Given the complexity and diversity of soft rock, the existing constitutive model is difficult to accurately describe the accelerated creep stage of its creep model. To more accurately describe the creep process of the carbonaceous phyllite, this study first summarizes the rheological properties observed in triaxial grading creep experiments. A nonlinear creep model, termed the Snishihara model, is established based on the traditional Nishihara model by incorporating fracture plastic elements and nonlinear damage plasticity. Based on this theoretical model, a three-dimensional finite difference scheme suitable for FLAC3D numerical solutions is derived. Additionally, the yield function, potential function, and three-dimensional stress state of the Snishihara viscoelastic constitutive model are modified. Finally, using the secondary development environment provided by Visual Studio 2015, the custom constitutive model is completed in FLAC3D. The non-accelerating and accelerating creep stages of the Snishihara model are then analyzed to verify the model’s accuracy and reliability. The results show that the numerical simulation aligns well with the indoor creep test results regarding strain increments and creep change curves. This confirms the validity and applicability of the Snishihara creep constitutive model proposed in this study, as well as its secondary development application.

The mechanical characteristics of coarse-grained materials exhibits a significant dependency on the stress path, and the large-scale true triaxial tests of complex three-dimensional stress paths are rarely reported. To investigate the influence of stress paths on the deformation and strength characteristics of coarse-grained materials, large-scale true triaxial isotropic consolidation and drained shear tests were carried out under two stress paths: constant minor principal stress 3 and intermediate principal stress coefficient b (constant 3 and b), and constant mean normal stress p and intermediate principal stress coefficient b (constant p and b). The results indicate that the stress path has a significant influence on the stress-strain relationship. As the intermediate principal stress coefficient b increases, the stress-strain curve under the stress path of constant 3 and b gradually ascends and becomes steeper, and the volumetric contraction deformation gradually enlarges. However, under the stress path of constant p and b, the stress-strain curve descends and becomes more gradual, and the volumetric change initially compresses and then expands, accompanied by an increase in the volumetric change rate, and the dilatancy deformation gradually increases. The intermediate principal strain undergoes a transition from dilation to compression as the intermediate principal stress coefficient, b increases from 0 to 0.25. When b exceeds 0.25 and reaches 1.00, the intermediate principal strain consistently exhibits compressive behavior, and the compressive deformation gradually increases. The stress path exerts a pronounced influence on the strength of coarse-grained material. Under the constant 3 and b stress path, the strength progressively enhances with the increasing intermediate principal stress coefficient, b. Conversely, a progressive diminution of strength is observed under the constant p and b stress path as b increases. The peak internal friction angle exhibits a positive correlation with the intermediate principal stress coefficient b, and the peak internal friction angle of the constant p and b stress path is higher than that of the constant 3 and b stress path. Additionally, lower consolidation stresses yield greater peak internal friction angles. The failure stress ratio exhibits a monotonic decrease with diminishing decrement rates as the intermediate principal stress coefficient b increases from 0 to 0.75, while lower consolidation stresses correspond to higher failure stress ratio. When the intermediate principal stress coefficient varies between 0.75 and 1.00, the failure stress ratio basically remains constant. The stress path exerts a significant influence on the deformation and strength characteristics of coarse-grained materials under three-dimensional stress states. Accurate understanding of these mechanical characteristics is crucial for scientific implementation of earth-rockfill dam engineering.

3D printing technology has demonstrated broad application prospects in rock laboratory experiments due to its advantages such as rapid prototyping and complex structure reproduction. However, the poor mechanical properties of printed specimens limit their practical engineering applications. To effectively enhance the mechanical properties of 3D printed sandstone-like specimens, this study employed GS19 type sand and furan resin-based sandstone-like specimens as research objects. Specimens were treated with different cycles of enzyme-induced calcium carbonate precipitation (EICP) solution infiltration. The evolution of mechanical properties was quantitatively analyzed through uniaxial compression tests, while scanning electron microscopy-energy dispersive spectroscopy and Fourier transform infrared spectroscopy were used to reveal the mechanical enhancement mechanism of EICP technology at microscopic scale. Results indicate that the compressive strength and elastic modulus of the specimens progressively increased with infiltration cycles. After four EICP solution treatments, the specimen strength increased by 51.92% and the elastic modulus increased by 35.57% compared to the control group, with failure modes and crack propagation characteristics resembling those of natural weakly cemented sandstone. Microscopic analysis revealed that EICP technology promoted the formation of “resin-calcium carbonate” composite cementation phases within specimens. This dual mechanism of enhancing intergranular cementation strength and pore-filling to reduce structural defects significantly improved mechanical performance. The findings expand the application potential of 3D printing technology in rock experimentation.

