The waste tire-faced retaining wall is an ideal way to effectively utilize waste tires. However, vertical modular waste tire-faced retaining walls cannot withstand high-intensity seismic action. Therefore, the geogrid striped-reinforced method is proposed to improve its seismic performances. According to the soil-structure dynamic similarity system, a shaking table test model of the geogrid striped-reinforced waste tire-faced retaining wall is designed. This test model can consider the influence of seismic intensity, seismic wave, length of geogrid reinforcement, spacing of geogrid reinforcement and the slope of the wall. The seismic response characteristics were analyzed, such as the tire-faced wall and the backfill acceleration, the lateral displacement of the wall, the settlement of the backfill on the top of the wall and the dynamic soil pressure on the back of the wall. In addition, the characteristics are compared with those of the shaking table test of the waste tire-faced retaining wall (unreinforced). The results show that geogrid striped-reinforced tire-faced retaining walls significantly improve the seismic response characteristics of unreinforced retaining walls, and improve the seismic performance of the tread retaining wall. Thus, the geogrid striped-reinforced vertical waste tire-faced retaining wall can be used as an ideal wall for engineering application.

The purpose of this paper is to investigate the effect of shield underpass construction on the deformation of the U-shaped trough structure and stratum of the high speed railway roadbed. The proposed Shijiazhuang rail transit line No.4 underpassing the Beijing-Shijiazhuang high-speed railway roadbed is selected as the engineering background. The stratum and structural model test materials are formulated based on the geometric similarity ratio, and the test monitoring system is designed. A 1 200 mm small shield is employed to carry out 2 sets of laboratory model tests for different distances from the vault of the shield tunnel to the bottom of pile in the roadbed. The results show that: with the increase of the distance from the shield tunnel vault, the settlement of the strata decreases, the influence range of the shield construction on the strata is about 2 times of the diameter of the tunnel, and the significant influence range is one time of the diameter of the tunnel; with the increase of the burial depth, the settlement of the strata under the structure caused by the shield construction decreases, and the maximum difference of settlement is 10% at the distance of 0.5 and 1.0 times of the diameter of the tunnel respectively from the shield tunnel vault; a dehiscence occurs between the U-shaped trough and the adjacent strata, and the percentage of settlement of the strata occurring during the shield tail release phase is greater than 74%. It is recommended to use grout filling material with good water retention and certain early strength after the completion of segment assembly to control the settlement deformation. Meanwhile, deep hole grouting should be carried out to timely fill loose ground pores so as to increase the compactness.

After the excavation unloading of underground engineering rock mass, the geostress in surrounding rock mass varies with the spatial position in a gradient form. The wave impedance of rock varies in a gradient form under the action of gradient geo-stress, and then affects the propagation and attenuation characteristics of stress wave. In order to investigate the effect of gradient geostress on the propagation characteristics of rock stress wave, the stress wave propagation tests were carried out on the red sandstone long specimen under nine stress gradient conditions by using the self-developed rock stress wave propagation test system with gradient static stress. By analyzing the variation laws of rock stress wave velocity and wave impedance with stress gradient, empirical models between stress wave amplitude and propagation distance, propagation time and stress gradient were established to explore the mechanism of stress wave propagation and attenuation affected by gradient stress in red sandstone. The results show that under the same stress gradient condition, the stress waveform change little with the increase of propagation distance, but the stress wave amplitude decrease gradually. With the increase of stress gradient, the stress wave velocity and wave impedance in each measuring point section of rock increase, but the increase rate gradually slows down, the wave impedance difference ratio between adjacent measuring point sections increases rapidly first and then decreases slowly. The stress wave amplitude decreases exponentially with the increase of propagation distance and propagation time, with the increase of stress gradient, the spatial-and time-attenuation coefficients of amplitude show a trend of “rapid increases at first, then slow decreases”. As the stress gradient increases, the stress wave amplitude at the same measuring point decreases rapidly first and then varies slowly, and at the low stress gradient stage, the further away from the free end, the faster the amplitude attenuation rate of the measuring point.

