The coupling action of wetting-drying and freeze-thaw is one of the basic forms for the severe climate conditions in high and cold regions, which has significant influences on the normal operation of water supply canals in expansive soils in these regions. Based on the field expansive soils of water supply canals in the northern area of Xinjiang Uygur Autonomous Region, the fissures tests under the unidirectional wetting-drying and the cyclic action of coupling wetting-drying and freeze-thaw are conducted respectively to investigate the influences of freeze-thaw process on evolution rules of fissures in expansive soils. The evolution characteristics of the internal fissures in expansive soils with various cycles can be obtained through the CT scanning and 3D reconstruction. The test results show that the internal fissures in the samples under the two conditions both develop from the surface towards the internal part and the regional distribution of fissures can be obviously captured. A comparison of the evolution characteristics of the internal fissures under different test conditions indicates that the freeze-thaw process in the coupling cycle of wetting-drying and freeze-thaw has significant influences on the focal depth of the fissures, and the induced fracture is macroscopically characterized by the transformation from non-horizontal long fissures into short ones. In addition, it is found that the development of fissures in expansive soil samples under the action of various cycles can be successfully evaluated by using the volume fraction of the fissures.

In order to obtain the relationship between the characteristics of acoustic emission (AE) and the unloading state of surrounding rock during roadway excavation, AE monitoring test was carried out during excavation unloading with three different unloading rates using thick-walled cylinder roadway surrounding rock samples made of cement mortar. The self-developed tunnel test system of excavating and unloading of small roadway surrounding rock is equipped with AE system and deformation monitoring system to collect AE time-frequency signals and deformation information of samples during the unloading process. In addition, the b values were also studied. The results show that: 1) The test process has four stages: loading stage, maintenance stage I, unloading stage and maintenance stage II. These four stages correspond well to the cumulative ringing curve. 2) In the loading stage, there are many high frequency and high amplitude signals, and the distribution range is wide. The original crack in the sample is obviously compacted under the pressure of three-direction. After that the high frequency and amplitude signals reduce obviously and the distribution range is also reduced, the sample continues to be compacted under three-direction, and the deformation is small. In the third stage, the signals with high and low frequency and their amplitude increase. The strain rate is high and the deformation increases rapidly, the unloading effect is more significant accordingly. In the maintenance stage II, the high-amplitude signals continue to increase, and the deformation of the inner and outer sides increase continuously with lower speed, when the unloading speed is slow. 3) The bigger the unloading speed is, the higher the amplitude of "steep rise" appears in the ring cumulative curve of the unloading stage, the earlier the amplitude of "steep rise" appears in the ring cumulative curve of maintenance stage II, the smaller the amplitude is. 4) With the increase of unloading rate, the magnitude of b value tends to decrease and varies steadily as a whole, which indicate that the proportion of big AE events are increased, and the fracture scale is larger.

Most of researches related to the immobilization mechanisms of stabilized/solidified heavy metal contaminated soils were depended on the mechanical tests and chemical analysis. Thus, proposing a physical method that is un-destructive, time-saving, convenient and cost-effective is the urgent issue to be solved currently. In this paper, a series of laboratory tests including electrical resistivity test was performed to investigate the microstructural mechanism of cement-fly ash stabilized heavy metal contaminated soils. Furthermore, the prediction model of unconfined compressive strength was established based on the electrical resistivity. The tests results presented that with increasing of curing time, electrical resistivity, pore water electrical resistivity, structure factor, as well as shape factor of the specimens increased, and the anisotropy coefficient decreased. While increasing the heavy metal ions concentration leaded to an obvious reduction in electrical resistivity, pore water electrical resistivity and shape factor, and the anisotropy coefficient increased gradually. The SEM results was consistent with that of the electrical resistivity, which confirmed the feasibility that revealing the microstructural mechanism of the stabilization/solidification of heavy metal contaminated soils by electrical resistivity method. Finally, a good linear relationship between the electrical resistivity and unconfined compressive strength can be obtained. It is suggested that the electrical resistivity can be used as the index to evaluate the mechanical properties of the stabilized soil.

Based on the self-developed multi-field coupling testing system for coalbed methane exploitation, a series of physical simulation experiments of gas drainage under different spacing distances between boreholes was conducted to investigate the evolution of gas pressure and gas flow rate during drainage process. The results show that the gas pressure at different measuring points in the coal seam drops faster with the decrease of the distance from the drainage pipe. With the influence of the drainage superimposed effect, the gas pressure drop rate of areas between the drainage pipes is greater than that of the outside under the same distance condition, and the pressure difference rapidly reaches the peak value after the start of drainage, and then drops. The closer to the drainage pipe or with the increase in the borehole spacing, the less the effect of the superimposed effect is. There exists a power function relationship between pressure gradient and instantaneous flow rate on the section of the coal seam perpendicular to the borehole. With the increasing amounts or decreasing spacing of boreholes, the gas pressure of the same position drops faster, the instantaneous flow rate decreases, the cumulative flow rate increases, and the value of coal seam permeability coefficient “ɑ” increases, indicates that the gas drainage efficiency is higher. It can provide theoretical basis for reasonable arrangement of boreholes.

Based on scaling and translation properties of wavelet function, a nonlinear constitutive model of shear stress-displacement of bolt interface is established to simulate the anchoring interface stick-soft-slip mechanical characteristics. The proposed model overcomes the disadvantage in complexity of stage analysis by tri-linear bond-slip model. The equations of load-displacement curves of the bolt is deduced based on the mechanical differential equations of load transfer of the bolt, and corresponding numerical calculation methods and steps are proposed. The displacement, axial force and shear stress distribution along the anchorage length under different pullout displacements are obtained, and the complete load-displacement curve is obtained through a practical example. The calculated load-displacement curves as well as the distributions of the interfacial shear stress and the bolt axial stress are in good agreement with test results. The method can accurately reflect the force transmission mechanism of the bolt under different loads, and simulate the whole process from elastic working state to the plastic slip of the bolt. The influence of the anchorage length, axial stiffness and the constitutive parameters of the anchorage on the anchorage effect are obtained through the analysis of parameters.

