In order to explore the pore structure characteristics of phase change energy storage backfill and their influence on the strength deterioration of backfill, a composite phase change material was prepared with butyl stearate as the phase change material and expanded perlite as the adsorption medium. Cement and tailings were mixed to prepare backfills with different additive amounts of the composite phase change material. The strength and structure characteristics of the phase change energy storage backfill with different addition amounts were obtained by using the methods of CT (computer tomography) scanning, MRI (magnetic resonance imaging) analysis, and uniaxial compression test, and the influence mechanism was analyzed. The results show that: i) The porosity of phase change energy storage backfill increases gradually with the increase of the addition amount. The macropore porosity increases approximately linearly, the proportion of macorepores increases gradually, and the pores approximate to sphere. ii) With the increase of the amount of phase change material, the connectivity of the backfill increases, the pore throat length increases, and the number of macropores increases. The pore throat coordination number is concentrated below 5, and the fractal dimension decreases first and then increases significantly, resulting in complex pore distribution. iii) With 5% additive amount, the uniaxial compressive strength of backfill decreases by 30.2% due to the increase of macropores and pore connectivity. With 10% additive amount, the pore size distribution becomes uniform and the uniaxial compressive strength decreases by 48.9%.

Series of unconfined compression and indirect tensile tests were performed to study the effects of curing age and delayed compaction on strength and stiffness of magnesium oxychloride cement (MOC) solidified sludge, analyzing the strain rate effect of solidified sludge under constant multi-strain rate loading and alternative strain rate loading. The obtained results indicate that the tensile strength, compressive strength, elastic modulus, failure strain, and total strain energy of sludge increase with an increased strain rate. The compressive strength, tensile strength, and total strain energy of MOC solidified sludge has an upward trend with strain rate, while the elastic modulus and failure strain have no apparent change. The faster the strain rate is, the more distorted and rugged the fracture surface is. With the prolongation of curing age, the compressive strength and total compressive strain energy are significantly increased, while the tensile strength and total tensile strain energy are reduced slightly. Significantly, different calculation methods adopted would result in different strain rate sensitivity coefficients for curing age. The delayed compaction would reduce the tensile strength, compressive strength, elastic modulus, total strain energy, and strain rate sensitivity coefficient of solidified sludge. The stress jump induced by a sudden change of strain rate is enlarged owing to the increase of applied stress, and it first increases and then levels off as the strain is extended. The strain rate sensitivity coefficient shows that solidified sludge is less sensitive than pure sludge, and the tensile strength is more sensitive than the compressive strength. The obtained results can provide an essential theoretical reference for the strain rate effect analysis of solidified sludge.

In order to investigate the failure characteristics and damage law of sandstone under the combined dynamic and static loading conditions, three-dimensional movement combination loading test system with separate Hopkinson compressive bar is applied to perform impact tests on sandstone samples under the condition of multiple combinations of dynamic and static state and different loading rates. CT scan and digital core technology are also used to observe the failure diagrams of different sections in the sandstone sample, and the three-dimensional reconstruction diagram and crack density in the damaged sample are obtained. The failure forms and failure mechanism of sandstone under different stress conditions are studied, and the effects of axial pressure, confining pressure and strain rate on the crack density of sandstone are explored. The test results show that the dynamic failure of sandstone is typical tensile splitting failure under the action of conventional dynamic impact. Under one-dimensional combined dynamic and static loading, the dynamic failure mode of sandstone is typical compression and shear failure, and the interior presents a conjugated double curved compression and shear surface. Under three-dimensional dynamic and static loading, the dynamic failure mode of sandstone is also compression and shear failure, but the internal failure surface is circular (cone) shape. The dynamic failure mechanism of sandstone under different loading conditions is analyzed. Under different loading conditions, the crack density of sandstone increases with the increase of strain rate. The application axial compression and confining pressure restrict the crack generation and growth rate, and the confining pressure has a greater limitation on the crack generation than the axial pressure. It is explained from the perspective of compressive strength and compensation space for crack generation of sandstone. The influence of strain rate on the crack density of sandstone at 0, 200, 400 m and 600 m underground is analyzed quantitatively by using axial compression and confining pressure to simulate the in-situ stress. To produce the same crack density, the strain rate of sandstone at 600 m underground is about 3.4 times that of sandstone without in-situ stress. The quantitative relationship between damage variables and strain rates under different loading conditions is established from the perspective of crack density. The research results can provide a reference for quantification of different explosion stress waves and the development degree of internal crack in rock during blasting mining without facing surface.