An alkaline water-enriched environment reduces the bearing capacity of surrounding rock, inducing uneven deformation, collapse, and roof fall in roadways, thereby threatening construction safety. Sandy mudstone specimens immersed in alkaline water environments were studied. The strength deterioration and pore structure evolution of these rocks after immersion in solutions with different pH values were investigated using uniaxial compression (UC) tests and nuclear magnetic resonance (NMR) tests. A correlation function between the pore fractal dimension (D) and uniaxial compressive strength Rc was established, revealing the mesoscopic pore structure evolution characteristics and macroscopic strength degradation mechanism of sandy mudstone in alkaline water environments. The results show that: (1) As pH increases, micropores (pore size r1≤0.01 m) and mesopores (0.01 m < r2≤1.00 m) in sandy mudstone gradually expand and develop into macropores (pore size r3>1.00 m), leading to a reduction in Rc. (2) The pore structure of sandy mudstone in alkaline water environments exhibits a multifractal characteristic, and its mechanical strength is closely related to the internal pore structure and distribution morphology. (3) With increasing pH, the number of axial fracture surfaces gradually increases, with numerous secondary cracks developing near the main fracture surface and pronounced spalling occurring. (4) The dissolution effect of alkaline water and the water-absorption swelling behavior of sandy mudstone are the fundamental mechanisms responsible for the evolution of mesoscopic pore structure and macroscopic strength degradation. The findings contribute to improving deformation control and disaster prevention for soft rock roadways in alkaline water environments.

Peaty soil, a distinct category of soft foundation soil, exhibits unique physical and mechanical properties that are strongly influenced by its microstructure. Its high water content, organic matter content, low strength and permeability often result in significant engineering challenges. Enhancing the mechanical strength of peaty soil has thus become a central focus in geotechnical engineering. Using slag-based geopolymer to synergize with cement for solidification, the mechanical properties of peaty soil before and after stabilization were examined through unconfined compressive strength and direct shear tests. The mechanisms of improvement were further analyzed through microscopic techniques, including scanning electron microscope (SEM), X-ray diffraction (XRD), mercury intrusion porosimetry (MIP), and Fourier transform infrared spectroscopy (FTIR). The results demonstrate that all three alkali activators contribute to the enhancement of the mechanical strength of the peaty soil, with NaOH showing the highest activation efficiency. Cement stabilization of peaty soil improves shear strength by reducing pore space and strengthening interparticle bonding via ion exchange, hydration product crystallization, and the formation of CaCO3 and calcium silicate hydrate (C-S-H). Four stages i.e., dissolution activation, ion exchange, gel formation, and structural reorganization are identified in the reaction process of activated slag improving peat soil. The alkali activator facilitates the dissolution of the slag’s vitreous phase, promoting ionic polymerization that leads to the formation of calcium-alumino-silicate-hydrate (C-A-S-H) gel. Simultaneously, organic functional groups in the peaty soil engage in ion exchange, forming CaSiO3 precipitates and establishing a “calcium bridge” structure. These reactions collectively contribute to the formation of a dense composite matrix, thus enhancing compressive strength. Grey relational analysis reveals that compressive strength is most strongly correlated with pore area, while shear strength shows the highest correlation with the shape factor. Modified soil specimens undergo five dry-wet cycles, with a minimum strength loss rate of 27%. These findings provide a theoretical foundation for the partial replacement of cement with alkali-activated slag in peaty soil stabilization, contributing both to soft soil improvement and the valorization of industrial byproducts. Furthermore, these results offer valuable insights for ground improvement in peat-rich regions, such as Yunnan, China.

Deep coal-rock masses exhibit pronounced creep characteristics. Under moderate dynamic impact loads (impact energy levels ranging from 103 J to 104 J), creep-affected coal-rock masses undergo time-delayed instability failure. This presents certain difficulties for the assessment and early warning of engineering hazards. To address this issue, a self-developed creep-impact testing system was employed to investigate the mechanical response characteristics of creep-prone coal under moderate dynamic impact loading. The study analyzed the influence of moderate dynamic impact loading on the creep failure of coal bodies and discussed the mechanism by which such loading accelerates the failure of creep-prone coal. Test results indicate: (1) Increasing the magnitude and frequency of moderate dynamic impact loads accelerates the transition of coal mass from isochronous creep deformation to accelerated creep deformation, while simultaneously reducing both the time required for coal mass to enter the accelerated creep stage and the corresponding stress threshold. (2) At low stress levels, coal exhibits only a hardening effect under creep conditions. However, under moderate dynamic impact loading, the coal demonstrates a hardening-damage effect, ultimately inducing a macroscopic failure mode dominated by tensile fracture. (3) The concept of “impact damage stress” in creep coal bodies was proposed, indicating that moderate impact dynamic loads accelerate damage and failure in creep coal bodies when stress levels reach a minimum of 0.4c. This stress state may be regarded as the “impact damage resistance strength” of creep coal bodies. The research findings provide theoretical support for the early warning and prevention of rockburst disasters in deep mines.

