In order to investigate the energy evolution and failure characteristics of single fissure carbonaceous shale under drying-wetting cycles, intact and fissure carbonaceous shale samples with fissure angles of 30°, 45°and 60° were fabricated respectively. MTS815 rock mechanical test system was used to conduct triaxial compression tests under different drying-wetting cycles. The influence of drying-wetting cycles on the strength, failure mode and energy evolution of single fissure carbonaceous shale were studied. The results show that the elastic energy and dissipated energy at crack initiation stress, damage stress and peak stress present exponential relationships with drying-wetting cycles. The elastic energy and dissipated energy at crack initiation stress and dissipated energy at damage stress are less sensitive to drying-wetting cycle, while the sensitivities of elastic energy at damage stress, and elastic energy and dissipated energy at peak stress are relatively high. The failure mode of carbonaceous shale is dominated by drying-wetting cycle and fissure angle, in which the drying-wetting cycle is the main controlling factor, and the fissure angle is the secondary controlling factor. It is found that tensile shear failure occurs in dry rock sample with fissure angle of 30°, while the dry rock sample with fissure angle of 45°and 60° are subjected to shear failure. With the increase of the number of drying-wetting cycles, the macroscopic length of the main crack increases, the density of secondary cracks increases, and the failure mode transforms to shear-tension composite failure. With the increase of the number of drying-wetting cycles, the energy storage level at crack initiation stress K_ci and the energy storage level at damage stress K_cd increase gradually. The higher the energy storage level at crack initiation and damage stress is, the more likely the crack initiation and rock damage occur. K_cd can be used as an warning indicator of rock failure. A larger K_cd indicates that the rock is more vulnerable to failure.

A numerical simulation study on the rainfall infiltration characteristics of unsaturated fractured soil was performed based on COMSOL Multiphysics software. By discretizing the fracture and the matrix into finite elements, a discrete fracture-porous medium model was established to fully simulate the fracture flow, matrix flow and fracture-matrix flow exchange in the soil. The upper boundary of the fractured soil was simulated by using the concept of “air element”. This method can describe not only the phenomenon of preferential infiltration for rainwater along the fractures at the beginning of rainfall, but also the phenomenon of rainwater flowing away along the surface when the rainfall is greater than the infiltration of the fractured soil. By simulating the low-permeability fractured soil at a depth of 2 m below the ground surface, the influences of the geometric characteristics of the fracture, the hydraulic properties of the matrix, the previous moisture condition and the rainfall intensity on the rainfall infiltration process of the unsaturated fractured soil were investigated. The results show two main seepage processes in the unsaturated fractured soil: firstly, water flows preferentially along the fracture; secondly, water is continuously imbibed into the matrix from the fracture, and the matrix imbibition inhibits the development of preferential flow in the fracture. Compared with the geometric characteristics of the fracture, the hydraulic properties of the matrix have a greater influence on the seepage flow in the unsaturated fractured soil. Increasing the saturated permeability coefficient of the matrix may change the seepage flow dominated by the fracture flow into that dominated by the matrix flow. The unsaturated properties of the matrix as well as the initial water content of the fractured soil change the soil moisture storage capacity, thereby accelerating or delaying the time for rainwater to infiltrate into a certain depth. The rainfall intensity has an influence on both the infiltration rate and infiltration amount in the soil. When it exceeds the infiltration capacity of the fractured soil, the excess rainwater flows away along the surface, and the infiltration rate across the section tends to stabilize with time.

To explore the anisotropy mechanism of joint shear strength from its damage characteristics, shear strength and failure characteristics were quantitatively analyzed. The study found that the shear strength and failure characteristics of joint show similar anisotropy with its morphology, and their anisotropy is weakened with the increase of normal stress. In addition, the influence of the inclination and height characteristics of the sawtooth joints on the shearing behavior was investigated. The analysis shows the angle and height characteristics of asperities have a positive correlation with shear strength, and the angle characteristics also determine the locality of asperities contact state during the shear process. The rough joint surface is composed of many microscopic asperities. The essential reason for the anisotropy of shear strength is the different contact area on the joint surface in different shear directions and the difference of asperities height and dip angle in the contact area. With the increase of normal stress, the failure of more and more micro convex bodies on the joint surface evolves from sliding wear to shear failure, and the difference of failure volume in different shear directions gradually decreases, which leads to the gradual consistency of the energy required for shear, which is the main reason why the anisotropic characteristics of shear strength are weakened by normal stress.

Particle shape is one of the most important characteristics of particles and its correlation with mechanical properties has attracted great attentions. In order to systematize the influence of particle shape on the mechanical properties of aggregates, the research results of physical characterization, numerical simulation and laboratory test of particle shape were summarized and analyzed. The results show that the definition method of particle shape based on three scales (form, angularity and surface texture parameters) is the most ideal method to describe the particle shape characteristics. The numerical method can simulate the shape of a single convex particle and the arrangement structure of a convex particle assembly, but the characterization of concave particles by numerical methods is not accurate enough. There are correlations of particle shape with pore structure, natural angle of repose, limited void ratio and particle size. The permeability coefficient of the aggregate is largely controlled by the shape characteristics of the particles under the same gradation and porosity. Due to the difference in interparticle locking and the number of contact points, the particle shape in uniform particle size affects the small strain strength, peak strength, residual strength of the aggregate and the dilatancy. Finally, the problems in particle shape research are discussed and future research directions are proposed.

In order to investigate the anchoring effects of the bolts on the jointed rock mass and its influences on crack propagation, the unanchored and anchored rock-like specimens with joint angles of 15o, 30o, 45o, 60o, 75o and 90o were made. The MTS816 was used for uniaxial compression test, and the acoustic emission (AE) and digital image correlation technology (DIC) were used to monitor the crack growth. In addition, the particle flow code software PFC3D was used to study effects of different anchoring angles on crack propagation. The results show that the peak strength, peak strain and cracking stress of the anchorage unit are improved compared with the unanchored specimens. The existence of bolt reduces the stress intensity factor during the crack initiation and propagation of the tensile wings, which can limit the initiation and propagation of tensile cracks effectively. Moreover, the occurrence of shear cracks can be inhibited. The process of the tensile crack propagation can be divided into initial stage and acceleration stage. The displacement of the characteristic points of the specimens with bolts is smaller than specimens without bolts. According to the PFC3D simulation results, the anchoring effect is the most obvious when the anchoring angle ? is 45o. With the increase of the anchoring angle ?, the development degree of the tensile wing crack first increases and then decreases. The failure mode of the prefabricated fracture specimen changes from shear failure to tensile-shear compound failure, and then to shear failure. The results can provide a reference value for analyzing the stability of rock engineering.