Carbonate sand, a geomaterial with special structural and mechanical properties, is the main material of the South China Sea Island Reef blow-fill project. A series of drained triaxial tests was carried out on carbonate sand that had previously been extensively researched. This paper presents a modified model referred to as the Shen Zhujiang's elastoplastic model that is capable of predicting the experimental data and simulating the stress-strain behaviour of carbonate sand over a wide range of confining pressures and regions. A functional relationship between the tangent modulus and the ratio of tangential volume to the stress ratio of carbonate sand is proposed. These functional relationships enable the model to describe the characteristic transitions from contractive to dilative shear response of carbonate sands as the stress increases under low confining pressure. Meanwhile, predictions based on the modified models are obtained and compared with the experimental data of the carbonate sand. It is found that the characteristic dilatancy and contraction of carbonate sand, as well as the stress-strain softening curve can be reasonably described compared with the drained triaxial tests data to illustrate the modified model capability for describing the shear behaviour of carbonate sands. For the un-drained triaxial tests, the pore pressure coefficient was introduced into the expression of pore pressure, and the results show that the data is well simulated. Compared with the existing Sun Jizhu's elastoplastic model, the proposed model expression is easy to understand and the parameters can be obtained easily.

In karst water-enriched tunnel, due to the complexity of water-rock coupling effect, the evolution of pore structure in rock mass is unstable and will induce the deterioration and instability of rock mass, which puts forward higher requirements for the research of the stability of underground permeability rock mass. Firstly, to investigate the effect of osmotic pressure on the karst pore structure, osmotic test, acoustic emission test, mercury injection test, SEM test and fluorescence photometric test were carried out on limestone from Maixi tunnel, Guiyang, treated with different osmotic pressures. Then, the mass loss rates, P-wave patterns and frequency spectrum, pore distribution curves and scanning electron microscope images were obtained. Results show that the mass loss rates of limestone show the tendency of "slow increase-rapid development" with the increase of osmotic pressure, and the demarcation points of the two phases is 6 MPa. With the increase of osmotic pressure, the amplitude of head wave gradually decays, and the waveform curve first decays quickly and then tends to be stable early, and the development degree of wave tail gradually slows down. Also, it is found the peak of dominant frequency gradually decays and evolves from single peak to multiple peak and has excellent quadratic function with osmotic pressure, and that the wave frequency develops from (relatively) high frequency to low frequency. Besides, the macropores and mesopores formed by dissolution play a decisive role in the permeability of limestone. With the increase of osmotic pressure, the accumulation of mercury increases exponentially. Meanwhile, the mineral contents including CaCO3 and SiO2 change exponentially. Compared with chemical dissolution, mechanical dissolution is more sensitive to time and dominates the formation of pores. In additional, the established water-rock contacting dissolution model indicates that the drag force of osmotic water depends on its flow velocity, while the increase of osmotic pressure can enlarge the osmotic pressure gradient, thus increasing the flow velocity.

A large-scale 1:200 three-dimensional dynamic model of the slope blocks combined with complex structural planes of Baihetan hydropower station was designed. By inputting seismic waves of different amplitudes, the responsive displacement time history and residual deformation distribution of the block under strong earthquake were studied. Residual deformation distribution reveals the failure mode of the slope block and the failure category and characteristics of the slope block were evaluated. The test results show that the seismic stability of engineering slope block was good under the design conditions. In the process of seismic overloads, the residual deformation occurred firstly in the trailing edge crack, and the residual deformation decreased gradually from the two triangular parts near the edge toward the inner slope. The triangular parts are the areas prone to failure first during earthquake. The seismic failure mode of the slope block is the double-slip failure along the bottom and side. Based on the residual deformation and statistical results after earthquake, the allowable design residual deformation should be less than 15 cm and the first obvious mutation. After the residual deformation of the engineering slope block exceeded 30 cm, the displacement mutation is strengthened, and the slope block may slide completely. Slope anchoring should pay attention to increase the tensile strength of trailing edge crack and the shear resistance of the bottom surface of block.

Fissure grouting is a general method in the fissure reinforcement field in earthen sites. The expansibility of slurry and the mechanical compatibility between concretion and site soil are the key factors on whether the grouting measures can fulfill the role of sealing and bonding. The swelling and fluidity tests of SH-(CaO+F+C) slurry with nine different mix ratios and water cement ratios, and three different water-cement ratio slurries of the mixture ratio CaO:F:C = 3:2:5 were selected as optimum slurry. The best curing age was determined by the concretion strength test with continuous interval age. And then the thermal expansion coefficient of the grout concretion and the site soil were obtained by the thermal expansion test and the unconsolidated undrained triaxial test. The failure modes, stress-strain curves, shear strength parameters and elastic modulus of three kinds of specimens, which are grout concretion, grout concretion - remolded site soil and remolded site soil were also obtained. The results show that the thermal expansion coefficient of grout concretion and site soils are the same in magnitude. The mechanical behavior of three kinds of grout concretion and remolded site soil are consistent. Among them, the grout concretion with a mixture ratio of CaO:F:C = 3:2:5 and the water-cement ratio of 0.50 has the best mechanical compatibility with the remolded site soil.

The macroscopic mechanical behaviors of coral sand, such as high compressibility, shear and creep are closely related to the characteristics of coral particle crushing. The strain rate effect of coral particles is of great significance to study the strength and deformation characteristics of coral sand under different loading forms . The effects of loading strain rate on particle crushing strength, crushing mode and fractal dimension were studied by applying different rates of 0.1-50 mm/min on about 300 coral particles in crushing tests. The analysis results show that the crushing strength of coral grain follows the Weibull distribution law, and the characteristic crushing strength increases nonlinearly with the increasing of strain rate. Due to the increase of loading rate, the main splitting failure of coral particles appears prior to the surface grinding and local fracture, and the corresponding load-displacement curves show a transition of "multi-peak" pre- and post-peak. The fractal dimension and crushing energy density of broken coral grain are also rate-dependent and the positively linear relationship with the logarithm of strain rate indicates that the dissipated energy and damage degree develop with loading rate.

Researches show that the particle morphology has a significant effect on the macro-mechanical properties of gravel aggregates. Accurately describing the morphological characteristics of particles is the foundation work for studying the macro-mechanical behaviour of materials from a micro-mesoscopic level. Firstly, the quantitative characterisation parameters of particle morphology were discussed based on the three-scale morphological characteristics of particle shape, angularity and surface texture. Then, the distribution rule of these parameters was analysed for different grain groups. Finally, the relationship between the particle morphology characteristics and the stacking porosity was discussed. The main conclusions are 1) The proposed quantification parameters with the perimeter as the core characterisation can better reflect and distinguish the particle morphology, as well as its shape, angularity and surface texture; 2) According to the statistical analysis of 1 000 particle samples in each group, the parameters describing the particle morphology, shape, angularity and surface texture all conform to a logarithmic normal distribution; 3) The comprehensive angularity index and surface texture index of particle aggregates present a good mapping relationship with the stacking porosity in both loose and dense states, and they are the main controlling factors on stacking porosity.