The shear strength of the soil-rock mixture-bedrock interface is a key parameter for the stability of the slope and engineering design. In order to explore the influence of the block stone size on the shear mechanical behavior of the soil-rock mixture-bedrock interface, a series of large-scale laboratory shear tests of the soil-rock mixture-bedrock interface with different block stone sizes was carried out. The results show that the failure mode of the contact interface is not significantly affected by the size of the block stone and the normal pressure, and both are characterized by strain hardening. However, the normal pressure will weaken the impact of the block stone size effect. As the size of the block stone increases, the shear strength and the shear strength index (φ, c) first increased and then decreased. There are both a positive size effect and a negative size effect. The size effect has a limited influence on the angle of internal friction φ , which fluctuates around 29º. However, the size effect has a significant influence on the apparent cohesion c. The specimens with different block sizes show obvious shear contraction behavior under different normal pressures, and the normal pressure can enhance the shear contraction characteristics of the soil-rock mixture. The soil-rock mixture-bedrock interface is a weak potential slip surface. With the increase of the block size coefficient, the weak failure surface tends to transform into the soil-rock mixture.

In order to study the mechanical properties and anisotropy characteristics of layered phyllite, rock mechanical tests of phyllite were conducted under different bedding inclinations and different confining pressures. The anisotropic mechanical properties in strength, deformation, brittleness and failure modes of phyllite samples were comparatively analyzed. Some findings were as follows. (1) As the bedding inclination β  increases, the strength and deformation characteristic curve of the rock sample is U-shaped; and the strength and plasticity increase, the anisotropy gradually decreases and stabilizes with the increase of confining pressure. (2) The anisotropic Saeidi criterion and the modified Ramamurthy criterion can well predict the failure strength of rock specimens under different bedding inclinations. (3) A comprehensive brittleness index of phyllite samples at different bedding inclination under different confining pressures is proposed based on the stress–strain curve characteristic and energy equilibrium. The brittleness index is lower, and the shear failure is more likely to occur when the bedding inclination β  is about 45º. The relationship between brittleness and the failure mode is revealed, and the decreasing order of brittleness is concluded as follows: tensile splitting along bedding plane > tensile splitting through bedding plane > shear along bedding plane > shear through bedding plane. (4) The failure modes of phyllite are related to the bedding inclinations and confining pressures. Under uniaxial compression conditions, complex networks of fractures are formed easily after splitting failure of rock samples; and under high confining pressure conditions, a single shear failure plane is generally formed along the bedding plane or through multiple planes after rock failure.

Different types of unsaturated soils are widespread in nature. Due to the differences in grain size and mineral composition, the shear strength characteristics of the unsaturated soils are significantly different over a wide suction range. In order to establish a unified model for predicting the shear strength of different soil types over a wide suction range, a systematic investigation was carried out to examine the effect of soil type on the shear strength of unsaturated soils. An improved shear strength model is proposed based on the advantages of two traditional equations. The proposed model is verified by comparing the published experimental data and the predicted results of the proposed and the existing equations. The conclusions are as follows: (1) the shear strength for different types of soils over a wide suction range can be roughly classified into two types: peak type and steady growth type; (2) the proposed model demonstrates good agreement between the predicted and measured shear strength results over a wide suction range; (3) the physical meaning and the recommended values of the model parameters for different soil types are given.

Understanding the vertical penetration process of submarine pipelines is of fundamental significance for the reasonable assessment of the initial pipeline embedment during installation and safety and stability during service. Geotechnical centrifuge model tests, combined with discrete element method modelling were conducted to investigate penetration resistance evolution characteristics and mesoscopic mechanism of submarine pipeline in sand with different densities under the real stress level. The test results show that for the medium-dense sand, the pipe penetration resistance-embedment curve shows a hardening trend. On the other hand, for the dense sand, the pipe penetration resistance-embedment curve exhibits a periodic softening trend; moreover, the deeper the embedment is, the greater the degree of softening becomes. Discrete element modelling results demonstrate that the difference in pipe penetration resistance-embedment curves is induced by different soil movement and failure modes when pipe penetrates in sands with different densities. The pipe penetration resistance evolution characteristics are closely related to the formation and development of shear band in sand. The evolution characteristics of penetration resistance with embedment should be fully considered when evaluating pipe embedment in dense sands using the current design specification for submarine pipelines. More specifically, the pipe embedment should be reasonably estimated based on the upper and lower limits of calculation results, when the preliminarily estimated pipe embedment is larger than one tenth of the pipe diameter.