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 order to reveal the fracture development and damage evolution law of marble under the influence of temperature, the deformation and failure characteristics and acoustic emission activity of marble under the conditions of 30, 60, 90, 120 ℃ and 150 ℃ were studied by using triaxial servo testing machine and acoustic emission testing system. The results show that the triaxial compression failure of marble under high temperature goes through four stages: compaction and elastic deformation stage, plastic deformation stage, ductile failure stage and instability failure stage. The higher the test temperature is, the longer the compaction deformation stage and ductile failure stage are, and the rock gradually transforms from elastic brittle failure to elastic-plastic failure. The peak strength, elastic modulus and internal friction angle of marble decrease with the increase of temperature. The acoustic emission activity of rock and the correspondence between AE location and macroscopic crack decrease with the increase of test temperature. By analyzing the characteristics of acoustic emission signal amplitude, impact, ringing count, acoustic emission energy, peak frequency and b value in the damage evolution process of marble, it is found that the amplitude fluctuation of acoustic emission is small and the temperature sensitivity is poor, but the peak frequency and b value parameters can better characterize the crack development inside the rock. The peak frequency has the highest temperature sensitivity, which is favorable as the prediction index of rock failure and instability. In the process of triaxial compression failure, tension-shear composite cracks appear in marble, of which shear cracks are dominant. High temperature inhibits the initiation of internal cracks in rock mass. The higher the test temperature is, the weaker the tensile failure and the stronger the shear failure are. The test results shed light on the study of rock thermal damage mechanism and engineering thermal damage monitoring and prediction.

The creep acoustic emission (AE) location tests of red sandstone were carried out, and the evolution of AE source along loading direction and perpendicular to loading direction was analyzed. On this basis, the fractal characteristics of source zi value and ri value were studied with fixed time window and time step. The results show that when the creep stress is equal to or greater than the damage stress, the spatio-temporal evolution law of AE source reflects the process of micro-cracks initiation, nucleation, expansion and coalescence. The initiation of micro-cracks mainly occurs in the stage of decelerating creep. During this process, the micro-cracks evolve from the end surfaces and the outer wall to the center of the specimen along the loading direction and the vertical loading direction, and the fractal dimension of zi values and ri values generally decrease. The nucleation process of the micro-cracks appears in the stage of steady creep, and nucleation area locates in the center of the specimen. Then, with the increase of the creep time, the micro-cracks step into the expansion stage, in which the evolution of the micro-cracks is manifested as the expansion process from the center of the specimen to the direction of the end surface and the outer wall, and the fractal dimensions of the corresponding zi values and ri values are at low levels, which can be used as a marker for the process of micro-cracks nucleation and expansion. The coalescence of micro-cracks occurs in the accelerated creep stage. During this process, the micro-cracks are mainly distributed in the axial upper and lower part of the specimen along the loading direction, while in the vertical loading direction, the micro-cracks are distributed all over the radius of the specimen. However, the fractal dimension of source zi value and ri value both increase slightly with the increase of creep time, which can be used as a symbol of micro-cracks coalescing process. The findings in study can provide theoretical and experimental support for exploring rock creep micro-mechanism and creep failure prediction method.

In order to investigate the influence mechanism of reservoir pressure and gas desorption on permeability of coal measure shale gas reservoir, effective stress-permeability models of matrix and fracture for reservoir in non-equilibrium desorption state were proposed on the basis of the kinetic diffusion of gas in the matrix. Subsequently, the permeability models of reservoir in non-equilibrium desorption state and uniaxial strain condition were set up. The effectiveness of permeability models was discussed by analyzing results of the field and laboratory experiments on samples as well as permeability of reservoir. The permeability model in non-equilibrium desorption state can provide a better fit to the experimental values of permeability. A smaller pore volumetric modulus leads to a more rapid rebound in the matrix permeability; and a greater fracture compressibility yields a more drastic reduction in the fracture permeability. Afterwards, the influence mechanism of various parameters on reservoir permeability was studied. The results indicate that the matrix permeability first decreases and then rebounds, fracture permeability increases gradually with the decrease of matrix pressure when the fracture pressure decreases to a certain value. Matrix permeability increases and fracture permeability decreases gradually with the decrease of fracture pressure when the matrix pressure drops to a certain value.