Ferrous sulfate(FeSO4)was used to stabilize chromium-contaminated soils. Leaching test, alkaline digestion test and sequential extractions test were conducted to investigate the variation of chromium speciation and leaching properties of FeSO4 treated chromium-contaminated soils along with Fe(Ⅱ)/Cr(VI) molar ratio and curing time. Results showed that the leaching concentration(hexavalent chromium and total chromium) and hexavalent chromium content of the soil were decreased significantly with Fe(Ⅱ)/Cr(VI) molar ratio and curing time. When Fe(Ⅱ)/Cr(VI) molar ratio was 3, the Cr(VI) and total Cr leaching concentrations were lower than the identification standards for hazardous wastes: identification for extraction toxicity(GB/T5085.3—2007) of China. Furthermore, When Fe(Ⅱ)/Cr(VI) molar ratio was 10, the amount of Cr(VI) in soil was below the threshold allowed by Environmental quality standards for soils(GB15618—2008) for industrial and commercial reuse of China (Cr(VI)<30 mg/kg). And the civil reuse of soil (Cr(VI)<5 mg/kg) was only achieved with Fe(Ⅱ) /Cr(VI) molar ratio of 20. Sequential extraction procedures revealed that majority of acid soluble fraction of chromium was converted to reducible fraction with FeSO4 stabilized. The leaching concentration of Cr(VI) increased as exponential function with the amount of Cr(VI) increases in soil. The leaching test (TCLP, SPLP) could not reliably evaluate the stability effect of chromium.

Lateral and torsional movements of the pile group caused by eccentric lateral loading, which would lead to different movement directions of the foundation piles. Therefore, the movement directions of the foundation piles is a key factor to affect the group effect occurred in the pile group. By analyzing the relationship in motion of double piles subjected to the eccentric lateral loading. It was found that the angles between the movement directions of the leading pile and the trailing pile with their connection line, η and θ, were in the range of 0°≤η≤90° and ?90°≤θ≤90°, respectively. Experimental and numerical data showed that the reduction factor was closely related to η and θ, which was used to quantify the pile-soil-pile interaction between two piles. There were a range for the combinations of η and θ in which no interaction occurred between them. Once the pile-soil-pile interaction existed, the influence of the interaction on the trailing pile was typically more significant than that on the leading pile. The concept of p-multiplier, which originally used to define the group effect in laterally loaded pile groups, was adopted to quantify the group effect on pile groups subjected to eccentric lateral loading. The p-multiplier of each pile can be calculated by multiplying cumulatively the reduction factors from all other piles in the group. An empirical formula was proposed to calculate the reduction factor according to the theoretical, experimental and numerical results. Then a calculation method of generalized p-multiplier was proposed. Two examples were employed to demonstrate the rationality of the method.

In order to further study the drawing test mechanism of the ribbed geomembrane, the mesoscopic properties of interface between ribbed geomembrane and sandy soil under different working conditions (ribbed height and temperature), the displacement and velocity field distribution of particles within the liner system were studied using laboratory drawing test and particle image velocimetry(PIV) analysis. The results show that different ribbed heights and temperatures have significant influence on the stability of the interface, and the ultimate drawing resistance increases with the increase of ribbed height and the decrease of temperature. The mesoscopic analysis shows the sandy soil near the interface of ribbed geomembrane and sandy soil can form indirect impact area. The sandy soil displacement and velocity within the indirect impact area are larger than that of surrounding sand particle, this can be explained by that the rib squeezes the left side of the rib or moves it as a whole during the drawing process. Moreover, the indirect impact area between the ribs interacts with each other, and thus causes strengthen effect on sand particles. Experimental results and PIV analysis indicate that temperature and ribbed height have significant effects on the indirect influence zone, so that the two factors have significant influence on the interface stability of ribbed geomembrane and sandy soil.

Based on series of laboratory tests including cyclic freeze-thaw, water immersion and cyclic drying-wetting, an experimental study has been performed to analyze the intrinsic change in the compressive strength and appearance of dredged sludge solidified with reactive MgO-fly ash, a green and environmental-friendly binder, and clarify the climate change-induced durability evolution. The test results indicate that the dredged sludge solidified with reactive MgO-fly ash has outstanding performance in resisting the damage from cyclic freeze-thaw, water immersion and cyclic drying-wetting. The durability performance of reactive MgO-fly ash solidified sludge is significantly enhanced as the MgO-fly ash content and mass ratio of MgO to fly ash increase. The strength behavior of MgO-fly ash solidified sludge is largely weakened by water immersion, cyclic dry-wet and cyclic freeze-thaw, and this causes a smaller compressive strength than that of samples cured under standard condition. As the number of freeze-thaw cycle increases, the compressive strength tends to decrease gradually. The compressive strength of samples immersed in water is initially reduced, and then changes within a limited interval. The compressive strength of damaged samples decreases with the increasing drying-wetting cycles. Based on the test results, an intrinsic microscopic mechanism model is proposed for the durability evolution of dredged sludge solidified with reactive MgO-fly ash.

To study the mechanical characteristics of unloading rate damage effect for Nanyang undisturbed expansive soil, the unloading-reloading undrained triaxial shear test was conducted with three unloading rates (0.02 kPa/min, 0.2 kPa/min and 2 kPa/min) through the GDS triaxial test system. The effect of over-consolidation ratio and consolidation state have also been studied. Test results showed that the shear mechanical properties of expansive soil are related to the initial unloading rate. Under the same axial strain, the deviatoric stress is smaller with lower initial unloading rate. The slope of stress-strain curve changed significantly before and after the principal stress direction rotation. Under the same consolidation state and over-consolidation ratio, pore water pressure increased first and then decreased. Peak pore water pressure strain decreases as the unloading rate increases. As the unloading rate decreases, the tangential modulus of the undisturbed expansive soil decreases monotonously. The higher the initial unloading rate, the greater the undrained shear strength. After the initial unloading damage of expansive soil sample, the shear strength obtained by reloading conventional triaxial elongation test was lower than that of non-damage soil sample under both isotropic consolidation and K0 consolidation. The damage degree SD obtained by the failure strength qf of expansive soil greatly underestimated the damage degree of the damage rate to expansive soil. It is recommended to use the strength of peak pore water pressure qumax to design and calculate the slope of expansive soil. The mechanical properties evolution of undisturbed expansive soil with unloading damage rate is influenced by the axial strain and cracking character.

