compression index and swelling index was also analyzed. The results show that the value of Cc/Cs of Shanghai normally consolidated clay ranges from 4.8 to 6.9, with an average value of 5.8. While the value of Cc /Cs for over-consolidated clay ranges from 3.3 to 5.2, with an average value of 4.3.

The opening of joints between segments of the underwater shield tunnel has a great influence on tunnel waterproofing and the structural safety in operation period. Based on the joint monitoring data of two typical monitoring sections in the structural health monitoring system of Nanjing Yangtze River Tunnel, through the multiple linear regression analysis method, the effects of water level and temperature on joint opening in different stratum combinations were analyzed. And the variation of joint opening with water level was simulated by finite element method. The numerical simulation results are in good agreement with the regression analysis. The results of the above analysis show that the longitudinal joint opening and the transverse joint opening have a good linear relationship with water level and temperature as follows: when the tunnel crossing the permeable stratum of sand for the upper half tunnel section and the impermeable stratum of sandstone for the lower half section, the opening of the longitudinal joint increases with the increase of water level and is relatively less affected by temperature; when crossing the whole permeable sandy stratum, the opening of the longitudinal joint decreases with the increase of the water level and is relatively less affected by the temperature; regardless of which of the two strata the shield tunnel crosses, the transverse joint opening increases with the increase of water level and decreases with the increase of temperature.

The right bank underground caverns of Wudongde hydropower station are dense and large-scale, and the geological stratification is complex, which results in severe deformation and failure problems during excavation. Based on the analysis of the characteristics of geologic framework, in-situ stress data and excavation consequence of underground caverns, a numerical model was established to investigate the stability of the right bank underground powerhouse during excavation unloading by using the discrete element method (DEM). In the numerical modelling, we studied the principal stress fields and deformation fields, and analysed the deformation mechanism of the structural planes with a steep dip angle. Then, a high-precision microseismic (MS) monitoring system was established to evaluate the stability of the surrounding rock during excavation. Their assembling rules of MS events and energy ratio Es /Ep were analysed to identify the potential failure regions and deformation mechanism. The combined numerical simulation and MS monitoring indicated that under the excavating unloading, the failure of rock masses was observed near the layered rock with the steep dip angle where MS events assembled and the energy released, which results in surrounding rock failure. The deformation of surrounding rock was largely induced by stress-driven tensile failure which was controlled by structural planes. The results provide important references for excavation and reinforcement of the right bank underground caverns at Wudongde hydropower station.

In valleys buildings are commonly tightly located. However, the interaction between sedimentary valley and compact buildings under ground motion is currently not clear. Therefore, in this paper, the indirect boundary element method (IBEM) is used to analyze the valley-buildings effect under the plane SV waves. The results show that there exists significant and complex dynamic interaction between the sedimentary valley and buildings, and the seismic response of the buildings in the valley illustrates rapid alternation of strong and weak amplitude. The interaction characteristics strongly depend on the crucial factors such as incident angle of seismic wave, sedimentary valley group of medium and depth of the sedimentary valley. Overall, the buildings group has a large effect on the seismic response of shallow alluvial valley, and the seismic response can decrease by 50% for high-frequency wave incidence. It’s necessary to consider the effect of the buildings when we evaluate the seismic amplification effect in the actual valley, and the anti-seismic design standard of the buildings located at the focus of seismic waves should be properly improved.

The footing placed on the top of a slope is common in geotechnical engineering practice. However, few studies were performed to study the bearing capacity and the failure mechanism of footing-slope systems. In this study, discontinuity layout optimization (DLO) is adopted to study the effects of geometry of the slope, soil properties and footing location on the limit load and the critical collapse mechanism. Evaluations are provided concerning the recommended estimating methods in present codes both at home and abroad. The results show that the ultimate bearing capacity decreases as the increasing slope height and slope angle, but the effects can be neglected when the slope height exceeds a critical value; the ultimate bearing capacity increases as the increasing soil strength. The failure slip becomes shallower with high soil cohesion, while the failure slip develops deeper with the increase of friction angle. The bearing capacity increases with the normalized footing distance of from the crest of slopes, and there is a critical normalized distance that makes the effect of slopes can be negligible. The specified value for the minimum distance suggested by the China Code for design of building footing is smaller than the critical distance with high soil strength and large slope angles, and the effect of slopes on the bearing capacity should be considered; conversely, the recommended designed distance tends to be conservative for low soil strength and small slope angles. In AASHTO code, the ultimate bearing capacity values for cohesiveless soil are reliable, but only face failure mode is considered; the recommended design chart is contrary to theoretical solution in cohesive soil. The normalized footing distance from the crest of slopes is underestimated in AASHTO code.