Electroosmotic reinforcement of low-permeability soft clay is one effective soft soil foundation reinforcement technique. However, excessive power consumption and uneven reinforcement effect limit its application. Chemical electroosmosis is an advanced technique which can improve the uniformity of the treated soft ground and reduce the electroosmosis time and power consumption by injecting sodium silicate (Na2SiO3) and calcium chloride (CaCl2) solutions in the late stage of electroosmosis. Model tests on chemical electroosmosis of soft ground were carried out. The overall water discharge and drainage rate during electroosmosis and chemical electroosmosis processes were measured. The energy consumption factors (such as, soil sample resistance, current, etc.) and reinforcement effects (such as, water content, soil sample strength, etc.) influenced by injection positions were analyzed in detail. Combined with scanning electron microscope (SEM) and inductively coupled plasma-mass spectrometry (ICP-MS), the microscopic mechanisms of reinforced soil sample were discussed. The results show that the relative optimal way of chemical electroosmosis is that CaCl2 solution and Na2SiO3 solution are injected simultaneously into the anode and middle of the soil sample. Compared with traditional electroosmosis method, the water discharge of chemical electroosmosis increases 25.5%, and the shear strength value increases 168.8% under the experimental conditions. The disadvantages of traditional electroosmosis such as uneven water content of the anode and cathode soil after drainage can be improved.

A series of tensile tests was conducted on gravelly soil based on a new self-developed soil tensile device. The different gravel contents and fiber contents were considered in the test. It is found that the tensile strength of gravelly soil decreases with the increase of gravel content, and the tensile strength and ultimate tensile strain of gravelly soil increases significantly after polypropylene fiber is added. The tensile strength and ultimate tensile strain of fiber-reinforced gravelly soil are positively correlated with the fiber content. However, as the gravel content in the gravelly soil increases, the effect of fiber incorporation on its tensile strength is significantly reduced. Scanning electron microscope analysis shows that the friction of interface between the fiber and the soil particle is the main reason for the increase of the tensile strength of the fiber-reinforced gravelly soil. For the pure clay specimen with a gravel content of 0%, the tensile strength of the soil is dramatically improved because there are only type-I fibers at the fiber/soil particle interface. As the gravel content increases, the proportion of type-II fibers at the fiber/soil particle/gravel interface increases, and the effect of fiber incorporation on the improvement of its tensile strength is markedly reduced. Finally, based on the test results of 60 specimens, a multivariate regression model for the tensile strength of fiber-reinforced gravelly soil is proposed, which can quickly predict the tensile strength of gravelly soil with different gravel contents and fiber contents. The test results can provide references for the anti-cracking design of core wall of high earth core wall dams.

The performance evolution of subgrade soil under the coupling effect of wetting-drying cycle and dynamic load is very important for the safety and stability of the subgrade. For this purpose, a series of electrical resistivity tests was carried out on the compacted silty clay samples subjected to different wetting-drying cycles and dynamic loads by using AC two-electrode method. Firstly, the effect of wetting-drying cycles sequentially coupled with dynamic loads on the electrical resistivity of samples was investigated. Then, a mathematical relationship was established to quantitatively characterize the dynamic behaviors of samples based on electrical resistivity. Finally, the effectiveness of the electrical resistivity method in evaluating the state of compacted soil was discussed. Results showed that under the action of wetting-drying cycles coupled with dynamic loads, the electrical resistivity of the sample decreased greatly with the increase of the number of wetting-drying cycles or the amplitude of wetting-drying cycle, but the reduction of the electrical resistivity of samples only caused by dynamic load decreased gradually. The test results can provide some reference for the evaluation of the state of subgrade soil under wetting-drying cycles sequentially coupled with dynamic loads by using electrical resistivity method.