The shaking table comparison test was carried out to study the dynamic response of the tunnel through fault under earthquake. In this study, the test materials, choice of the similarity ratio and the dynamic loading conditions were introduced. Test results show that stress on the waist is obviously greater than that on the top and bottom of the tunnel lining under earthquake, which means the waist needs to be strengthened in design. The tunnel lining has a certain acceleration magnification effect. In addition, the main Fourier spectrum energy of the lining is concentrated in the middle and low frequency, and the acceleration and lining strain responses will change abruptly when the structure gets close to failure triggered by earthquake. The development of lining cracks is closely related to their location. Cracks at the bottom of the lining develop from the inside to the outside; however, cracks at the waist develop from the outside to the inside. The smaller the angle between the fault and axial direction of the tunnel is, the more obviously the cracks develop. The ultimate tunnel failure modes consist of longitudinal tensile crack and joint dislocation. This research results provide a reference for the relevant design of the tunnel through fault.

Using Rowe’s stress dilatancy equation to considering the influence of particle breakage energy is a main method to express the dilatancy property of coarse-grained soils. However, the calculated breakage energy may be decreased during shearing or even be negative, which violates the irreversible thermodynamic theory. The reason lies in the friction factor, namely the critical state stress ratio Mc, is too larger during the entire shearing process. The breakage energy was closely relate with the evolution of particle breakage. An equation was proposed to express the relationship between the particle breakage factor and stress/strain state. A reduction scheme was suggested to replace the Mc by a new friction factor M. The newly calculated breakage energy proved to increase during shearing and approach to a stable value, which is reasonable and in accordance with the evolution of particle breakage. Then, the remaining unknown parameter dEb/pdεs of the dilatancy equation obtained by consolidated drained triaxial test was found to show a significant linear relationship with M. Therefore, the dilatancy equation was finally determined and verified by a series of test data from different rockfills.

In the rock-like materials, two types of cracks can be observed when the specimens are subjected to compressive loads: tensile cracks and shear cracks. Peridynamics is a novel meshfree method used to solve the fracture problems. To simulate crack propagation of rock specimens under compression, Mohr-Coulomb criterion and maximum principal stress criterion are implemented into the non-ordinary state-based peridynamics, which are used respectively to simulate the shear failure and tensile failure of materials. The extended non-ordinary state-based peridynamics (ENSPD) is efficient for modelling the crack propagation and coalescence under complex loads. The program codes of ENSPD are compiled using Fortran 90. Firstly, the numerical simulation of a three-point bending test is regarded as a benchmark example to prove the accuracy of proposed method. The crack branching phenomenon under dynamic loads is modeled using ENSPD without any additional branching criterion and the present results are in good agreement with the previous experimental and other numerical results. A Brazilian disk of rock-like materials containing an inclined pre-exiting crack is then simulated. The predicted crack growth paths and the estimated fracture toughness well coincide with the previous experimental results. Finally, a rock-like specimen containing two coplanar pre-existing cracks under uniaxial compression is modeled using ENSPD. Compared with the experimental results, it can be found that the proposed numerical method has the ability to simulate and predict the initiation, propagation and coalescence of cracks in rock-like materials.

Biochar is increasingly used in soil improvement, pollutant remediation and other fields for its porosity, high specific surface area and strong adsorption properties. The addition of biochar can alter the pore structure of the soil, and subsequently change the water retention behaviour. Among many influencing factors, the biochar content and particle size have a great influence on the improvement effect. In order to study the effect of biochar content and particle size on water retention behavior of biochar-clay mixture, the soil-water characteristic curve in the wetting process was obtained for biochar-clay mixture with different biochar contents (0%, 5%, 10% and 15%) and particle sizes (>74, 40-74, 20-40 μm and <74 μm) at high suction range (3.29-286.7 MPa) by using the vapour equilibrium technique for controlled suction. The micro-pore structures were analyzed by scanning electron microscopy (SEM) and mercury intrusion porosimetry (MIP) tests. The results show that: 1) When the particle size of biochar is small, the water retention capacity of biochar-clay mixture improved obviously with the increase of biochar content; With the increase of biochar particle size, the water retention capacity of mixture is slightly affected by the amount of biochar. 2) When the content of biochar is small, the soil-water characteristic curves of biochar-clay mixtures with different particle sizes are basically the same, with the increase of biochar content, the influence of biochar with small particle size on the water retention capacity of biochar-clay mixtures has appeared out gradually. 3) The evolution of microscopic pore structure of biochar-clay mixtures further explains the influence mechanism of biochar content and particle size on its water retention capacity.

Crack interaction in materials is an important research area of fracture mechanics that plays a fundamental role in various fields, but few studies have been conducted on the interaction between internal cracks of brittle spherical material. Based on the 3D-internal laser-engraved crack(3D-ILC) technology, two 45° embedded cracks were generated internally in the spherical glass, and uniaxial compression test was performed. The crack propagation, failure morphology and interaction of the cracks were analysed in this experimental research. The distributions of KⅠ, KⅡ, KⅢ were also analysed based on the M-integral, and the results were consistent with experiments. The results show that: 1) The interaction between two cracks occurs in the test. Due to the shielding effect between the double embedded cracks, the crack growth presents obvious anti-symmetric characteristics. 2) The wing crack growth surface presents both smooth area and tear area, corresponding to the I-II crack and III crack respectively. 3) The simulation result is consistent with the experimental observation. The 3D-ILC technology introduced by the paper can provide experimental and theoretical basis for the study of fracture theory of brittle spheres with parallel internal cracks.

The characteristics of dynamic water pressure in transient saturated zone of red sandstone residual soil subgrade is the key to deeply understand the mechanism of repeated mud pumping under the continuous rainfall condition. Based on the undrained dynamic triaxial test, the dynamic water pressure characteristics of the subgrade transient saturated area are studied by simulating the train loading frequency and the drainage boundary conditions under the condition of continuous rainfall. The influence of fine content on dynamic water pressure in transient saturated zone of red sandstone residual soil subgrade is emphatically analyzed, it is found that the critical value of fine content is about 25%. When the fine content is less than 25%, the dynamic water pressure increases with the increase of fine content. When the fine content is greater than 25%, the dynamic water pressure decreases with the increase of fine content. The concept of equivalent intergranular contact void ratio is introduced to clarify the physical significance of the critical value of fine content. An empirical model of dynamic water pressure in transient saturated zone of red sandstone residual soil subgrade is established based on the experiment and the sensitivity of model parameters is analyzed. The model is helpful for engineers and technicians to predict the dynamic water pressure in the transient saturated zone of railway subgrade under continuous rainfall.