Taking the well vegetated slope soil in the Touzhai of Yunnan province as case study, a 3D macropores structure model is reconstructed based on the CT scanning and digital image processing technology. Then the soil skeleton is taken as the boundary of seepage field, and the bounce-back scheme is set to simulate the interaction between soil skeleton and water. The lattice Boltzmann method is applied to study the seepage behavior in the macropores of well vegetated slope soil. The results show that the preferential flow is formed in the channel with good longitudinal connectivity, and the flow velocity in the channel is larger than that of other regions. In the closed macropores or these with poor connectivity, the water conductivity is small and the flow velocity almost equals to zero. The flow velocity in the channel increases gradually from the wall of macropores to the center of the channel, and the relationship between flow velocity and distance to the center of channel is approximately quadratic parabola. Along the depth direction, the section with larger macropores porosity has the faster average seepage velocity of Uz. Overall, there is the similar trend for the section macropores porosity and the average seepage velocity, which indicates that the distribution of macropores in the well vegetated slope soil influences significantly on the seepage characteristics along the depth direction.

The structural effect of fractured rock mass on equivalent mechanical parameters is studied by introducing a new synthetic rock mass technique based on the discrete element method (DEM). Through the controllable change of structural parameters in the three-dimensional (3D) rock mass structure network, the corresponding synthetic rock specimens are produced, and numerical simulations are carried out. The results indicate that equivalent mechanical parameters decrease with the increase of fracture density. Particularly, the equivalent elastic modulus is the most sensitive to the change of fracture density. With increasing the fracture dip angle, the equivalent elastic modulus increases and its standard deviation decreases. While strength parameters decrease first and then increase to form a U shape with increasing the dip angle. The equivalent mechanical parameters decrease with the increase of fracture size. Meanwhile, their standard deviations increase with the increase of the fracture diameter, and especially the uniaxial compressive strength is the most sensitive index. Moreover, the dispersions of the fracture size and dip angle have little influence on the equivalent mechanical parameters. This study can provide certain references for studying the structural effect of the fractured rock mass on equivalent mechanical parameters.

Coupling and systematic development methods of three dimensional geographic information system (3DGIS) and finite element method (FEM) have recently become the research hotspot in interdiscipline of geotechnical engineering and GIS. By explorating and programming of open source libraries such as MeshPy, GRASS GIS, vtkPython and Scipy, and using Python as "glue", the platform framework of seamless coupling between GIS and 3D FEM was eventually built. Under the constraint condition of geometry, Delaunay properties, angle, area and volume, MeshPy's Triangle and TetGen library were used to generate the triangle and tetrahedron grids for finite element numerical calculation and the insert algorithm of constrained Delaunay triangulation (CDT) and the grid quality control method were expounded. Combining with related module and algorithm provided by vtkPython, the surface model of a coal mine dump was established and a new interpolation method considering both lineament and triangle barycenter coordinate was put forward, which can effectively eliminate the elevation distortions in certain areas. The program including Duncan-Chang E-μ model and nonlinear solution algorithm based on mid-point incremental method was developed and was verified by the numerical simulation of indoor large-scale triaxial shear test. Finally, taking Zhundong coal mine as an example, the 3D finite element numerical calculation about the heaping process of the northern dump was carried out. Combined with the spatial distribution of material height and thickness in GRASS GIS, a preliminary research on the displacement variation of dump slope was accomplished. It is shown that the vertical displacement is mainly controlled by material height and thickness. The maximum horizontal displacement mainly appears in slope edges of each step, which is consistent with the actual situation.

To study the effect of temperature and water pressure on heat and mass transfer in buffer material, taking the compacted Gaomiaozi bentonite as an example, coupled thermo-hydro equations are derived from potential-based control equations of unsaturated soil, with the consideration of evaporation. We modify the smoothed particle hydrodynamics (SPH) algorithm to get coupled solutions with varying parameters during the calculation. Parameters of each single representation elementary volume will be updated in real time according to the current state. Results indicate that: Because calculation parameters are closely related to soil’s state, variation of parameters influence significantly on the results. The heat from nuclear waste can spread through the buffer in a short time, water migration rate is relatively slower. Increasing of buffer’s temperature will accelerate the migration of water. In comparison, water pressure has smaller effect on temperature distribution.

Contact problem is one of the common non-linear problems. How to simulate the deformation and strength characteristics of the contact surface and realize the real simulation of the contact problem between deformers is a difficult problem in this field. Based on the physical meaning of the 2D (2 dimensional) contact problems, their equivalent complementary models are established in normal and tangential directions respectively. Then a system of non-smooth equations is employed to describe the complementary models by Fischer-Burmeister (FB) function in the NCP function, which can be solved by the conventional Newton algorithm. In addition, good accuracy of the solution can be obtained by a small number of integration points based on the Gauss integration. In order to improve its accuracy and remove its discontinuity, a procedure of the contact surface at the Gauss point is employed in the surface-to-surface contact model, in which the accuracy of the solution can be controlled by adjusting the number of integration points. The procedure is easy to understand and convenient to implement. Based on this, a 2D contact finite element model is established, and the feasibility and effectiveness of the method are verified by some engineering examples. The results show that the proposed procedure has a higher accuracy and more realistic reflection of the actual problem than ABAQUS finite element method.

