The deep rock mass is in a three-dimensional stress field, and the surrounding rock of the borehole or roadway is in an obvious true triaxial stress state due to the excavation-induced disturbance. Extraction technology is widely used as an important measure for gas control in coal mines. The plastic zone around a borehole is a good channel for gas flow. Therefore, it is important to study the characteristics of the plastic zone of the surrounding rock for the optimal layout of the extraction borehole. To study the plastic zone characteristics of drilling borehole surrounding rock, we analyze the stress distribution of surrounding rock based on the generalized plane strain theory, and compare the applicability and accuracy of several commonly used strength criteria (e.g. MC, Mises, DP, MLC, SMP). The results show that the MLC criterion can better characterize the true triaxial strength of rocks. Based on the MLC criterion, we derive a formula for the radius of the surrounding rock plastic zone, and analyze the effects of deviatoric stress, intermediate principal stress, internal friction angle, cohesion and radius of drilling hole on the permeability-increasing circular area. It is found that the permeability-increasing radius increases with the increase of deviatoric stress, decreases first and then increases with the increase of the intermediate principal stress, decreases with the increase of the internal friction angle or the cohesion of rock, and increases linearly as the radius of drilling hole increases. This research results can provide an important reference for the parameter design of engineering technology such as roadway support and coal seam gas drainage in practical projects.

It is quite common that deformation occurs when earth and rock materials are wet. Generally, initial strain method is more popular than the theory of unsaturated soil in calculating the wetting deformation. The wetting tests are usually carried out by single line method or double line method. In this paper, the wetting tests and relevant calculation methods are used to calculate the wetting deformation and related influence parameters, and a new approach in calculating wetting deformation and its influence is put forward, which mainly involves: (1) under a certain stress state, wetting deformation actually includes two components, one is induced by the ‘soft’ effect of the soil’s stiffness, the other is the ‘real’ additional deformation induced by wetting. (2) It is unsuitable to directly subtract the deformation of ‘dry soil’ from that of the ‘wet soil’ at the same stress state in the double line test. The subtraction should be calculated by using the tangent modulus under the same stress state. (3) Based on the generalized potential theory and the dual properties of stress and strain, one can transform the effect of the wetting deformation into an equivalent stress. Although this is an approximate method, it is simple and easy to quantify the wetting effect. This idea can also be used to investigate the effect of rheological deformation and the deformation caused by particle breakage and drying-wetting cycles.

Based on the triaxial compression test under different confining pressures of red sandstone, the failure characteristics at the outside surface of rock samples is monitored by using digital image correlation technique, and the failure images of outside surface of rock samples at different stages are obtained. The images showing the failure of the rock samples are analyzed and the fractal dimension calculated by box-counting method was used to characterize the variation of cracks on the outer surface of rock samples. It is found that the fractal dimension of cracks is proportional to the degree of damage in the rock samples. In addition, at the final stage of the failure in the rock sample, the fractal dimension of cracks appears an abrupt change, demonstrating the failure characteristics of rock samples. Combined with the mineral composition and mesostructure distribution characteristics of the red-bed soft rock, the cascade failure model is introduced to generalize the random connection model and failure mode of the structure. Furthermore, the redistribution principle of "capacity-load" of internal nodular particles under loading, and the initial load transfer path of nodular particles after failure are also analyzed. Results show that the red-bed soft rock under loading presents a multiple stage failure process, i.e. cascade failure. The cascade factor is continuously propagated and increased through the cascade paths. Thus the cascade failure becomes more obvious than before. Combining the deformation and damage process of rock failure and the interval characteristics of its axial stress-strain curve, the cascade failure classification criterion of rock failure is revealed, which has a certain reference value for expanding the research in this field.

Two sets of model tests using the conventional vacuum preloading(CVP) method and the air-booster vacuum preloading(AVP) method, were implemented to strengthen the ultra-soft dredger fills and compare their reinforcement effect. The microstructure characteristics of ultra-soft dredger fills after two methods of reinforcement were investigated using scanning electron microscope(SEM) and mercury intrusion porosimetry(MIP), respectively. A comparative analysis was performed with respect to variables microstructure characteristics of ultra-soft dredger fills, and quantitative analysis was conducted by Image-Pro Plus graphics processing technology under the two reinforcement methods. The model test results show that the reinforcement effect of AVP method is better than that of CVP method. The microstructural characteristics of soil reinforced by two methods were analyzed from three aspects: the morphology of soil skeleton particles, the contact form of soil skeleton particles and the pore space between soil skeleton particles. Compared with CVP method, the microstructure quantitative analysis results indicate that the number of pores, flattening degree K and shape coefficient ff of soil are larger, whereas the porosity and fractal dimension D of soil are smaller after reinforcement in AVP method. However, the directional probabilistic entropy Hm has no obvious change rule, and the peak of the aperture distribution increment is on the left, indicating a greater number of small pores by using AVP method. Furthermore, it is proved that the reinforcement effect of the AVP method is superior to that of CVP method from the microscopic aspect.

Freezing and thawing process, as a weathering process, is very common in the Northwest China. It repeatedly changes the microstructure and physical properties of the soil, and strongly affects the interaction between soil and structure. In view of the influence of freezing and thawing on the mechanical properties of salinized soil-structure contact surface, the shear tests under unconsolidated and undrained conditions were applied, considering the effects of freezing and thawing cycles, salt content, matrix suction and other factors on the mechanical properties of unsaturated chloride salinized soil-steel interfaces. The results showed that when the salt was not contained in the soil, the mechanical parameters (e.g. cohesion and internal friction angle) of the interface increased at first and then they decreased with the increase in the number of freezing-thawing cycles. When the salt was contained in the soil, the cohesion of the interface between the salinized soil and steel decreased, but the internal friction angle slightly increased with the increase in the number of freezing-thawing cycles. Before and after freezing and thawing, the mechanical parameters of the interface decreased at first and then increased with the increase in salt content, including threshold values of salt contents. Before the freezing-thawing process, the mechanical parameters of the interface have a salt threshold value of 8%. When the number of the freezing-thawing cycles increases, the threshold changes. The suction force of the matrix at the interface generally decreased with the increase in cycles of freezing-thawing process and it eventually stabilized. It firstly decreased and then increased with the increase in the salt content. The salt threshold of matrix suction is approximately 10%. The shear stress-shear displacement of the interface presented two stages including the linear elastic deformation and strengthening stages. When the vertical load is small, the shear stress-displacement curves show weak hardening characteristics, and no obvious strain-softening phenomenon occurs.Therefore the applicability of the shear stress-shear displacement modes at the interface were evaluated and it is found that the Gongpaz model can be well matched with the experimental results. Thus the mechanical model of chlorine salinized soil-steel block interface was established and its reliability was also verified by the experimental data.

As a common fracture model in the rock slope, the gentle dip fracture system is significant in geological engineering for slope evolution and stability evaluation. In this paper, based on the characteristics of stress field, the rock shear tests were carried out by applying the particle flow program, revealing the formation and evolution mechanism and impact factors of the gentle dip rupture system in the slope. The results show that: (1) the gentle dip angle system in the rock slope is a set of Riddle low angle shear rupture system, which consists of high-order sub-sexual geese fractures and low-order sub-conjugate shear fractures; (2) the evolution of the fracture with a gentle dip angle can be divided into several stages, showing that the propagation of the fracture system firstly occurs at a set of echelon rupture and then the conjugate shear cracks of the low order begins to expand and connects with original echelon ruptures; (3) the crack coalescence mode is different under different confining stress. The crack develops along the tips of the echelon rupture and finally links under a low confining stress, and it develops through the middle part of the original rupture under a moderate confining stress, and the coalescence will occur through a new set of echelon ruptures in the high confining stress; (4) the morphology of the rupture surface in the rock after the shear tests is closely related to the confining pressure, showing that the rupture surface is flat but the roughness is large under the medium and low confining pressure; under high confining pressure, however, the fracture presents curved surface, and it is smooth but the roughness is small.

In order to study the failure strength and crack propagation characteristics of fractured rocks, the uniaxial compression experiments on rocks with single fissure were carried out by MTS rock mechanics system. Acoustic emission(AE), ultrasonic testing and camera recording were monitored simultaneously to study the relationships between stress-strain behaviors, acoustic emission characteristics, the propagation and coalescence of cracks. The results show that the variations of crack initiation stress, peak strength(PS) and modulus are consistent with the inclination angle ? of fissure, which decreases firstly then increases later. Crack initiation stress and peak strength are affected greatly by ?. The stress-strain curves of fractured rocks rise stepwise before peak strength attributed to the stress transference and stress release (stress drop) induced by intermittent accelerated propagation of cracks. The stress drop is accompanied with energy release, modulus deterioration and sharp increase in ring counts of acoustic emission. With the increase in ?, the stress corresponding to the first stress drop and sharp increase in ring counts of acoustic emission gradually increases. Besides, the ring distribution gradually shifts to peak strength. The initiation and propagation of cracks result in the attenuation of wave amplitude and velocity, and the amplitude decrease occurs earlier than that of stress drop. The velocity decrease of rock sample with ??= 0o reaches up to 50% before peak strength. With the increase in ?, the stress increases at the moment corresponding to the velocity drop, and the drop rate of wave velocity before peak strength decreases gradually. When ? is small, crack propagation is highly developed before peak strength of rock. It shows that with the increase in ?, crack development before peak strength subsides. Thus the accelerated crack propagation and convergence transfer to post-peak stage. When the crack length is extended to a critical value, the expansion rate will increase rapidly. The threshold of crack length gradually decreases as ? increases.