China’s annual hydraulic dredging produces a great amount of slurry with high water content, while yard disposal usually needs several years to complete the self-weight sedimentation. It is crucial to explore the intrinsic relationship between the evolution of self-weight sedimentation rate and the initial water content (w0), the liquid limit (wL) and the initial height of slurry surface (H0) for calculating the self-weight settlement accurately. By establishing the database of slurry surface height vs. time curves derived from laboratory cylinder tests, a mathematical model for the height of slurry surface-time was developed, and the changing law of intrinsic relationship between settlement and sedimentation rate with the initial state of slurry and H0 was studied. The study indicates that the sedimentation rate firstly increases and then decreases with the increasing sedimentation time. For slurry with higher w0/wL, a higher sedimentation rate is observed at the hindered settling stage, while a lower sedimentation rate is found at the self-weight consolidation stage. The time required for completing 90% to 95% of slurry sedimentation is less than 50% of the final settlement time. Meanwhile, the degree of sedimentation completion is introduced, an empirical relationship between the initial slurry state and sedimentation rate at different degrees of sedimentation completion is established, and then a method for calculating the slurry surface settlement is proposed. The calculated values of mud surface settlement are approximately 1.0−1.1 times the measured values. This method can calculate the settlement rate at different degree of sedimentation completion according to the measured settlement-time curve, the w0 and the wL, and estimate the stable self-weight settlement quickly and accurately, which offers a quantitative analysis for optimizing the design of storage capacity.

Based on the wave theory of unsaturated porous elastic media, a wave equation for unsaturated porous thermo-viscoelastic media is established by considering the viscosity of the soil skeleton and thermal effects, and utilizing the Kelvin-Voigt viscoelastic model, the equations of motion for unsaturated porous viscoelastic media, and the generalized thermoelasticity theory. By introducing displacement potential functions for each constituent phase, the dispersion characteristic equations for body waves in unsaturated porous thermo-viscoelastic media are derived. Numerical calculations are performed to analyze the influence of thermophysical parameters, such as relaxation time, thermal conductivity, medium temperature, specific heat capacity of the solid phase, and saturation, on the wave velocity and attenuation coefficient of thermo-viscoelastic waves. The results show that for every 0.5×10⁻³ s increase in relaxation time, the velocities and attenuation coefficients of the P1 wave increase by up to 2.93% and 44.51%, respectively; those of the S wave increase by up to 17.49% and 51.32%, respectively; while the velocity and attenuation coefficient of the T wave decrease by up to 25.4% and 20.3%, respectively. The thermal conductivity coefficient only affects the T-wave velocity and attenuation coefficient. For every 1 J/(m·s·K) increase in thermal conductivity, the T-wave velocity increases by 21.62% and its attenuation coefficient decreases by 6.83%. For every 20 K increase in medium temperature, the P1-wave velocity increases by approximately 0.3% and its attenuation coefficient decreases by approximately 0.2%. An increase in the specific heat of the solid phase causes the wave velocities of both the P1-wave and the T-wave to gradually increase. Saturation has a significant effect on the wave velocities of P1, P3, and S waves. When the saturation decreases from 0.99 to 0.4, the wave velocities of the P1 and S waves increase by up to 15.6% and 6.5%, respectively, and the wave velocity of the P3 wave increases by up to 4.4 times.

In the field of underground engineering, the traditional isothermal cycle scheme employed in TBM-assisted thermal damage rock breaking technology is hampered by high energy consumption, which severely limits its economic viability and future application. To address this challenge, investigating the rock degradation effect induced by a novel variable temperature-water cooling thermal loading method is of significant value for promoting energy efficiency and practical implementation of this technology. Through uniaxial compression, Brazilian splitting, and penetration tests, this study systematically investigated the evolution in penetrability of extremely hard granite subjected to three cycles of variable temperature-water cooling at different high temperatures (T1, T2, T3). The analysis focused on the evolution of rock failure patterns, morphology, compressive and tensile strength, and penetration rate. The results indicate that as the initial temperature T1 increases, thermally induced microcracks evolve from a sparse to a networked distribution, with damage initiated solely through intergranular crack propagation. The failure mode transitions from tensile splitting to oblique shear, accompanied by a brittle-to-ductile shift in failure morphology, with 500 ℃ identified as the critical transition point. Both the rate of strength reduction and the rate of penetration increase peak within the T1 range of 500−600 ℃, beyond which they gradually decelerate and stabilize. When T1 exceeds 600 ℃, the rate of penetration increase diminishes compared to that at 600 ℃, although the absolute penetrability remains superior to the room-temperature control group. While the second high-temperature stage T2 (200−600 ℃) further reduces rock strength and increases microcrack density and crushed zone volume, its influence on failure mode and morphology is less pronounced than that of T1. A significant enhancement in penetration rate is observed specifically when T2 is 200 ℃. Therefore, for TBM excavation in extremely hard rock formations utilizing variable temperature-water cooling assistance, a cyclic scheme with an initial high temperature of 600 ℃ followed by a subsequent temperature of 200 ℃ is recommended.

