The construction of salt cavern underground storages in China originated from the Jintan salt mine in Jiangsu province, achieving success and establishing the first technical standard system for salt cavern storage construction in China. However, the geological conditions of other salt mines in China differ significantly from those of the Jintan salt mine, such as NaCl grade, buried depth, salt rock thickness, and interlayer characteristics. Therefore, the construction of salt cavern storage faces numerous challenges, and the “Jintan mode” cannot be directly replicated. Focusing on the geological design of salt cavern storage, solution mining engineering, and gas injection and brine discharge schemes, this paper firstly summarizes the “European and American mode” of salt cavern storage construction and the basic characteristics of China’s “Jintan mode”, which basically developed from the “European and American mode”. Then, based on research and engineering practices in salt cavern storage construction of other salt mines in China, several typical “XX modes” are put forward, adhering to the principle of “one solution design for one region or one salt mine”. The “XX modes” differ from the “Jintan mode”, with great breakthroughs and changes. They represent technical solutions and new ideas to deal with the geological complexities of salt mines in China. Finally, the key theories and technologies that need to be researched further in the next step to cope with geology challenges of salt mines are prospected.

Particle movement in rock fissures, influenced by dynamic water action, often leads to infiltration damage. To address the random and hidden nature of natural rock fissures, a fissure channel with a rough joint surface was constructed using a three-dimensional Weierstrass-Mandelbrot fractal function. A transparent and refined fracture channel was obtained using 3D printing technique, and the movement of particles in the fissures under the action of moving water was studied using microfluidic control instruments. The influences of the degree of fissure roughness, the ratio of particle size to fissure width (it is referred to “particle-to-gap size rate” in text), and the seepage pressure on particle initiation and transport were analyzed to deduce the possibility of infiltration damage in rock. The results show that of particle movement in rock fissures is related to three factors: the degree of fissure roughness, the water pressure, and the particle-to-gap size rate. Larger fissure roughness or larger particle-to-gap size rate makes particle movement less likely, reducing the possibility of seepage failure. At the same time, the water pressure is the dominant factor in particle movement, the impact of particle-to-gap size rate on particle movement is related to water pressure. There is a specific water pressure critical value. Below this value, the influence of particle-to-gap size rate on average movement speed is not obvious. Beyond this value, the influence increases significantly. The critical water pressure value is related to the degree of fissure roughness. The rougher the fissure, the higher the critical water pressure.

During an earth pressure balance shield tunneling in water-rich coarse-grained stratum, muck spewing or discharge stagnation often occurs. Only foam conditioning cannot improve the fluidity and anti-permeability of poorly graded muck due to foam dissipation and loss. Therefore, bentonite slurry must be added. Understanding the large shear deformation of slurry-foam-conditioned poorly graded sand under chamber pressure can optimize the conditioning scheme to ensure safe tunneling. In this study, a series of pressurized vane shear tests on slurry-foam-conditioned poorly graded sand was conducted. The results show that increasing the bentonite injection ratio (BIR) raises the initial void ratio and saturation of conditioned sand under pressure, resulting in a decrease in effective stress, peak strength, and residual strength. However, the shear-rate dependence of shear strength and effective internal friction angle are found to be insensitive to BIR. Moreover, increasing BIR reduces the yield stress of the conditioned sand after large deformation. A comparison of the large shear deformation under pressure of three types of conditioned soils with similar workability at atmospheric pressure indicates that increasing the slurry or foam injection ratio can enlarge the void ratio and reduce the effective stress under pressure. Notably, slurry-foam-conditioned sand exhibits lower effective stress and shear strength than foam-conditioned sand. The suitable conditioning parameters for testing poorly graded sand are FIR=20% and BIR=10%–15%. Microscopic images reveal the dual weakening effect of bentonite on the shear strength, including the bentonite adsorption on the surface of sand particles to reduce inter-particle friction and the deceleration of foam degradation, leading to a uniform distribution of small-size bubbles in pores and effectively weakening the soil skeleton strength.