Soil has significant spatial variability, and an important indicator for describing the spatial variability of soil is the autocorrelation distance. This paper proposes a method based on soil microstructure simulation to calculate the autocorrelation distance of soil. For generating the numerical model of the microstructure of a soil more reasonable, the traditional quartet structure generation set (QSGS) method is modified, by considering the particle size distribution information of the soil. The input parameters required for the modified quartet structure generation set (MQSGS) method can be obtained by combining the scanning electron microscopy test and the particle size analysis test of soils. Based on the generated numerical model of soil microstructure, the 2-point spatial autocorrelation function can be calculated and the autocorrelation distance of soil can be obtained by curve fitting for the 2-point spatial autocorrelation function. The study shows that soil microstructure generated by the MQSGS method is more similar to the real soil structure in nature than that generated by the QSGS method; the autocorrelation distances calculated from the microstructure numerical model generated by the MQSGS method are slightly smaller than those obtained by the QSGS method.

Pore water chemistry will change the intergranular force and directly affect the microstructure, the physical mechanical properties are further affected for clays. To investigate the effect of pore solution on the compressive properties and microstructure of remolded nature clay, one-dimensional compression test and mercury intrusion porosimetry (MIP) were carried out on the Ningming expansive soil which saturated with the NaCl solutions. The results show that with the increase of osmotic suction, the particles change from uniform dispersion to flocculation structure and forming intra-aggregate pore and inter-aggregate pore. Thus, the salt content specimens exhibit a double pore-size distribution. The initial void ratio of pre-consolidated sample (20 kPa) decreased with the increase of osmotic suction, leading to the increase of consolidation yield stress. However, the osmotic suction had little effect on the compression and swelling index. The permeability decreased with the increase of vertical pressure. When the vertical pressure is lower than 200 kPa, the permeability increased first and then decreased with increasing osmotic suction, but as vertical pressure is greater than 200 kPa, the effect of osmotic suction on permeability was negligible. It is found that the permeability under low pressure is controlled by the coupling effect of microstructure and density. The increase of osmotic suction leads to the increase of macropore and permeability, while permeability is decreased with the increase of density.

During the gestation and development of landslides, due to the uneven effect of external conditions, local stress concentration occurs in the landslide. Once the concentrated shear stress is greater than the peak shear strength of sliding zone, the strain-softening behavior of sliding zone appears. The redundant shear stress will be transferred and be borne by the adjacent sliding zone, then the progressive failure in slope will happen. The progressive failure process of the thrust-type gently inclined shallow landslide is divided into five stages considering the back edge thrust. The progressive failure mechanical models for different stages are established based on the coordination of mechanical balance and deformation. The critical loads and criteria for different progressive failure stages are given by the comprehensive analysis of the progressive failure process of the landslide. The shear strength of sliding zone in different stress states is re-characterized, the stability coefficient of each progressive failure stage of the landslide is given, which makes the physical meaning more clear. Finally, the established mechanical model is applied to a simple shallow landslide, and the quantitative relationship among the back edge thrust, slope displacement, shear stress distribution and stability coefficient is achieved. The model well reflects the space-time characteristics of the mobilized shear strength of sliding zone in the progressive failure process of landslides. By comparing with the calculation results of unbalanced thrust method, results show that it is dangerous to adopt peak strength in engineering design, but uneconomical to adopt residual strength. It is necessary to use progressive failure theory in engineering design of landslide.

Tensile cracking characteristics of gravelly core material are crucial for in-depth knowledge of hydraulic fracturing, longitudinal cracks and abutment transverse cracks of high earth core dams. However, the existing research is still not deep enough. Using the self-developed uniaxial tensile test device, a series of uniaxial tensile tests was carried out on the core materials with different gravel contents. Based on the test results, the mechanism of the cracking failure of the gravelly core material was analyzed. The following conclusions are obtained: for the samples under the maximum dry density and the optimum water content, the tensile strength and tensile strain of the gravelly core material decrease linearly with the increase of gravel content; the tensile stress-strain curves of all samples can be described in a piecewise exponential relationship, i.e. positive and negative exponential function for the test curves before and after the ultimate tensile stress, respectively. A series of consolidated-drained triaxial tests was also conducted, the relationship between the tensile strength and the strength index of each sample was analyzed. The tensile strength has a good linear relationship with its cohesion. This relationship can be used to estimate the tensile strength of the gravelly core material without test conditions. The relevant test results can provide reference for the anti-crack design of the actual earth core rockfill dam.

The microstructure of soil is an important internal factor that determines its engineering properties. However, preparing suitable samples for microstructure analysis and quantifying structural parameters have always been challenging. A method of vacuum freeze drying-glue injection for preparing sandy soil samples is introduced. This method can remove water and maintain the samples’ original structure for further processing. A series of sample images with different grain sizes were taken by SEM and optical microscope, respectively. Based on the self-developed soil microstructure quantitative analysis system SMAS, the images were quantitatively analyzed. A quantitative index system composed of soil particle number, particle perimeter, apparent porosity, shape factor, equivalent diameter, fractal dimension and shape ratio were proposed and used to quantitatively evaluate the microstructure characteristics of sand soil samples. The results show that the vacuum freeze drying-glue injection method for preparing samples has the advantages of convenient operation and reliable results, which can effectively maintain original structured sand. In comparison, SEM is more suitable for the microstructure analysis of fine sand, while optical microscopy is more suitable for medium/coarse sand. SMAS can effectively identify soil particles and pores in microstructure images, providing an efficient tool for quantitative analysis of the microstructure characteristics of sand or other soils.

Aiming at the optimal installation of horizontal drainage channels in hydraulic fill reclamation projects, one dimensional consolidation model with drainage channels installed in the foundation which considering the non-uniform distribution of initial excess pore water pressure was established. The analytical solution was obtained using the finite sine Fourier transform method. After verifying the correctness of the solution, the optimal position of the drainage channels under different time factors was solved and also given in the diagram form. The results show that drainage channels should be installed at the depth of soil layer with great excess pore pressure at the early stage of consolidation, and turn to 2H/3 (PTIB) or H/2 of soil (PTPB) at the later stage of consolidation (H is the depth of foundation). In order to save the consolidation time when the degree of consolidation reaches 90%, the optimal installation depth of horizontal drainage channels was 0.52H, 0.72H, and 0.62H for the single-drainage condition with the distribution of initial excess pore pressure in the inverted triangle, triangle and trapezoidal pattern, respectively. For the double-drainage condition, the optimal installation depth of horizontal drainage channels was 0.42H, 0.58H, and 0.46H, respectively. A practical example shows that the foundation with optimal depth of horizontal drainage channels has higher consolidation degree (6% and 54%) than that with drainage channels installed in 0.5H and without drainage channels installed.