To investigate the deformation characteristics of loess under different stress paths, a bi-directional triaxial apparatus was used to perform constant q compression tests (q is deviatoric stress), constant p shear tests (p is spherical stress), and constant stress ratio tests. The cross-influence mechanism of spherical stress and deviatoric stress on volumetric strain and shear strain was systematically analyzed. The experimental results indicate that the spherical stress not only generates volumetric strain, but also results in radial strain being greater than axial strain (about 5.79 times) due to anisotropy, leading to negative shear strain; while the deviatoric stress generates shear strain, it is also accompanied by approximately 1%―3% shear-induced volumetric shrinkage. This indicates that the deformation of loess is jointly dominated by spherical stress and deviatoric stress, forming 4 types of cross stress-strain relationships: p-vp, q-vq, p-sp, and q-sq (where vp and vq are the volumetric strains caused by spherical stress and deviatoric stress, respectively; sp and sq are the shear strains caused by spherical stress and deviatoric stress, respectively). Given that traditional models are difficult to describe such cross relationships, this study extends the K-G model framework (where K is the bulk modulus and G is the shear modulus) by adding cross moduli J1 and J2 on the basis of the original bulk modulus Kt and shear modulus Gt, quantifying the volumetric strain caused by deviatoric stress and the shear strain caused by spherical stress, respectively. By introducing the stress ratio hardening coefficient, a four-modulus incremental nonlinear constitutive model capable of reflecting the anisotropy and shear shrinkage characteristics of loess is developed. By solving the model parameters and comparing them with experimental data, the accuracy of the model in predicting the mechanical response of loess under different loading paths was validated. The research provides new methods and theoretical foundations for a deeper understanding of the mechanical behavior of loess and its engineering applications.

To explore the suitability of soft rock residual soil as backfill material for geogrid-reinforced embankments under the humid and rainy climate region, this study investigated the drainage performance of geogrid-reinforced improved soft rock residual soil structures under different rainfall conditions. A large-scale rainfall infiltration model test device was designed to study the rainfall infiltration patterns, infiltration rates, and drainage efficiency of different drainage design schemes under conditions of short-term intense rainfall and short-term extreme rainfall and long-duration rainfall. The results show that: (1) The rainfall infiltration can be divided into three stages: unpressurized infiltration, pressurized infiltration, and saturated infiltration. These structures exhibit low infiltration capacity and poor drainage performance. The saturated hydraulic conductivity stabilizes around 0.28 mm/min. (2) Graded gravel drainage layers significantly improved the drainage performance of the soft shale residual soil. Medium gravel drainage layers perform best, followed by fine gravel, while coarse gravel perform relatively worst. The performance is closely related to the particle composition of the fill material. (3) At rainfall intensities of 20, 30 mm/h, and 40 mm/h, the graded gravel drainage layers increase drainage volume by 21.7%, 36.7%, and 34.6%, respectively. Replacing unidirectional geogrids with composite drainage geogrids increases drainage by 7.4%, 10.1%, and 4.89%, respectively. The drainage duration is extended by 1.5, 2.5 h, and 4.5 h. Both graded gravel drainage layers and composite drainage geogrids effectively improve the drainage performance of geogrid-reinforced soft rock residual soil structures.

To reveal the degradation mechanism of the mechanical properties of grout-treated joints in high-temperature environments, high-temperature treatments were applied to sandstone joint samples grouted with different water-cement ratios, followed by direct shear tests to analyze the effect of high temperature on the shear performance of grout-treated joints. The results indicate that an increase in grout strength is associated with higher peak shear stress and smaller displacement. In contrast, high-temperature treatment yields less noticeable post-peak softening, delayed failure, and increased plastic deformation. Both increased grout strength and roughness help improve shear strength. However, with increasing temperature, the cohesion of the rock-grout interface and the compressive strength of the rock face decrease non-linearly, indicating that the grout reinforcement effect weakens with higher temperatures. High-temperature-induced debonding and softening at the rock-grout interface are the two main mechanisms responsible for shear strength degradation. Based on the three-dimensional morphological parameters of Grasselli, a shear strength criterion for grout-treated joints considering the high-temperature effect is proposed. Comparison with experimental data and literature shows that this criterion can accurately predict the peak shear strength of grout-treated joints after high-temperature exposure, providing a theoretical basis for the stability evaluation of rock masses in high-temperature environments or after fire exposure.

Soft rock tunnels in red beds frequently encounter anchor failure issues. Improving the mechanical properties of the red-bed soft rock-grout interface is highly significant for ensuring the stability of anchor support. This study employs laboratory direct shear tests in combination with digital image correlation (DIC) technology to investigate the mechanical properties (peak shear strength p, residual shear strength r, and shear stiffness ks) of the grout-rock interface (GRI) among three typical soft rocks from red beds (red sandstone, mudstone, and grey sandstone) and two grout materials (ordinary Portland cement (OPC) and early high-strength cement (EHC)) under varying curing periods (6 h, 1 d, 2 d, and 7 d). The shear behavior and DIC-based failure mode are analyzed for red beds GRIs. The following key findings were observed: (1) As normal stress increases, both the shear strength (p) and shear stiffness (ks) of the GRI also increase, following the trend: red sandstone > mudstone > grey sandstone. The cohesion values of the red sandstone, mudstone, and green sandstone with the EHC grout-rock interface at 2 days are 2.4, 0.9, and 1.2 MPa, respectively. The corresponding internal friction angles are 57.6°, 38.0°, and 27.0°, respectively. (2) With an increase in curing age, both p and ks at the GRI increase non-linearly. EHC exhibited superior bonding performance compared to OPC, with p reaching 7.9 MPa at 6 h and 90% of the 7-day p being achieved at 2 d in red sandstone conditions. (3) The OPC-bonded specimens primarily exhibit grout failure near the GRI, whereas the EHC-bonded specimens show rock failure in red sandstone and mudstone conditions, and adhesive failure at the interface in grey sandstone. (4) An empirical model for the shear strength of the GRI in red bed soft rocks is proposed and validated, providing a framework for the rapid and reliable evaluation of anchorage strength in red bed soft rock tunnels.