The consolidation coefficient is a crucial parameter for settlement calculation and stability analysis of soft foundations. Existing in-situ testing methods for the consolidation coefficient have the disadvantages of time-consuming and low accuracy. Based on the penetration mechanism of piezocone penetration test (CPTU) and the dissipation pattern of excess pore water pressure at the cone shoulder, the formation, development, and dissipation processes of excess pore water pressure at the CPTU cone shoulder are described using the theory of circular cavity expansion and the axisymmetric consolidation equation. By incorporating the automatic differentiation capability of neural networks, the axisymmetric consolidation equation is embedded into a deep neural network. The physical information constraints of the neural network are formed through the loss functions of physical equations, boundary conditions, and initial conditions. At the same time, the CPTU pore pressure test data serve as a data-driven term. Consequently, with the minimization of the excess pore water pressure loss function as the optimization goal, a physics-informed neural networks (PINNs) model is established for inversely analyzing the in-situ consolidation coefficient using CPTU pore pressure test data. The effectiveness of the PINNs model in inversely analyzing in-situ consolidation coefficient is verified through example analysis and inversion validation using existing centrifuge test data. The robustness of the PINNs model is also analyzed using CPTU pore pressure test data. The results indicate that the proposed PINNs model can effectively use CPTU pore pressure test data to rapidly and accurately invert the site in-situ consolidation coefficient. Due to the integration of physical mechanism constraints, the model requires only a small amount of training data and exhibits strong robustness and generalization performance against noisy pore pressure test data, providing an effective approach for accurate, rapid, and reliable testing of the in-situ consolidation coefficient.

The phenomenon of wave dispersion and its causes are a hot topic in the study of rock physics and acoustics of porous media. Based on the Biot theory and Darcy-Brinkman law, Maxwell-Brinkman and Kelvin-Brinkman viscoelastic porous media mechanical models were proposed, and the dispersion relationship of compression waves was established. The dispersion and attenuation laws of waves were revealed, and the effects of stress relaxation and creep on wave propagation velocity and inverse quality factor were analyzed. At low frequencies, the dispersion and attenuation of compression waves predicted by the Kelvin- Brinkman model and the Darcy-Brinkman-Biot model are consistent. While, at high frequencies, the dispersion and attenuation of compression waves predicted by the Maxwell-Brinkman model agree well with those by the Darcy-Brinkman-Biot model. The reliability of the physical model was verified using experimental data from the Gulf of Mexico. Through parameter sensitivity analysis, the method for optimizing medium parameters based on measured data was discussed.

Measuring the real contact characteristics between joint surface walls is challenging due to their inherent characteristics, which closely relate to the hydro-mechanical behaviors of rock mass. To quantitatively investigate the real contact area of rock joints under normal stress, rock engraving systems were used to manufacture a series of rough rock joints. The real contact area of rock joints was investigated using a multi-type pressure-sensitive film (MS, HS, and HHS), and an adaptive threshold method was utilized to recognize the scanning image of the film. Simultaneously, the contact stress transformed by the scanned film was also quantified. The evolution of the real contact area affects the mechanical characteristics of rock joints, characterized by the emergence, expansion, and coalescence of contact patches. The real contact area ratio increases hyperbolically with normal stress, and an empirical model can effectively predict the evolution characteristics. Joint roughness coefficient JRC has a positive correlation with initial growth rate of contact area. The contact points oscillate with increasing normal stress, and their growth rate has a negative correlation with the fractal dimension. Even at low normal stress (0.5 MPa), the highest contact stress can exceed 600 times the nominal normal stress. Distribution of the contact stress becomes more inhomogeneous with the increase of normal stress and roughness. The impact extent of roughness at high normal stress is approximately 20 times greater than at low normal stress. The experimental results can provide a reference for the constructive of the constitutive model and validation for numerical simulations

With the increasing attention paid to three-dimensional numerical limit analysis, there is an urgent need to develop a new Drucker-Prager (DP) criterion suitable for geomaterials under conventional triaxial stress condition. Yet, an exact DP criterion for geomaterials under conventional triaxial stress condition does not exist. Instead, an approximate equal-area-circle DP-3₁ criterion has been used traditionally, which is relatively safe. This study developed a new DP-3₂ criterion for geomaterials under conventional triaxial stress condition based on the tri-shear energy yield criterion. The theoretical formulation was derived to determine the highest point of the criterion (i.e., the tangent point between the criterion and the Mohr-Coulomb criterion). Then, the conventional triaxial DP-3₂ criterion was established through the highest point. Thereafter, this new criterion was used to determine the ultimate load of soil under conventional triaxial condition and slope stability analysis. The ultimate load of soil under conventional triaxial condition determined by the DP-3₂ criterion was found to be about 87%–97% of the measured value. Moreover, the maximum ratio of ultimate load computed by the DP-3₂ criterion to the DP-3₁ criterion was 1.19, and it increased with decreasing confining pressure, increasing cohesion c, or increasing internal friction angle φ . The factor of safety (FOS) of soil slopes determined by the DP-3₂ criterion was approximately 1.01–1.04 times that determined by the DP-3₁ criterion. Furthermore, the difference increased at larger slope angles. These results suggest that the DP-3₂ criterion is suitable for numerical limit analysis of geomaterials under conventional triaxial stress condition.