In order to study the effects of different cut depths and cut spacings on cutting force and specific cutting energy acting on a conical pick, the laboratory linear rock cutting tests were carried out on carbonaceous slate and rhyolite clastic rock based on the TRW-300 true triaxial electro-hydraulic servo system and self-designed innovative pick loading platform. Experimental results indicate that there were four types of failure modes on a conical pick in rock cutting, and there was a synergistic effect between two adjacent picks. The mean cutting force linearly increases with the increasing cut depth and cut spacing. The specific cutting energy decreases with the increasing cut depth in a power-function way. The specific cutting energy decreases first and then increases with the increasing ratio of cut spacing to cut depth, and there is a minimum value that is optimal, where it is far less than energy under an unrelieved cutting groove. The optimum value of the ratio of cut spacing to cut depth is 2 or 3. Based on the results of laboratory tests, the pick arrangement on the cutting head of the cantilever roadheader is optimized, and applied to the field industrial test of non-explosive tunneling of roadheader, which significantly increases tunneling efficiency and reduces tunneling costs.

Basic friction coefficients are important parameters affecting the shear strength of structural surface in rock masses. The basic friction coefficient determined by conventional laboratory test methods is influenced by mineral composition, temperature, and other factors. To systematically reveal the basic frictional properties of structural surfaces, the macro and micro friction coefficients of sandstone were investigated separately, and their relationship was established. First, mineral components and mechanical parameters of the sandstone were determined by X-ray diffraction and nanoindentation tests. Second, the friction coefficient in the macro scale was examined by tilt test and direct shear test, and the direct shear test was carried out on flat structural surface specimens dimensioning 10 cm×10 cm×5 cm subjected to constant normal stresses of 1, 2, 3, 8, and 12 MPa. The results show that the friction coefficient decreases in the logarithmic form as the normal stress rises, whereas it increases in the logarithmic form as the shear rate grows. Then, the friction coefficient in the micro scale was examined using nano-scratch test, and the results show that the friction coefficients of feldspar minerals first decrease and then remain unchanged with increasing load under low load conditions, and the friction coefficients of quartz minerals first decrease and then increase with increasing load. In addition, the friction coefficients of both minerals show an increasing trend with increasing shear rate under low load conditions, and remain stable under high load conditions. Finally, based on the macro-micro friction coefficient test results, the rate- and state-dependent friction (RSF) law was used to establish the linear regression equation linking the basic friction coefficient of red sandstone and the mineral friction coefficient, and the reliability of this empirical relationship was verified by the direct shear test with errors ranging from 0.17% to 0.91%. This study provides a new idea for the determination of the basic friction coefficient.

The seepage field of circular cofferdam in permeable anisotropic soil layer is analyzed in present study. First, the seepage region around circular cofferdam is divided into three regular parts. The series solutions of water head distribution in three regions in cylindrical coordinate system are obtained by using method of separation of variables after coordinate transformation and homogenization of boundary conditions. And then an analytical solution of steady seepage field of circular cofferdam in permeable anisotropic soil is obtained by utilizing the orthogonality of Bessel function. The solution is verified through the comparison with the results obtained from the finite element software Plaxis2D. Based on the analytical solution, the expressions for seepage quantity and exit gradient are deduced and compared with the results obtained by different methods. Finally, the water pressure under steady seepage is calculated and analyzed. The results show that the seepage quantity, exit gradient and the water pressure obtained by analytical solution are in good agreement with numerical results as compared with results obtained by other methods; the seepage water pressure decreases with the increase of the ratio of cofferdam radius and vertical and horizontal permeability coefficients, and then gradually approaches a stable value.

To overcome the squeezing effect of pipe pile construction, a drilling with a prestressed concrete (DPC) pipe pile with drilling into hole, synchronous sinking pile and post-grouting was developed. It was a type of the rock-socketed and non-squeezed pipe pile. In this paper, the deformation and pressure change in the soil around the pile, caused by the drilling unloading and side post-grouting, were tested by in-situ tests. Based on test results, the change rules of unloading and grouting effects along the horizontal and vertical directions were analyzed, revealing the disturbances of soil around the pile. Moreover, a method for estimating the disturbances of DPC pipe pile construction was developed based on the circular hole shrinkage theory. The study shows that the disturbances mainly manifests unloading and grouting effects. Drilling unloading causes a shrinkage of the hole wall of 3.5−18.9 mm, and subsequent pile side post-grouting restore the shrinkage deformation by 28% to 50%, partially eliminating the impact of drilling unloading effect. In addition, the deformation and pressure change of soil around the pile caused by drilling unloading can be estimated by the modified circular hole shrinkage theory. Compared with the squeezing effect of displacement piles, the unloading effect of DPC pipe piles is significantly smaller, more advantages in application in urban areas with dense buildings and pipelines.