In order to study the influence of non-penetrating cracks on rock failure characteristics under dynamic loading, splitting Hopkinson pressure bar(SHPB) is used to conduct impact tests on sandstones with different number and developed depth of non-penetrating cracks. The dynamic failure characteristics are analyzed based on fractal theory and energy dissipation principle. The results show that dynamic peak stress is more complicated in different number and developed depth of cracks. Peak strain increases with the increase in the number of cracks, and it increases at first and then decreases with the increase in the developed depth of cracks. Dynamic elastic modulus increases at first and then decreases with the increase in the number of cracks, but it decreases with the increase in the developed depth of cracks. The failure modes of specimens under different working conditions are mainly divided into three types, i.e. tension strain failure, tension strain-shear composite failure and crushing failure. The number of cracks has more significant influence on the failure mode than the developed depth of cracks. The fractal dimension, energy dissipation per unit volume and energy absorption rate of the rupturing rocks increase with the increase of the number and developed depth of cracks. Moreover, the fractal dimension and dissipation energy per unit volume are both approximately linear correlated with energy absorption rate.

The mechanical behavior of the interface between the structure and soil has become one of the hotspots in geotechnical engineering. A series of direct shear tests on the loess-cement mortar interface was carried out to study the interfacial shear properties under different normal stresses and water contents of soil. The results show that: (1) the shear stress-horizontal displacement curve of the loess-cement mortar interface presents strain-softening characteristics, including the pre-peak, post-peak and residual strength regions, with the normal stress of 50 and 100 kPa and the water content of 9.2% and 13.1%. While it presents a strain- hardening state when the normal stress is 200 and 300 kPa. When the water content of the soil is 17.1% and 20.8%, the shear stress-displacement curves at the loess-cement mortar interface show the strain-hardening effect with different vertical stresses. (2) The shear strength envelope of the loess-cement mortar interface is in accordance with the Mohr-Coulomb criterion. The shear strength of the loess-sand block interface meets the Mohr-Coulomb criterion. As the water content increases, the interface cohesive strength decreases significantly, while the frictional strength increases slightly, and the overall shear strength decreases. When the soil moisture content increased from 9.2% to 20.8%, the interface cohesion decreased from 41.5 to 3.6 kPa, and the interface friction coefficient increased from 0.50 to 0.58. The nonlinear model is used to fit the shear stress-displacement curve of the interface. The results show that the nonlinear model can better reflects the shear behavior of the loess-mortar block interface.

Based on the Yangzhou Shouxi lake tunnel, the load of segments and structural strain etc of shallow buried shield tunnels with super-large diameters were constantly monitored in the field. Variation rules of earth pressure, and the strain of steel bar were analyzed. Then, the theoretical earth pressures at the segment of the tunnel were calculated based on the overburden load method and Terzaghi’s earth pressure method. These theoretical results were compared with the measurement values, analyzing the applicability of different earth pressure calculation methods under different buried depths. The results show that shield rectification has a great impact on the value of tunnel segment load and its spatial distribution, and the impacts lasts until the stability period. The earth pressure on shield segments decreases and tends to be stable gradually after the construction of shield tunnel. The change trend of steel bars’ strain is basically the same as the earth pressure, but the former lags behind the latter for reaching the stable state. When the earth pressure reaches the stable state, the measured value is about 48%-60% of the theoretical value. The earth pressure value calculated by the Terzaghi’s earth pressure method is much closer to the measured value than that of the overburden load method. Thus, the study in this paper can provide guidance for the load design of shield tunnel segments.

The self-developed multi-field coupling test system is applied in the physical simulation of coal-bed methane (CBM) drainage in boreholes along seam, considering different stress concentration conditions caused by mining activities. The evolution of coal-bed gas pressure, gas drainage rate at boreholes, stress sensitivity coefficient, and permeability at stress released and concentrated zones during coal-bed methane drainage process, and the initial stress regions are analyzed. The experiment results show that (1) In the drainage process, the coal seam gas pressure decreased rapidly at the early stage, and then declines steadily. The gas drainage flow rate is much larger in regions closer to the borehole than far from it. In the periphery of the borehole, the gas pressure gradient increases at first and then decreases a certain period later. (2) During the mining process, the coefficient of stress concentration increases, while the permeability of coal seam and coal-bed methane flow decrease. The minimum drainage amount of coal-bed methane is located at Region 1 characterized with maximum stress concentration, while the maximum drainage amount is located at the stress released zones with minimum stress state. (3) The stress sensitivity coefficient in the stress concentration zone, which is characterized as the slowest decline rate of dimensionless permeability, is higher than that of stress released zones and initial stress regions.

College of Petroleum Engineering, China University of Petroleum, Beijing 102249, China

The suction caisson foundation can be used as the anchoring foundation of the tension leg platform(TLP), which is mainly affected by the uplifting load. The pullout failure mode is the key to analyze the uplift bearing capacity of the suction caisson foundation. But studies on the mechanism of the ultimate uplift bearing capacity are still few and not thorough enough. In this paper, the uplift bearing capacity of the suction caisson foundation are sutided based on by a series of laboratory model tests. Two uplift failure modes, including the top and bottom tension failure mode, are derived. The test results show that when the top tension failure mode occurs, the negative pressure at the top and bottom of soil plug is approximately equal. The bearing capacity increases with the increase in the negative pressure at the top of soil plug. In addition, the uplift bearing capacity of caisson is made up of its self-gravity, the frictional force of inside and outside wall and the tensile force at the top of caisson. When the bottom tension failure mode occurs, the bearing capacity increases with the increase in the negative pressure at the bottom of caisson. In this condition, the uplift bearing capacity of suction caisson foundation is made up of its self-gravity, soil plug gravity, the frictional force at outside wall and the tensile force at the bottom of caisson. When the load reaches the ultimate bearing capacity, the negative pressures both at the top and bottom of soil plug are less than the undrained shear strength Su and the reverse bearing capacity coefficients at the top and bottom of the caisson are both less than 1.0. Considering the top negative pressure in the caisson, the calculation method of the uplift bearing capacity of the suction caisson is proposed, and the uplift bearing mechanism of the suction caisson foundation in soft clay is revealed. These can provide a reference for the analysis of the uplift bearing capacity and engineering design of the suction caisson foundation.

The rheological properties of rock are of great significance to the analysis of long-term deformation and stability of slopes. Based on the sampling of construction site at a highway in Guizhou province, the graded unloading rheological tests under two stress paths including constant axial stress and unloading confining pressure, and equal proportional unloading of confining pressure and axial stress were designed. The test results show that under the two stress paths, the rheological trend of the mudstone specimen under unloading is basically the same. Under the failure stress, the lateral deformation of the mudstone sample increases rapidly and it exceeds the axial deformation, and the lateral expansion is more significant. Compared with the constant axial stress unloading condition, the equal proportional unloading requires a much longer time, and the axial deformation of the mudstone sample rebounds. However, axial and lateral deformation account for a large proportion of the total deformation, which is about twice of that in unloading of constant axial stress. It shows that under the equal proportional unloading of confining pressure and axial stress, the predictability in failure of rock sample is smaller, but its damage effect is much greater. The proportional design experiment refers to the actual slope excavation gradient is designed to carry out the proportional unloading test, which has great reference for the long-term stability analysis of the actual slope engineering.

The shale reservoir is usually characterized as many natural fractures and weak planes. A large amount of self-supporting shear fractures appear in the hydraulic fracturing zones. These self-supporting shear fractures play an important role in the development of shale gas. Thus the evaluation on the flow conductivity of these shear fractures can provide great guidance for the development of the shale gas. However, the common evaluation formula of the flow conductivity for shear fractures, which is based on the Darcy flow equation, is not applicable when the fluid flow rate is high. Therefore, it is necessary to develop a new evaluation method suitable for assessing the flow conductivity of the self-supporting fractures, and provide the support for the accurate evaluation of the flow conductivity of the self-supporting fractures in the shale after hydraulic fracturing activities. The shale samples of Longmaxi formation are selected in the experiments. The shear tests are carried out in the self-developed shear-seepage coupling experimental system and the self-supporting shear fractures are generated. The nitrogen flow test is carried out under the stress-seepage coupling conditions. A set of pressure and gas flow data are obtained. The flow characteristic equation of gas flow in the self-supporting fractures is described based on the non-linear Izbash’s law. Thus the formula for calculating the flow conductivity of self-supporting fractures is established. Based on the test data, the flow conductivity of self-supporting fractures is computed. Furthermore, the theoretical gas flow amount is obtained. It is found that the theoretical flow rates and the measured flow rates are in a good agreement. The flow conductivity of the self-supporting fractures under different shear displacements and confining pressures are tested. The experimental results show that the flow conductivity of self-supporting fractures is affected by the shear displacement, roughness etc. At the confining pressure is in the range of 10-40 MPa, the flow conductivity of self-supporting shear fractures are in the range of 0.1 to 1 D?cm. The new method proposed in this paper can provide an effective way for the evaluation of the flow conductivity of the self-supporting fractures when the flow rate is high.