Soft clay has high initial stiffness at very small strain. Its stiffness decreases nonlinearly with increasing strain and the degradation relationship depends on previous stress history. The above small strain stiffness characteristics play an important role in surrounding ground deformation induced by underground excavation. Based on that, the intergranular strain (IGS) elastic model and the critical state anisotropic boundary surface model are combined to establish an effective stress elastoplastic model considering the small strain stiffness characteristics of clay. The proposed model can memorize soil deformation history through IGS and characterize stress direction change through variation in the angle between IGS and strain increment. It can also realize the stress path-dependent small strain stiffness evolution simulation through path-dependent stiffness evolution equation. This work further proposes the convert method between the parameters of the IGS model and those of the Overlay model (i.e., the small-strain module of the widely used HS-Small model), while perform a comparative analysis between the two models. The IGS model is applied and examined by analyzing boundary value problem of deep excavation case studies. The results show that (1) the IGS model can reasonably reflect the influence of previous stress history on the small strain stiffness of clay and the ground deformation induced by excavation. (2) When the previous stress history amplitude is small, both IGS and Overlay models can effectively reflect the effects of previous history on the degradation relationship of small strain stiffness. (3) With the previous stress history amplitude increasing, Overlay model could underestimate the small strain stiffness under the stress path with a small angle from the previous path.

The issues of severe soil and water loss and weak erosion resistance in the Loess Plateau region are primarily attributed to the low strength and poor water stability of loess. Soybean urease-induced calcium carbonate deposition is a green and low-carbon biological soil stabilization technology, in which the calcium source critically influences the mechanical and hydraulic properties of the solidified soil. Three calcium sources, calcium chloride, calcium acetate, and calcium nitrate, were selected to prepare cementation solutions for treating loess, with a control group established for comparative analysis of the solidification effects. A series of macro- and micro-tests were conducted on the loess solidified with different calcium sources to clarify the development patterns and interrelationships among macro- and micro-indicators, including strength, failure characteristics, disintegration evolution, calcium carbonate content (CCC), mineral composition, and micromorphology, under varying cementation solution concentrations and curing periods. The results demonstrate that the unconfined compressive strength (UCS) of calcium nitrate-solidified soil reached 1 857.24 kPa, approximately 1.91 times higher than that of the control group. The calcium chloride-solidified soil exhibited the best disintegration resistance, with disintegration time extended by up to nearly 3.52 times. The UCS and disintegration time of the solidified soil showed linear and exponential relationships with CCC, respectively. The calcite crystals induced by soybean urease reshape the microstructure of loess through cementation, filling, and encapsulation, enhancing particle compactness and overall integrity, thereby significantly improving the strength and water stability of loess. This study provides robust support for selecting the optimal calcium source in soybean urease-solidified loess.

Current research on the mechanical properties of plant root systems in soil reinforcement is mostly limited to isolated root groups or localized root samples. To explore the mechanical mechanism of root-soil interaction in the entire root system, a 3D reconstruction model of the complete root system of Moso bamboo was developed through holistic modeling of the Moso bamboo root system. Using ABS-like photosensitive resin, a material possessing mechanical properties similar to real roots, a complete root system model was fabricated by 3D printing technology. Large direct shear tests were conducted to analyze the mechanical properties of the root-soil composite. The results indicated that the morphology and structure of the root system significantly influences the mechanical properties of the root-soil composite. At cross-sections with sufficient root density, the shear load can be collectively borne by the entire root system. The robust root structure facilitates the mobilization of the entire root network to resist shear loads. The root system can enhance both the cohesion and internal friction angle of soil. In granite residual soil, higher vertical stress promotes close contact between the roots and soil, resulting in an accelerated increase in shear strength. Based on the experimental results, an empirical formula for calculating the shear strength of the complete root-soil composite system was proposed. The findings provide scientific evidence for further elucidating the mechanical interaction between roots and soil and introduce a novel, controllable, and repeatable experimental method for studying soil reinforcement effects of roots.

In geotechnical engineering projects related to energy underground structures, nuclear waste disposal and storage, and landfill waste disposal, the thermal exchange between soil and structures can lead to significant changes in the volumetric deformation and shear characteristics of the soil. Establishing a reasonable constitutive model to reflect the thermo-mechanical coupling properties of the soil is of great importance. Firstly based on thermodynamic principles, this paper introduces a new dissipation function and free energy function, deriving a yield surface equation that includes three shape parameters. Further considering the thermo-elastic strain of saturated clay, the pre-consolidation pressure, and the thermal-dependent behavior at the critical state, a thermo-mechanical coupled bounding surface constitutive model for saturated clay is established using a non-associated flow rule. Finally, the predictive capability of the model was validated through heating-cooling cycle tests and temperature-controlled triaxial drained/undrained compression tests. The results indicate that within the temperature range of 0−95 ℃, when the overconsolidation ratio is no greater than 12, the model can reasonably reflect the thermo-mechanical coupling effects in saturated clay, with the strain hardening and volumetric contraction of normally consolidated soil under temperature effects, as well as the strain softening and dilatancy characteristics of over-consolidated soil, being adequately described.