The numerical manifold method (NMM) has numerous advantages for discontinuous deformation analysis. In this study, by combining NMM with the basic concept of crack tip function, we conduct hydraulic fracture simulation and analyse the failure process without the concepts of Heaviside function and level set. To avoid the error caused by different positions of the crack tip in an element, singular cover functions are added for every physical cover near the crack tip within a certain range. One example is selected to show the influence of inner water pressure on the stress intensity factor (SIF). When the force on the crack is considered, the influence of each factor on the SIF is quantitatively analysed, and then hydraulic fracture of branch crack is simulated. The results show that the improved results are in good agreement with the analytical solutions. It is found that the error is often large when the force on the crack is ignored. However, when the force on crack is considered, the error gradually decreases with increasing either crack length or mesh density. The progressive failure of the branch crack shows that the method is feasible and is of great practical value.

The dynamic caustic method and ABAQUS software were used to investigate the propagation behaviour of pre-cracks with different angles under the linear charge explosion. The results showed that when the pre-crack was 0°, the wing crack at pre-crack far-end initiated and propagated deviating from the borehole, but the wing crack at pre-crack near-end initiated and propagated toward borehole due to explosive stress wave. The stress concentration at the pre-crack far-end was higher than that at the pre-crack near-end, but the arresting toughness of the wing crack at pre-crack far-end was lower than that at the pre-crack near-end, and thus the wing crack was much easier to propagate. When the pre-crack angle was 90°, the pre-crack in the column area of the borehole gradually changed from open to close under explosion stress wave. The explosion stress wave transmitted through the pre-crack, then compressive-shear stress concentration and mode I-II fracture which was mainly mode II fracture appeared at the pre-crack tip. Then, the anti-wing cracks initiated by the reflection stress wave of the boundary of the model, and propagated along the horizontal direction. In the end area, the wing crack initiated by the reflected tensile wave of pre-crack, and tensile stress concentration at the tip of the pre-crack led to an approximate mode I fracture. Then the wing crack propagated gradually along the explosive stress wave propagation.

The optical fiber testing technology is stringing the fiber Bragg grating (FBG) by naked fibers in order to construct the multi-point sensing testing system, which has the advantages such as higher precision, stronger anti-interference ability, higher spatial resolution and continuous data acquisition. In this paper, the fiber grating sensing technology was applied to monitor the glass fiber reinforced polymer (GFRP) anti-floating anchor, and the strains at the interfaces of anchor-anchorage body, anchorage body- surrounding rock, and within anchorage body were measured, achieving the multi-interfacial and full-length measurement of GFRP anti-floating anchor. The results showed that the fiber grating sensing technology was able to accurately record the strain change of GFRP anti-floating anchor in the entire duration of the pull-out test. Additionally, the results revealed the distribution law of axial force and shear stress within the anchorage body and at the interfaces of anchor-anchorage body, anchorage body-surrounding rock with the change of applied load and depth. The transfer depth of the load and the influenced range of the shear stress were different at different interfaces. Generally, the measurement technology and sensor embedding method have many advantages, showing broad prospects in geotechnical engineering research and applications.

Earth pressure coefficient at rest of soil, K0, varies due to different stress conditions. Besides, it interacts with other physical and mechanical properties of the soils. Based on the geological survey of Jinan subway project, the horizontal pressure of the thick alluvial cohesive soil from Yellow River in northwest of Jinan was tested by the in-situ soil pressure tester, K0 stepped blade (KSB). To verify the test results, other test methods are also adopted, such as the lateral loading test, indoor K0 test by side pressure gauge as well as common calculation formula taking internal friction angle and plasticity index of cohesive soil into consideration. According to the comparative analysis of the results by different methods, the results can be obtained as follows. In the form of exponential interpolation, test results from KSB are superior in presuming the in-situ earth pressure coefficient at rest. The cohesive soil in the Yellow River alluvial layer can be approximately in normal consolidation saturated state inside the range of 0-20 m. K0 changes between 0.4 to 0.7 and decreases with the increase of depth. The value of K0 is on the high side with Jaky formula’s result and on the low side with indoor test. The calculation results obtained by Jaky formula are close to that of measured by KSB when introduce equivalent angle of internal friction and internal friction angle obtained by consolidated quick shear test.

Due to the significant size effect in the direct shear test of coarse grained soil, and the incapability of the general indoor direct shear apparatus to adjust the size of the sample, it is difficult to obtain the shear strength parameters of coarse grained soil. Therefore, an assembled direct shear apparatus which can be employed to change the size of the sample is introduced. This direct shear apparatus is composed of an output device, a displacement measuring device, a reaction frame, and a shear box. By changing the number of steel rings nested in each shear box, the diameter of the sample can be set to be 6.18, 10, 15 and 20 cm. By changing layers of shear boxes, the height of the sample can be set to be 5, 10, 15 and 20 cm. The values of diameter and height of the sample can form 16 combinations in total. On this basis, abandoned slags from the diversion tunnel of Xuecheng Hydropower Station were selected as samples, and then 5 groups of parallel direct shear tests were carried out to verify repeatability of the direct shear apparatus and reliability of the test results. By changing the diameter D and height H, a series of direct shear tests was conducted to study the size effect of the sample, and to determine the reasonable size of the sample.