In order to analyze the deterioration effect of mechanical properties of the fractured rocks strengthened by grouting in a long-term immersion in pressurized water, the long-term immersion tests of reinforced fractured rock samples considering the influence of water pressure. These tests are based on the earlier research of strengthening the fractured rock mass using carbon fiber reinforced cement-based composites mixed with grouting materials. Besides, the shear tests were also carried out in different immersion periods. The results show that: (1) a significant change trend of the shear stress-shear displacement occurs for the fractured rock samples reinforced by grouting and immersed in the pressure water in the long-term, with the shear stiffness gradually decreases, and the shear displacement gradually increases when the shear strength reaches to its peak. (2) With the longer immersion period, shear strength of reinforced fractured rock mass shows a sharp degradation trend at first, and then a gentle degradation. After 6 cycles of soak periods (i.e. 150 days), the degradation rate of the maximum shear strength under different normal stresses is in the range of 32.6%-39.9%. It means that the internal friction angle and cohesion is reduced by 19.3% and 39.4%, respectively. The shear strength parameters in the first four soaking periods (0-90 d) account for about 90% of the total deterioration range. (3) Dissolution, corrosion, lubrication, softening of the fractured rocks occurs in the pressurized water. Besides, the gel in the slurry is softened by the water molecules entering into the interior of the slurry, which result in a gradually deterioration in mechanical properties of fractured rocks and grouting. Meanwhile, cementation between the slurry and fracture surface, and bond performance between the slurry and carbon fiber are gradually weakened. As a result, shear performance of grouting reinforced fractured rocks gradually deteriorates, and shear failure surface gradually developed to cementation surface between the slurry and fracture surface. The experimental results in this paper can provide reference for the long-term evaluation on stability analysis of fractured rock mass reinforced by grouting method.

To explore the macro-mechanical characteristics and damage evolution mechanism of frozen rocks, the CT scanning images of each scanning layer of rock samples are obtained by using CT real-time scanning under uniaxial compression at 20 ℃, ?5 ℃ and ?10 ℃. The grayscale histogram and binary images are generated based on self-programming program in Matlab. The safety, damaging and fracture zones in the scanning layers of the rock sample are defined using the relationship between H value and density. Combined with thermal effect, water softening effect and ice frost heaving pressure on rock sample, the damage evolution law of frozen rock is studied quantitatively. The experimental results show that (1) during the process of temperature drops from 20 ℃ to ?5 ℃ and then to ?10 ℃, the unfrozen water content decreases and the frozen ice content increases. Correspondingly, the rock sample strength increases and the resistance to deformation enhances and the overall stability also enhances. The CT scanning images show that the development of cracks slows down and the rock sample damage gradually weakens; (2) the damage evolution of rocks under uniaxial compression generally experiences different stages including crack initiation, closure, propagation, influx and finally the formation of macroscopic main cracks. Thus the deformation of rock samples increases and damage occurs at different temperatures; (3) due to the end effect, the damage degree of the scanning layers at middle is stronger than that of the end. Besides, the damage formation of the external scanning layers is earlier than that of the internal scanning layers. Damage marginalization occurs and it expands to the inside of rocks.

The boreholes in frozen coal mines of western China mainly cut through the Cretaceous and Jurassic rock formations characterized as water-rich and weakly cemented strata. The rock mass in the frozen wall undergoes a complete freeze-thaw process under the coupling effect of temperature field and crustal stress field. In this paper, considering the coupling effect of the temperature field and the geostress field in the process of freezing and thawing, the Cretaceous water-rich and weakly-bonded red sandstone is selected to see its mechanical properties based on experiments. Meanwhile, the deterioration mechanism of Cretaceous weakly cemented red sandstone induced by freeze-thaw process as well as the influence of crustal stress during the freeze-thaw process are analyzed. During freeze-thaw process, the confining pressures are set to be 0, 2, 4, 6, 8 MPa separately, the freezing temperatures are ?5, ?10, ?15 ℃ separately, and the thawing temperature is 20 ℃ for all samples. The results show that the Cretaceous weakly cemented red sandstone is very sensitive to freezing and thawing process because of its very poor cementation and low strength. After only one freeze-thaw cycle, the uniaxial compression strength decreases 28.39%. The crustal stress during the freeze-thaw process enhances the binding forces of pores and cracks of Cretaceous weakly cemented red sandstone, therefore the freezing fully develops to the secondary micro-cracks. Due to the further development of freezing effect, the frozen-heave force in the red sandstone increases, strengthening the damage of red sandstone. Therefore, the mechanical properties of red sandstone after melting are further reduced relative to the non-confined freezing and thawing condition. The experimental results in this paper is helpful for the design of shaft lining of borehole located in the frozen coal mines in western China.

The accumulative landslide of Yushu Airport Road, induced by the Yushu M7.1 earthquake, has a slope of about 10 degrees and the size is 317 m × 482 m × 19.8 m in length, width and thickness direction. It is composed of three layers, i.e. an overlying layer consisted of gravel soil, a sliding zone consisted of pebble soil and a bedrock layer. In this paper, the large-scale shaking table model test is carried out to study the slope’s capacity to withstand the vibration load after the earthquake and the contribution of the vertical component of the earthquake to the stability of the slope, and its dynamic response characteristics and failure mechanism are also analyzed. The results show that the permanent deformation of accumulative formation’s landslide under strong earthquake is an important factor causing seismic geological hazards. With the increase of seismic intensity, the foot of slope firstly breaks down and settles. The arched fissures develop in the middle of the slope and the subsidence occurs. Furthermore, a series of tensile cracks and shear fractures occur at the top of the landslide and they advance towards the waist of the slope, showing typical tractive landslides. Peak ground acceleration(PGA), dynamic soil pressure and acceleration spectrum are positively correlated with the intensity of the input seismic wave and the landslide elevation. The PGA amplification coefficient exhibits distinct nonlinear characteristics, and its variation trend decreases with the increase of seismic load intensity. The vertical component of seismic wave has a slightly greater influence on the PGA amplification coefficient of landslide than the horizontal component.

The crack propagation under external load is generally non-uniform in space, which causes derivative anisotropies of the material. In this context, rock materials can be discretized into a large number of randomly distributed material points connected by force bonds. Based on the directionality of the force bonds, the local elasticity tensor is discretized into a certain number of directional tensors, and the relationship between the force bond modulus and the macroscopic parameters is deduced theoretically. An anisotropic elastic damage constitutive model is developed by considering the fracture effect of the force bonds, In order to simulate the mechanical behavior of brittle to ductile transition of medium porosity rocks in conventional triaxial compression tests, a damage suppression function is introduced into force bond fracture effect. The rationality and effectiveness of the model are verified by simulating the conventional triaxial experiments of Tennessee marble and Indiana limestone and comparing with the experimental data.

It is an innovation to use pebble cushion with furrow as the foundation of a large-scale immersed tunnel. In order to investigate the deformation characteristics of the pebble cushion subjected to engineering load under water, static load model tests are carried out in seven working conditions. The compression modulus of pebble cushion under various working conditions, the relationship between load and settlement are obtained. The effects of cushion thickness, ridge and furrow dimensions, preloading load and gradation on the mechanical properties of pebble cushion are analyzed. Unlike the gravel cushion, the settlement of pebble cushion initially changes nonlinearly with vertical load. With the increase of load, the settlement rate of the pebble cushion gradually decreases. The relationship between load and settlement tends to be linear after the cushion is compacted to a certain extent. The larger maximum size of the pebble results in a larger compression modulus of the cushion and a smaller settlement. The pre-compression leads to a significant reduction in the subsidence of the pebble cushion and an increase in compression modulus when compressed again. As the distance between the furrows increases and the compression modulus of pebble cushion decreases, the settlement increases. With increasing the thickness of the cushion, the settlement and compression modulus of the pebble cushion increase slightly. The research results can provide reference for the design of immersed tunnel cushion.

The hydraulic sorting effect in the dredger filling process results in uneven distribution of grains with different sizes in the calcareous sand foundation. Because of the formation of fine-grained layer, composed of silt and fine-grained sands with different contents of fine-grained particles, which is uneven distributed and very heterogeneous, it may lead to the bearing capacity and uneven settlement problems of the calcareous sand foundation. In this paper, to study the effect of fine-grained particles content on mechanical properties of calcareous sands, a series of triaxial shear tests under consolidation drainage conditions are carried out on fine sands with different contents of silty-sands. The results show that (1) the dilatancy property and peak deviator stress of soil samples decrease gradually when the content of fine particles increases under the same confining pressure. (2) All these fine sands with a dry density of 1.40 g/cm3, show the strain softening characteristics. The worst stability occurs in samples with 10% content of silty-sands, and it strengthens with the increased content of silty-sands. (3) The particle occlusion effect, makes a non-negligible apparent cohesion in the calcareous fine-grained sands. However, a significant decline in the cohesion occurs due to the decreased particle occlusion effect with the increase in silty-sand content when the fine particles content is less than 40%. Results in this paper can provide references for foundation treatment and slope stability analysis of fine-grained sand layers.

This study conducted a series of indoor and outdoor tests on Quelo sand in Angola, which has the characteristics of water softening and collapsibility. Four combination schemes with different levels of tamping energy(1 000, 2 000 kN·m) and different working conditions(natural, optimal moisture content) are adopted to carry out comparative consolidation tests. The variation rules of dry density, void ratio, heavy dynamic penetration number DPT(N63.5) and collapsibility coefficient of Quelo sand before and after dynamic compaction are analyzed. Furthermore, Suggested values of the impact depth, the effective reinforcement depth and the correction coefficient α of Quelo sand are proposed under different dynamic compaction programs. The experimental results show that although the physical mechanics indexes of Quelo sand can be significantly improved with the lower tamping energy level under the condition of humidification; however, its collapsibility can not eliminated obviously. Only when the tamping energy is high enough, the collapsibility of Quelo sand can be further eliminated. And there is a threshold of tamping energy in the dynamic compaction for eliminating the collapsibility of Quelo sand under humidification conditions.