As the main raw material for marine engineering construction, coral sand with high calcium carbonate content is porous and fragile. In order to further accelerate the construction of marine engineering and improve the strength and stability of geo-structures, using geogrid to reinforce coral sand is a potentially effective means. In present study, a series of triaxial tests was conducted to investigate the effects of the number of geogrid layers, initial water content, and confining pressure on the strength and deformation of geogrid-reinforced coral sand. The results indicate that the geogrid reinforcement can obviously improve the mechanical properties of coral sand. With increasing number of geogrid layers, the overall strength of the reinforced coral sand gradually increases. The stress-strain relationship shows a general hardening trend, and the lateral bulging is also restrained effectively. Similarly, the introduced pseudo-cohesion increases linearly with increasing number of geogrid layers, and the internal friction angle decreases slightly. As the initial water content increases, the strength of geogrid-reinforced coral sand decreases slightly, and the pseudo-cohesion changes little. However, the internal friction angle decreases by about 4º compared with the drying condition. The particle breakage of geogrid-reinforced coral sand is greatly affected by confining pressure in triaxial case, and the relative breakage rate below 400 kPa confining pressure is mainly less than 3%. Additionally, the lateral and axial additional stresses are considered and calculated by analyzing the geogrid-sand interfacial characteristics. These results further enrich the understanding of the mechanism of geogrid-reinforced coral sand.

Soil particles in cold regions are deposited and arranged along the dominant direction due to gravity, forming transversely isotropic frozen soils. Without considering the effect of the deposition angle between the deposition direction and the load direction, the deformation characteristics and bearing capacity of the actual geotechnical engineering in cold regions may be misestimated. However, the effect of deposition angle on the mechanical properties of frozen soil has not been explored in the existing literature. In response to this problem, the uniaxial compressive tests under different temperature conditions were carried out in this paper to examine the effect of deposition angle on the mechanical behaviors of frozen soil. With the developed sample preparation mould, frozen soil samples with four different deposition angles δ (δ = 0º, 30º, 60º and 90º) were prepared. Uniaxial compression tests on these frozen soil samples were carried out at four different temperature conditions T (T = −5, −10, −15 ℃ and −20 ℃). The significant effects of T and δ  on deformation mode, failure behaviors and uniaxial compressive strength of frozen soil are analyzed. The uniaxial compressive stress-strain curves of frozen soil at certain T and δ are normalized, and the slope variation rule of the softened section is also analyzed. According to the above analysis, the deformation modes of the frozen soil under the effect of T andδ  are divided into three deformation modes, i.e., I, II and III. According to the test results, it can be observed that as T decreases and δ  tends to 60º, the deformation mode of frozen soil tends to transition from deformation mode I to deformation mode III, and the failure model tends to transition from the expansion failure mode with the X-shaped shear band to the single shear plane failure mode with a smaller failure range. The uniaxial compressive strength of frozen soil increases with the decrease of T, and shows a trend of first decreasing and then increasing with the increase of δ .

The joints within the rock mass control the stability of rock engineering. Most of the existing peak shear strength criteria are empirical models, and the sampling effect and anisotropy of shear strength of rock joints have been rarely reported. A total of 5 groups of matched resin molds were prepared based on 3D laser scanning technology and 3D printing technology. A batch of matched mortar joints was duplicated using these molds. A total of 25 groups of the direct shear test were conducted under five constant normal load conditions. A novel theoretical peak shear strength criterion incorporating the newly developed roughness parameters was proposed based on the modified relationship between contact area ratio and threshold apparent dip angles. The proposed criterion was simple and no additional fitting constant was introduced in addition to the roughness parameters. Comparisons between the new criterion and the existing empirical criteria in the literature showed that the new criterion had better accuracy in predicting the shear strength of rock joints. In addition, it was found that the new model was slightly affected by the sampling interval and could capture the anisotropic characteristics of the shear strength of rock joints.