In the future, the extraterrestrial human activities, such as resources exploitation and base construction beyond earth need the aid of geotechnical engineering technology. Currently, there are only two approaches for humans to obtain the rock samples beyond earth: sample-return activities by spacecraft and meteorite survey. Meteorites are rare, expensive, small in size and arbitrary in shape, so it is difficult to process them into standard rock samples required by mechanical testing & simulation (MTS) and other traditional macroscopic rock mechanical tests. In this paper, a novel technique for measuring mechanical property of small-size meteorites is developed based on microscopic rock mechanics experiments (micro-RME) and statistical probability models. Firstly, the composition, content and distribution of NWA13618 meteorite rock-forming minerals were obtained by TESCAN integrated mineral analyzer(TIMA). Then, the nanoindentation technology was used to carry out a large number of indentation tests to obtain the multi-point elastic modulus. After that, the mechanical parameters of four main minerals in meteorite NWA13618 were derived by using Gaussian mixture model, and the elastic moduli of olivine, pyroxene, Fe-Ni and feldspar were 116.73, 101.77, 87.24 GPa and 70.74 GPa, respectively. Finally, the macroscopic cm-scale elastic modulus of NWA13618 meteorite determined by the homogenization method Mori-Tanaka model was 90.48 GPa according to the mineral content and mechanical properties. The novel micro-rock mechanical experiment technique and scale upgrading method proposed in this paper provide theoretical basis and technical means for predicting the mechanical properties of L4 parent asteroid.

To improve the liquefaction resistance of calcareous sand foundations, microbially induced calcium carbonate precipitation (MICP) technology combined with fiber reinforcement technology was proposed to treat the calcareous sand in the South China Sea. Based on dynamic triaxial tests, the dynamic behaviors of MICP and fiber-treated calcareous sand were studied. The dynamic strain, dynamic pore pressure, cyclic stress-strain response and dynamic elastic modulus were analyzed. Then, the strengthening mechanism of MICP and fiber on the mechanical properties of the treated calcareous sand was explored from the microscopic point of view, based on the scanning electron microscope (SEM) test results. The results show that: (i) MICP could improve the deformation resistance and liquefaction resistance of calcareous sands. Compared with the untreated calcareous sand samples, the dynamic strain and dynamic pore pressure of calcareous sand treated by MICP decreased by 95.74% and 92.46%, respectively. (ii) The addition of fibers further improved the reinforcement effect of MICP. Compared to the MICP-treated samples, the dynamic strain and dynamic pore pressure of MICP and fiber-treated samples decreased by 74.32% and 74.18%, respectively. (iii) MICP and fiber reinforcement technologies improved the deformation resistance and liquefaction resistance of calcareous sand subjected to cyclic loading by reducing the cyclic activity strength and energy dissipation, increasing the dynamic elastic modulus and reducing the decay rate of the dynamic elastic modulus. (iv) The results of the SEM test showed that MICP and fiber reinforcement had a synergistic effect on the improvement of the mechanical properties of calcareous sands. The incorporation of fibers provided more spots for bacterial adhesion and promoted the formation of calcium carbonate crystals, which not only increased the bonding strength between sand particles, but also enhanced the restraint of fiber nets by fixing fibers and sand particles together.

To investigate the description and evolution of the damage state of rock-like brittle materials, the physical meaning of each elastic modulus method parameter based on the strain equivalence hypothesis and the limitations of the model application are discussed. The modulus change during the triaxial cyclic loading and unloading test of limestone is studied. Moreover, the defects of the unloading modulus substitution method and the statistical damage evolution model in damage evolution analysis are discussed. The results show the existing elastic modulus method can only be used to reflect the damage evolution process of rock under uniaxial compression, and the unloading modulus substitution method cannot correctly describe the damage state and its evolution law. In addition, the statistical damage constitutive model can only be regarded as a theoretical self-consistent solution under the numerical range [0, 1] of the statistical damage evolution model. Based on the above research, a damage characterization variable and its evolution model considering the effects of damage strain threshold are proposed. Additionally, the constitutive model below the damage strain threshold and the damage constitutive model above the damage strain threshold are established, respectively. The sensitivity of model parameters is also analyzed in this study. The final results show that the proposed model and method can not only reasonably explain the damage mechanism of rocks under triaxial compression, but also accurately simulate the full stress-strain process, which is rationable and feasible.

The loess was stabilized using geopolymer (GP). Triaxial compression tests were conducted on stabilized loess with varied GP contents subjected to wet-dry cycles. The degradation law of the shear strength of the stabilized loess after varied wet-dry cycles was evaluated and an empirical model for predicting the shear strength was proposed. The chemical composition of the hydration products, the microstructure and pore size distribution of stabilized loess were investigated by X-ray diffraction (XRD), scanning electron microscopy (SEM) and mercury intrusion porosimetry (MIP) tests. The degradation mechanisms of GP stabilized loess under wet-dry cycles were discussed based on the experimental results. The experimental results show that compared with untreated soil, the shear strength of stabilized soils is significantly improved with the increasing GP content, i.e. the cohesion and internal friction angle increase by 260% and 43%, respectively. The shear strength of stabilized loess decreases with the increasing ratio of porosity to GP volumetric fraction ( ) in a power function. It indicates that GP stabilization can remarkably improve the durability of loess under wet-dry cycles. The stabilized loess with 10% and 15% GP can maintain over 75% of their original shear strength, but those with 5% GP shows evident deterioration in shear strength after nine wet-dry cycles. The wet-dry cycling has greater impact on the degradation of peak deviatoric stress and cohesion than that of internal friction angle. An empirical model was proposed and validated for predicting the degradation in shear strength of the GP stabilized loess under wet-dry cycles, considering influence of the GP content, confining pressure and the number of wet-dry cycle. The experimental results of XRD, SEM and MIP show that the main hydration products of GP are calcium silicate hydrate (CSH) and calcium aluminosilicate hydrate (CASH), which fill the soil pores and enhance the bonding between soil particles. Due to this reason, a denser microstructure develops and the cohesion of the stabilized loess increases, which consequently improves the shear strength of the GP stabilized loesses. Moreover, the wet-dry cycle results in the expansion of soil pores and the formation of new fissures, which destructs the bonding between soil particles and reduces the shear strength of the stabilized loess.