To expand the application of biochar in civil engineering, biochar was used to partially replace cement in biochar-cement soil. The permeability characteristics and mechanism of biochar-cement soil with different biochar contents and curing ages were analyzed using the falling head permeability test and scanning electron microscope (SEM) test to determine the optimal biochar content. The permeability test results showed that the permeability coefficient of cement soil decreased first and then increased with the increase of biochar content. The lowest permeability coefficient was observed when biochar replaced 1% of the cement. Additionally, the permeability coefficient of cement soil decreased nonlinearly with the increase of curing age, and biochar had the greatest influence on permeability of cement soil in the early stage. Meso-structure analysis revealed that biochar did not participate in chemical reactions within the cement soil. Biochar improved the anti-permeability performance of cement soil by filling the pores between soil and cement particles, making the mixture more compact.

This study investigates the creep damage characteristics of weakly cemented soft rock induced by excavation and unloading in an underground chamber, focusing on samples from the western mining area. A series of multi-stage unloading confining pressure creep tests on weakly cemented soft rock was conducted using GDS HPTAS. Under different water conditions, the creep characteristics and creep rate characteristics of weakly cemented soft rock were studied under confining pressure unloading. Based on the reciprocal steady creep rate method, the mathematical exponential function relationship between long-term strength and water content of weakly cemented soft rock was determined. By introducing the Riemann-Liouville fractional order integral operator and the negative time exponential damage evolution variable, nonlinear damage Abel clay pot was defined, and a six-element fractional order creep damage constitutive model, including elastic elements, viscoelastic damage elements, viscous damage elements and nonlinear viscoplastic damage elements, was established. Based on the experimental results, the Trust-Region algorithm in Matlab was used to identify the model parameters and analyze the sensitivity. The results show that the established model has clear physical significance and high agreement with the test values. It can accurately reflect the creep damage characteristics of weakly cemented soft rock under excavation and unloading confining pressure in underground engineering, which is significant for ensuring the long-term stability of underground engineering.

To investigate the effect of various intermediate principal stresses on rockburst incubation and failure during tangential stress gradient loading, a series of true triaxial rockburst model experiments on quasi-rock material specimens under three intermediate principal stress conditions was conducted using a self-developed gas-liquid combined loading apparatus. A high-speed camera and an acoustic emission monitoring system recorded the test data. This paper compared and analyzed the macroscopic failure differences of the specimens and the evolution characteristics of acoustic emission parameters. The results indicate that the failure load, instantaneous burst intensity of rockburst, total mass of rockburst debris, mass of debris of each particle size, and scale of specimen failure increase with higher intermediate principal stress. During loading, specimens under high intermediate principal stress show weakened acoustic emission activity in the early and middle stages, revealing that the high intermediate principal stress suppresses the damage and crack development. In the later stages, the acoustic emission ringing count and energy release of the specimens under high intermediate principal stress become more concentrated, and the energy peak and cumulative values increase. The acoustic emission b-value drops sharply to a lower level, and the proportion of high main frequency amplitude points increases, indicating that high intermediate principal stress exacerbates the damage and crack development. The number of large-scale and shear cracks increases, leading to more violent instantaneous energy release, more severe failure, and higher intensity of rockburst.

Traditional anchoring agents often fail to provide effective bolt support for soft rock mass. To address the issues of poor anchoring performance and easy attenuation of anchoring force in soft rock mass, a physical anchoring device is proposed. This device aims to enhance the support effect of bolt and ensure continuous stability of anchoring ability. This study combines numerical simulations and laboratory experiments for systematic research. Numerical simulation results show that within the range of design parameters, higher inverted teeth on the anchoring device results in greater anchoring force. A smaller tooth angle decreases the anchoring force. More rows of inverted teeth increase the anchoring force, while the spacing of inverted teeth has no significant effect on the anchoring force. Based on numerical simulations, the optimal parameters for the inverted teeth of the anchoring device were selected, and drawing experiments were carried out. Experimental results show that the physical anchoring device maintains anchoring force over a long period. The three-row inverted teeth anchoring device provides higher anchoring force compared to traditional anchoring agents. This study helps to solve the core problem of poor anchoring performance in soft rock conditions, and provides a valuable reference for improving bolt support in soft rock mass.