Rainfall has induced many landslides of soft rock in Northwest China. To reduce the loss caused by this kind of disaster, based on the old landslide of Caijiapo town, two groups of centrifuge model tests with different hydraulic actions were conducted through the combination of geological analysis and simulated rainfall centrifuge model test, and the influence of hydraulic action on the initiation mechanism of the slope was discussed. The results show that: i) The soft rock slope is destructed by hydraulic action in a multi-staged and progressive way, and the deformation process includes the initial deformation stage, the accelerated deformation stage, and the flowing failure stage. ii) The pore water pressure obviously accumulates due to the seepage and diffusion of water towards the structural plane and the surface. Water mainly accumulates in the tensile cracks of steepened micro-geomorphologic area and areas with severe compressional deformation at the foot of slope. iii) In a short time, the increment of pore water pressure increases with time, causing the sliding mass fluidized. Therefore, in the process of disaster risk prevention and control, it is necessary to strengthen the monitoring of pore water pressure growth rate in fractured area.

To investigate the evolution of the pore structure and the relationship between the pore structure and the macroscopic mechanical properties of the expansive soil in the seasonally frozen area, nuclear magnetic resonance (NMR) tests were carried out on the Jiamusi undisturbed expansive soil. Besides, the scanning electron microscope (SEM) test was also adopted to distinguish the pore structure of the soil. Meanwhile, the influence of the consolidation pressures and the freeze-thaw cycles on the pore structure of the expansive soil were studied, and the relationships between the evolution of the pore structure and mechanical properties of the expansive soil were further explored. The results show that: i) Cracks developed in the Jiamusi expansive soil induce a trimodal characteristic of the T2 time distribution curve. The distribution of the pore size of the undisturbed sample behaves significantly differently with the varied consolidation pressure, including the three stages of the pressure is less than, slightly greater than and much greater than the pre-consolidation pressure. As the consolidation pressure is increased, the adjustment rate of the pore structure decreases. The void ratio of the soil decreases with the increased freeze-thaw cycles. The increased freeze-thaw cycles will lead to the increased proportion of mesopore and the decreased proportion of the macropore, while the micropores are not sensitive to the variation of the freeze thaw cycles. ii) The stress-strain curves vary from strain-softening to strain-stable with the increased freeze-thaw cycles and the failure modes of the expansive soil change from brittle failure mode to plastic failure mode. The unconfined compressive strength decreases exponentially with the increased freeze-thaw cycles. The shrinkage characteristic of the expansive soil is enhanced significantly under the freeze-thaw cycles. iii) The variation of the pore structure exhibit the linear trend with the mechanical characteristic. The effects of freeze-thaw cycles on the pore structure of the expansive sole are consistent with that on the macroscopic behaviors. The results obtained in this study can be used to establish a quantitative relationship between pore structure and engineering performance of the expansive soil.

The decomposition of natural gas hydrate generates excess pore pressure, which will lead to the deterioration of the engineering properties of deep-sea energy soil and cause geological disasters such as submarine landslides. Considering the effect of deep-sea energy soil permeability, a modified Grozic-Nixon excess pore pressure model is established. The Storegga submarine landslide, as the research case, is compared with the modified Grozic-Nixon model and Xu-Leonid model, and the research results are analyzed. Based on the modified model, further research is carried out to reveal the time-varying law of excess pore pressure and the effect of different parameters on the excess pore pressure. The results show that the maximum value of excess pore pressure calculated by the modified model reduces by about 21.5% compared with the original model. The excess pore pressure first accumulates and then dissipates, and the accumulation rate gradually decreases, while the dissipation rate first increases and then decreases. The effects of hydrate decomposition rate and deep-sea energy soil permeability on the excess pore pressure have the most significant correlation to excess pore pressure, and the maximum excess pore pressure at selected conditions can differ by 6 times at most.