Based on the Hertz-Mindlin with bonding (HMB) contact model in EDEM, the modeling method of the rock discrete element model was investigated. First of all, all of the bonds in HMB rock model must be “weak bond”, i.e., the tensile strength of the bond is less than its critical normal stress. Then, the regression relationships between the elastic modulus , Poisson’s ratio , tensile strength of rock model and the particle and bond parameters were established, respectively. According to the three regression relationships, a mathematical model that can fit the real-rock’s E, , and through the HMB model was proposed. For the compressive strength of rock, when the critical stress ratio ( = / , is critical normal stress, and is critical tangential stress) ranges from 1 to 2, is influenced by , and the ratio of critical normal to tangential stress. Especially, when the is closed to 1.5, the failure mode of the rock model presents a better agreement with the experiment result. In addition, when the is constant, a linear relationship between the and can be clearly observed. In conclusion, a HMB model based modeling method that can fit the elastic modulus, Poisson’s ratio, tensile strength, compressive strength and the failure form of rock sample was proposed, and the detail steps were also given. Finally, an example was conducted to verify the proposed modeling method. The results showed that the rock models established through the proposed method can fit both the static and dynamic mechanical properties of rock.

Water vapor supply can induce frost heave of sand filler. Based on the self-developed moisture migration and frost heaving tester, the variations of sand moisture content, temperature and frost heave with freezing time under gaseous water recharge were studied. The change characteristics of sand microstructure during freezing were examined by stereomicroscope. By combining with the grey correlation theory, the correlation between the macroscopic index, i.e., frost heave and the microscopic parameters was analyzed. Notable frost heaving was observed under water vapor supply, and the frost heave reached 3.45 mm after 7 days of freezing. After sandy soil frost heaving, the proportion of large pore area increased and the proportions of small and medium micropore area decreased, and the change trend of pore number proportion was opposite to that of area proportion, while pore abundance value changed little, pore orientation angle distribution in each interval tended to be more uniform, the probability entropy of pore orientation present an overall upward trend of oscillation, and the pore fractal dimension showed a trend of decrease. In addition, the grey correlation theory is used to establish the correlation between the frost heave and the average pore size and other microscopic parameters, which provides an insight into the microscopic mechanism of sand frost heaving under water vapor supply.

The traditional methods for determining the soil-water characteristic curve (SWCC) in the full suction range always takes several months. To better guide the engineering practice with the theory of unsaturated soil mechanics, the rapid determination of SWCC is particularly important. In this study, a rapid measurement method for SWCC in the wide (full) suction range was developed with the combination of pressure plate method, parallel filter paper method and dew point water potential meter method, and the proposed method was used to test two different soils. Results showed that: i) The experimental data of the parallel filter paper method and that of the serial filter paper method were in good agreement, so the parallel filter paper method can be used instead of the serial filter paper method to shorten the measurement time. ii) When the measured data meet the general shape of SWCC (or meet the requirements of engineering design), the number of tests can be reduced appropriately to improve the measurement efficiency. A complete SWCC can be determined by the proposed combined method with only 10 to 12 points. iii) The measurement method proposed in this study can shorten the measurement time of SWCC testing from several months to 7–10 days. The combined method proposed in this study achieved the rapid and accurate determination of SWCC in wide (full) suction ranges, which is expected to make the SWCC measurement become a routine geotechnical test.

In recent years, tunnel engineering is developing towards super-long and deep, multi-field and multi-phase coupling, thus, which easily induce the structural instability during tunnel construction and operation. Aiming at the mountain tunnel, underwater tunnel and urban tunnel, this paper expounds the research status of structural model test system, and points out the existing issues. The results show that: mountain tunnel model test takes into account passing through the adverse geological rock, and the loading conditions such as high ground stress, high karst hydraulic pressure and high seismic intensity. However, it’s hard to simulate the high stress of surrounding rock due to the size limitation of the apparatus. The model test of mountain tunnel is mainly based on plane strain model, which lacks the simulation of active fault with various dislocation forms under large-scale true three-dimensional stress condition. The model test of underwater tunnel has achieved stable high water pressure loading, but it is still unable to simulate the true three-dimensional underwater environment of the coupled stress field and seepage field. Additionally, the visibility of the chamber is poor, the seepage of the tunnel is only be characterized by the water inflow at the entrance, which makes it difficult to obtain the deformation law of the surrounding rock. Urban tunnel is affected by train dynamic load, pavement load and adjacent construction disturbance. At present, the model test usually considers the vibration response of the tunnel under a single load, and often ignores the structural characteristics such as segment joints and cracks. Thus, it is necessary to design novel and efficient local loading device to analyze the ultimate deformation characteristics of weak parts of segment to ensure the safety of subway operation. Finally, the bottleneck issues of tunnel engineering model test are summarized, and the internal visualization of the apparatus needs to be improved based on transparent similar materials. Aiming at the adverse geological conditions under high geostress conditions, a three-dimensional static-dynamic coupled model test system with expandable size is developed. Combining with the micro low-power wireless monitoring sensors and 3D visualization results display platform, a tunnel engineering structure model test system suitable for complex environment is established.

Soft rock engineering involves many important engineering fields such as mining, hydraulic engineering, transport and national defense. With the increase of mining depth and the development of tunnel engineering, a large number of tunnels and roadways need to pass through soft rock formations, in which the problems such as high geostress and broken and weak surrounding rocks are prominent. Large deformation disasters of soft rocks pose serious threats to engineering safety and cause enormeous economic losses. In this paper, the research progress on soft rock support in China is first reviewed, and the research status of technology for soft rock control for large deformation hazards is summarized in the following aspects. (1) Passive support methods represented by improved rigid support, retractable support and compound lining. (2) Reinforced active support technology using high-strength bolts and cables. (3) Soft rock modification technology dominated by grouting modification. (4) Soft rock reinforcement with pressure relief as the core idea. (5) Compound support methods. Furthermore, the development of different supporting technologies and methods are elaborated, and the applicable conditions, advantages and disadvantages of different supporting methods are analyzed. It is usually difficult to meet the demand of large deformation control of soft rock relying on a single support method. Therefore, it is urgent to solve the problems of the prevention and control of large deformation disaster of soft rock to realize the efficient collaborative control among different supporting measures and achieve the real-time accurate monitoring of deformation and stress fields. Finally, based on the above research results, the development tendency of support technology for soft rock with large deformation hazards is prospected and the countermeasures are proposed.