How to estimate the seismic response of shield tunnels reasonably has been a significant issue in both industry and academia. Compared with the transverse seismic response, the longitudinal seismic response is more complex. The shield tunnel is modeled as a Timoshenko beam on Winkler foundations, considering both residual axial force and additional axial force due to longitudinal seismic. A theoretical model was presented to consider the longitudinal and transverse stratum displacements. The longitudinal seismic response of shield tunnels was solved using the finite difference method. The theoretical model and calculation method were validated through case studies of stagger-assembled shield tunnels with a 6.2 m diameter. Considering axial force increased the overall stiffness of shield tunnel, resulting in decreased internal force and deformation response. The proposed method was degraded to the traditional one when the axial force is neglected. The influences of residual axial force, seismic wavelength, seismic incidence angle and foundation reaction coefficient on the longitudinal seismic response were further explored. With the increase of residual axial force, the overall stiffness of shield tunnel increased, and the peak seismic response of shield tunnel decreased. When the incident angle is less than 45°, the influences of residual axial force and foundation reaction coefficient on the peak response were more significant. With the increase of the foundation reaction coefficient, the tunnel deflection and the discontinuous deformation between joints increased. Wavelengths between 20 m and 100 m may lead to greater opening and dislocation between joints. These studies can provide theoretical support for the longitudinal seismic design of shield tunnels.

During hydraulic fracturing in hot dry rock, elevated temperatures weaken the fracture surface strength of the thermal reservoir rock, resulting in fracture slip behavior. This weakening phenomenon is closely related to thermal variations in cohesion and internal friction angle. Therefore, understanding the mechanical properties of granite at high temperatures is crucial for the stability assessment of thermal reservoir rocks during hot dry rock development. Variable angle shear tests (45°, 55°, 65°) were conducted on coarse and fine-grained granite under different high-temperature conditions. The study analyzed the impact of elevated temperatures on the cohesion and internal friction angle of granite with different particle sizes. In conjunction with scanning electron microscopy and low-field nuclear magnetic resonance tests, a further analysis was conducted on the influence of thermo-mechanical coupling on the evolution of microcracks and pore structures in granite. The results indicate that the cohesion of granite initially increases and then decreases with the increase of temperature. The variation in the internal friction angle is minimally influenced by temperature. The threshold temperature for the transformation of granite cohesion is identified as 300 ℃. Prior to the threshold temperature, the cohesion increase in coarse-grained granite is approximately 5.73 times greater than in fine-grained granite, and the effect of temperature strengthening is more significant on coarse-grained granite. As the temperature increases, the form of microcracks in granite transitions from intragranular cracks to intergranular cracks. The nuclear magnetic resonance (NMR) T₂ spectrum exhibits multiple discontinuous peaks. With increasing temperature, compaction affects the large pores in coarse-grained granite, leading to a reduction in relaxation effects. Conversely, thermo-mechanical coupling has a limited effect on the development of large pores in fine-grained granite. These findings are expected to provide valuable insights for assessing the stability of reservoir rocks in hot dry rock geothermal energy development.

Typhoons frequently occur along the southeastern coast of China, and offshore wind turbines often suffer from multiple cycling and reconsolidation during their service life. However, there is currently limited research on effects of cycling and reconsolidation. To address this, a series of centrifuge tests was conducted on two types of large-diameter monopile foundations in soft clay with different amplitudes, aiming to reveal and quantify the impact of episodic cycling and reconsolidation effect on key design indicators such as cumulative displacement and pile head stiffness of large-diameter monopiles. The test results show that under the same cyclic load amplitude, after two episodes of cycling and reconsolidation, the maximum stiffness of the pile head of monopile foundation can increase by 1.5 times compared to the initial stiffness. This significantly reduces the cumulative displacement of the monopile, with a maximum reduction of 63%. Therefore, if the episodic cycling and reconsolidation effect on the subsequent cyclic response of a monopile is not considered in the design process, and the cumulative displacement of a monopile under each extreme typhoon is directly accumulated, it will seriously overestimate the cumulative displacement of the pile, leading to a conservative design.