The precise acquisition of deep rock mechanical parameters is required for the safe and efficient development of deep rock engineering. In response to the problems of difficult and costly coring, little data from laboratory tests and large discreteness, a method based on mm-indentation tests is developed to accurately obtain the elastic modulus of tight sandstone. In this paper, five indenters commonly used in mm-indentation test are used. A large number of mm-indentation tests on tight sandstone in the Tarim oilfield are performed by controlling the maximal load and loading/unloading rate as test variables. The results show that Brinell spherical indenters can accurately measure the elastic modulus of tight sandstone based on the Oliver-Pharr theory. In addition, the influence of test conditions such as Brinell indenter diameter, loading/unloading rate, and maximal load on test results is analyzed, and the optimal test conditions are clarified. The stress distribution in the stressed area is analyzed using FEM software, and the applicabilities of representative volume element and Oliver Pharr theory under various test conditions is discussed. Compared with the traditional micro/nanoindentation test, the proposed method has the advantages of rapid testing speed, simple sample preparation, and low cost, and it can be applied to a large number of drill cuttings returned from drilling sites as well as core samples with fracture development that are difficult to make standard samples, providing strong support for the efficient development of deep resources and the safe construction of deep rock engineering.

In order to investigate the influence of the different thicknesses of soft soil layers on the dynamic response characteristics of a single pile under different types of seismic waves, the dynamic response characteristics of acceleration, horizontal displacement, bending moment, and pile foundation damage were analyzed by shaking table test. The test results show that under the seismic wave, the restraint behavior of soil around the pile is significantly affected by the thickness of the soft soil layer. The magnifying effect of pile shaft acceleration in soft soil is the most significant. The acceleration magnification factor of the pile top is positively correlated with the thickness of the soft soil layer. The horizontal displacement of the pile top reaches the maximum when the thickness of the soft soil layer is the maximum. The maximal bending moment of the pile appears in the soft soil layer and increases with its thickness. For different soil thicknesses, the maximal bending moment of the pile shaft is less than the design value of bending capacity, and the pile has good integrity. In the seismic design and calculation of pile foundation, the seismic capacity of pile foundation in soft soil layer should be strengthened, and a variety of seismic waves should be selected for seismic checking.

Based on the traditional Knothe time model hypothesis, a new hypothesis is proposed and an improved Knothe time model is constructed. The theoretical analysis shows that the improved Knothe time model conforms to the variation law of surface point subsidence and subsidence velocity. Based on the improved Knothe time model, a calculation expression of the maximum surface subsidence velocity is deduced, and a method to determine the model parameters is given. The monitored values of maximum surface subsidence velocity caused by mining in 20 mining areas are used to compared with the theoretically predicted values. It shows that the predicted value is in good agreement with the monitored value, and the relative standard deviation is only 2.1%, which verifies the accuracy and reliability of the present model. It is important to note that the maximum surface subsidence velocity is affected by mining height, mining speed, loose layer thickness, bedrock layer thickness and other parameters. The maximum surface subsidence velocity increases linearly with the increase of mining height and mining speed, decreases nonlinearly with the increase of loose layer thickness and bedrock layer thickness, and its sensitivity to the change of bedrock layer thickness is higher than to that of loose layer thickness.

At present, many studies have been done on the compression properties of calcareous fine sand. However, there is also calcareous coarse sand extensively existing in practical engineering. Therefore, it is important to investigate the compression characteristics of calcareous coarse sand. In present study, single particle crushing tests and one-dimensional oedometer tests of calcareous coarse sand with a single particle size were carried out using particle strength tester and automatic large-scale consolidation instrument. The effects of particle size and specific gravity on single particle strength and compression properties of calcareous coarse sand were studied. The single particle crushing test results show that the characteristic stress of single particles increases with the increment of specific gravity. The single particle strength, which can be normalized by the characteristic stress of single particles, is effected by particle size greatly. A curve for survival probability of single particle can be created based on a Weibull distribution. The oedometer test results show that the fractal dimension used to describe the gradation of sand after tests increases with the increase of the particle size; and the relationship between the Hardin’s crushing index and the plastic work is a power function. There is an approximate linear relationship between the yield stress of calcareous coarse sand with a single particle size and the characteristic stress of single particles.