With regard to undrained analysis of saturated clay foundation for its ultimate bearing capacity under rapid loading, this study has proposed the calculation of the unconsolidated-undrained (UU) strength instead of employing the consolidated-undrained (CU) strength parameters directly due to overestimated results. The formula for predicting UU strength has been deducted based on the CU strength parameters referring to the reports of geological investigation, and the profile of the UU strength , found increases linearly with depth, has been built up. A calculation method for the bearing capacity of this foundation type is therefore proposed. The basic idea thereof is to use the value of at the average depth of the slip plane in the calculation, and a dimensionless parameter, which plays the key role, is introduced to determine the maximum depth of the slip surface. The accuracy and the precision of this parameter, as well as the proposed method, has been validated via a large number of comparative calculations with the finite element limit analysis method in this study.

As a typical geological structure, the layered composite structure of surrounding rock is formed by weak interlayer and hard brittle rock, and it significantly affects the stability of tunnel surrounding rock. In the past, the research on composite rock with weak interlayer focused on uniaxial, biaxial or conventional triaxial stress paths, and the mechanical properties and failure characteristics of composite rock at the free face of tunnel under true triaxial stress path were lack of analysis and discussion. In this study, the influence of the thickness of weak interlayer on the mechanical response and failure characteristics of surrounding rock at the free face is discussed based on the composite rock samples with different thicknesses of weak interlayer. The results show that: (1) The thickness of weak interlayer significantly affects the peak stress and strain of composite rock samples. With the increase of thickness, the sliding deformation of rock blocks above the weak interlayer gradually increases, and the compression deformation of the weak interlayer decreases gradually. (2) With the increase of thickness of the weak interlayer, the failure mode of the rock element near the free face of the composite rock sample gradually changes from mixed tension shear failure to tension failure, and the number and failure range of macro cracks gradually decrease, while the rock element far away from the free face gradually changes from shear failure to basically undamaged fracture. (3) The failure areas of composite surrounding rock of side wall with different thicknesses are concentrated on the weak interlayer and the surrounding rock above it, while the surrounding rock below the weak interlayer is basically stable. In terms of stress distribution, with the increase of thickness of the weak interlayer, the maximum compressive stress gradually transfers to the deep weak interlayer, and the tensile stress area gradually decreases, while the depth of the tensile stress area gradually increases.

Tilt test is a common method to obtain the basic friction angle of rock. However, the existing tilt test technology has not formed a recognized standard, leading to an obvious difference between the test results of the same kind of rock materials. To solve this problem, an inclinometer was developed to measure the basic friction angle of rock based on the existing tester structure and testing mechanism. A series of multi-scale specimens with dimensions ranging from 50 mm to 150 mm in length and 25 mm to 75 mm in height can be tested. The high-precision rotation control of angular speed ranging from 6(°)/min to 60(°)/min can be realized through the servo motor and the high-speed ratio reducer. The corresponding test parameters, such as specimen’s sliding distance and inclination angle, can be monitored and recorded in real time. Then, the operation flow of testing basic friction angle was put forward. The main factors that have obvious effects on the experimental results were analyzed, such as specimen size and processing technology, angular velocity, test results criterion and the dryness of the structural plane. In this paper, the basic friction angles of 11 specimens, including granite and diabase, with discontinuity size of 100 mm×100 mm were tested, and it provided a reference for the test specification of inclinometer.

Directional drilling technology (DDT) originated from oil drilling industry. After many years of development, DDT has been extended to geological investigation and disaster management for tunnel engineering, aquifer reconstruction at the bottom of coal mines, pumping channel construction for power storage stations, municipal pipeline laying and other fields. Combined with the advantages of DDT for long-distance accurate crossing, ground grouting technology can achieve safe and efficient management of underground engineering disasters, and has a broad application prospect. This paper systematically reviews the development history, research status and key technologies of DDT. Firstly, the development history of DDT is introduced. Secondly, the scientific research plan, academic paper publication, relevant patent authorization, and application exploration of DDT are analysed. The analysis results show that the DDT in China has reached an internationally advanced level of technological independence and broad-range application, but there are still some core technical problems that need to be resolved, such as thousand-meter long-distance fast drilling, accurate drilling guidance in complex geological conditions, and accurate grouting in underground engineering. Focusing on the construction requirements of strategic major projects, this paper proposes four functional modules, i.e., forecasting, drilling, grouting and detecting, and analyses and discusses six aspects of ultra-long directional drilling and grouting (i.e., sensing prediction while drilling, rapid drilling and sticking prediction, precision guidance technology, remote gel controllable grouting material, pressure stabilizing and controlling sectional grouting process, and detection system with drilling). This paper can provide reference for the development of DDT, and for the prevention and control of disasters in underground engineering.

Typhoon rainstorms occur frequently in the southeastern coastal areas of China. The infiltration of rainwater under typhoon rainstorm conditions is an important cause of landslides in hilly mountains. Therefore, it is of great significance to study the migration law of rainwater infiltration wetting front under different geological environmental conditions. The temporal and spatial changes of rainfall infiltration were analyzed based on the monitoring data collected by the in-situ test of a landslide. A comprehensive rainfall infiltration model was established, which takes into account the nonuniform distribution of the initial water content of the soil and the slope environment. Then this model was verified by comparing to field test results and the predictions from existing classic models. The results demonstrate that: 1) At the initial stage of rainfall infiltration, the increasing slope of water content at different depths was steep. With the increase of time, the slope gradually slowed down. When the accumulated amount of rainfall reached 44.4 mm or more, the shape of the water profile changed from "Z" to reverse "S". As the rainfall intensity decreased, the type of rainwater infiltration process changed from water infiltration to non-water infiltration. At the later stage of infiltration, the growth rate of water content at different depths tended to be constant, and a stable infiltration stage was reached. 2) The migration speed of the wetting front increased with increasing rainfall. At the beginning of the rainfall, the rainwater infiltration rate was large. The migration speed of the wetting front was fast, and it decreased with the increase of soil depth. 3) The results of wetting front depth change over time calculated based on the improved rainfall infiltration model are consistent with the field monitoring test results, indicating that taking into account the slope angle and the inhomogeneity of the initial water content distribution can improve the accuracy of the predictions of the Green-Ampt model. The research results can have a certain significance for establishing an effective early warning model for landslides induced by the typhoon rainstorm.