To investigate the creep characteristics of fractured rock mass in coal measure strata under anchorage, uniaxial creep tests were carried out on anchored, unanchored fractured and intact specimens at different fracture dip angles (15°, 30°, 45°, 60°, 90°). The results show that the instantaneous strain of the unanchored fractured specimens is greater than that of the intact specimens under the same stress loading level. The difference between the anchored fractured specimens and the intact specimens gradually decreases as fracture inclination increases. With the increase of stress loading level, the steady-state creep variable of each rock specimen also increases. The steady-state creep rate of fractured specimens initially decreases and then increases with the increase of fracture dip angle. However, the steady-state creep rate of anchored fractured specimens is lower than that of unanchored fractured specimens under the same stress level. Significant differences in crack penetration development exist between unanchored and anchored fractured specimens at low fracture dip angle. Based on the CVISC creep model, parameter inversion was performed, and the trends of Maxwell and Kelvin correlation coefficients with stress loading level were obtained. The stress-strain isochronous curve method and the unsteady rheological discriminant method were employed to study the long-term strength variation of fractured specimens. The research results provide an experimental basis for long-term stability analysis and bolt support schemes of coal mine roadway.

Quantitatively characterizing damage in granite under thermal conditions is a significant challenge. This study approaches the issue from an energetic perspective by measuring the temperature during the unsteady-state heat transfer process of granite specimens at elevated temperatures, specifically within the range of 600 °C. The evolution of thermophysical parameters during the regular state stage was systematically analyzed using a cooling method. This investigation explores the characteristics of granite undergoing unsteady state heat transfer, focusing on energy absorption, release, and dissipation. It reveals the intrinsic connection between performance damage and energy evolution laws of granite under thermal action. The findings reveal that granite specimens in different high-temperature states undergo natural cooling in two distinct phases of specific heat change during the regular state stage: steady change and significant change in specific heat. As the initial high-temperature state decreases, the energy dissipated during the whole unsteady state heat transfer process also decreases. In the same cooling mode, the larger the input energy, the larger the proportion of dissipated energy. The amount of dissipated energy in the unsteady state heat transfer process strongly correlates with the deterioration of the macro-mechanical parameters of granite after high-temperature thermal action. The amount of the dissipated energy and the maximum thermal shock factor have a significant linear relationship.

The sandy loess in the desert-loess Plateau transition zone exhibits obvious collapse deformation when it encounters water. The collapsibility of the sandy loess in Changqing Oilfield, widely distributed in Jingbian, northern Shaanxi province, is seriously influenced by the collapsibility of sandy loess. It is urgent to reveal the collapsibility characteristics and mechanism of the sandy loess in Changqing Oilfield construction. Therefore, taking the sandy loess of Jingbian Q₃ in the transition area of Jingbian Desert-Loess Plateau in northern Shaanxi as the research object, the basic physical properties and material composition of the sandy loess were analyzed using laboratory basic physical properties, X-ray diffraction and collapsibility tests. The collapsibility characteristics, influencing factors and rules were clarified. On this basis, the microstructure, pore size distribution, directional frequency and abundance changes before and after the collapse of sandy loess were explored using scanning electron microscopy test and pore and fissure image recognition analysis. The collapse mechanism of the sandy loess was revealed from a microscopic perspective. The results show that the Jingbian Q₃ sandy loess has collapsibility. The collapsibility coefficient first increases and then decreases with the increase of axial pressure, and it gradually decreases with the increase of dry density and moisture content. The collapsibility coefficient peaks at the axial pressure of 150 kPa. Jingbian Q₃ sandy loess is mainly composed of quartz, albitite, muscovite and calcite. The grain morphology is mostly angular or subangular, with overhead arrangement structure and overhead pores, and point-to-point contact. Clay cements are mostly distributed in the contact areas of skeleton particles. The collapse of the overhead pore structure in Jingbian Q₃ sandy loess is the essence of its collapsible deformation, providing the main space for collapse. Subsidence deformation, caused by the coagulated structure formed by a small amount of debris particles wrapped in the clay cement, contributes to the increment of collapsible deformation. The research results provide a data basis for evaluating the collapsibility of sandy loess in the construction area of Changqing oil and gas field engineering.