Soils usually experience multiple drying-wetting cycles in nature, which can have a significant impact on the engineering properties of soils. Laboratory tests were conducted to investigate the effect of drying-wetting cycles on the initiation and evolution of cracks in clay layers. A clay specimen was prepared and was subsequently subjected to five drying-wetting cycles. The evolution of surface cracks during the drying-wetting cycles were monitored. The effect of drying-wetting cycles on the geometric characteristics of crack patterns was analyzed by the digital image processing software, and four geometric parameters, including the surface crack ratio, the total crack length, the average crack width and the intersection angle, were selected for the quantitatively analysis. The results show that: i) The desiccation and cracking behavior can be significantly affected by the applied drying-wetting cycles. The cracks develop sequentially during the first drying process and the crack pattern exhibits typical hierarchical characteristics, while this phenomenon disappears in the subsequent drying-wetting cycles. ii) The edge of the cracks can break down and heal into "shallow ridges" during the wetting path. After the drying-wetting cycle most cracks occur at the original position, but the direction of cracking is different during different drying processes. iii) The drying-wetting cycles may cause the edges of the cracks to become rough and induce more tiny cracks in the soil specimen. iv) Drying-wetting cycles can advance the occurrence of the cracks in the specimen, increase the total length of cracks, decrease the average width of the cracks, but induce limited changes in the crack ratio. v) With the increase of drying-wetting cycles, the pattern of the intersections of the crack network gradually change from the T-junction to the Y-junction, while the intersection angle between cracks gradually changes from 90° to 120°.

Fracture mechanics is an important branch of solid mechanics. The three-point bending test is a classic test in the field of mechanics. Most of previous studies on three-point bending tests with cracks focus on either through-wall or surface cracks, while limited studies have been performed on internal cracks. Based on the 3D-internal laser-engraved crack (3D-ILC) method, without any damage on the surface, the internal crack was created. Experiments were performed on specimens with horizontal internal cracks at different depths and intact specimens. The stress-induced birefringence, failure load, fracture characteristics, fractography size and fracture morphology were analyzed. The distribution of KI, KII, KIII around crack front was obtained by numerical simulation. The results show that: i) The internal crack changes the stress birefringence law of the intact sample, the moire is semi-circular at the upper end of the crack, and the color difference is significant. ii) The existence of internal cracks significantly reduce the failure load of the specimens. Compared with that of intact specimen, the failure loads of the single crack specimens with depths of 10, 20, and 30 mm and the double and triple crack specimens with depth of 20 mm, decrease by 11.8%, 19.5%, 78.4%, 30.9% and 35.5% respectively. iii) The fracture morphology of the specimens containing cracks with depths of 10 and 20 mm are similar to that of the intact specimen, all of which is belonged to dynamic crack-branching fracture, showing the characteristics of mirror atomization feather plain grain. The mixed mode I-II-III fracture occurs in the specimen with single internal crack with a depth of 30 mm, the characteristic of fracture is a smooth wing-like extension. iv) A fitting formula of the relationship between the fractography size and the crack depth is obtained. v) The KI, KII, KIII are inversely proportional to the depth of internal crack. The farther away the crack is from the center of the specimen, the smaller the value of KI and the larger the value of KII and KIII. The cracks on the left or right side have no effect on the value of K of the middle crack. This study shows that 3D-ILC is an important tool for studying internal cracks and mixed mode in fracture mechanics.

Immersed tube tunnel is adopted to cross the deepwater channel in both the Hong Kong-Zhuhai-Macao Bridge and the Shenzhen-Zhongshan Bridge. In traditional theoretical analysis, the foundation was usually simplified as a set of unrelated springs, the continuity of soil was ignored. The previous models were widely used in primary design calculation for its simplicity. In this paper, the cyclic effect of tidal load is considered, the pipes of the immersed tube tunnel are equivalent as Timoshenko beams placed on the Vlasov double-parameter foundation. Based on these assumptions, a calculating formula for vertical deformation of the immersed tube tunnel is derived, and then compared with the Timoshenko beam calculation model based on the Winkler foundation to verify the rationality of the proposed calculation method. Taking Yongjiang immersed tube tunnel project as an example, the vertical displacement of pipe joint in immersed tube tunnel under the influence of tidal load is analyzed, and the theoretical calculation results of the two models are compared with the field monitoring results. The results show that the proposed model based on the hypothesis of Vlasov double-parameter foundation can better describe the vertical displacement of pipe joint in immersed tunnel, which is more consistent with the monitoring data. The research results might have certain guiding significance for the design and calculation of immersed tunnel.