In order to tackle the problem of large deformation at the crossover point of deep large section soft rock tunnel, this paper takes the crossover point of kilometer deep tunnel in Wanfu coal mine as an engineering example and applies microscopic negative Poisson’s ration (NPR) steel anchor cable in tunnel support for the first time. The deformation mechanism of the surrounding rock at the intersection of deeply buried roadways is analyzed, and a control countermeasure with microscopic NPR steel anchor cables as the core is put forward by using a combination of laboratory test, theoretical analysis, numerical simulation and field test. The mechanical properties of microscopic NPR steel anchor cables are studied by static tensile test in laboratory. The results show that the whole process of microscopic NPR steel anchor cables is characterized by high constant resistance, uniform tension, no yield platform, no obvious necking effect at break. Based on theoretical deduction and the maximum influence range of surrounding rock at T-type tunnel intersection, a calculation model of the maximum influence range of T-type tunnel intersection supported by microscopic NPR steel anchor cable is established. Through numerical simulation, the failure evolution process of T-shaped tunnel intersection is reproduced, and compared with the support effect of common microscopic NPR steel anchor cable. On-site support application test and monitoring are carried out. It verifies that the long-short microscopic NPR steel anchor cable combined support strategy has good control effect on large deformation of surrounding rocks at the intersection, which provides new support materials and means for safety and stability control of surrounding rocks at the intersection.

This paper presents the results of the inclinometer monitoring and back analysis using three-dimensional finite elements in the scale pit deep excavations. Scale pit is a deep circular excavation construction with a depth of −27 m and a total diameter of 26 m. The soil layer above from the final excavation is dominated by a sandy soil layer. The retaining wall construction is a secant pile reinforced with waler beams at each excavation stage. The hardening soil model was used as soil constitutive model. The results of the back analysis show that the trend of wall deflection is close to the results of the inclinometer measurements. The maximum deflection that occurs in the wall is 5.1 mm. The effective cohesion c′ of cemented soil in the case is about 50 to 200 kPa, depend on the depth of cemented sand layer. NSPT is a number of blows for a split-barrel sampler to penetrate 30 cm into the soil during Standard Penetration Test. The soil modulus in sand soil as a result of the modeling is equivalent to 1 400 NSPT to 2 000 NSPT (unit: kPa), while for cemented sand, it is equivalent to 7 000 NSPT. This study also investigated the effect of waler beam installation and external loads on the deformations and forces that acting in the secant pile. In this case, the external load has a dominant effect in the result of wall deflections. Meanwhile, the waler beam has no significant effect for reduce the wall deflection.

Rock chips are the direct product of rock-machine interaction, and are also often used to evaluate tunnel boring machine (TBM) rock breaking efficiency and optimize TBM tunneling parameters. In-situ sieve tests of rock chips for different lithologies were conducted based on Lanzhou water source construction project and Longyan Wan’anxi water diversion project. The particle size distribution was obtained. From the perspective of energy conversion of disc cutter rock-breaking, a novel evaluation index of TBM rock-breaking efficiency based on newly added surfaces theory was proposed. Based on the particle size distribution and the statistics of TBM tunneling parameters, the correlations among the index of newly added surfaces theory, specific energy and coarseness index were investigated. The advantages of the index of newly added surfaces theory in reflecting the fragmentation degree of rock chips and accurately evaluating the rock-breaking efficiency of TBM were clarified. The regression analysis was carried out to examine the correlation between the newly added surfaces theory index and TBM tunneling thrust, the ratio of disc cutter spacing to penetration (s/p), the optimal tunneling thrust force and s/p value range of TBM under the conditions of hard rock (grade II surrounding rock) and soft rock (grade III surrounding rock) were obtained respectively. The research results showed that the index of newly added surfaces theory not only conformed to the principle of rock fragmentation, but also evaluated the rock-breaking efficiency of TBM accurately. The more broken the rock chips, the larger the index of newly added surfaces theory, and the higher energy consumption of the TBM. At this time, the rock breaking efficiency of TBM is relatively low. The index of newly added surfaces theory has a good linear correlation with specific energy and coarseness index. The more broken the rock chips, the more energy consumption of the TBM. The larger the index of newly added surfaces theory, the smaller the coarseness index. Under the condition of soft rock, the specific energy of TBM tunneling is lower than that of hard rock, however, the fragmentation degree of rock chips is higher than that of hard rock. The index of newly added surfaces theory decreases first and then increases with the rise of tunneling thrust and s/p. Under the conditions of hard rock and soft rock, when the optimal tunneling thrust interval and s/p interval of TBM are 7 400−7 700 kN, 13.9−14.4 and 1 000−1 500 kN, 8.0−8.5 respectively, the index of newly added surfaces theory is small and the rock breaking efficiency of TBM is high.

