The deficiency of artificial sand in medium-sized particles leads to poor workability. A solution is the partial substitution of medium-sized artificial sand with natural sand particles of the same size range. Therefore, the effect of the particle size of replaced particles on the mechanical behavior of sand samples is an issue worth studying. This study investigates the substitution size effect (SSE) for the first time, defined as the influence of substituted particle size on the mechanical behavior of particles mixtures. Sand mixtures were prepared by substituting artificial sand particles of different size ranges with spherical glass beads of corresponding sizes, while keeping particle shape and grading parameters constant. Three pairs of mixtures with varying substitution sizes were subjected to direct shear tests, supported by macro- and microscale analyses. The findings confirm and quantitatively characterize the SSE and identify particle orientation anisotropy as a key factor in its underlying mechanisms. This mechanism, explored through the lens of interlocking mechanisms via visualization of discrete-element method (DEM) simulation, offers the potential to provide valuable insights into how alternative sand can be effectively used to substitute partial artificial sand particles as a new aggregate with better workability in the future construction industry.

This study investigates the creep behavior of sandstone under various damage conditions through true triaxial tests and submicroscopic mineral degradation analysis. X-ray diffraction (XRD) and scanning electron microscopy (SEM) were employed to characterize submicroscopic mineral degradation, while creep rate, failure modes, long-term strength, and yield strength were determined from true triaxial creep tests. The results show that in its pristine state, sandstone's creep failure is primarily influenced by clay minerals. After reaching the long-term strength threshold, plastic fractures initiate within these minerals and propagate stably. As stress increases, transverse fractures occur, and at elevated temperatures, sandstone experiences microstructural compaction, reducing yield strength. Under the combined effects of high temperature and freeze-thaw cycles, quartz exhibits hydrophobic properties, while other minerals remain hydrophilic, leading to microcrack formation mainly along feldspar. These microcracks coalesce into shear cracks, significantly degrading long-term and yield strengths. The interaction between feldspar, calcite, and acidic hydrogen ions triggers pore invasion, shifting the primary crack propagation mechanism and further weakening the sandstone. A fractional derivative viscoplastic (FDVP) model, incorporating the Mogi yield criterion, was developed and demonstrated a good fit with the experimental data.

Tunnel face instability prediction represents a critical technical challenge in underground engineering, particularly during tunnel boring machine (TBM) excavation under complex geological conditions. This study proposes the Multi-modal Gaussian Cross-Attention Fusion (MGCAF) algorithm, which integrates physics-constrained Gaussian processes with cross-attention mechanisms to achieve intelligent tunnel collapse prediction. The MGCAF framework reconstructs the traditional prediction paradigm by treating earth pressure balance chamber pressure as the primary prediction target rather than an input parameter, while incorporating first-principles constraints of TBM cutting mechanisms into kernel function design. The algorithm employs a dual-pathway architecture that fuses TBM operational parameters through temporal modeling, processes geological radar images via deep feature extraction, and achieves cross-modal information fusion through physics-constrained cross-attention mechanisms. Dynamic kernel optimization enables real-time adaptive parameter adjustment through multi-source gradient feedback. Validation results based on the Yinsong Water Diversion Tunnel project (20 km length, 9 collapse events) demonstrate that the algorithm achieves high-precision prediction with R-2 = 0.8330, successfully predicting major collapse locations with approximately 20-m accuracy. Comparative analysis against baseline methods (Transformer, Gaussian Process, Random Forest, XGBoost) indicates that MGCAF exhibits superior performance in engineering reliability (0.95) and ROC-AUC (0.765) metrics. Generalization testing on the 2025 Los Angeles Wilmington Sewage Outfall Tunnel confirms the algorithm's cross-domain applicability. Ablation experiments reveal that the cross-attention mechanism serves as the primary performance driver, while uncertainty quantification provides interpretable risk assessment for TBM operations in heterogeneous geological environments.

Cemented clay is commonly used in ground improvement, but its mechanical properties are challenging to characterize due to the combined effects of cementation and fabric. This study presents a hypoplastic constitutive model that incorporates cementation strength and anisotropic fabric to simulate the mechanical response of cemented clay under both monotonic and cyclic loading. A transformed stress technique is adopted to modify the nonlinear failure criterion. A cementation state variable is introduced to capture its influence on soil strength and stiffness, together with evolution laws describing the degradation of cementation and structure. In addition, a structural factor combining structural degree, overconsolidation ratio, and cementation strength is introduced to characterize structure-related effects induced by stress history. Based on anisotropic critical state theory, a fabric tensor is incorporated to represent consolidation history and fabric evolution. The model is further extended to cyclic loading by incorporating intergranular strain improvement. Overall, the proposed model successfully captures the combined effects of cementation, structure, overconsolidation, anisotropy, and cyclic loading on the mechanical behavior of cemented clay. Comparisons between model simulations and experimental data demonstrate the model’s accuracy and predictive capability.