Deterioration of lining is a common phenomenon in tunnel engineering that threatens the safety of operating tunnel. For tunnel lining deterioration in rheological rock mass, an analytical model is established which the rheology effect of surrounding rock and deterioration of lining are considered. The surrounding rock displacement and support pressure of tunnel during its service life are obtained. And then the correctness of analytical model is verified by numerical simulation. Subsequently, the sensitivity analyses of deterioration coefficient of lining, thickness of lining, support time, and rheological properties of surrounding rock are carried out by the proposed analytical model. Finally, the service performance of tunnel is discussed. Research shows that under the rheological effect of surrounding rock, support pressure increases with time, while the mechanical properties of support structure decrease with time. Therefore, the time which support structure reaching its yield strength is shortened. Time history curves of support pressure and bearing capacity of lining are then obtained. Based on this, a model for predicting the service life of operating tunnel is established which the rheology effect of surrounding rock and support performance deterioration are considered. Finally, it is stated that the interaction between support and surrounding rock is the core for the long-term safety of operating tunnel.

Peak pressure of borehole wall is an important parameter for blasting parameter optimization and dynamic response analysis of non-fluid-solid coupling explosion shock. Based on theoretical analysis of the interaction between underwater explosion shock wave and borehole wall, a simplified calculation model for the peak pressure of borehole wall in water-coupling blasting is preliminarily determined. Using the fluid-solid coupling dynamic finite element numerical analysis method, the peak pressure of the borehole wall under a variety of water-coupling contour blasting charging conditions is studied, and the results are compared with the theoretical calculation results. The correction coefficient reflecting the influence of underwater shock wave propagation in the confined field and the superposition of transmission and reflection in the interaction between underwater shock wave and borehole wall on the peak pressure is determined. The results show that the characteristics of explosive, decoupling coefficients, and medium conditions of rock borehole wall have significant influence on the peak pressure of borehole wall. Through statistical analysis of the relationship between the decoupling coefficients and the correction coefficients under different conditions of explosive type and the wave impedance of rock, it is found that the correction coefficient increases approximately linearly with the increase of the decoupling coefficient. Based on the calculation model obtained by theoretical analysis and the method for determining the correction coefficients, a simplified method for calculating the peak pressure of borehole wall in water-coupling contour blasting is proposed. The method can fully reflect the comprehensive influence of explosive performance, charge conditions and rock medium conditions on the peak pressure of borehole wall.

In order to explore liquefaction resistance and related test technology of hydraulic fill coralline soils at artificial island and reef engineering sites, the phenomenon of unsteady state saturation in the liquefaction tests of hydraulic fill coralline soils was pointed out, and a cyclic triaxial liquefaction test method was proposed based on optimized saturation method and membrane compliance correction technology. Four groups of high-degree compaction samples with different gravel contents prepared by hydraulic fill coralline soils in an artificial island site were taken into the study, and the corrected liquefaction resistance curves of the studied materials were obtained. By comparing with the ground motion conditions of historical liquefaction sites, the occurrence of in-situ liquefaction was reproduced, and the rationality of the test results was verified. The study points out the liquefaction risk of the actual island and reef engineering sites in the South China Sea, analyzes the differences of the influence of gravel content on the liquefaction resistance of continental gravel soils and hydraulic fill coralline soils, proposes a gravel content based modified formula for the liquefaction resistance of hydraulic fill coralline soils, and provides a simplified estimation method for the liquefaction resistance of hydraulic fill coralline soils.

The blasting induced seismic waves are generally composed of compressional wave (P-wave), shear wave (S-wave), and Rayleigh wave (R-wave), however, wave-type and seismic components are not differentiated in the attenuation law and safety criteria for the current blast vibration studies. In this study, a method of wave-type discrimination is used for the seismic wave prediction based on polarization direction. Using theoretical analysis and numerical modelling, the blasting source characteristics and the radiated wave-types are investigated for different shapes of explosive charge. Combined the results of the site blasting experiments, the wave-type and seismic components induced by three typical blast-holes are analyzed and three blast holes include the single vertical blast-hole, the smooth blast-hole, and the slope pre-splitting blast-hole. The source characteristics and acting mechanism are then discussed for different blast-holes. The dominant wave-type at special location is predicted for three blast types. The research results indicate that the blasting source of the vertical blast-hole can be viewed as a delay superposition of the short explosive column. All the P-, S-, and R-waves contribute to the ground surface vibration from the vertical blast-hole. With the increase of the blasting-target distance, it is found that the S-wave gradually deviates from its dominant radiation direction, while the P-wave mainly contributes to the horizontal radial vibration, and the R-wave dominates the vertical vibration. Because the horizontal smooth blast-hole and the slope pre-splitting blast-hole are both contour blast-holes, the two blast holes have a similar acting mechanism, in which the main acting force is the loading from the normal surface. The S- and R-waves are the dominant seismic wave types within the blasting contour surface, whereas the role of the P-wave is negligible. Besides, the R-wave becomes the dominant wave-type as the blasting center distance increases; however, the contribution of P-wave outside the contour surface cannot be ignored.