Investigating the dynamic response and damage condition of variable-section single pile in seismic subsidence site, taking Xiang’an Bridge as the engineering background, the dynamic response of variable-section single pile under 0.10g–0.45g ground vibration intensity was investigated through large shaking table tests, and the damage degree was evaluated and analyzed. Results are as follows: Influenced by the seismic subsidence characteristics of the soil layer, the top horizontal displacement, acceleration, and bending moment of the variable-section single pile tend to increase with the increase of vibration intensity, while the base frequency of the variable-section single pile gradually decreases. The maximum bending moment of the variable-section single pile occurs at the seismic subsidence soil stratum demarcation, where the ground shaking strength of 0.30g exceeds its flexural capacity. At a ground vibration intensity of 0.20g, the top horizontal displacement, acceleration, bending moment of the variable-section single pile increase significantly, and the base frequency of the pile decreases abruptly. Based on the system damage theory, the damage condition of variable-section single piles in seismic subsidence sites can be divided into three stages: stable stage, intensified damage stage, and plastic failure stage. At the end of the test, bending plastic damage occurred at the location of the variable section.

Mining deep coal resources involves high stress geological environment and rock excavation unloading engineering environment. Research on the re-bearing capacity characteristics and damage mechanism of unloading-damaged rock is crucial for revealing the instability and rupture behavior of deep rock body. The damaged sandstone specimens were prepared by controlling the unloading point during triaxial loading. Transverse and axial strains of the specimens were monitored using the RMT-150C rock mechanics loading system and the Donghua DHDAS stress-strain acquisition system during triaxial unloading and uniaxial reloading. The test results show that according to the wave velocity differences before and after loading, the unloaded rock body can be classified into three categories: compact rock body, low-loss rock body and high-loss rock body. With the increase of unloading stress level, the damaged sandstone transitions from plastic-elastic-plastic deformation to plastic-elastic deformation, and finally to elastic-viscous deformation. The crack volume strain curve of the compact rock body during uniaxial reloading is basically the same as that of the standard sandstone. The elastic deformation stage of the low-loss rock body is obviously shortened, while the high-loss rock body fails after minimal deformation. The compact rock body exhibits Y-type diagonal shear damage characteristics. The low-loss rock body develops and penetrates through longitudinal cracks and fissures formed during the unloading phase, leading to master cracks causing damage. The high-damage rock body develops X-type diagonal shear damage on the basis of Y-type diagonal shear rupture. A mechanical analysis model of cracks under uniaxial compressive stress was established. The relationship between rock crack fracture angle  and crack angle  and friction coefficient f of crack surface was elucidated.

The reconstruction of rock mass fracture is crucial for accurately constructing fracture rock mass model and obtaining its mechanical properties. To ensure the reconstructed fractures fit the natural state, this paper first proposed the D-n_max threshold formula and modified the Weierstrass-Mandelbrot (W-M) function used in rock mass fracture simulation. Barton’s 10 profile lines were digitized using binarized gray images, and Hurst values for 10 roughness levels were determined using the equi-difference method, trial and error method and comprehensive analysis method. The new parameter Spp, representing the magnitude feature, is obtained by Fourier transforming the 10 profile lines. The parameter Spp is then combined linearly with the standard deviation of the angle σ_i to construct a new joint roughness coefficient (JRC) calculation formula. The optimal scaling constant G value corresponding to the target JRC value is obtained using the dichotomy method, establishing the mapping relationship between JRC and the W-M function. The W-M function considers random terms and realizes the differentiation of reconstructed fractures. The roughness quantification research and laboratory tests are conducted to verify the validity of the new parameter Spp, the new JRC formula, and the mapping relationship. This study provides a new approach for expanding the method of obtaining fracture in rock mass, and lays the foundation for studying the mechanical properties of fractured rock mass.

Deep excavated expansive soil slopes are influenced by multiple factors, including precipitation, groundwater, geological structure, and support measures, resulting in complex spatiotemporal heterogeneity in deformation. Using the slope of Taocha canal segment in the middle route of the South-to-North Water Diversion Project as a case study, the spatiotemporal deformation characteristics were analyzed. Data mining methods, including variational modal decomposition, weighted multiscale local outlier factor, and clustering analysis, were applied. The temporal variation of deformation was analyzed, and spatial distribution characteristics were identified. The influence mechanisms of factors such as precipitation and groundwater on the trend, periodic, and fluctuating components of deformation were explained. Deformation measuring points were clustered, and potential sliding surfaces and sliding bodies were speculated. The deformation mechanism was discussed, and reinforcement measures were proposed. The results indicate that slope deformation exhibits significant trend changes, as well as seasonal and intermittent fluctuations. Deformation in the lower part is relatively large and gradually decreases upwards. The depth of the significant deformation in the upper part is 3 m, located within the climate-influenced layer. Deformation in the central part is influenced by groundwater fluctuation and dense fissure zones, with a significant deformation depth of up to 11 m. Deformation in the lower part is limited by the retaining system and occurs only in the shallow layer. Perched water in the upper part is replenished by rainwater, causing large fluctuation depths and resulting in deep deformation in the middle and upper parts, and there is a certain degree of deformation within the depth of 16.5 m. The potential sliding surface is a polyline. Due to the influence of groundwater, densely fissure zones, and retaining system, the leading edge is approximately horizontal. Drainage wells should be installed to drain groundwater to reduce the impact of groundwater fluctuations on the swelling-shrinkage deformation of expansive soil. The research results can provide technical support for the operation management and reinforcement disposal of deep excavated expansive soil slopes.