The global cement industry contributes approximately 8% of carbon dioxide emissions. Utilizing low-carbon wastes such as red mud, steel slag, and fly ash to replace cement is of great significance for carbon neutrality. To address the issue of their low reactivity in geotechnical engineering applications, this study proposes an alkaline-sulfate synergistic activation method. Techniques such as backscattered electron spectroscopy and X-ray diffraction were employed to reveal the microstructural characteristics and reaction kinetics. Experimental results indicate that the concentration of alkaline solution, temperature, dissolution time, solid-to-liquid ratio, and sulfate concentration significantly affect the dissolution of Si, Al, and Ca from the waste materials. Kinetic analysis confirms that the dissolution process follows the internal diffusion mechanism of the shrinking-core model. Among the elements, Si has a lower activation energy under alkaline conditions, resulting in higher dissolution reactivity, while Ca has a higher activation energy, making its dissolution rate more temperature-sensitive. Molecular dynamics simulations show that in the NaOH-Na2SO4 system, Si and Al exist in the form of and Al(OH)4⁻, respectively. This study provides a theoretical foundation for the development of low-carbon geotechnical materials and supports the low-carbon transformation of the construction industry.

The dissolution process induced by reactive fluid flow in rock fractures significantly impacts geotechnical engineering safety, while the dissolution mechanisms of inclined fractures under gravitational effects induced by solution density differences remain unclear. This study systematically investigates the dissolution patterns and permeability evolution of fractures with different inclinations through pore-scale numerical simulations and visualization experiments. Results indicate that buoyancy-driven convection caused by solution density differences generates “vortex-like” flow structures during inclined fracture dissolution, where buoyancy convection along the fracture length dominates channel development, whereas gravitational effects in the vertical direction can be negligible. A criterion for the transition of dissolution patterns was established using the Richardson number (RiII): When RiII > 10 (buoyancy-dominated regime), increased inclination promotes wormhole growth and reduces the required injection volume for breakthrough; when RiII ≤ 10 (forced convection-dominated regime), dissolution patterns are governed by injection velocity, manifesting as compact dissolution, wormhole dissolution, or uniform dissolution. A theoretical model for optimal injection velocity incorporating gravitational effects was developed, enabling accurate predictions across varying inclinations. This research provides a theoretical guidance for seepage control in underground engineering within soluble rock formations and offers critical insights for safety assessments in CO₂ geological storage, in-situ leaching mining, and related applications.

A rockburst is a dynamic disaster induced by excavation and multi-source dynamic disturbances, characterized by the sudden release of accumulated elastic strain energy in deep engineering rock masses, leading to the bursting and ejection of surrounding rock. In line with the national strategy of building a strong transportation network, field measurements were conducted to capture the vibration characteristics of key tunnel boring machine (TBM) components and the surrounding rock during the construction of a deep-buried TBM-driven railway tunnel. The TBM vibration exhibits a significant surging characteristic, with particularly pronounced vibrations at the shield tail. The measured vibration acceleration ranges from −10g to 10g. At the thrust shoe, the primary vibration direction is parallel to the tunnel axis, with a vibration velocity of 1.5–2.5 cm/s and a frequency of 25–35 Hz. As the distance from the tunnel face increases, the peak particle velocity (PPV) and peak particle acceleration (PPA) of the surrounding rock exhibit a general attenuation trend in the axial, radial, and vertical directions. Accordingly, true triaxial tests on high-pressure hard rock under wide low-frequency disturbances were conducted to examine the effects of TBM low-frequency disturbance on the mechanical properties and disturbance-induced damage behavior of granite. The results demonstrate that the disturbance strength of granite decreases as the stress amplitude increases. For the area affected by the TBM thrust shoe, the disturbance strength increases with both frequency and minimum principal stress. Furthermore, low-frequency disturbance applied in the direction of the minimum principal stress promotes the development of tensile fractures, thereby reducing the bearing capacity of the rock.

In this study, a cyclic dynamic damage constitutive model (referred to as the Damage model) was developed within the finite difference software FLAC3D by incorporating the Mazars damage variable into the Drucker-Prager (D-P) constitutive model. Using this proposed model, a numerical model of a subway station in an inclined liquefiable site was established. Dynamic analyses were conducted under various working conditions, considering different earthquake intensities and ground inclinations, to investigate the damage distribution and evolution characteristics of the subway station. The results indicate that the damage value increases with increasing earthquake intensity and ground inclination, and damage is likely to occur at the connections between walls, slabs, and columns. In an inclined liquefiable site, the subway station experiences simultaneous uplift, sliding, and rotational deformation due to the horizontal movement of the liquefied soil layer and the action of earth pressure. When the peak ground acceleration is 0.3g and the surface inclination angles are 1°, 2°, and 3°, the inter-story drift ratios of the subway station are 1/294, 1/234, and 1/217, respectively. Furthermore, the damage value on the earth-facing side wall is significantly greater than that on the opposite side wall, indicating that the station structure is more prone to damage in inclined liquefiable sites than in horizontal liquefiable sites. The comparative analysis of the liquefaction zone distribution, deformation, and damage characteristics of the subway station provides a theoretical basis for the design and construction of subway stations in inclined liquefiable sites.