With the development of industrialization and urbanization in loess regions, the problem of soil pollution caused by acid, alkali solutions in industrial wastewater, domestic sewage and rainwater with changed pH value is becoming more and more serious. Secondary disasters such as secondary subsidence may occur to different degrees under the load of surface buildings after the foundation soil is polluted. In order to find out the influence of acid-base pollution on the compression index at different pressure zones of Q3 loess, consolidation tests were carried out on intact loess samples contaminated by different concentrations of solutions. The results show that comparing the distilled water, the total deformation of acid-contaminated soil increases and the pressure section of the maximum deformation decreases; the total deformation of alkali-contaminated soil decreases and the pressure section of the maximum deformation increases. The compressibility of soil samples will change from medium to high compressibility with the increase of compressibility coefficient after acid solution contamination, but the compressive coefficient of alkali contaminated soil reduces somewhat. Compression coefficient, compression modulus, and volumetric compression coefficient of acid contaminated samples are more sensitive in medium and low pressure range less than 400 kPa, while those of the alkali contaminated samples are more sensitive under the high pressure range beyond 400 kPa. On this basis, the stress-strain relationship under confining condition is analyzed. It is found that the stress-strain curve of contaminated loess under different contamination conditions is in good agreement with the Guanry relationship.

The instability characteristics of fractured rock masses under unloading conditions can be better understood depending on compression-shear failure experiments on fractured rock samples. In this paper, a series of normal unloading tests on rock specimen with a single fracture were conducted under pre-peak shear stress. The deformation of rock samples and energy evolution were analyzed. The results show that the normal stresses at failure under normal unloading condition are larger than those of direct shear loading test, and the shear resistance of fractures decreases. The shear displacement increases with the decrease in normal stress. The normal deformation ratios(K) under normal unloading conditions increase with the increase in the contour area ratios(Rs) of rock fracture surfaces. The total energy decreases at first, and then increases during normal unloading process. This phenomena can be used to predict the imminent unloading instability of fractured rock masses. The results in this paper have certain references for understanding the unloading instability failure of fractured rock masses.

The unsaturated shear strength of fly ash is necessary for precise stability analysis of ash dams in unsaturated cases. However, less studies can be taken for references in this research field currently. A series of triaxial tests of unsaturated fly ash were carried out by controlling compaction, matric suction, and net confining pressures, and the effects of compaction and matric suction on the stress-strain curve and strength parameters of fly ash were discussed. The results show that the stress-strain curve of fly ash has no obvious peak and shows strain hardening characteristic when the matric suction is low. When the matric suction is gradually increased, the stress-strain curve has an obvious peak and presents the strain softening characteristics. If the compaction degree is larger, the strain softening characteristic is more obvious and the total cohesion and the effective internal friction angle are bigger. The total cohesion of the unsaturated fly ash increases with the increase in the matric suction, and the increase rate is gradually slower and tends to be stable. The effective internal friction angle of the fly ash sample with different matric suctions has little change, and is nearly equal to the effective internal friction angle of the saturated sample. As the matric suction increases, the contribution of the internal friction angle to the shear strength of the fly ash sample becomes smaller and smaller. The shear strength characteristics of unsaturated fly ash obtained in this paper have theoretical significance and engineering practical value for the design and stability analysis of ash dams.

Polyurethane is a kind of polymer, which can be used to improve the strength and integrity of material by filling pore and cementing soil particles. This paper used polyurethane to solidify the sea sand. The unconfined compressive strength tests and triaxial shear tests were designed to analyze the strength and deformation characteristics of solidified sea sand under different gelation time and different contents of curing agent. The experimental results show that with the increase of cementing time, the strength of solidified sea sand increases. More than 80% of the stable strength can be achieved at the solidification time of 3 hours. Moreover, the higher the mass ratio, the better the solidification effect. The polyurethane foam is filled into the pores, which increases the bite force between particles and enhances cohesion. The resistance of particle rearrangement increases, making the ability of sea sand to resist deformation become stronger, increasing the peak strength and changing the static properties.

The seepage prevention and reinforcement of fractured rock mass is one of the key technical problems that affects engineering stability and safety. In this study, the microbial reinforcement technology is applied to grouting reinforcement of fractured rock mass on the basis of the previous technology of microbial reinforcement of sandy soil. The results show that the confining pressure before the reinforcement has a great influence on the hydraulic opening of the fractured rock sample. After the reinforcement, the fracture is filled with calcium carbonate sediments and the seepage of rock sample changes from fissure seepage to pore seepage, and the unit seepage flow and seepage coefficient are controlled by confining pressure and seepage water pressure. After the reinforcement of microbial grouting, the percolation rate per unit time of fractured rock samples decreases by 80.31%-90.04%, and the percolation coefficient can reach the order of 10-6 cm/s. In the process of grouting and reinforcement of fractured rock samples, CaCO3 deposit induced by microorganism during grouting reinforcement of fractured rock samples have good cementation effect, and the splitted rock samples are cemented as a whole to achieve the effect of reinforcement and seepage prevention. The research results can provide a good reference for grouting reinforcement of fractured rock mass.

Expansive soil is featured with expansibility, multi-crack and over consolidation. When exposed to natural conditions, it vulnerable to the drying-wetting cycle effect of rain and evaporation, thus the shear strength of expansive soil will decreases with time, resulting in expansive soil slope instability. A series of ring shear tests is conducted on weak expansive soil sampled from Jinmen, Hubei province under the condition of drying-wetting cycles to investigate the variation of peak strength and residual strength. Results show that the shear strength of the sample is related to the normal stress, both its peak strength and residual strength increase with the increase of normal stress. Meanwhile, the greater the normal stress is, the smaller the shear displacement is when the sample reaches the residual strength. As the drying-wetting cycle increases, the peak strength of the expansive soil obviously attenuates, while the residual strength of expansive soils varies slightly, but it is not obvious, which can be approximately considered stable as stable. The peak cohesion and residual cohesion of expansive soils after three dry-wet cycles are almost the same. The difference between the peak internal friction angle and residual internal friction angle is always 2°, which is largely unaffected by the number of drying-wetting cycles.

An incremental method is proposed for analyzing the load transfer mechanism and deformation of pre-stressed anchor and soil nailing composite supporting system in the layered excavation process by considering the application of pre-stress as a loading step as well as assuming that the potential slip surface is a plane. Analysis of the influence of anchor pre-stress and setting position shows that with the increase of anchor pre-stress, the inclination of potential slip surface decreases gradually, and the contribution to the overall stability of foundation pit increases first and then decreases. Further, the contribution of the pre-stress becomes more significant as the position of the anchor becomes lower. The validity of proposed method is verified through comparisons among predictions using proposed method, numerical simulations and field investigations.

Three dimensional geological modeling is one of the basic tasks for the construction of "smart city", which involves urban-level geological range from hundreds to thousands of square kilometers. Because the urban geological model spans many geological units and the geological conditions are complex, the traditional three dimensional geological modeling methods have insufficient consideration of correct stratigraphic connections across different geologic units. A unified stratigraphic sequence method based on stratigraphic sedimentary sequence can identify the special condition such as stratum loss and stratum reversal, thus both the abnormal and the normal cases can be identified, avoiding incorrect interpolation in 3D geological modeling. By adding zero-thickness layers, the stratigraphic sedimentary sequence of all drilling holes can be completely consistent, and the match-up of stratum between drilling holes are constructed. In this paper, three special cases of stratum loss, reversal and repetition are discussed in detail, and this method is verified in different examples. The results can effectively make a clear judgment on the connection problems of various geological conditions in the three dimensional geological modeling in the urban. It can reduce the incorrect stratum connection in the three dimensional urban geological modeling with complex geological conditions and many geological units, and it provides efficient support for the development of "smart city".

Through the acoustic emission monitoring on two loading modes i.e. Brazilian splitting and uniaxial compression of limestone, the b-value characteristics of rock rupture and the impact factors are discussed. The results show that cumulative calculation of b-value has a high fitting degree and small error. When the step length is 5 dB, the overall change trend of b-value with time under different threshold values is similar. It shows that the calculation of b-value is more reasonable with the step length of 5 dB and threshold value of 40 dB, which can reduce the influence of non-rock fracture signals on the variation of b-value. Additionally, based on the fluctuation characteristics of the dynamic b-value, it can be seen that under the splitting load, the damage evolution of limestone can be divided into three stages: (0-40%)?c, (40%-90%)?c, and (90%-100%)?c. Under uniaxial compression load, the damage evolution of limestone can be divided into two stages: (0-80%)?c and (80%-100%)?c (?c is the peak strength). At different stages of rock damage, it presents different levels of damage in the rocks. In the crack propagation stage, compared with uniaxial compression load, the dynamic b-value increases steadily during crack growth process under Brazilian splitting load condition, and the dynamic b-value fluctuates greatly during the whole failure process. Through the experiments in this paper, it shows that the variation of b-value depends on the structural property and failure modes of the limestone fracture surface.

Rock mass quality rating index is the basis of rock slope stability evaluation. In revised Standard for engineering classification of rock mass, the calculation formula of quality index of slope engineering rock mass(QISERM) is added. In the formula there is a correction coefficient F3, which reflects the relationship between the dip angles of the main discontinuity and the slope. According to the standard, the value of F3 is determined by the difference between the dip angles of the main discontinuities and the slope. The paper discussed the flaw existing in the aforementioned method in determining F3, and gave the modification suggestions by applying the difference between the dip angle of the main discontinuity and the apparent dip angle of the slope in the dip direction of the main discontinuity. In addition, the QISERM of 13 357 examples were calculated using the two methods suggested by the standard and this paper, and then the difference of [BQ] between these two methods was analyzed. It shows that when the correction coefficient of discontinuity type and extension ? is equal to 0.6, the difference is less than 40 and 78.3% of them is less than 5; when ? is equal to 1.0, the difference is less than 60 and 68.7% of them is less than 5. The method proposed in this paper is more complete and more accurate than the new standard method in theory, and it can provide reference for the calculation of QISERM.