The columnar dangerous rock mass is well developed in the Three Gorges Reservoir area. The change of water level accelerates the deterioration of the pedestal rock mass of the columnar dangerous rock mass and increases the risk of its crushing failure collapse. The potential surge disaster threatens the safety of shipping. According to the survey data of relevant dangerous rock masses in the Three Gorges Reservoir area, a dynamic observation system and experiment platform for particle column collapse was constructed, and physical model experiments on granular column collapse surge were carried out. The experiment results show that the crushing failure mode of dangerous rock mass is similar to that of the composite movement of collapsing-sliding of the experimental granular column. The staged motion of the granular column can be analyzed by the velocity of gravity center, which can represent the velocity of the particles. Formulas are derived by the nonlinear regression to estimate the water entry velocity of the particle and the maximum amplitude of surge. The Froude number is the main sensitive factor of the formula. Compared with the prediction formula of rigid block subsidence, the experimental formula in this paper is more suitable for crushing failure mode and has higher prediction accuracy. This research will provide the technical support for the prediction of the surge caused by the instability of the columnar dangerous rock mass in the reservoir area.

Based on Mohr-Coulomb and modified Drucker-Prager constitutive model, the inversion method of mechanical parameters of similarity model is performed using digital speckle correlation method and finite element method (DSCM-FEM). The similarity model test is carried out, the digital speckle correlation method is used as the experimental observation method, the surface displacement field is analyzed as the measured value during the loading of the model, and the displacement field obtained by the finite element method is used as the simulation value to construct the inverse problem model. The mechanical parameters of similar models are reversed by the optimization algorithm based on the Mohr-Coulomb and the modified Drucker-Prager constitutive model respectively. The test results reveal that the measurements and simulation displacements in the speckle area obtained by the proposed DSCM-FEM mechanical parameter inversion method match both numerically and regularly. The inversion constitutive model based on the modified Drucker-Prager model has higher inversion accuracy.

The dynamic response of geosynthetic reinforced embankment under traffic moving load has attracted increasing attentions in engineering field. A 3D model of geosynthetic reinforced embankment was established by using the ABAQUS finite element software in this paper, which was used to analyze the dynamic stress and deformation of geosynthetic reinforced embankment under moving load. The traffic load was simulated by two moving rectangular plane loads. Fortran subroutine was developed to control the amplitude, range, and speed of the moving load. The equivalent linear viscoelastic model was developed to simulate embankment fill to reflect the viscoelasticity of embankment fill. The geogrid was simulated by T3D2 truss element. The infinite element was used to reduce the boundary effect caused by model size at the boundary. The numerical model of geosynthetic reinforced embankment under moving load was established without considering drainage consolidation. Based on the results of the existing literatures and the results of this paper, the cross-section deformations of the geosynthetic reinforced embankment and the stress at the embankment top surface were compared and verified. The dynamic stress distribution in the longitudinal section and transverse section and the vertical dynamic stress distribution characteristics of the geosynthetic reinforced embankment under moving load were also analyzed. The results showed that the dynamic stress and deformation decayed rapidly within the range of 1.0 m on the embankment top surface and gradually transitioned to an equivalent uniform load with small amplitude. At the same depth, the attenuation coefficient of dynamic stress under the wheel load was the smallest, followed by that at the center of double wheel loads, and the attenuation coefficient at the outer edge of the wheel load was the largest.