Current theoretical studies of ground settlement induced by small radius curved tunnelling mostly consider the shield as a continuous whole and do not consider the effect of shield articulation. Therefore, the over-excavation gap caused by the change of excavation path of small curvature shield cannot be correctly evaluated. Firstly, based on the shield articulation position and the geometric position relationship between the shield and the small curvature tunnel excavation path, the calculation formulas were obtained for the over-excavation gap and articulated angle at different articulation positions of the shield machine during the excavation of the small curvature tunnel. Then, the mirror image method and Mindlin solution were used to solve the ground settlement caused by the combined effect of over-excavation ground loss, shield tail ground loss, excavation face uneven thrust, shield shell uneven friction and slurry pressure at the shield tail during articulated shield. Finally, engineering monitoring data were used to compare and verify with the theoretical solution of this paper, and good agreement was obtained. In addition, parametric analysis was carried out for the tunnel turning radius, front shield length, shield articulation angle and over-excavation gap. The analysis results show that ignoring the effect of shield articulation will overestimate the ground loss and result in large predictions of ground settlement. As the turning radius decreases, length of the front shield, shield articulation angle and over-excavation gap increases, the ground settlement increases, but the change of its value has a small effect on the settlement in front of the excavation face and a larger effect on the settlement behind the excavation face. Behind the excavation face, the longitudinal surface settlement tends to increase and then decrease as the distance from the excavation face increases, when the turning radius is taken to be small and the front shield length, shield articulation angle and over-excavation gap are taken to be large.

In order to address the problem of groundwater infiltration and inner water exosmosis induced by the hydraulic deterioration of lining joints in pressurized shield tunnels, the seepage characteristics and mechanical responses of tunnel lining and surrounding stratum under localized leakage conditions are analyzed through both numerical and analytical method based on the hydraulic aperture theory. The results show that the analytical solutions derived from the classical image method for a multi-point leakage case match well with the numerical solutions, which verifies the effectiveness of the proposed numerical model of localized leakage. There is a significant difference in the characteristics of lining-stratum interaction induced by the localized groundwater infiltration and inner water exosmosis. The former leads to a reduction of the localized pore water pressure, resulting in a more compressive interaction between lining and stratum, a reduction in the lining axial force but an increase in the bending moment and further inducing an outward deformation at the leakage region. While the latter is the opposite. The influence of multi-point localized leakage on the seepage field and lining response has a coupling effect, and a special hydraulic interaction situation where the groundwater infiltration and inner water exosmosis may co-exist when the inner water pressure approaches the stratum hydraulic head. The localized leakage behavior is largely affected by the stratum permeability coefficient. Under composite stratum conditions, the induced lining response is not significant when the localized leakage occurs in a highly permeable stratum. The seepage flow refraction phenomenon exists at the stratum interface when the leakage is in the lower stratum with low permeability, while the upper stratum with high permeability acts as either a water supplement or drainage source.

The investigation on failure patterns and paths of cracked rock slopes under excavation unloading is one of the hot issues in the slope engineering field. Accurate identification of the potential failure path is of great significance for excavation safety and support design of the slope. A theoretical method for crack propagation discrimination was embedded into the numerical simulation based on the fracture propagation analysis method, and the quick simulation of initiation, propagation, and coalescence of discontinuous cracks in rock masses was realized through stress intensity factor calculation at crack tip, crack propagation pattern recognition, crack initiation angle derivation, and crack propagation and coalescence. The proposed simulation method was used to analyze the crack propagation mechanism and the failure path of a highway cutting slope under multi-stage excavation unloading. The results show that during the multi-stage excavation from top to bottom of the slope, the crack initiation first occurs at the slope shoulder, and then a dominant propagating crack is formed through tensional propagation. With the downward excavation of the slope, the dominant crack gradually propagates downward the slope in tensile-shear mixed pattern and coalesces with existing cracks, forming a step-path failure in the middle and upper parts of the slope. When the crack propagates to the lower part of the slope, the crack propagation pattern transforms from tensile-shear mixed pattern to shear pattern, and finally the potential sliding body slides out along an arched shear surface at the slope toe. This study reveals a composite failure pattern of the cracked rock slope under multi-stage excavation unloading, which includes the tensile-shear mixed propagation pattern with a step-shaped path in the upper part and the shear propagation pattern with an arched path in the lower part, and can provide new ideas for the support design and construction stability control of rock slopes.