A high risk will be raised when the groundwater level changes dramatically during the construction of shield tunnel. However, the impact of fluctuation of groundwater table is seldom considered in the construction of shield excavation in current studies. Besides, the deformation and stress analysis of tunnel lining are paid even less attention. The non-uniform convergence deformation at the periphery of the tunnel cavity is introduced into this paper, considering the fluctuation of groundwater level when the shield tunnel is being constructed. The complex functions are applied in calculating the deformation of ground and lining and stress caused by shield tunnel excavation under the influence of the variation in groundwater table. The presented method is proved to be reliable by comparing the theoretical analysis results with the numerical simulation results. In addition, sensitivity analysis of parameters such as the groundwater level depth, thickness of lining, radius of tunnel is carried out. The results show that the radial displacement of the lining presents a stem-upwards "apple" shape. When the groundwater level decreases, the radial displacement gradually transforms from "duck egg" shape to "apple" shape that tends to be more and more obvious. The circumferential displacement of the lining likes an inclined "apple". With a decline of groundwater level, the range of the oblique "apple" expands. Under the influence of lining thickness, tunnel radius etc., the distribution of radial stress shows the same variation rules as mentioned above. With the increase in these two parameters, the distribution of radial stress gradually changes from the horizontal “duck egg” to “ ” shape. As the groundwater level drops, the shape of radial stress shrinks towards the center. The circumferential stress of the lining decreases slightly with an increase in thickness of the lining. While it decreases obviously with an increase in the tunnel radius. As the groundwater level drops, the distribution of circumferential stress gradually changes from the vertically placed "duck egg" shape to the "8" shape. The tangential stress of the lining shows the same variation rules under the influence of lining thickness and the tunnel radius. The distribution of tangential stress shows a stem-downwards "apple" shape. The research results can give a reference to the shield construction when the groundwater level changes obviously.

Taking the soil of the Great Wall of Ming Dynasty at Gulang segment as the research object, its mechanical strength and durability are studied by adding polymer material SH into remodeling samples under ultraviolet aging, freeze-thaw cycle, and salt and alkali resistant tests. In addition, scanning electronic microscope(SEM), energy dispersive spectrum(EDS) analysis and grain size analysis are used to study the mechanism of rammed earthen sites reinforced by SH material. The results show that the unconfined compressive strength of the solidified sample after ultraviolet aging test often increased at first and then decreased with the aging time. However, when the solid content of SH is very high in the remodeling samples, the strength of reinforced sample showes continuous increase trend after UV-aged tests. With the increase in cycling of freeze-thaw test, the unconfined compressive strength of the sample gradually decreases, and the mass loss rate gradually increases. The sample with SH solid content of 0.8% shows excellent frost resistance performance. Compared with raw soil, the salt and alkali resistant performance of samples reinforced by SH materials are improved obviously. SH materials mainly make the soil particles closely arranged and strengthen the connectivity through agglomeration, hydrogen bonding, bridging and wrapping. Thus it improves various properties of the soil at archaeological sites. which has good durability.

In order to realize the accurate and quick evaluation on the surrounding rock mass in front of tunnel face during the construction process, in this paper, a dynamic classification method for tunnel surrounding rock, which is based on traditonal BQ classification method, is proposed depending on machine learning and reliability algorithm . The machine learning tool is selected as the least squares support vector machine(LSSVM), and its parameters are optimized by the bacterial foraging algorithm(BFOA) to construct a nonlinear mapping relationship between the hierarchical index group and the surrounding rock level. The grading index group is made up of parameters include geological advance prediction results and the strength rebound value of the face surface, which are easy to be acquired during the construction process. Furthermore, the reliability theory is applied to verify the randomness problems that may exist in the results of some grading indicators. By constructing the functional function of reliability calculation through machine learning results, the surrounding rock classification with probability meaning are realized. To verify its feasibility in some sections based on calculation results, the new dynamic grading method is applied in Zhenfengling tunnel and its applicability is proved by automated monitoring data. The results show that the classification method can effectively realize the dynamic grading calculation of surrounding rock during construction, which provides a new idea for the dynamic design process of tunnel construction.

Due to the complex construction process, many risk factors and high collapse accident rate of tunnel constructed by drilling and blasting method, the application of traditional fault tree analysis (FTA) method is limited by the probability, the logical relationship, and the fault state of events only considering the binary-state. In this paper, a method for evaluation of collapse possibility of construction tunnel by drilling and blasting method based on T-S fuzzy fault tree is developed to reduce construction risk and provide decision-making basis for prevention. In this method, fuzzy numbers and T-S fuzzy gates are introduced into fault tree analysis. T-S fuzzy gates are used instead of traditional logic gates to describe the relationship between events, which reflects the fuzziness of fault mechanism and event connection, and reduces the difficulty of establishing the fault tree. Fuzzy numbers are used to describe the fault probability and degree of the bottom event, which overcomes the problem that traditional fault tree relies too much on probability and can not consider the influence of the fault state of events. This method not only can calculate the tunnel collapse possibility by using the priori fuzzy possibility of the bottom event and the actual fault degree in construction, but also can guide the risk control work according to the result analysis of bottom event importance. Two engineering examples are analyzed. Results show that the proposed method is more in line with the engineering practice than the traditional fault tree method. It can evaluate the possibility of tunnel collapse more scientifically and rationally and determine the key risk factors. It can be used as a decision-making tool for tunnel construction safety risk analysis and management.

The resettlement of residents in the nearest possible areas is always applied in the Three Gorges Reservoir region, China. But engineering disturbances and heavy rainfall often induce the landslide instability. This paper discusses the impact factors and formation mechanism of the Jinjiling landslide in Jiangdong district, Wushan county, Chongqing, China. It shows that the cutting and filling of slopes lead to loading of the trailing edge and removal of material from the leading edge. More importantly, the filling of channels with surface soil blocks the surface water discharge channel along the landslide body, promoting the transformation of surface water into groundwater. Therefore the groundwater level in the Jinjiling landslide rose substantially under heavy rainfall, resulting in a large deformation on approximately August 1, 2018. After emergency treatment, including groundwater pumping from the landslide body, the landslide deformation slowed considerably. It is analyzed that the groundwater rise caused by filling of the landslide body is the key factor of the landslide. The numerical simulation shows that the stability coefficient of the engineering decreases obviously after the disturbance of engineering, and the seepage field in the corresponding filling area changes obviously. The seepage is intensified, characterized as the increase in the pore water pressure, hydraulic gradient and total water head. Based on Darcy's law and principle of effective stress, the increase of seepage force and decrease of shear strength induce landslide deformation.

The permeability coefficient function of unsaturated soil is difficult to be measured and a long time is required. Wetting front advancing method(WFM) can be applied to measure the hydraulic conductivity function(HF) of unsaturated soil in a short time. However, this method relies on an artifical recognition of wet front, which has certain limitations. In addition, the measurement accuracy of WFM is not clear and needs to be verified. In this paper, impacts of parameters include of initial water content, critical water content of wetting front, rainfall intensity, location of sensor etc on the accuracy of WFM is analyzed. The Seep/W software is used to simulate the infiltration process occurs in the homogeneous soil column. The data is analyzed by WFM to calculate the permeability coefficient of the soil. Simulation results are compared with the input permeability coefficient(i.e. the real solution). The calculation accuracy of the wetting front method is evaluated and the source of error is discussed. The results show that WFM can obtain relative high accurate calculation results. Using the characteristic moisture content at the wetting front to calculate wetting front advancing rate, it overcomes the limitations of original wetting front advancing method with artificial observation and greatly expands the applicability of WFM. The sensor spacing has no direct effect on the calculation accuracy of the permeability coefficient function of wetting frontal advance method. The lower the initial water content, the greater the rainfall infiltration rate, and the greater the span of the permeability coefficient function. Based on the analysis in this paper, the following suggestions are proposed for test design of WFM. The soil column with a length of 30-40 cm and 3-4 sensors are suggested. Any initial water content is acceptable and the dry sample is the best choice. To avoid the surface ponding problem, the rainfall intensity should be smaller than the saturated permeability coefficient of soil.

In order to overcome the shortcomings by using the traditional finite element method, this paper presents a new method in the stability analysis of soil-rock slopes considering the random distribution and proportion of rock blocks. Based on the rock block proportion and gradation of soil-rock slope, the soil-rock mixture model is randomly generated for the rock block proportion in the range of 10% to 60%. Each rock block proportion model takes into account eight different block distribution locations. Finally, the generated model is imported into OPTUM G2 to establish the model of soil-rock slopes. The stability of soil-rock slopes is analyzed by finite element limit analysis method, and the calculated results are compared with those obtained by two kinds of equivalent strength parameter models. The results show that the maximum and minimum limits of safety factors of soil-rock slopes with the same rock block proportion vary greatly due to the difference in the spatial distribution of blocks. The plastic zone in the soil-rock slopes is no longer circular-shape, but shows three typical expansion modes of "around stones", "distributary" and "inclusion". The safety factors of soil-rock slopes obtained by using two kinds of equivalent strength parameter model and random rock block model are quite different. The research results can provide reference for the design and construction of soil-rock slopes.