The researches on sinking characteristics of jacked pile in clayey soil are relatively few. In this study, variations of pile sinking resistance, axial force and unit skin friction during jacked pile sinking were studied based on fiber Bragg grating sensing technology, considering the different pile end forms (such as open-ended and close-ended pile) and different pile diameters. Jacked pile sinking model tests were performed to determine the influences of pile ending form, pile diameter and pile sinking depth on pile sinking resistance, axial force and unit skin friction. The results shows that the requirement of FBG technology shows superior performance in testing sinking resistance of jacked piles in clayey soil, which can accurately reflect the penetration characteristics of jacked piles. The end resistance of open-ended and close-ended piles accounts for 66.7% and 59.5% of the piling resistance, respectively when sinking pile ends. The effects of open soil plug effect and pile diameter on pile sinking resistance can not be neglected. The distribution of the axial force of the pile shaft is not changed due to the pile end forms. The axial force transmission performance of pile with a diameter of 140 mm is better than that with 100 mm pile diameter. The decline rate of the axial force of the two piles is 40.8% and 34.2%, respectively. The maximum unit skin friction of the inner and outer tube piles of the open-ended pile is 1.67 kPa and 4.83 kPa, respectively, which are smaller than that of the close-ended pile. When the sinking depth is 90 cm, the maximum unit skin friction of the pile increases with increasing the pile diameter, and the larger pile diameter at the same depth results in more obvious degradation of the unit skin friction of the pile.

The present theoretical analyses about high fill culverts structure stress state mainly focus on the earth pressure acting on the top of culverts, but do not consider the effect of the skin friction between the culvert side walls and the adjacent backfill. In this paper, a mechanical model of culvert-soil under load reduction condition was proposed and formulas for calculating the side wall friction and the foundation contact pressure were also deduced. The theoretical method was compared and calibrated with the numerical simulation. Results show that the soil-side wall friction linearly increases with the depth of the culvert side wall and the increase of fill height. The soil-side wall friction under the load reduction condition is greater than that under the non-load reduction condition at a certain fill height. Although the vertical earth pressure on the top of the culvert is reduced due to the load reduction measure, the horizontal earth pressure on the side of the culvert is increased, as a result, the base pressure is not reduced. It is unreasonable to analyze the culvert performance under load reduction condition without considering the effect of the friction acting on the culvert side wall in present theories.

Considering the deficiency of the nonlinear consolidation calculations of vertical drains foundations solved by approximate equivalent method, and based on the assumption of plastic drainage plate elliptical column, the horizontal permeability coefficient and compression of foundation soil are deduced using the soft soil consolidation theory. An analytical solution for the nonlinear consolidation is given by the nonlinear differential equation theory, which considered the drainage plate under the influence of compressive modulus with nonlinear variation of the consolidation process. By comparing with the results in Indraratna (2005), it is considered that the larger the ideal vertical drains foundations preload load (the larger Nu) is, the greater the error obtained by the approximation method in Indraratna et al. (2005), and this error can be up to 16% compared with the results in this paper when the Cc/Ck=1.8 and Nu=5. The results reveal that the analytical solution proposed in this paper is aligned with the exact numerical solution, and the theory suggested is more effective in engineering application.

To improve the stability of sandy slope and achieve ecological slope protection, a kind of eco-ester-polymer named as Aqua-Dispersing-Nano-Binder (ADNB) was used to reinforce sandy soils. Properties of molecular weight and thermal stability of ADNB were tested through gel permeation chromatography (GPC) method and thermogravimetric (TG) method. The effects and the regularity of strength improvement of sandy soils were studied by designing and carrying out tests of unconfined compressive strength (UCS) and direct shear under the condition of different ADNB dosages. Finally, the disintegrating test and the slope scour model test have been performed to investigate the improvement effect and regularity of both water stability of sandy soils and the anti-erosion ability of slope surface, and the improvement mechanism of sandy soils and ecological slope protection was further analyzed. Results show that ADNB can improve the strength and water stability of sandy soils, and the erosion resistance of sandy slope surface is improved. With the increase of ADNB dosage, the UCS and cohesive force of sandy soils are increased, in which the increase of cohesive force is 46%-77% while internal friction angle has no significant change; the peak disintegration rate and the disintegration coefficient of sandy soils decreases, besides, the occurrence time of the maximum disintegration rate is more and more delayed; the erosion rate of slope surface decreases about 68%-86%. The application effect of slope protection engineering shows that ADNB performed well on stabilizing soils with a long time and promoting the vegetation growth. As a result, the long-term ecological protection of slope is achieved.

The accuracy of the structural seismic responses largely depends on the calculation of the Rayleigh damping coefficients in the dynamic time and history analysis. By analyzing the seismic responses of a stratified site under different earthquake actions, the results in time domain and frequency domain based on commonly used damping coefficients are compared with each other. A new method for calculating Rayleigh damping coefficients for time domain analysis with hysteretic damping ratio is proposed. The influence of different calculation methods of Rayleigh damping coefficient on the seismic responses of a single-story two-span underground subway station is discussed based on the proposed method. The results show that the average error of the seismic responses of the site is minimum between the new method and the frequency domain method. The results of the modified full Rayleigh damping agree well with the new method for the seismic responses of the underground structure. The calculation method for Rayleigh damping coefficients proposed in this paper has the advantage of high precision and simple operation, can be widely used in seismic response analyses of site and underground structures.

The anisotropic properties of soft clays are closely related to the sedimentary environment and stress history. Based on a theoretical formula of anisotropic undrained shear strength which take soft clay characteristics into account, this paper analyzes the soft clay stability against basal heave with slip circle method with consideration of strength reduction on the failure surface. The considered strength reduction formula can be expressed as , which means shear strength on the failure surface equals to the undrained shear strength times the cosine value of effective internal friction angle. Stability against basal heave of excavation is studied. Based on a number of examples, the influences of geometrical factors such as the excavation depth H, embedded depth D, and relative distance of the lowest support to the pit bottom He/H are analyzed. Strength factors such as the effective internal friction angle , anisotropic strength ratio k and slope of heterogeneous shear strength versus depth are also investigated. Finally, the applicability of the formula is verified by the engineering example. This paper provides a new idea for the stability analysis against excavation basal heave.

Energy pile is an energy-saving and emission-reduction technology, combining the functions of structure load supporter and shallow geothermal heat exchanger. There has been application in the world in recent years. Normally, the comprehensive thermal conductivity coefficient of soil is measured through ground source heat pump technique. However, this value is unsuitable to calculate the heat exchange efficiency of energy pile accurately. Field tests and numerical simulations on the soil thermal conductivity coefficient were obtained in energy pile group with low cap located on Henan Polytechnic University. The influence of heating time, heating power, flow rate, pile length and the arrangement of energy pile in groups on the soil comprehensive thermal conductivity coefficient are analyzed. It shows that the line heat source analysis method, based on the ground source heat pump testing, is unsuitable for the calculation of soil comprehensive thermal conductivity coefficient . Hence, it is necessary to develop one testing and analyzing method which can considering pile diameter, etc.