In order to make up for the shortcomings of traditional low-strain detection methods in identifying the integrity of high-cap pile foundation and the pile length information, a novel vertical uneccentric excitation method and an axial multi-point traveling wave decomposition of a series of velocity responses along the shaft pile are proposed in this study. Furthermore, two-dimensional and three-dimensional finite element models are used in this study to verify the feasibility of the method in identifying the integrity of high-cap pile foundation and pile length information. Finally, a series of influencing factors such as pile size, capsize, sensor setting locations, spacing, and other parameters are analyzed. The research results show that this method can eliminate the influence of the superstructure complex vibration characteristics and high-frequency noise after processing the initial complex speed response. Hence, a clear and identifiable response curve of the bottom reflection of the pile and effectively predicting the unknown pile length can be obtained. It can be concluded that this study can break through the limitations of the existing non-destructive testing methods and has reference significance for the integrity judgment of the existing high-cap pile foundation and the pile length prediction in actual engineering.

The investigation of low frequency ultrasonic guided waves propagation in anchored rock bolts is of great significance to selecting optimal guided wave frequencies for non-destructive testing (NDT). The propagation of 0-100 kHz guided waves through anchored rock bolts of varying grouting structures has been investigated by determining their frequency domains and time-domains using theoretical calculations and numerical simulations, respectively. The numerical simulation and theoretical calculation results of the anchored rock bolt models with finite grouting thicknesses agree well with the test results in related literature, which verifies the validity of the numerical and theoretical calculation methods presented in this paper. Afterward, the propagation of low frequency guided waves through anchored rock bolts with infinite grouting thicknesses was investigated using the validated theoretical calculations and numerical simulations. The results indicate that the grouting thickness and properties of the grouting material significantly influence the propagation characteristics of the guided wave. Though the optimal guided wave frequencies in anchored rock bolts with finite thicknesses exhibit sufficiently low attenuation and a great propagation distance, they are not suitable for field NDT testing due to the significant attenuation in anchored rock bolts with infinite thicknesses. The attenuation of low frequency guided waves increases as the elastic modulus of grout increases; therefore, accurate field NDT testing can only be conducted within eight hours after the initial pouring of the grout.

To determine the distribution of the active earth pressure of the cohesionless soil under the rigid retaining wall rotating around the base (i.e., RB) mode, the discrete element simulations concerning different post-fill widths were carried out. The simulation results show that the active earth pressure distribution in the RB mode is different from the parabolic distribution of the active earth pressure in the translation (T) mode. When the soil behind the wall is in an active limit state, multiple parallel slip lines are formed inside the soil. An oblique differential element method based on slicing of slip line is proposed based on the numerical results. Based on the proposed method, the soil behind the wall is divided into several oblique differential elements, and the theoretical formula of active earth pressure under RB mode is derived according to the static equilibrium condition. In the case of infinite soils, the theoretical formula is consistent with the Coulomb's active earth pressure formula of triangular distribution. In the case of finite soils, the active earth pressure is piecewise linear distribution along with the depth. In the case of wall soil without friction, the theoretical formula will degenerate into Rankine's active earth pressure formula. Finally, the rationality and effectiveness of the theoretical formula are verified by comparing with the results of discrete element simulation, previous theoretical methods, and model tests.

To evaluate the retardation capability of buffer materials in an underground repository for high-level radioactive waste, it is necessary to understand the performance of densely compacted bentonite block and joint combination under coupled thermo-hydro- mechanical (THM) condition. For this purpose, a testing apparatus for coupled THM responses of combined bentonite blocks and joints is developed. It consists of five parts, including the model test box, the combined block-joint sample, the temperature control system, the hydraulic control system, and the data measurement and acquisition system. Two compacted blocks composed of the Gaomiaozi (GMZ) bentonite and one joint filled with sand and bentonite powder were used to make the composite sample. To examine the feasibility of the design of this apparatus, the test was carried out under a thermal boundary condition of 90 ℃ and a hydraulic boundary condition of 0.1 MPa for 266 days. Based on test data of temperature, volumetric moisture content and expansive force, the coupled THM responses and their spatiotemporal evolution rules were obtained. A new definition formula of joint-healing- degree was proposed on basis of variation of dry density in test sample. The joint-healing-degree was calculated through the measured change in joint thickness during the test. Regression analysis was conducted according to the relationship between the change in joint thickness and the coupled THM responses including temperature, volume moisture content and expansive force, and the regression formula was accordingly set up. It has shown that this apparatus can be used in the indoor physical simulation test of buffer materials for deep nuclear waste disposal.