To analyze the vertical vibration characteristics of a solid pile with vibration isolation layer, a vertical dynamic impedance analytical model of the vibration propagation medium and the solid pile was established. The model was based on the Rayleigh-love rod model and Novak plane strain theory, considering the radial and longitudinal non-uniformity of the vibration propagation medium around the pile and the lateral inertia effect of the pile. Utilizing the Laplace transfer method, complex stiffness transfer method, and impedance function transfer method, the analytical solution of dynamic impedance at the solid pile head under arbitrary vertical load in a viscous damping medium was derived. The rationality of this solution was verified by comparing with the existing analytical solution. It is found that the amplitude of the pile head dynamic stiffness has an obvious influence on the dynamic response at the pile head, indicating a high correlation. Further, the effects of different thicknesses and longitudinal depths of the rubber layer and different types of vibration isolation material on the pile head dynamic stiffness are investigated. The results show that in the frequency range of 1-80 Hz: (1) The dynamic stiffness curve shifts to a lower (higher) frequency as the rubber layer thickness increases (decreases), with the thickness variation having an insignificant effect on the feature frequency and the dynamic stiffness; (2) The amplitude and oscillation frequency of the dynamic stiffness curve increase (decrease) significantly as the rubber layer longitudinal depth increases (decreases); (3) When the vibration isolation material is softer than the surrounding soil, the amplitude and oscillation frequency of the dynamic stiffness curve increase as the shear modulus of the vibration isolation material decreases; (4) When the vibration isolation material is harder than the surrounding soil, the amplitude of the dynamic stiffness curve slightly increases as the shear modulus of the vibration isolation material increases, and the oscillatory property disappears.

A theoretical model of the defective pile-beam system has been proposed. A three-dimensional continuous medium model is adopted for the surrounding soil, while a Rayleigh-love rod is employed for the pile shaft to account for the transverse inertial effect. The pile defect is simulated using segments with a radius different from the normal one. The impedance function recursive method, in conjunction with ring soil pile theory (RSPT) and amended impedance function transfer method (AIFTM), is used to obtain the impedance function at pile head. The beam is simulated using Timoshenko beam, and transient excitation is applied at the pile-beam connection. The analytical solution for the dynamic response of the pile-beam system in the frequency domain is obtained, and a semi-analytical solution in the time domain is derived using the discrete Fourier transform. The obtained semi-analytical solution is compared with experimental and finite element method results to validate its reasonability. It is found that the pile-beam system is more suitable for excitation pickup at the pile-beam connection, although the influence of pile-beam parameters must also be considered. Finally, precautions for using low-strain testing on the pile-beam system are studied through parameter analysis.

This study focuses on the calcareous cemented conglomerate on the side wall of the Zhoukoudian Ape Man Cave site in Beijing. We analyzed the change patterns of cracks in grotto rock under natural conditions using statistical analysis of continuous monitoring data from 12 rock fracture measurement points for 7 years and environmental monitoring data. The results indicated that: (1) Under natural conditions, rock fracture width exhibited seasonal and periodic variations, which can be predicted using autoregressive integrated moving average (ARIMA) model. The ARIMA (2, 1, 1) model has the highest fitting accuracy. (2) The width of rock fractures showed a negative correlation with ambient temperature and relative humidity, with a stronger correlation with temperature. Changes in fracture width lagged behind changes in temperature and relative humidity, with a higher lag correlation and shorter lag time for temperature compared to relative humidity. (3) The degree of correlation and lag correlation between rock fracture width and ambient temperature followed the trend of room-temperature＞ high-temperature＞low-temperature＞extremely-low-temperature. The lag time followed the trend of extremely-low-temperature＞low-temperature＞high-temperature. Additionally, more stable relative humidity of the surroundings indicated a higher correlation. (4) At a constant, the width of rock fractures increased gradually over time, following an approximate exponential growth or asymptotic distribution. (5) At the same temperature, more stable relative humidity corresponded to a slower increase in rock fracture width over time. Under natural conditions, lower ambient temperatures led to a rapid increase in rock fracture width over time. These findings enhance our understanding of the expansion patterns of cracks in grotto rock under natural conditions. Moreover, they provide a reference framework for predicting crack changes in the preventive conservation of ancient rock sites, caves, and stone carvings in temperate monsoon climate regions.