When shield is driving along a curve alignment or during deviation correction, the asymmetrical thrust will generate an additional bending moment at the head of the tunnel, which will cause construction problems such as longitudinal deformation of the tunnel and dislocation between rings. Based on the theory of elastic foundation beam, the shield tunnel is simplified as a Timoshenko beam in Winkler foundation, an analytical model for evaluating the additional response of the shield tunnel induced the asymmetric thrust is established, and the analytical solutions of longitudinal deformation and internal force of shield tunnel are deduced. Then the correctness and applicability of the analytical solutions are verified based on the finite element method, and the sensitivity of key parameters in the analytical model is further analyzed. Finally, the influence range of additional bending moment and the second order effect of shield thrust are discussed. The research results show that the proposed analytical model is reliable and has good applicability for evaluating the additional response of shield tunnel under asymmetric thrust. The influences of foundation stiffness and tunnel stiffness on the longitudinal deformation of the tunnel are more significant than the internal force. The influence range of the additional bending moment is more sensitive to the changes of the foundation stiffness, and an exponential attenuation relationship is found to exist between the two. The shield thrust improves the longitudinal bending stiffness of the tunnel, and the second-order effect produced is relatively low. Under the condition of lower bending stiffness of tunnel and foundation stiffness, the second-order effect of axial force is enhanced.

In order to study the long-term creep law of extremely soft coal rock, the uniaxial and triaxial creep tests of extremely soft coal rock subjected to single-stage loads were carried out by using the self-developed triaxial creep test system. The following results and conclusions are obtained. 1) In the uniaxial long-term creep test, decay creep stage, steady-state creep stage, and accelerated creep stage occur successively during 232 hours, and the cumulative creep strain is as high as 3.45%, which is 10.5 times of the instantaneous deformation. The strain rate for the total steady-state creep stage is as high as 8 ?10 /h, the maximum accelerated creep rate reaches up to 0.043/h, and the strain rate of total creep process is distributed in a U-shape. 2) In the triaxial long-term creep tests subjected to the same axial pressure (0.96 MPa), the ability of resisting long-term deformation of extremely soft coal rock increases continuously with the increase of confining pressure from 0 to 0.6 MPa, which is shown as follows: the creep strain decreases significantly, the strain rate of steady-state creep stage decreases by order of magnitudes, the duration before creep failure increases obviously, the ratio of creep to instantaneous strain decreases dramatically, the intensity of accelerated creep failure decreases markedly. The creep strain and deformation rate are especially sensitive to the confining pressure change in the range of 0?0.2 MPa. 3) The accelerated creep stage of extremely soft coal rock is characterized by “gradual” time dependent instability, which is significantly different from the “abrupt” accelerated fracture instability of ordinary rock. 4) By connecting a new nonlinear viscoelastic element considering the concept of accelerated creep start-up time with Burgers model in series, a nonlinear viscoelastic plastic creep model is established to describe the three creep stages of extremely soft coal rock subjected to single-stage load. Then, the creep parameters were identified by employing the Levenberg-Marquardt optimization algorithm. The fitting curves are highly consistent with the experimental curves, verifing the validity of the proposed model. These findings can provide reference for theoretical analysis of nonlinear large creep deformation and support design of extremely soft rock.

To explore the influence of acid or alkali contamination on the shear strength of natural loess, the intact loess samples were first immersed in various concentrations of HCl and NaOH solutions. Then, triaxial shear tests, scanning electron microscopy tests, and mercury intrusion tests were carried out, and the soil chemical composition as well as liquid and plastic limits were measured, so as to assess the influence of acid or alkali contamination on the shear strength, microstructure, chemical composition and plasticity of loess. The results indicate that with the increase of the acid concentration, the peak of soil stress-strain curve attenuates, the shear strength deteriorates, the cohesion decreases exponentially, and the internal friction angle is relatively stable. As the alkali concentration increases, both the peak of soil stress-strain curve and the shear strength increase, the cohesion enhances markedly, and the internal friction angle enlarges slightly. Acid contamination breaks the soil particles, dissolves the cementing material, blurs the boundary between particles and pores, and increases the number and size of pores between skeleton particles and those between clay particles. Alkali contamination leads to the scaffold pore collapse of the soil, and the secondary cementation balances the local damage and strengthens the structural connection. Besides, the content and size of pores between skeleton particles decrease while more pores with a smaller size form between the clay particles. After soaking in acid solutions, the cation content in the soil increases notably, the calcium carbonate content decreases sharply, and the liquid and plastic limits are reduced. After soaking in alkaline solutions, both Al³⁺ and calcium carbonate in soil increase slightly, other cations decrease gently, and the liquid and plastic limits increase. Based on the analysis of test results, the microscopic mechanism of the evolution of the shear strength of loess contaminated by acid or alkali was summarized. Acid or alkali contamination resulted in mineral dissolution, ion exchange, and adjustment of particle and pore structure in the soil, destroying the initial structure of the soil and facilitating the formation of new structures. It is the damage of the initial structure or the new structure formation taking the dominance that decides the comprehensive effect improving or impairing the shear strength of the soil.

The line heat source method of soil moisture monitoring requires higher heating power. When the intensity of the heat source is unstable, the monitoring results are easily affected. To solve the shortcomings of the line heat source method, a mobile point heat source (PHS) method for soil moisture monitoring is developed in this study. This method uses the correlation between the thermal physical properties of soil and the moisture content to indirectly identify the moisture content by monitoring the cooling law of the PHS in the soil. Firstly, the heat transfer law of the PHS in the soil with evenly distributed moisture content was examined through numerical simulation. According to the characteristics of the cooling curves of the heat source, the moisture content discrimination index η was defined, and then the fitting relationship between the moisture content and the discrimination index was established. In addition, parameter analysis was carried out on the factors affecting the monitoring sensitivity. Then, numerical simulation of the PHS cooling law was carried out in the soil with unevenly distributed moisture content, and the moisture content discrimination index was calculated according to the time history curve at the cooling stage of the measuring point. After that, the moisture content distribution was inverted by the fitting formula of the moisture content and the discrimination index. The results show that the moisture content distribution inverted by the PHS method is in good agreement with the actual value, which verifies the theoretical feasibility of the method. Finally, the feasibility of the method is further verified by model tests.

With the rapid development of mountainous highways and high-speed railways in China, geological disasters caused by new tunnel excavation, such as landslides, are widespread. Meanwhile, the diseases caused by landslides in existing tunnels are also increasing, resulting in significant harm to the tunnel construction and operation. In this paper, the academic research status, existing problems, and the development prospects associated with the tunnel-landslide system were summarily all over the world. First, the relative spatial location relationship and deformation characteristics of the tunnel-landslide system were systematically investigated. Second, the detailed analyses of the current status and prospects of research on tunnel-landslide interaction from five aspects geological survey, theory, model test, numerical simulation and field monitoring. Then, the control and protection techniques of the tunnel-landslide interaction were expounded from landslide reinforcement, tunnel reinforcement, and monitoring and prediction technology, and the corresponding shortcomings in the existing research and the aspects that still need to be discussed were marked. Finally, it is recommended to carry out further research on the landslide soil plasticity, nonlinear contact, earthquake and rainfall multi-factor coupling effects, the development and utilization of centrifugal model tests, the applicability of constitutive models, and the fine modeling of tunnels. Also, the impact zones of tunnel excavation should be further optimized, and the corresponding novel control and protection technologies should be developed. On this basis, a new type of monitoring technology system linked and shared for tunnel-landslide can be thus established. This paper provides new perspectives and essential data for academic research on tunnel-landslide system engineering.