Earthquake-induced liquefaction poses a significant threat to tunnel structures. Particularly, shield tunnel crossing sandy stratums with different liquefaction susceptibilities could suffer more severe seismic damages near the soil interface. In this paper, a three-dimensional numerical study was carried out to investigate the seismic response of a shield tunnel passing through saturated sandy strata with two different relative densities. Firstly, a practice-oriented two-surface plasticity sand model was employed to model the sandy soil and was validated by shaking table experiments on a tunnel structure embedded in liquefiable soil. Secondly, a deformable force-displacement link model for circumferential joints between each successive segmental ring was employed to model the interactions between segmental rings. The approach was validated using the results of two loading experiments on model segmental linings from the literature. Finally, the 3D numerical model was established considering various relative densities of soil, peak input accelerations, and the dip angle of the interface. The results indicate that the tunnel’s horizontal displacements due to seismic excitations are coupled with the liquefaction-induced vertical uplift displacements, and the tunnel’s deformation is not simultaneous in the two soil strata, resulting in twisting distortion of the tunnel structure. The uplifts of tunnel change rapidly and are increased by the rising of the dip angle near the soil interface. Also, the bending moments suddenly change, and the shearing/tensile displacements of joints increase remarkably, which confirms that the seismic design of shield tunnel segments near the soil interface is a critical issue.

Crack propagation experiments were conducted on the prefabricated cross-fissured rock specimens to study the crack initiation, propagation, and coalescence processes. The effects of the angle between the primary fissure and the axial load and the angle between the primary and minor fissures on the crack initiation stress and the coalescence stress were analyzed. The experiments were modelled and calculated with a hybrid finite-discrete element program, i.e., 3D Y-HFDEM code parallelized by a graphic processor. The transition of rock damage from continuous to the discontinuous medium was achieved. Crack categories and damage modes were identified and the phenomenon that was difficult to find in the experiments was captured. It is shown that the number of tensile cracks in the crack coalescence zone increases as the angle between the main fissure and the axial load increases. The crack initiation and coalescence stresses are proportional to the angle between the main fissure and the axial load. The damage mode of the rock changes from tensile damage to shear damage as the angle between the primary and minor fissures increases. Cross fissures increase the degree of rock fragmentation. Damage caused by mixed tensile-shear cracks initiating and propagating at the tip of the primary fissure dominates the rock damage and is the main control fissure leading to the loss of bearing capacity of the rock mass. The hybrid finite-discrete element simulation software GPGPU parallelized 3D Y-HFDEM IDE is advantageous in study of rock crack propagation to capture damage and fracture types that are difficult to detect in the laboratory, and can be a powerful tool for study of rock crack propagation.

Sand-bearing rock is a special rock formed in the high-level key layer before water inrush and sand burst disaster. Its strength and mechanical properties are different from ordinary rock, which determines the stability of the high-level key layer. It is found that the fractured rock and sandy rock with different fracture angles have different characteristic stresses, and with the increase of fracture angle, the initiation stress, damage stress and peak stress of fractured rock and sandy rock increase, and the bimodal stress first increases and then decreases. The characteristic stresses of sandy rock under the same fracture angle are less than those of fractured rock, indicating that the sand body has a weakening effect on the characteristic stress of rock. From the perspective of failure form, the fractured rock is easy to show wing tensile fracture, the sandy rock is easy to form tensile fracture under the condition of low fracture angle (30º) and shear fracture under the condition of high fracture angle (60º), indicating that the sand body has shear effect on the rock after entering the rock fracture. At the same time, the mechanical model of sand filling is established, and it is pointed out that the reason why the strength of sandy rock is less than that of fractured rock is that the sand body reduces the friction coefficient of rock. Based on the cumulative acoustic emission ringing counts, the amount of rock damage is defined and the mechanism of sand-bearing rocks causing the disaster is analysed. The on-site sand-bursting disaster can be divided into four phases: 1) elastic deformation phase; 2) fracture expansion phase; 3) sand and energy storage phase; 4) sand-bursting energy release phase. Finally, PFC2D is used to verify the differences between fractured rocks and sandy rocks, and the energy evolution laws of different types of rocks are analyzed. The research results can be used as precursory information identification of water inrush and sand burst phenomenon in coal mine roof, and help to guide the safe production of water inrush and sand burst face.