In order to further study the effect of shear rate on the strength and deformation characteristics of calcareous sand, direct shear tests are carried out on dry calcareous sand under different shear rates in this paper. With the increase in shear rates from 0.1 mm/min to 2.4 mm/min, the shear strength of calcareous sand decreases at first and then increases in the tests, and a similar change trend of the internal friction angle is also found. Meanwhile, the measured critical shear rate is 1.6 mm/min for dry calcareous sand. Under a low normal stress condition, the shear dilatancy phenomenon in calcareous sand sample is more likely to occur with an increase in the shear rate. Under a relatively high normal stress condition, overall shear shrinkage decreases gradually with the shear rate increases from 0.1 mm/min to 1.6 mm/min. When the shear rate continues to increase from 1.6 mm/min to 2.4 mm/min, the maximum shear shrinkage increases gradually. The micro-mechanism of loading rate effect of calcareous sand varies under different normal stress conditions. Under the low stress conditions, the loading rate effect of calcareous sand is mainly caused by the dislocation, displacement and rearrangement of particles, while the particle breakage plays an important role at higher stress levels.

Good sealing performance is the fundamental prerequisite for a safe operation of a salt cavern gas storage engineering. However, the complicated geological conditions, the lack of practice in building gas storage, and the incomplete theoretical knowledge will lead to various leakage risks for the gas storage in the salt cavern. According to engineering practices and statistics on accidents worldwide, and combined with the unique engineering geological conditions and operational states of the salt cavern gas storage, three major leakage factors (i.e. geological factors, engineering factors and human activity) and four leakage types were summarized in this paper. The potential leakage types in underground salt cavern gas storage include near-horizontal leakage of gas caused by insufficient interlayer sealing, upward movement of gas through the strata caused by the breakthrough and failure of caprock, natural gas leakage along the wellbore due to its insufficient integrity, the gas flow to the faults owing to the connectivity between the interlayers and faults. Based on their respective leakage characteristics, corresponding preventive measures were proposed to prevent the occurrence of gas leakage accidents and its large-scale spread. In view of the fact that the development of salt cavern gas storage in China is temporarily at a developing period, this research results have certain reference and guiding significance for safety construction of deep salt cavern gas storage.

Grouping structural planes based on their occurrences is an important way in analyzing the structure characteristics of rock mass. Traditional classification methods usu CUI Xue-jie, YAN E-chuan, CHEN Wu ally greatly rely on geological experience of researchers, which is lack of objectivity. The powerful clustering methods, however, also have drawbacks. In this paper, based on variable length string genetic algorithm, an improved K-means clustering method was proposed, making the automatic clustering of discontinuity occurrence of rock mass available. The essence of the proposed method is to select appropriate initial cluster centers for K-means algorithm applying the genetic algorithm. It overcomes the limitation that the K-means algorithm is usually affected by the initial cluster center and easily converges to the local optimal solution. The application of variable length strings in the improved algorithm of classification, however, can not only automatically determine the number of the optimal structural plane groups during the clustering process, but also provide optimal grouping results. In addition, a new mutation algorithm is proposed based on the occurrence data. It is realized in C++ and is applied in analyzing the occurrence of structural planes in an underground water-sealing rock caverns located at Zhejiang province, China. Reasonable classification results are achieved, proving the applicability of the proposed approach in this paper.

Coral soil is mainly composed of calcareous sand and reef limestone. A series of in-situ static loading tests was conducted in the prestressed high strength concrete(PHC) piles located at the coral soil, which were used to clarify the interaction between PHC piles and coral soil while loading. Fiber bragg grating(FBG) sensing technology was applied to collect strain values along pile shaft during tests, and then pile axial forces were calculated. Base on the values of axial forces, the piles bearing capacities have been analyzed. Experimental resutls show that: (1) the responses of pile-soil are in the linear elastic stage during these tests; (2) the bearing performance of pile is significantly related with its penetration depth in the soil. When the penetration depth exceeds 15 times of the pile diameter, the resistance at the tip of the pile reaches its peak; (3) the installation style of the pile hardly affects its bearing capacity in coral soil, but it significantly affects the penetration energy; (4) the penetration depth and energy affects the degree of fragmentation of the coral soil, which constrains the bearing capacity of pile. Based on in-situ testing, it can be seen that in the coral soil, when the penetration depth is less than 15 times of the pile diameter, the penetration depth is positively correlated with the resistance at the tip of the pile, and is inversely related to the friction resistance at the side of pile .

In order to establish the relationship among microscopic properties, loading rates and macroscopic responses of cemented geo-materials, the parallel bond model was employed to simulate the physical and mechanical characteristics of the cemented materials based on the discrete element method(DEM). Moreover, a series of undrained triaxial numerical tests were performed with different microscopic parameters (number, cohesion, Young's modulus, internal friction angle of bonds, and void ratio) and loading rates (1, 0.1, 0.01, 0.002 mm/min). Based on three parameters include residual strength, peak strength and the corresponding axial strain, the rate-dependent behaviors of the cemented geo-materials under different microscopic parameters were discussed. Furthermore, based on the numerical test results, BP neural network algorithm is used to establish an intelligent model for predicting the rate-dependent behavior of the macroscopic properties in cemented materials. The results show that: (1) the cemented material has a significant loading rate sensitivity characteristics, presenting a significant increase of the peak strength with the increase of loading rate, and it has a semi-logarithmic linear correlation. However the residual strength and axial strain at peak strength are less sensitive to the loading rates; (2) the rate-dependent behavior of the cemented material is mainly caused by the fragmentation of internal bonds. During the whole shearing process, the evolution of average bond breakage percentage per unit strain shows a similar trend as that of the deviatoric stress. In addition, the average bond breakage percentage increases with the increase of loading rates; (3) the proposed BP neural network model, which considers the influence of microscopic parameters and loading rates on macroscopic responses, can reasonably describe the rate-dependent behavior of the macroscopic properties in cemented materials with the relative error around 10%.

Accurate determination of statistical characteristic values (e.g. mean and standard deviation) of geotechnical design parameters is an important prerequisite for geotechnical reliability analysis and design. In this paper, under the condition of satisfying the accuracy of statistical design of geotechnical design parameters, a new method is proposed to determine the minimum test data quantity of geotechnical engineering. The relative error and relative variability index are defined to measure the accuracy of the statistical characteristics of geotechnical design parameters. The influence of the static cone penetration test data on the calculation accuracy of the statistical characteristic value of effective internal friction angle of sand soil is systematically analyzed. Moreover, the minimum cone penetration test data quantity is determined based on the relative error and relative variability index. The results show that the relative error of mean value of internal friction angle of sand estimated indirectly from static cone penetration test is small. The relative variability index of sand internal friction angle decreases with the increasing in the data quantity of static cone penetration test. The ratio of uncertainty caused by insufficient cognition to total variability decreases with the increasing data quantity of static cone penetration test. When the allowable relative variability is less than 0.2, the variability of sand internal friction angle is in the range of 5% and 20% COV, which satisfies the predetermined requirement of minimum static cone penetration test data quantity of 10-100. If the relative variability is allowed to be less than 0.3, the minimum static cone penetration test data quantity is 5-43. In addition, when estimating the geotechnical parameters, the uncertainty of empirical regression model has a significant influence on the minimum data quantity. The minimum data quantity of static cone penetration test increases with the increasing uncertainty of empirical regression model. Therefore, the test data should be collected as widely as possible and the calculation model should be selected with higher accuracy to obtain more accurate statistical characteristic values of geotechnical design parameters.

The deformation of expansive soil slope of the South-to-North Water Transfer Project is significantly affected by the dry-wet cycles such as rainfall and evaporation. The characteristics of expansion-shrinkage deformation and vertical displacement deformation of the expansive soil at canal top under dry-wet cycling are revealed, which provides a basis for the stability evaluation and treatment plan of the slope. Based on the experimental results of the expansion-shrinkage characteristics of expansive soil, combining the field observation data such as vertical displacement, water content, rainfall, evaporation and temperature monitoring, the correlation between the displacement of the canal top and the influence factor is analyzed, and a model for vertical deformation of the expansive soil at canal top under the action of dry-wet cycling is established. The results show that the expansion-shrinkage factor composed of water content, rainfall and evaporation are negatively correlated with the vertical displacement of the canal top. The vertical displacement component is smaller due to the expansion-shrinkage factor. The temperature factor is positively correlated with the vertical displacement of the canal top, and has a great influence on the vertical deformation of the canal top, which is the main factor affecting the deformation of the canal top. The slope excavation(time-effect factor) has a certain influence on the vertical deformation of the canal top at the initial stage of construction, but little influence on the vertical deformation of canal top after excavation.

Based on the monitoring data of the foundation pit project in the adjacent metro tunnel in operation, the relationship between the lateral displacement in the deep soil and the deformation of the adjacent metro tunnel during the excavation process of the foundation pit has been studied. The dangerous nodes and key affected areas during the excavation of foundation pit are identified and discussed. It shows that the construction of retaining support structure and the precipitation at the early stage of foundation pit excavation have an initial displacement effect on the stratum and adjacent metro. The long-term unsupported exposure of the retaining structure results in the rapid growth of the lateral deformation of the foundation pit. The excavation of foundation pit has a spatial effect. The lateral deformation is greater at the center than the corner of the foundation pit. The one-way excavation easily causes the superposition of the displacement field and the stress field of the soil in the post-excavation area, causing the maximum deformation of the adjacent tunnel to be moved to the back excavation area; When the excavation depth of the foundation pit is close to that of the adjacent metro, the tunnel structure produces significant horizontal displacement and convergence deformation with “transverse duck egg” type , but the vertical displacement fluctuates little; The lateral displacement curve of the deep soil presents the stepped belly-shape, and the maximum horizontal displacement of the soil is in linear with the tunnel deformation in a small range. However, with the increase in the lateral displacement, the deformation of tunnel deviates from the fitting curve and tends to the superlinear increase, and this should be paid more attention in engineering.