To reveal the mechanical characteristics and failure modes of stratified rock under combined static-dynamic loads, a split Hopkinson pressure bar device was used to apply one-dimensional load on stratified sandstone specimens. Meanwhile, digital image correlation technique was used to monitor the fracturing process in real time. Fracture evolution and energy consumption characteristics of different dip angles were further summarized. The results show that: under the combined static-dynamic loads, the dynamic strength of stratified sandstone specimens with holes initially increases and then decreases with the increase of bedding angle. The failure modes of the specimens can be divided into four types. The energy consumption characteristics of different dip angles under combined loads show that when the bedding dip angle is 45°, the strength of sandstone specimens and the proportion of absorbed energy reach the maximum values. In underground engineering, reasonable arrangement of breaking position and the angle between the direction of impact load and bedding can significantly improve the efficiency of rock burst, reduce the adverse impact of dynamic disturbance on surrounding rock, and enhance the stability of deep rock engineering.

The influence of different initial water contents, cold end temperatures and dry densities on sand water migration was studied using the self-developed water migration and frost heaving testing equipment. The influence of these three factors on frost heaving force and frost heaving capacity and the position of ice peak were determined. The results show that the initial water content and cold end temperature have obvious influence on the soil water migration and frost heaving effect. The water content increases from 0% to 10%, the peak water content increases by 5.00 times. The lateral frost heaving force and the frost heaving amount are increasing, and the ice front position moves up to 2.5 cm in height. The cold end temperature reduces from ?5 ℃ to ?15 ℃, the peak water content increases by 4.38 times. The lateral frost heaving force and the frost heaving amount are increasing, and the position of the ice peak moves up to 2.6 cm in height. Dry density has relatively insignificant influence on water migration and freezing characteristics of specimens. The results show an overall trend of slightly larger increases in specimen water content, lateral frost heaving force and frost heaving amount at smaller dry densities, and the position of the ice peak is concentrated in 2.2 ?2.5 cm in height. The prediction formulas of frost heaving force and frost heaving amount are put forward for different influencing factors, which can provide a reference for understanding the water migration law in sand under water vapor recharge and reasonably preventing frost heaving.

Rainfall-induced landslides are the most common geological hazard in the world, and the effect of preferential flow infiltration has been neglected for a long time regarding the slope reliability analysis under rainfall infiltration. The preferential infiltration under rainfall conditions is solved numerically by Comsol Multiphysics, and the slope factor of safety is calculated using an infinite slope model. An improved Cholesky decomposition method is applied to generate spatially correlated random fields, and the slope reliability during rainfall is analyzed using the Monte Carlo method. Combining the deterministic and reliability results to compare the variation of slope factor of safety between the homogeneous infiltration and the preferential infiltration during rainfall, it can be found that: (1) the preferential infiltration model is safer when the rainfall intensity is low, while the homogeneous infiltration is more stable at higher rainfall intensity; (2) the spatial variability of infiltration parameters is the crucial factor of slope failure in homogeneous infiltration, while the slope instability of preferential infiltration is mainly caused by the rapid advance of the wetting front; (3) for the study of preferential infiltration model, the slope has a higher probability of failure when the hydraulic exchange intensity between the matrix domain and the preferential domain, while smaller hydraulic exchange intensity may affect the slope failure at the bottom.

Rock rupture refers to the process of crack initiation, propagation and coalescence. In order to study the dynamic propagation and evolution process of internal cracks in rock subjected to deformation and failure, industrial CT was used to conduct phased observation and scanning of the rock rupture process, and a three-dimensional rock fracture model was constructed by vectorization of CT image stack. The characteristic parameters of the crack structure were statistically analyzed to quantitatively characterize the crack propagation during the rock rupture process. On this basis, the local failure morphology characteristics on crack propagation path were extracted and the rock and mineral identification experiment was combined for meso-scale analysis. The research results show that the three-dimensional fracture propagation process can be quantified using parameters such as fracture volume V, surface area S, and fractal dimension D, and the parameters experience a change law of "basically unchanged–small increase–surge". Based on the CT slice images, the crack area can characterize the local crack propagation characteristics of the rock, and it corresponds to the expansion and evolution of the three-dimensional cracks at the same stage. The mesostructure of the rock has a great influence on the crack propagation, which form the extension around the gravel, through the gravel, and bifurcation when encountering gravels. The research results will provide a research foundation for rock instability failure and disaster warning of engineering rock mass.

In the reliability analysis of slope stability, the deterministic analysis method is usually used to calculate the safety factor to evaluate the stability of slope. However, the inherent spatial variability of rock mass properties cannot be considered and described adequately in traditional deterministic method, resulting in the inaccurate calculation of slope failure probability. Based on Hoek-Brown criterion and random finite difference method (RFDM), the reliability analysis of slope stability and random response of pile are discussed in this paper considering spatial variability of rock mass. The uniaxial compressive strength and material constant for the intact rock are regarded as random field variables and geological strength index GSI is assumed to be a random variable. The results show that the spatial variability of rock mass parameters has a significant effect on slope failure probability and pile response. Ignoring the spatial variability of rock mass parameters will overestimate slope failure probability and the mean value of the maximum bending moment of anti-slide pile, and underestimate the mean value of displacement at pile head. The results can provide design guidance for slope reinforcement as well as layout optimization of anti-sliding piles.

To explore the environmental durability of heavy metal contaminated soil remediated by enzyme induced carbonate precipitation (EICP) technology, acid soaking, freeze-thaw tests, and rainfall tests were carried out on the zinc and lead contaminated soil after EICP remediation, respectively. The durability and influence of the zinc and lead contaminated soil remediated by EICP technique under different environmental conditions and the corresponding influence regularity were discussed in this paper. The results showed that under different concentrations and types of acid solutions, the leaching amount of heavy metal ions in exchangeable and carbonate bound forms in the zinc and lead contaminated soil after EICP remediation decreased with pH value, with the content of ions in carbonate bound forms decreasing and the content of ions in exchangeable forms gradually increasing. It was also found that the stability in sulfuric acid solution was greater than that in nitric acid solution. As the number of freeze-thaw cycles increased, the leaching amount of exchangeable ions in the zinc and lead contaminated soil remediated by EICP technique increased, while the content of ions in the carbonate bound form reduced. Under the condition of heavy rain, Zn²⁺ and Pb²⁺ were mainly released within the first 20 minutes and migrated from top to bottom. All the results demonstrate that the heavy metal contaminated soil remediated by EICP technology has a good durability under acid soaking, freeze-thaw cycles, and heavy rain.