Submarine landslides often occur on gentle cohesive slopes with inclinations of less than 5°, which cannot be adequately reproduced using traditional limit equilibrium methods. The shear band propagation method, considering the strain-softening behavior of cohesive soils, provides a rational explanation for such failures. Previous shear band propagation methods mostly focused on planar or symmetrical S-shaped slopes, where the symmetric propagation of shear bands is simplified as a unidirectional problem. However, most natural slopes are characterized by asymmetric profiles. A theoretical method for asymmetric shear band propagation is proposed, where the evolution of shear stress and related variables is solved based on deformation compatibility along the entire slip surface to assess slope stability. Using the asymmetric Gaussian slope as an example, the progressive failure process triggered by local disturbances is investigated. The results show that when the shear band length exceeds a critical value, catastrophic failure occurs. As the curvature variation of the slope decreases, the critical gravity shear stress at the steepest point of the potential slip surface approaches the peak undrained shear strength of soils. Based on these findings, a simplified failure criterion is proposed and its general applicability and accuracy are verified through finite element analysis. A case study of the St. Niklausen landslide in Switzerland is conducted to evaluate the prediction performance/predictive capability of different methods. Limit equilibrium methods and conventional symmetric shear band propagation methods that approximate the slope as a symmetric shape overestimate slope stability. The proposed asymmetric analysis method can reasonably reproduce the occurrence of the St. Niklausen slide, with predicted safety factors and shear band locations closely matching finite element results. The developed simplified failure criterion provides quick stability evaluations with a conservative bias.

In landslide stabilization engineering, discretely arranged anti-slide piles establish a continuous resisting surface by mobilizing horizontal soil arching effects between adjacent piles. To elucidate the distribution mechanism of landslide thrust under the inter-pile horizontal soil arching effect and establish a rational calculation method, a finite element analysis was first conducted to investigate the transmission and distribution characteristics of landslide thrust. The results demonstrate that a portion of the landslide thrust continues to propagate into the soil ahead of the arch under the soil arching action. Beyond a critical depth, the load transfer coefficient toward adjacent anti-slide piles stabilizes, indicating a convergence in load redistribution patterns. Subsequently, based on the Mohr-Coulomb strength criterion, the position of the failure surface at the arch foot is determined. Following the principle of prioritizing load distribution to the adjacent anti-slide piles, the height-to-span ratio of the soil arch is established. A calculation method for load transfer coefficient of landslide thrust to the piles under inter-pile horizontal soil arching is then established using the limit equilibrium method. The validity of this method is verified through comparative analysis with results of numerical simulation and model test. Finally, a parametric analysis was performed by selecting the cohesion and internal friction angle of the soil and the pile spacing. The results indicate that when the pile spacing is less than 3 times the pile diameter and the soil cohesion and internal friction angle are sufficiently high, the landslide thrust behind the piles can be considered entirely transferred to the anti-slide piles. The findings provide a theoretical basis for determining the landslide thrust on anti-slide piles, thereby improving the accuracy of pile internal force and deformation analysis as well as the reliability of engineering design.

Rockburst is one of the critical scientific challenges demanding urgent resolution in deep mineral resource exploitation and underground engineering construction. Based on elastic strain energy and the rock integrity failure criterion, combined with the secondary stress distribution of surrounding rock after deep-buried tunnel excavation, this study establishes a multi-factor integrated criterion model for evaluating rockburst propensity. Results demonstrate that the proposed criterion systematically synthesizes key parameters including rock mass tangential and radial stresses, rock integrity, Poisson’s ratio characteristics, and secondary stress redistribution effects, while specifically integrating three essential factors governing rockburst initiation: stress concentration state, structural integrity condition, and material brittleness property. Based on the intensity variation of rockburst activity, the classification system is established with four distinct grades: none, weak, moderate, and intense, with three classification thresholds determined at 2.5, 8.1, and 25.7, respectively. The sensitivity characteristics of each parameter were also discussed. The results show that the ratio of tangential stress to compressive strength of the rock mass has the greatest influence on the classification of rockburst intensity. When this ratio exceeds 0.75, intense rockburst is more likely to occur; when it approaches 0.40, moderate rockburst is more likely to occur. Empirical investigations incorporating case studies of representative deep tunnel projects across China demonstrate that the criterion’s predictive outcomes exhibit strong alignment with actual rockburst manifestations while maintaining robust consistency with classical rockburst criteria, thereby demonstrating enhanced predictive accuracy and broad applicability. The developed energy-stress synergistic rockburst criterion offers an innovative methodology for strain-type rockburst prediction in deep tunneling engineering.