Soft rock tunnel surrounding rock deformation exhibits significant time-dependent characteristics, potentially causing cracking and damage of tunnel linings during operation. This study focuses on a highly deformable mudstone section in a water diversion tunnel in Xinjiang, and proposes a support scheme with a high compression cushioning layer between initial support and secondary lining to ensure the long-term safety of the tunnel. The existing cushioning layer support scheme was initially subjected to on-site monitoring and structural forces analysis. Subsequently, numerical simulation methods were used to optimize the cushioning layer support parameters. Finally, the optimized and original schemes were compared to analyse their respective support effects. (1) Monitoring of the existing support scheme reveals that, with the installation of a 5 cm polyethylene cushioning layer at a density of 90–100 kg/m³, the surrounding rock pressure reaches 0.36 MPa, indicating the compression phase of the cushioning layer. The non-uniformity of lining force is evident, suggesting potential for optimizing the cushioning layer material and thickness. (2) Optimization of cushioning layer support parameters indicates that if the stress of buffer layer platform is too high, it cannot fully absorb energy, and if too low, it cannot effectively limit surrounding rock deformation. Both scenarios result in insufficient energy absorption and low lining safety. Increasing the cushioning layer thickness gradually reduces lining damage degree, but the reduction rate diminishes over time. For this project, the optimal cushioning layer support is achieved with a platform stress of 0.5 MPa, a compression ratio of ≥0.6, and a thickness of 10 cm. (3) Comparative analysis indicates that the optimized cushioning layer support reduces the maximum principal stress on the secondary lining by 20%–30% compared to the original scheme, alleviating stress concentration in the lining and ensuring the long-term stability of the tunnel support structure.

The liquefaction-induced diffusion and redistribution of the excess pore pressure in inhomogeneous strata may lead to pore water concentration in local areas beneath low permeability layers and cause shear strain localization and delayed failure. In this study, the nonlocal peridynamics (PD) theory is introduced as a novel regularization technique to model this phenomenon, overcoming the mesh-size dependency problem associated with the classical finite element method (FEM). The computational model couples PD and FEM for the solid and pore fluid phases, respectively. Liquefiable sand is modelled using a unified plastic model for large post-liquefaction shear deformation of sand (CycLiq). After validating the proposed method, the seismic response of an idealized one-dimensional sloping site with a low-permeability interlayer is analyzed using various discretization resolutions. It is demonstrated that the proposed method for liquefaction-induced strain localization analysis is insensitive to spatial discretization theoretically and numerically. At the same time, the parametric study shows that a higher location and a smaller permeability coefficient of the interlayer could lead to a greater lateral displacement of the stratum induced by shear strain localization.

Based on the wave propagation theory in saturated porous media and saturated frozen porous media, this study investigates transmission and reflection of P₁ waves at the interface between saturated soil and saturated frozen soil. Using the Helmholtz vector decomposition principle and considering the boundary conditions at the interface between saturated soil and saturated frozen soil, analytical solutions for the transmission and reflection amplitude ratios of P₁ waves from the saturated soil layer to the interface with saturated frozen soil layer is obtained. Through numerical calculations, the influences of incident frequency, incident angle, permeability of the saturated soil medium, cementation parameters, temperature, and contact parameters of the saturated frozen soil medium on the transmission and reflection amplitude ratios at the soil interface were analyzed. The research findings indicate that the propagation velocity of the incident P₁ wave is significantly lower than that in the saturated frozen soil medium. Therefore, a critical angle arises when the P₁ wave propagates from the saturated soil medium to the saturated frozen soil medium, and the amplitude ratio undergoes a sudden change when the incident angle equals the critical angle. The permeability of saturated soil has a minor impact on the transmission and reflection amplitude ratios of wave but has a prominent effect on the reflected P₂ wave. Variations in temperature and cementation parameters of the saturated frozen soil medium significantly affect the transmission and reflection amplitude ratios of wave.