In order to realize the visualization of unsaturated preferential flow migration of soil, a column device was designed to carry out unsaturated seepage tests. The transparent soil and digital image processing technology were used to establish the relationship between the normalized pixel intensity and the saturation of transparent soil. On this basis, the effects of the connectivity of the preferential flow path and the rotation angle of the adjacent preferential flow path on the preferential flow migration in unsaturated soils were studied by conducting laboratory model tests. The results showed that it was feasible to characterize the saturation of unsaturated transparent soils based on the intensity of image gray pixels. The profile of fully connected preferential flow (O-O type) and upper connected preferential flow (O-C type) presented T-shape. The saturation of the central axis profile was obviously different from that of the central edge profile. In the lower connected preferential flow path (C-O type), the soil oil pressure and matrix potential could not make the fluid enter the preferential flow path to form the preferential flow, in which the change of infiltration was consistent with the uniform flow. The stable infiltration rate and the wetting front moving velocity of O-C preferential flow were 1.5 times and 1.4 times of those of the C-O type, respectively. A new preferential flow was formed in the area between adjacent O-C preferential flows. The soil in that area reached higher saturation and the growth rate of saturation decreased with the increase of rotation angle. When the preferential flow rotation angle was 90?, 60? and 30?, the stable infiltration rate was respectively 1.5 times, 1.3 times and 1.2 times that of the uniform flow. As the fluid was affected by gravity, the preferential flow with a small rotation angle only infiltrated horizontally along one side of the path. However, the moving velocity of the wetting front was 1.3 times, 1.4 times, and 1.5 times that of uniform flow, respectively. The interaction between adjacent preferential flows was weakened, so it was difficult to form the new preferential flow.

Cement-based 3D printing technology is increasingly popularized in civil construction and rock-like physical simulation, yet its cement-based 3D printing materials not only is limited by the challenges in quality control such as liquidity, shrinkage, and processability, but also is lack in basic formula in concrete material. This paper firstly establishes a set of assessing system based on six indexes and corresponding quality classification method considering the whole process of solid model forming for cement-based 3D printing, including fluidity, extrusion, setting time, buildability, compactness and deformation rate. The quality evaluation of the 3D printing solid model is realized in three aspects: printing process, stacking process, and solid accuracy. Then, through the two-stage orthogonal design and printing test with different threshold ranges, the basic mix ratio of cement-based 3D printing materials, which is ideal in comprehensive printing performance, is obtained. Finally, the uniaxial compression test is carried out for the printed cement-based 3D solid model and its rock-like mechanical properties are evaluated, which can provide a new type of rock-like cement-based 3D printing material. The specific mix ratios lay the foundation for the civil construction and physical simulation test of rock engineering based on cement-based materials.

Aiming at the disaster of rock burst in graben structural area of coal mine, based on the particularity of graben structure and the occurrence characteristics of overburden rock, the movement characteristics of overburden and stress evolution law of coal and rock mass in the graben structural region are analyzed and the corresponding mechanical model is established by means of field investigation, theoretical analysis and numerical simulation. The mechanism of rock burst in the graben structural region is consequently studied. At the same time, the evolution characteristics of microseismic energy and frequency before and after rock burst are analyzed, and the time sequence of microseismic precursory information of rock burst in graben structural region is obtained. The results show that: the sliding subsidence of wedge is closely related to the width of goaf. The existence of goaf in graben structural area will break the stable state of wedge and lead to more severe stability conditions. The local slip and dislocation of FD6 and FD8 faults lead to wedge sliding subsidence, and the hanging roof structure of graben structural area also has corresponding extrusion effect on the undeveloped coal mass. The same action provides high static stress conditions for the occurrence of rock burst, and the breaking of suspended roof structure provides dynamic load disturbance conditions for the occurrence of rock burst. The joint action of static load and dynamic load results in the upper drift and triple entry in the graben structure. The energy and frequency of microseismic show obvious deviation before the impact, and the daily average energy of microseismic is measured. There are obvious sudden drop and sudden rise. The anti-scour technology of “chain-brokening and consumption-increasing” is proposed based on the analysis results and field practice. It provides some theoretical reference value for occurrence mechanism, early warning and prevention and control of rock burst on working face in graben structural area.

The pressure balance method is adopted in calculating the stability against the inrush of excavation in the current criterion. The effects of soil strength and excavation size are not considered in this method. Therefore, the method gives a conservative evaluation of the situation approaching the critical artesian pressure. Aiming at these problems, firstly, the safety factor is defined according to the principle of shear strength reduction in this paper. The existing method considering soil shear strength along slip surfaces is improved by adopting effective stress analysis and considering the passive zone of excavation. Then, a new calculation method for inrush resistance is proposed based on the elastic plate theory. Subsequently, the results of the finite-element method with reduced shear strength are compared with those of above calculation methods under the plane strain condition. And the effects of the excavation length-width ratio for the improved and proposed methods are further analyzed. Finally, the analysis of an engineering practice is carried out. The analysis shows that compared with the existing methods, the improved and proposed methods in this paper can better reveal the trends of excavation stability against inrush with the change of the artesian pressure and improve the economy on the premise of ensuring the safety of engineering practice.

To explore the strength degradation mechanism of gas bearing coal, the uniaxial compression tests were performed on coal under different initial gas pressures by using self-developed gas-solid coupling test system, and the pore and fracture structures of gas bearing coal were characterized by SEM, high-pressure mercury injection, low-temperature liquid nitrogen adsorption and micro-CT scanning system. The mechanical and non-mechanical effects of different gas occurrence states on pores and fractures were described respectively, and the internal relationship between pore and fracture failure and the loss of macroscopic strength of gas bearing coal was revealed. The results show that the degradation degree of average uniaxial compressive strength of coal increases with the increase of initial gas pressure. The pore structure and pore size distribution of coal were characterized by multi-means, and it is found that the development of fracture structure of gas bearing coal is not obvious. The proportion of isolated pores is relatively large, and the connectivity between them is poor, which is not conducive to gas seepage. The method of jointly characterizing the pore structure and pore size distribution of coals can correct the errors caused by "shielding effect" of micropores and transition pores and the errors caused by the "compression effect" of coal matrix or pore and fracture failure caused by the sample size. Adsorbed gas leads to the fracture and failure of micro-elements through non-mechanical action, mechanical action of expansion stress and gas wedge effect of free gas on coal body. The mathematical model of mechanical property deterioration and macro strength loss of gas bearing coal based on the micro view angle was established. From the model, it is found that the action of gas leads to the shift of the center of the Mohr circle to the left, the envelope of Mohr-Coulomb strength to the right, and the cohesion becomes smaller, and finally it leads to the loss of macro strength of coal.