The generalized Hoek-Brown (HB) criterion has been widely applied to the computation and analysis of rock mass. However, the generalized HB criterion is a typical nonlinear failure criterion, so that it can’t be directly used in conjunction with shear strength reduction finite element method (SSRFEM) for the stability analysis of rock slopes. In response to this problem, the instantaneous linearization technique is investigated, in which for an arbitrary stress point, the instantaneous cohesion and internal friction angle corresponding to a tangent line with respect to the HB envelope can be obtained. Consequently, two instantaneous linearization strength reduction schemes are proposed: 1) Based on the stress field attained from the elasto-plastic analysis, the Mohr-Coulomb (MC) strength parameters of each stress point can be reduced, and the resulting approach is denoted as Point-IL-SSRFEM; 2) Since the strength parameters are required to be constant in each element for conventional finite element method, the attained instantaneous strength parameters for stress points in each element shall be averaged and then reduced, and the resulting approach is denoted as Element-IL-SSRFEM. Based on the calculation example of rock slope, the three-parameter strength reduction technique and the instantaneous linearization strength reduction technique are compared. The numerical results show that: compared to Element-IL-SSRFEM, the numerical accuracy and computational efficiency of Point-IL-SSRFEM are relatively high; Compared to the three-parameter strength reduction technique, the numerical accuracy of Point-IL-SSRFEM is higher and the computational performance of Point-IL-SSRFEM is more stable. Therefore, it is recommended to apply it to the stability analysis of rock slopes.

In regularly layered elastic media where the shear velocity of layer increases with increasing layer depth, the surface wave-fields are dominated by Rayleigh waves of the fundamental mode. Thus, it is very important for calculation of the dispersion curve of the fundamental mode in the surface wave tests. The matrix methods are often used to calculate the dispersion curves of Rayleigh waves in the layered elastic media. The dispersion curves could be obtained from the determinant of the matrix. However, the determinant must be solved using root searching techniques. To avoid the complex algorithm, based on the expression for calculation of the fundamental mode dispersion(Aki & Richards), the phase velocity is assumed to be the root-mean square of the shear wave velocity and/or Rayleigh wave velocity of layer weighted by the integral of the displacement shape functions of the fundamental mode. It can be known that the displacement shape functions of the fundamental mode in regularly layered media are highly correlated to those in the homogenous half space with the properties equal to the first layer. The displacement shape functions in the regularly layered media can be evaluated by calibrating the displacement shape functions in the homogenous half space with the shear wave velocity contrasts between layers. It is shown from the results that the algorithm of the proposed approach is simple compared to the matrix methods, and the accuracy is higher than that of the empirical half wavelength method.

Based on the slope system energy principle and FISH language embedded in FLAC3D finite difference method, secondary development of energy calculation in three-dimensional open-pit mine slope excavation model is carried out. Combining energy with catastrophe theory and taking the total dissipative energy of slope system as state variable, the energy dissipation criterion for slope system catastrophe is established. The analysis of sudden change in energy dissipation, plastic zone evolution, safety factor changes and key point displacement is done using the northwest slope of Dagushan open-pit mine as an example. It can be seen that the slope stability evolution obtained by the above analysis is consistent and the applicability of the proposed catastrophe criterion is verified. Results show that energy mutation eigenvalue is a sufficient and necessary mechanical property of slope instability after the ki step excavation. When the excavation is carried out to the 6-8th step, the local rock mass system of the slope is unstable, and the three steps of the 7th, 8thand 9th stage are seriously affected by the excavation disturbance, which is consistent with the actual excavation situation in the field. It provides some theoretical support for preventing the instability of local surrounding rock system during excavation disturbance of open-pit slope.

Ultimate bearing capacity analysis of the strip footing is a classical topic in soil mechanics. Based on the Hellinger-Reissner mixed variational principle and finite element method, a geotechnical elasto-plastic problem involving the Mohr-Coulomb model can be casted into a second-order cone programming (SOCP) problem in the finite element framework. A second-order cone programming based incremental finite element method named FEM-SOCP is then proposed. The proposed approach can avoid the complex stress integration algorithms and the smoothing treatment of the yield surface edges and corners, which are often necessary in the traditional elasto-plastic calculations. For the casted SOCP problem, the commercial solver MOSEK with the primal-dual interior point method can be used. The FEM-SOCP method is applied to the ultimate bearing capacity analysis of a shallow strip footing, with the associated and non-associated Mohr-Coulomb yield criteria considered, respectively. The numerical results show that the bearing capacity coefficients and bearing capacity obtained by the incremental loading FEM-SOCP method is consistent with the traditional FEM solution, but the yield zone attained is generally smoother compared to that obtained from conventional FEM.

The long-runout distance of rock avalanches is the largest accumulation range that sediment can reach, which has great significance in disaster warning and risk assessment. This paper describes a calculation model to predict the long-runout distance, which is of interest in connection with the motion of landslide and mass movement. By summarizing the research results of long-runout distance of rock avalanches and basing on dimensional analysis, it is considered that the potential energy of rock mass plays an important role on long-runout, and that the long-runout distance (L) is the function of potential energy. Then, experiments were performed with 4 different quartz granular materials, and masses of these granular materials with various initial volumes were released from rest in a chute. The front position of avalanches was recorded. The effect of mass volume(V), the slope angle of transport zone(?), fragments particle size(d) and maximum vertical travel distance(H) on the long-runout distance(L) were studied. The calculation model for long-runout distance on potential energy was established by stepwise regression method. The data of 38 landslide-avalanches triggered by the Wenchuan earthquake and other 17 typical landslide rock avalanches that are widely used internationally were taken to verify the calculation model. The results show that the transport distance calculation model established by potential energy of rock mass is reliable and that the transport distance calculation model is suitable for rock avalanche disaster warning and risk assessment in mountainous areas.