Soil-bentonite cutoff walls have been widely used for municipal sanitary landfill in United States. However, there are less engineering applications in China and domestic bentonites and stratum have been mixed and formed to cutoff walls under the consolidation of self-weight stress. Less information is known to how the porosity, permeability and compressibility of cutoff walls are influenced by the addition of bentonite. To solve this problem, the effects of bentonite content on the k, n and av of sand-bentonite backfills are investigated based on several improved flexible wall permeability tests and consolidation tests using Fujian Standard sand as the simulated stratum and three typical bentonite mixtures as additional material. The results show that when bentonites and sand are mixed into cutoff walls, there is a critical bentonite content Copt, where the n and the av are lowest. If the bentonite content is less than or equal to Copt, k decreases rapidly and av reduces slowly with the increase in bentonite content. As the bentonite content is larger than Copt, reduced rate k slows down and av increases rapidly with the increase in bentonite content. In theory, when the bentonite content is much lower, clay particles only fit within the sand pore space and without disturbing the sand pack pattern. However, after bentonite content reaches Copt, sand grains become disconnected and are suspended in the bentonite, resulting in increased porosity of the backfill increases with bentonite content, which may be the main reason of the optimum bentonite content at the macro level.

The rock mass is a natural damage geological body with both macroscopic and mesoscopic flaws. How to accurately assess the co-effect of these two types of flaws on the rock mass mechanical behavior is an important and difficult issue to be solved urgently. The Elastic-brittle model and Null model in FLAC3D code are adopted to describe the mechanical behavior of the intact rock and these two types of flaws respectively, and the superfine element division method is adopted to divide the numerical model to simulate the rock mass failure. Meanwhile a new method of reflecting mesoscopic flaws by rock porosity is developed to investigate the effects of the void ratio, the pre-crack dip angle and length on mechanical behaviors of the fractured rock mass. The proposed method is adopted to study the effects of these two types of flaws on the factor of safety(FS) and critical failure surface(CFS) of rock slope. The results show that the macroscopic flaws control the failure mode, peak strength and elastic modulus of the rock mass under uniaxial compression, and also govern the failure mode and FS of the rock mass slope. Although the mesoscopic flaws can not change the control effect of the macroscopic ones on the mechanical behavior of the rock mass, they do have some influences on this control effect. In sum, the mesoscopic and macroscopic flaws impact on the rock mass mechanical behavior with different action mechanisms.

In recent years, the number of large-area filling sites projects have been greatly increased in China. This can alleviate contradictions between supply and demand of construction lands. However, it also gives rise to many engineering problems. For example, subsidence of large area site increases negative skin friction(NSF) on pile shaft, which affects the pile safety. However, there have been limit studies on NSF of frictional piles occur in large area filling sites. Based on a field test, a single pile finite element analysis model is established, and the rationality of the model is verified by comparing with the measured data. Based on the verified numerical model, influences of weight and thickness of fill, load on pile head, surcharge loading and field consolidation time on skin friction of a single frictional pile are studied. The results show that fill thickness and load on pile head play major influences on neutral point position. An increased negative frictional force is associated with an increase in thickness and weight of fill, and a reduction in fill consolidation time. It is also linked to a reduction of load on pile head and increase in surcharge. Sensitivity analysis show that factors related to fill formation play a more significant role in NSF of pile shaft than the loading factors. The applicability of the common design methods used at home and overseas, calculating the NSF of frictional pile in large filling sites is discussed. The results can provide guidance for engineering application of frictional pile in large filling sites.

Accurately and vividly analyzing the propagation law of microseismic wave in different media of coal and rock can provide strong theoretical basis for active seismic exploration in the coal mine and the inversion on the media structure in the different regions. In this paper, based on two-dimensional media models in the coal mine, including the general homogeneous medium model of coal/rock, the multi-media model combing the coal and rock, the fault-bearing medium model of coal and rock and water-bearing two phase medium model of coal and rock, the wave field simulations of different media above are carried out using the high-order finite-difference method and the wave field characteristics are analyzed. The results show that the high-order finite difference method is powerful in accurately obtaining the wave field characteristics of P- and S-wave in different media models. Different wave field characteristics appear in different media models. In the multi-layered media of coal and rock, the reflection, transmission and transition of P and S wave occur at the interfaces of different media, and the transmissive wave, the transmitted converted wave, the reflected wave and the reflected converted wave appear. In the fault-bearing media model of coal and rock, not only the transmissive wave, the transmitted converted wave, the reflected wave and the reflected converted wave can occur , but also the diffracted wave can appear at the fault. In the water-bearing two-phase medium model of coal and rock, not only the fast P-wave and shear-wave, but also the slow P-wave can occur. The snapshots of the wave field at different time can further vividly describe the dynamic characteristics of the wave field, and they can provide the theoretical foundation for analyzing the micro-seismic wave field in the coal mine.

Excavation in the rock mass engineering is inevitably associated with the rock mass unloading. In this paper, the granite is selected as the test material, the rupture characteristics of the granite under linear unloading conditions are studied experimentally. The rapid dilation in the radial direction and the enhancement of the plasticity are found. Furthermore, the FJM (flat joint model) in the PFC (particle flow code), is applied to simulate the mechanical properties of the granite under unloading conditions. A new “stress/time step” unloading method, which can easily execute the simulation of the nonlinear unloading behavior, is used considering that the excavation of the rock mass in the practical engineering is mostly characterized as the nonlinear unloading behavior. Numerical simulation results show that the granite is seriously damaged when the axial pressure is increased and the confining pressure is unloaded. The load capacity was decreased sharply and no residual strength is found. As the initial confining pressure increases, the failure mode is transformed from the splitting to the shear failure. The faster the unloading speed, the slower the crack propagation is found. The research results can provide the reference for the study of the mechanical behavior and the practical engineering under unloading conditions.

In view of the continuous deterioration of strength parameters of rock and soil during the progressive failure of slope, the strain-softening model, instead of the traditional ideal plastic model, was applied. An improved vector sum algorithm, which considers the tensile-shear progressive failure, was proposed in this paper. Both the safety factor expression and the method of searching the most dangerous sliding surface were studied. Based on vector characteristics of sliding and force, the sliding trend and safety factor expression of vector sum were improved. Moreover, the most dangerous sliding surface of slope was searched based on the improved adaptive genetic algorithm. Results of several case studies show that, when the strain-softening model was adopted, the safety factor based on the definition of vector sum and the most dangerous sliding surface shape have dynamic changes and adjustment characteristics. The proposed algorithm in this paper, which is based on the tensile-shear strain softening model, expands analytic methods of slope vector sum to a certain extent.

A rational constitutive model is critical for numerical simulation of rheological behavior of debris-flow. It is also one of the key issues that determine the accuracy of numerical simulation of its dynamic process. The numerical models based on Bingham and Cross constitutive models has been observed failing to simulate the shearing thickening and thinning phenomenon of the rheological behavior of debris-flow that belongs to multi-phase mixture of fluid and debris. We discuss the numerical divergence problem of using the Bingham model with low shear strain rate. To simulate the dynamic flow process of diluted debris-flow, the 3D numerical model is set up, applying an alternative solution of the Herschel-Bulkley-Papanastasiou(HBP) constitutive model with the smoothed particle hydrodynamic(SPH) method. Comparing to the traditional two-dimensional numerical model based on the hypothesis of shallow water wave, the proposed method in this paper solves Navier-Stokes equations describing the debris-flow with the SPH in three-dimension, thus the velocity field and deposition pattern of debris flow can be simulated. The proposed method incorporates the HBP constitutive model, ensuring the numerical convergence, can perform well to reveal the non-linear variation of stress-strain relationship of debris flow at transition phase of plastic yield and large shear rate. The proposed method is verified by a flume experiment. It indicates that the simulation results match well with the experimental measurements.

Variable stiffness optimization design is always an important and difficult problem for large-scale pile-raft foundations in complex environments. In this article, a two-stage optimization method was developed using finite element analysis, According to the stress distribution on top of piles in the traditional design with uniform pile arrangement, the pile groups are divided into several sub-groups firstly. Then the adjustment coefficient of the number of piles is determined according to the relationships among the pile top stresses of every sub-group. Finally, by adjusting the spacing of piles to change the number of piles in each sub-group, the variable stiffness optimization design can be realized. This procedure is used to optimize the large-scale piled raft foundations with non-uniform superstructure loads in multi-layer soils. The results show that after optimal design, the differential settlement of the raft, the average overall bending moment and the differential stress at the top of the pile are significantly reduced. The method is simple in calculation and wide in application, and is not limited by complex soil conditions, non-uniform superstructure loads and the size and shape of the pile foundation.

To solve the problem that the positioning accuracy of the acoustic emission time difference localization algorithm is affected by many factors such as rock wave velocity in rock mechanics test, an unknown wave velocity acoustic emission localization algorithm based on particle swarm optimization is developed in this paper. The algorithm takes the rock wave velocity as an unknown value, establishing the objective function based on the least squares principle according to the picked time difference, applying the particle swarm optimization algorithm to solve the objective function and find the target position and wave velocity. The weight coefficient is an important factor affecting the accuracy and stability of the algorithm. Firstly, the optimal weight coefficient 0.729 8 is determined by numerical model, which can satisfy the accuracy and stability requirements of the algorithm. The numerical simulation results show that the calculation accuracy based on the selected weight coefficient is higher than the traditional wave velocity algorithm. In order to verify the algorithm in practical application, the lead-breaking test is carried out. It is proved that the algorithm is superior to the traditional wave velocity algorithm.