Based on the continuous medium model and the pile-soil interaction, the horizontal seismic response of a single pile subjected to vertically propagating S waves was investigated by regarding the single pile as a one-dimensional linearly elastic beam. The time-harmonic displacement of bedrock was introduced as the vertically propagating S waves, and the horizontal dynamic impedance function of the soil was derived by the governing equations of the plane strain model. Analytical solutions for the seismic response of the single pile subjected to vertically propagating S waves were obtained by subsuming soil impedance into the governing equation of the single pile and considering the boundary conditions at pile top and toe. The solution was verified by comparing it to the results of existing studies. Furthermore, as pile-soil modulus ratio increases, the minimum value of the kinematic response factor decreases. The kinematic response factor is not particularly sensitive to the large pile slenderness ratio and the soil material damping. For the horizontal amplification factor at the pile top, the increase of the pile-soil modulus ratio only suppresses the amplification at high resonance frequency, and the large pile slenderness ratio has the trivial effect on it. As the soil material damping increases, the amplification at resonance frequency gets considerably suppressed. The seismic response of the pile is obviously affected by the pile-soil modulus ratio only when the pile slenderness ratio is small, and it decreases with the increase of the pile-soil modulus ratio.

To study the active earth pressure of finite soil behind the retaining wall, the cohesionless soil behind the wall is taken as the research object. The fracture surface is assumed as the plane passing through the heel of the wall, and in the translational mode of the retaining wall, the soil behind the retaining wall forms an arc-shaped small principal stress arch. The soil behind the retaining wall is divided into several curve thin-layer elements by the stratification method along the small principal stress. Considering the inhomogeneity of stress distribution on the upper and lower surface of the element, a calculation method is proposed for the active earth pressure of finite soil retaining wall. The expressions of active earth pressure resultant force and the height of its action point are given, and the correctness of this method is verified. The results show that the curve thin-layer element method can accurately consider the complex stress condition of the element, and can better reflect the variation law of the active earth pressure of finite soil behind the retaining wall. The active earth pressure shows a nonlinear distribution along the wall height H, it firstly increases with the soil depth increasing, then decreases monotonically near the bottom of the wall. In parameter sensitivity analysis, the distribution of active earth pressure of retaining wall and the height of combined force applied point are analyzed with different width-height ratios of soil and wall back roughness. The results show that with the increase of width-height ratio n, the active earth pressure gradually increases, the curve of earth pressure distribution becomes more and more nonlinear, the height of resultant force application point gradually decreases, and it is always greater than . It tends to be stable when n is greater than 0.71, so 0.71 can be assumed as the critical width-height ratio of finite soil and semi-infinite soil. The active earth pressure decreases gradually with the increase of the frictional angle ; the curve of earth pressure distribution becomes more and more nonlinear, the height of resultant force application point increases gradually and is always greater than .

Seepage erosion can lead to the loss of particles in the soil with water flow, thereby changing the mechanical and hydraulic properties of the soil, and then causing the deformation and even damage of earth-rock dams in engineering. In this paper, the coupled CFD-DEM method is used to study the seepage erosion characteristics of gap-graded sandy soil, which is widely found in dam foundation, filter-base soil systems in earth-rock dams and gravel packing for sand control in oil and gas production, after reasonable simplification of its gradation. The effects of hydraulic gradient, confining pressure and fines content on seepage erosion in gap-graded sand soils were investigated through 8 groups of upward seepage simulation. Macroscopic phenomena such as the mass of fines loss, fines loss rate, soil surface settlement and microstructural changes were monitored in the tests and verified against previous indoor tests. The results showed that the effect of fines content on the mass of fines loss, peak fines loss rate and soil surface displacement from the numerical simulation was more significant than the effect of the confining pressure and hydraulic gradient in the tests. In addition, the analysis of microstructure demonstrated that the conversion of the soil force chain transfer structure occurred near the threshold of 25% fines content. The triaxial results indicated that the loss of fines also led to a decrease in peak strength and E50 modulu of soil.

The water softening of argillaceous sandstone in complex stress state will seriously reduce the bearing capacity of surrounding rock, resulting in the increase of failure range and deformation pressure of surrounding rock, and finally seriously threaten the safety of water tunnel during operation. To solve this problem, triaxial tests under different immersion time are carried out to investigate by using argillaceous sandstone from water delivery tunnel of Lanzhou water source project. The results are as follows. As the immersion time prolongs, the sensitivity of peak strength and residual strength to confining pressure decreases gradually. This may be attributed to the fact that the water reduces the sensitivity of peak strength and residual strength to confining pressure to a certain extent. In the early stage of immersion, the decreasing rate of residual strength with immersion time is significantly less than that of peak strength with immersion time, and the softening effect of water on the peak strength of argillaceous sandstone is more remarkable. The peak strength, residual strength and elastic modulus of argillaceous sandstone samples decrease with the increase of immersion time in a negative exponential relationship. With the increase of confining pressure, the influence of immersion time on elastic modulus gradually weakens, and the increase of confining pressure reduces the softening effect of water on the elastic modulus of argillaceous sandstone to a certain extent.

The true triaxial loading and unloading tests were conducted to obtain the characteristic stress, failure mode and energy evolution characteristics of the granite in an underground cavern. Results show that under the true triaxial loading and unloading stress paths, the failure modes of granite are both tensile and shear composite failure, and characteristics of high damage stress and brittleness are obviously observed. A new brittleness index is proposed to evaluate rock brittleness by using volumetric strain curve. The brittleness of granite under unloading condition is higher than loading condition. In the true triaxial loading test, the change trend of total energy with axial strain goes through three stages: slow increase, rapid increase, and steady increase. In the true triaxial unloading test, the dissipated energy increases rapidly at the moment of unloading, and its proportion in the energy distribution increases, which becomes the main energy consumption. The energy dissipation value of the granite in the loading test is obviously greater than that in the unloading test. It indicates that more energy is required for samples under the loading path to cause damage. More elastic strain energy can be released under the unloading path, which is more dangerous than the loading path.