Pipelines are widely used in the petroleum industry for oil and gas transportation, and the interaction between pipelines and seabed is crucial for assessing pipeline behavior during lateral thermal buckling. This paper presents a numerical investigation on the pipe-soil interaction behaviour using a recently developed sequential limit analysis approach for deep-water pipelines on soft clay. This approach, primarily based on upper-bound limit analysis, is capable of handling the extremely large deformation problems and implementing the strain softening and strain rate effects. During the analysis, the pipe was laterally driven forward with the elevation of its invert fixed, differing from existing work in which the pipe can dive or rebound during the lateral displacement. The evolution of vertical (V) and horizontal (H) soil resistance was examined, and the validity of the present model is demonstrated using comparisons with centrifuge data. V-H yield envelopes were derived during the analysis to investigate the influence of load path on pipe response. The equivalent friction ratio, H/V, was found to increase linearly with lateral displacement. This study aims to provide guidance for the controlled buckling design for deep-water pipelines, and show the potential of SLA in solving large displacement soil-structure problems in undrained clay.

Abaqus? is one of the most popular commercial finite element analysis(FEA) software, but the visualization function of post-processing module—Abaqus/CAE is relatively weak. Especially, it handles mass data with difficulties, such as the extremely slow refreshing, limited functionalities and unexpected crash of software. Therefore, this paper proposes a solution: develop a Python script—ODB2VTK+ to convert Abaqus output (ODB file) into VTK file format, then visualize the output by menas of ParaView. ODB2VTK+ provides space-based partition algorithm which can achieve good partition and minimizes data redundancy to reorganize the mass data into small pieces suitable for parallel visualization. The parallel programming on it achieves higher efficiency, and the configuration file is provided to facilitate parameter setting. Testing ODB2VTK+ on the mechanical FEA output of 3D rock models from microtomography which have complex pore structure with fine grids shows the conversion is completely correct. Then it is visualized perfectly with ParaView’s capability of parallel visualization and its functionalities of volume rendering, tensor glyph, etc. The immersive virtual reality(VR) with HMD is applied, it encompasses the user with computer generated 3D environment that allow user to walk through the internal pore structure for observing and analyzing directly the result of FEA. The proposed solution overcomes the limit of Abaqus/CAE in post-processing and shows that VR will be a popular tool in FEA post-processing. In addition, due to its good readability and extensibility, ODB2VTK+ can be easily modified to process more different problems in Abaqus. ODB2VTK+ is shared on GitHub (https://github.com/ Liu-Qingbin/ODB2VTKplus) for the convenience of Abaqus users.

Safety of tunnel construction is a hot topic for the tunnel passing through complex loose deposits. Based on Monte Carlo stochastic principle and digital image processing technology, a mesostructure model for the excavation of Luodala tunnel is established with considering contact surface element and tensile strength. The deformation, failure process and instability mechanism of loose deposits surrounding rock during excavation are discussed and verified by application of the model in practical engineering. The results show that the model can reflect the failure process of surrounding rock during tunnel excavation, and the displacement contour of surrounding rock presents obvious fluctuation and asymmetry. The failure mode of surrounding rock is mainly shear failure, and tension-shear composite failure is also observed in some part. The failure zone extends from tip of the margin rocks to the deep layers, forming wedge-shaped shear zone and tension-shear loose zone, which are prone to cause local instability under construction disturbance. According to the failure characteristics of loose deposits, the grouting reinforcement measures of surrounding rock are proposed. After the reinforcement, the grouting reinforcement ring has been formed around the tunnel with good soil-rock bond and the deformation of surrounding rock can be controlled and the support structure is stable during the excavation and support process. The resutls can provide new ideas for the design and construction of tunnels in similar loose deposits.

The spatial variability of shear strength parameters of rock and soil mass on slopes is structural. Aimed at studying the effect of spatial variability structure of rock and soil parameters on failure probability of slopes, the transformation relationship between range and correlation distance is deduced based on the mathematical intension of variogram firstly. The simulation method of structural cross-constraint random field is proposed to simulate the cross-correlated parameters random fields. In order to study the effect of shear strength parameters with different spatial variation structures, a slope calculation model of structural cross-constraint random field is established. The results show that the structural cross-constraint random field can reflect real formation parameters by considering random deviation, conditional data and different spatial variation structures. Compared with conditional parameter interpolation field, the proposed method is suitable for parameters with more complex spatial variation structures, i.e. directionality and anisotropy, and the generated data have less fluctuation. The reliability analysis also reveals that the failure probability of slopes is prone to be overestimated without considering the spatial structure parameters. The failure probability increases with the increase of the correlation coefficient when the correlation between the two factors is considered. It is easier to overestimate the slope failure probability under the scenario of negative correlation between c' and φ'.

As the buried depth of TBM-excavated tunnels increases, how to reduce the risk of TBM shield jamming becomes an urgent problem. Based on the Newton iteration and the finite element method, an inverse method for the identification of the interaction loads between shield surface and surrounding rock mass is proposed. The error between measured and calculated strain values of the observation points on the internal shield surface is set as the objective function, and the Moore-Penrose inverse is introduced to solve the inverse equation. Meanwhile, a finite partition strategy for the external shield surface is adopted, then the nodal loads of partitioned areas are defined as the inverse parameters, which is convenient for the control of the number of the inverse parameters. The loads within each partitioned region are obtained by the interpolation of the nodal loads, and the fitting of the arbitrary distributed loads acting on the external shield surface is realized. The results of numerical examples show that: the proposed inverse method is featured with high computational efficiency, and good immunity to observation error; the load distribution law and the magnitude of friction on the external shield surface can be identified effectively by performing the inverse analysis. To a certain extent, it can provide guidance for safe tunneling, jamming warning and accident treatment. Furthermore, the proposed method can also be extended to the low-dimensional cases and provide reference for the inverse analysis of other structure loads or physical parameters.

In order to study the influence of temperature on the mechanical properties of pile-soil interface, a setup of temperature-controlled pile-soil interface triaxial shear test system was developed. The new instrument improves the internal construction of the pressure chamber based on the conventional triaxial shearing instrument. A special structure named pile-soil interface was built in the soil sample and two independent temperature control systems were established for pile and soil, respectively. As a result, the mechanical characteristics of the pile-soil interface under non-isothermal conditions can be tested. A series of triaxial tests were carried out on the thermal-mechanical coupling characteristics of pile-soil interface of sand under different shear conditions and temperature levels. The temperature control accuracy of pile and soil, the shear stress of pile-soil interface, the pore water pressure variation under undrained conditions, and the volume change under drainage conditions were measured. The above results preliminarily verify the accuracy and reliability of the new temperature-controlled pile-soil interface shear test system. The new system has a simple structure and is easy to disassemble. It is also suitable for both temperature-controlled pile-soil interface shear test and conventional (temperature-controlled) triaxial shear test of soil samples. The relevant research results can provide technical support for understanding the thermal-mechanical coupling mechanism and constitutive modelling of the pile-soil interface.