The single particle shape is irregular in the rockfill, which is characterized as random strength and size effect. The development of rockfill particle model is the basis of particle flow simulation in laboratory tests of rockfill materials. Based on the micro-crack propagation mechanism of particle breakage, a random crack model with three-dimensional shape is carefully proposed in this paper. Furthermore, the parameters of model are analyzed and discussed. The model can reasonably simulate the behavior of brittle splitting, Weibull distribution and size effect of the rock strength. It is the premise of in-depth simulation analysis of particle flow simulation of damming material crushing and scale effect. The single particle numerical test shows that the particle shape has a significant influence on the breaking mode and particle strength under the same lithological material and fracture distribution. The more irregular the particle is, the greater the dispersion of particle strength.

Because of the progressive damage of slope slip, it is significant to study the local safety degree of the slope. In this paper an improved method on calculating the local zone safety index (ZSI) is established to describe the stability state of local rock and soil mass at different stages of elastic, plastic and damage. studied the safety degree space distribution of slope by dynamic time-history numerical simulation. Based on the numerical tests of the centrifuge loading on the static slope, the rationality of the improved local zone safety index is verified. The time-history dynamic numerical simulation of the slope affected by blasting is carried out, and the distribution modes of local safety index of slope under different blasting parameters are obtained. The results show that with the increase of blasting load, pore water pressure and blasting action time, the negative area of ZSI at the slope also increases and the sliding surface is extending. The velocity increase zone is located at the sliding body, the damage zones are mainly located at the regions affected by blasting activity and sliding zones. The slope sliding surface is obviously affected by pore water pressure under the blasting action. The blasting action time obviously affects the damage zone near explosion source. The method and conclusions from this paper is meaning for understanding the mechanism of slope progressive destroy with blasting effects.

To study the effect of fissures and soil-root interstices on hydraulic and mechanical characteristics of the soil, the FEM numerical tools are used to study the influence of fissure and soil-root interstices on the infiltration during the rainfall process. Then the influence of the rainfall on the reinforcement of roots is analyzed. The zonal strength reduction method is applied to simulate the direct shear tests of the root-soil composite before and after the rainfall, considering the impacts of lateral roots’ angles. The results show that the fissures and soil-root interstices become the preferential channels for rainfall infiltration. The depth of rainfall infiltration increases with the depth of fissures. And it is mainly controlled by the vertical depth of the major root, while the influence of the inclination angle of lateral roots is small. The rainfall infiltration depth with root interstices is increased by 93.3% compared with the case without the root in the soil. The root system can significantly improve the shear strength of the soil. The shear strength of the soil with 60° lateral roots is the greatest, followed by the condition with 45°, 30° and without roots. Rainfall infiltration not only reduces the strength of the soil, but also weakens the reinforcement effect of roots. Therefore, the shear strength of soil-root composite system is greatly reduced after the rainfall, resulting in the sliding of the vegetation-covered slopes in a shallow depth.

As an efficient foundation reinforcement technology, dynamic compaction technology has been widely used in various engineering constructions. However, most related theoretical research mainly focus on the tamping at a single location, and the interaction between adjacent tamping points under tamping at several locations simultaneously is rarely studied. In order to investigate the reinforcement impacts of multi-tamping under different arrangement forms of tamping location, the geometric nonlinear finite element method for large deformation and the constitutive model of “cap” in the framework of LS-DYNA are applied to analyze the degree of soil compactness under strong tamping at multi-locations. Firstly, based on the tamping experiment at a actual situation of site construction, the rationality of the model is verified by comparing the numerical results with the measured data for the lateral displacement in the soil around the tamping points. Moreover, considering two common arrangements of tamping points (i.e. rhombus-grid pattern and square-grid pattern), numerical simulations are carried out on the dynamic construction process, and compaction degree of the soil under different layouts of tamping points are studied. The results show that the compacted area between adjacent tamping locations is susceptible to multi-tamping effects, and the reinforcement degree of soil near the subsequent tamping location is better than that near the previous tamping location. Besides, the reinforcement of soil is better in regions applying rhombus-grid tamping points compared with that of square-grid pattern. In addition, by using the depth of the crater and the effective reinforcement depth, the soil affected by the strong tamping can be divided into three parts: the dilatation zone, the strong reinforcement zone and the slightly disturbed zone.

The composite element method(CEM), as one kind of the numerical simulation method, has very high efficiency in modeling. It is suitable for the dynamic design of rock slopes, which can realize quick adjustments of structural planes, reinforcement treatments and excavation surfaces in the model. The algorithms for structural planes and reinforcement treatments have been completed in the CEM. However, the simulation of the excavation surfaces and process has not yet been studied. In this study, the excavation model is set up by using the new algorithm and taking the CEM for simulating excavation surfaces for references. Moreover, the relevant program is coded with FORTRAN. Based on an example of rock slop excavation, the simulated results of the new CEM method are compared with the finite element method, and the validity of the proposed excavation simulation method is proved.

The influence of rock fragmentation on the trajectory of rolling rock on slopes is investigated using energy tracking method (ETM). Based on the impulse method, the ETM develops a multiple point energy conservation algorithm to improve the calculation accuracy. Compared with the traditional penalty function method, the advantage of the impulse method is that the ETM does not rely on penalty parameters, thus it improves the accuracy of simulation; ETM is a coupling algorithm of analytical solutions and numerical solutions, which improves the computational speed. Compared with traditional impulse methods, such as sequential impulse method and simultaneous impulse method, ETM can track the energy transfer between multiple blocks and multiple collision points during the collision process. Therefore, it can accurately simulate the high frequency collisions among hundreds and thousands of collision bodies. Failure pattern recognition simulation based on Weibull distribution is used to simulate the fragmentation of the rolling rock during the falling process. The results indicate that: (1) ETM is able to accurately simulate the three-dimensional trajectory of rolling rock on slopes; (2) when considering the rock fragmentation effect, the trajectory of the rolling stone fragments changes significantly compared to that of not considering the fragmentation of the rolling rocks. In this case, the horizontal lateral displacement increases obviously, so the slope protection scope needs to be enlarged horizontally. The effectiveness of the ETM and pattern recognition simulation in simulating falling rocks are verified in this paper, which is able to provide modelling tools for avoiding falling rock disasters.

The wall of borehole tends to deform under in-situ stress state, resulting in a change in the cross-sectional shape of the borehole. Accordingly the morphology of borehole after deformation also reflects the stress state of borehole. Based on the classical rock mechanics theory, the geometric shape of borehole under the plane two-dimensional stress is studied, and the relationship between the stress and geometric parameters of the circular hole after deformation is established. The stress solution based on the elliptical parameters of the borehole is realized. The measuring method and measuring technology of borehole elliptical parameters is proposed, forming a new in-situ stress measuring method based on borehole shape analysis. The conclusions are as follows: (1) under the action of plane two-dimensional stress, the geometric shape of the circular hole turns to be ellipse, and the relationship between the ellipse parameter and the stress is derived; (2) the parameters of ellipse can be resolved by using the diameter values in three different directions. Besides, the contact micro-optical aperture measurement technology is developed to measure the morphology of borehole. The feasibility of this new measurement technology is analyzed from the principle point of view. Furthermore, the possible problems of this technology in measuring the in-situ stress were discussed; (3) based on the indoor simulation tests, the feasibility of the technical principle is verified. The accuracy of measurement results is also discussed through error analysis and example calculation.

With the development of underground space, more and more attention has been paid to the special areas on borehole wall. In order to overcome the low resolution of traditional ultrasonic borehole imaging technology, a new borehole imaging method based on ultrasonic synthetic aperture technology is proposed to improve the resolution and image quality. Firstly, high resolution synthetic aperture focusing technique(SAFT) is used to synthetically process a large number of cross-section scanning echo signals, so as to enhance the reflection signal and suppress the interference signal, which makes the echo of the corresponding typical area at the rock wall more prominent. Then the sound amplitude and sound time parameters of the scanned cross-section echo signals are extracted. The principle of ultrasonic drilling image reconstruction based on correlation is proposed based on the synthesized characteristics of sound amplitude and sound velocity parameters. After superimposing the detection data of different depths, the borehole reconstruction image with more obvious focus area is formed. Finally, the method is applied in practical engineering, and its feasibility and accuracy has been verified.

Measurement on strain deformation of soils is one of the significant challenge in geotechnical testing technology. Double integration of acceleration measurement for displacement is a common method in understanding the shear strain response of soil in in-situ monitoring and model tests under dynamic loading, but it lacks validity and reliability study. In this paper, the sand-slope-rocker-mass vibration table model test was designed, besides the mass acceleration and time-history response of displacement under multiple working conditions and two seismic loads were measured. Four representative methods were applied to solve dynamic and permanent displacement and their reliability and deflection effects were discussed. The results show that high-pass filtering method, which is less affected by the integral methods, is reliable to obtain dynamic displacements. Comparing the integral displacement with the measurement values in different dynamic displacement conditions under two earthquake loadings, the average errors of peak values are less than 25%. However, the dynamic displacements calculated by double integration present obvious deviation, which is related to the function form and processing flow. To obtain permanent displacements, the integral methods consisting of high-pass filtering process are invalidated due to the loss in low frequency components of acceleration records. The permanent displacements calculated by double integration method are in good agreement with the measured values. The relative errors are 7% and 5% respectively under two seismic loading conditions. The base deflection presents less impact on the reliability of dynamic displacements in different integral methods. But it plays obvious impacts on the permanent displacements. The permanent displacements calculated by double integration are distorted induced by the influence of base deflection. Experimental methods and conclusions herein provide valuable scientific reference and useful suggestions in the selection of double integral methods of acceleration to obtain soil displacement .