This paper presents an extended hypoplastic constitutive model for describing the mechanical behaviors of cemented clay. The model introduces state variables representing cementation and structural degree to capture the structural effects induced by cementation. A coupled evolution law is proposed to govern the simultaneous degradation of cementation and structure. Additionally, a novel structural factor is developed to characterize the strain-softening behavior of cemented clay. The mathematical formulation is relatively concise and involves only eight parameters. These parameters can be readily calibrated using conventional triaxial and isotropic compression tests. The model is validated by comparing its predictions with results from isotropic compression and triaxial experiments. The comparison shows that the model accurately reproduces the experimental behavior of cemented clays across a range of cement contents and confining pressures. Overall, the proposed model serves as a reliable tool for numerical analysis, preliminary design, performance evaluation, and stability assessment in geotechnical engineering projects involving cemented clays.

The prominent multi-scale pore heterogeneity widely developed in deep-buried coal reservoirs, which seriously restricts accurate reservoir characterization and precise resource evaluation for deep coalbed methane exploitation. Taking deep-buried coal reservoirs of the Benxi Formation in the Ordos Basin as the research target, this study integrates data from low-temperature CO2/N2 adsorption and high-pressure mercury intrusion experiments. Segmented monofractal quantification and multi-fractal singularity analysis are further adopted. Fractal differentiation characteristics, scale effects of pores at different scales, and the synergistic control mechanism of multiple geological factors were indicated. Eight segmented single-fractal dimensions (D1~D8) are defined. The results indicate an obvious scale-dependent zonal distribution of pore heterogeneity in deep-buried coal. Micropores smaller than 1.2 nm possess the largest fractal dimension and the strongest heterogeneity. They act as the primary adsorption and storage space for coalbed methane. Mesopores ranging from 1.1 nm to 8 nm have the lowest fractal dimension with the most homogeneous structure, serving as major gas migration pathways. The heterogeneity of macropores gradually increases with pore size. A continuous full-scale pore size distribution curve is reconstructed. Coal reservoirs exhibit a typical bimodal pore structure dominated by micropores and macropores, with the micropore-dominated storage and macropore-dominated seepage. Key multi-fractal indicators including Δα, Δf and the Hurst index are used for quantitative comparison. Micropores display strong aggregation and weak interpore connectivity. Mesopores own superior connectivity, while their heterogeneity differs greatly between individual samples. Macropores feature moderate aggregation and connectivity. Coalification degree, organic macerals, clay minerals and industrial parameters are associated with the formation and differentiation of pore heterogeneity. Each factor differentially regulates the fractal evolution of pores across various scales.

Understanding the coupled thermo-mechanical behavior of anisotropic rocks under true triaxial stress states (i.e., thermo‑anisotropic‑mechanical, T‑A‑M) is essential for deep geological engineering. This paper develops a statistical damage constitutive model that integrates temperature effects, bedding anisotropy, and true triaxial loading within a unified framework. Based on the Lemaitre strain equivalence hypothesis, the Weibull distribution, and the SMP failure criterion, a novel anisotropic factor is proposed that explicitly captures the intrinsic matrix anisotropy, bedding-plane shear weakening, the U-shaped influence of the intermediate principal stress, and the suppression of anisotropy by confining pressure. Using the defective volume ratio concept, a nonlinearly coupled total damage variable is derived, requiring only a few physically meaningful parameters that can be calibrated from conventional triaxial tests. Extensive experimental validations under varying temperatures, confining pressures, deposition angles, and intermediate principal stress parameters demonstrate excellent agreement between theoretical predictions and experimental data. The model accurately reproduces temperature-induced strength degradation, U-shaped strength anisotropy, the non-monotonic intermediate principal stress effect. Parametric analyses further reveal the evolution of anisotropy and damage with strain and stress state and the brittle‑to‑ductile transition. The proposed model provides a robust, physically transparent tool for the design and safety assessment of rock engineering under complex thermo‑mechanical conditions.

Tunnel boring machine (TBM) rock breaking parameter optimization is a technical challenge in underground engineering. Traditional numerical simulation methods have limitations in computational efficiency and accuracy when dealing with multi-parameter coupling optimization under complex geological conditions. This study proposes a deep learning method based on the Ada-Attention mechanism for predicting and optimizing parameters such as confining pressure, penetration depth, and cutting tool spacing in TBM rock breaking processes. The method employs a hybrid attention architecture that combines global window mechanisms with local sliding windows, reducing the computational complexity of traditional self-attention mechanisms from O(n²) to O(n(w+α)). Additionally, the Newton-Gauss optimization algorithm is introduced to improve the softmax normalization process, enhancing numerical stability and convergence performance. The research constructs a prediction framework covering a temperature range from 25°C to 500°C, using 800 experimental samples for model training and validation. Experimental results show that the Ada-Attention model achieves R² values of 0.92, 0.93, and 0.94 for torque, rolling force, and specific energy predictions respectively, obtaining 2–10 times computational speedup compared to traditional Transformer architectures. Generalization capability validation demonstrates that the model exhibits high prediction accuracy in soft sedimentary rock and medium sandstone (R²>0.95), maintains moderate performance levels in hard limestone and crystalline rock (R²=0.80–0.90), while prediction accuracy decreases in complex geological environments such as composite formations and fractured rock masses. This method provides a feasible technical solution for TBM parameter optimization under complex geological conditions.