In existing studies on reinforced ballast beds, dynamic loads are predominantly applied as fixed-point excitation. This approach fails to replicate the stress deflection effect generated in substructure elements by actual moving train loads. Moreover, the subgrade structure is frequently omitted in such studies, resulting in an unclear understanding of how reinforced ballast beds influence the dynamic response of the subgrade. Therefore, this study employs a coupled discrete element method–finite difference method (DEM–FDM) to establish a ballast–subgrade coupled numerical model. The model is used to comparatively investigate the effects of ballast reinforcement on the responses of both ballast particles and the subgrade under moving train loads. The results indicate that ballast reinforcement increases the diffusion angle  beneath the sleeper by approximately 7º, thereby promoting the extension of force chains further into the inter-sleeper zone. This effectively mitigates stress concentration within the ballast bed. The increase in the total energy dissipation of the system after reinforcement is primarily supplied by damping energy, while the reduction in frictional energy helps alleviate wear among ballast particles. Since reinforcement enhances the overall energy dissipation capacity of the ballast bed, the dynamic stress at various depths within the subgrade is reduced. Specifically, the peak dynamic stress at the subgrade surface and the bottom of the subgrade bed is decreased by 13.4% and 2.2%, respectively. Compared with an unreinforced ballast bed, the variation in the rotation angle of the principal stress axis within the subgrade is reduced after reinforcement. This reduction helps alleviate subgrade settlement issues induced by the rotation of the principal stress axis.

Under the influence of ground motion loads, saturated soil is susceptible to liquefaction, leading to secondary disasters such as ground settlement, foundation failure, and sand ejection with water seepage. When performing dynamic response analysis of saturated media using the traditional central processing unit (CPU)-based serial numerical calculation method, the large number of coupled model coefficient matrices and the strong coupling between equations result in issues such as low computational efficiency and precision. To address the numerical computing bottleneck associated with ultra-large-scale complex fluid-structure coupling models under the traditional numerical framework, this study proposes a computational framework based on CPU and graphics processing unit (GPU) heterogeneous parallelism. By optimizing the parallel computing process for large-scale data, minimizing data transfer between the CPU and GPU, and employs a data packaging strategy to reduce unnecessary time consumption. By directly constructing the global matrix in compressed sparse row (CSR) compression format through a preprocessing non-assembly method, it avoids the significant memory consumption typically incurred by traditional assembly methods. This enables the program to compute ultra-large-scale finite element models with reduced memory resources, thereby lowering both memory and time costs. Leveraging an intrinsic CUDA function, a novel set of iterative solution schemes for equations was developed, and a parallel iterative solver tailored for multi-field coupling problems was constructed. For complex fluid-structure interaction dynamic response problems, the proposed parallel solver demonstrates an order of magnitude improvement over traditional serial computation methods. The computational capacity threshold surpasses tens of millions of degrees of freedom. Compared to ABAQUS software, the computational efficiency has increased by more than 15 times.

To address the limitation of traditional machine learning methods, which primarily focus on text data and lack the capability to recognize and analyze image data, this study proposes an improved densely connected convolutional networks (DenseNet) based soil layer classification model using key parameter curve images from cone penetration test (CPT) data. First, key parameter curve images were generated from CPT data and compiled into a dataset. Second, the Optuna optimization framework and the squeeze-and-excitation (SE) attention module were integrated into the DenseNet model. Evaluation metrics including loss function, accuracy, and receiver operating characteristic curve (ROC) were adopted to assess model performance. Finally, the improved DenseNet model was applied to practical engineering projects to validate its generalization capability. The results show that the proposed model achieved a recognition accuracy of 0.920 9 on the self-built CPT image dataset from the Yellow River alluvial plain in Shandong Province, demonstrating high accuracy and strong robustness. Compared with current mainstream deep learning models and the baseline DenseNet, the improved model exhibited superior performance in soil layer identification. The model was further validated using data from 50 boreholes across five regions (Binzhou, Dezhou, Dongying, Heze, and Liaocheng), achieving a stratification accuracy exceeding 0.82 in all cases. Compared with conventional dual-bridge CPT classification charts, the improved model demonstrated clear advantages. The proposed method offers an effective solution for soil layer classification and provides valuable insights for future research in this field.

Obtaining in-situ mechanical properties of lunar soil is a prerequisite for human scientific research and construction on the lunar surface. This paper systematically reviews the principles, functions, advantages and disadvantages of various in-situ detection methods, and summarizes the research achievements, existing problems and current development trends of related technologies in in-situ mechanical characterization technology employed in the Soviet Union’s Luna and the United States’ Apollo missions. Focusing on the research status of in-situ mechanical characterization technology for lunar soil in China, it elaborates on the preparatory research work carried out by relevant research institutions in the in-situ mechanical detection of lunar soil and extraterrestrial soil based on static cone penetration method. Due to the unique physical properties of lunar soil, extreme lunar environments, and constraints imposed by limited resources, achieving in-situ mechanical detection of lunar soil and precise exploration and interpretation at depths exceeding 3 m still requires further technological innovation and breakthroughs. In accordance with the development needs of China’s lunar exploration program, this paper identifies the critical technical challenges and key development directions that require urgent attention for in-situ mechanical detection of lunar soil. The future in-situ mechanical detection payloads for lunar soil will develop towards miniaturization, automation and intelligence, forming novel operation modes based on new principles and methods. At the same time, innovations will be made in structural design and machine learning technology will be integrated. AI will be empowered to establish precise interpretation methods for mechanical parameters suitable for lunar soil, thereby providing reference and technical support for the successful execution of China’s in-situ lunar soil mechanical exploration missions and achieving more accurate and deeper exploration objectives.