Global urbanization has led to massive generation of high-water-content waste slurry, creating serious environmental challenges. Conventional treatment methods are costly and unsustainable, while cement-based foamed lightweight soils typically exhibit low strength and limited CO2 sequestration. To address this issue, this study proposes a novel stabilization pathway by integrating a MgO–mineral powder–carbide slag composite binder with CO2 foaming–carbonation. The approach enables simultaneous slurry lightweighting, strength enhancement, and CO2 fixation. A series of laboratory tests were conducted to evaluate flowability, density, compressive strength, and deformation characteristics of the carbonated lightweight stabilized slurry. Microstructural analyses, including SEM and XRD, were used to reveal the formation of carbonate phases and pore structures. The results showed that MgO content strongly promoted carbonation, leading to denser microstructures and higher strength, while mineral powder and carbide slag optimized workability and pore stability. Orthogonal testing indicated that a mix with 25% mineral powder, 12.5% MgO, and 7.5% carbide slag achieved the best performance, with unconfined compressive strength up to 0.48 MPa after carbonation. Compared with conventional cement- or GGBS-based foamed lightweight soils, the proposed system exhibits superior strength development, improved pore stability, and enhanced CO2 sequestration potential. These findings demonstrate the feasibility of recycling high-water-content waste slurry into value-added construction materials while contributing to carbon reduction targets. This study not only provides a sustainable solution for waste slurry management but also offers new insights into the integration of CO2 mineralization into geotechnical engineering practice.

Rainfall is widely recognized as the dominant trigger of red-bed slope failures, yet most previous studies focus on slopes that have already experienced complete collapse. However, relatively few studies have investigated slopes undergoing progressive deformation without full failure, and potential threshold values for rainfall-induced weakening remain poorly constrained. This study analyzes the deformation mechanism of a red-bed highway slope in Niuchang Town, Zhenxiong County, Yunnan Province, China, through field monitoring, laboratory ring shear tests, and time-series comparison of rainfall and displacement. Inclinometer data revealed that displacement was mainly concentrated in the upper 10 m, consistent with ring shear test results showing more significant shear strength reduction at lower normal stresses (20–200 kPa, equivalent to 1–10 m depth). A time-series comparison between rainfall and surface displacement (0.5 m) further demonstrated their clear temporal correspondence, confirming rainfall infiltration as the dominant trigger of shallow deformation. Laboratory tests also showed that when water content reached about 30

Landslides are essentially processes of granular materials undergoing flow instability under shear stress, where variations in the viscosity of the granular system directly control the initiation, cessation, and acceleration of slope movement. Therefore, investigating the jamming phenomenon in granular assemblies is of great significance for understanding landslide mechanisms and the shear behavior of granular materials. In this study, a signal analysis approach combining fast Fourier transform and empirical mode decomposition was applied to ring shear tests, aiming to quantitatively characterize the transitions between jammed and unjammed states from a frequency-domain perspective. Experimental investigations on glass bead assemblies with different particle size distributions reveal that the density of the contact network and the inertial effects of particles govern the frequency ranges of shear-dominant mechanisms. For example, in specimen S-2, the dominant frequency of IMF4 was identified as 0.22 Hz, corresponding to large-scale periodic reconstructions of the force-chain network, which marks the transition of the system from an unjammed to a jammed state. This study not only uncovers the frequency-dependent evolution of shear-dominant mechanisms in granular systems with varying particle sizes but also introduces the concept of a critical frequency as a criterion for distinguishing between jammed and unjammed states. The proposed index provides new quantitative insight for predicting landslide failure and early warning of geological hazards, while also offering guidance for the stable control of industrial granular flows.

The corrosive effects of acid rain have a profound impact on the construction of underground tunnels, the extraction of water hydrocarbon compounds, resource development, and the long-term preservation of stone cultural artifacts. A hydro-chemo-mechanical constitutive modeling of cemented granular materials was developed using the Drucker-Prager strength criterion and the Weibull distribution as a basis. Three components of the granular material during chemical corrosion were focused on: depositional bond (DP) dissolution, diagenetic bond (DG) dissolution and grain dissolution. Under acidic conditions, the specimens exhibited progressively greater chemical damage with decreasing pH. Conversely, in alkaline solutions, damage progression was mitigated by the formation of new reaction products that filled surface fissures and pores. The chemical factor (Dc) was determined by considering the initial concentration and time of hydrogen and hydroxide ions. The degree of acidic damage or alkaline enhancement was obtained by calculating the chemical variables for chemical aqueous solutions at pH 3, 5, 7, 9 and 11, and the data showed that the acidic solution at pH 3 had the highest degree of deterioration. In addition, the constitutive model was compared with experimental data obtained by earlier researchers. The results show that the stress–strain behaviour of cemented granular materials, particularly sandstones, under different pH and peritectic conditions can be predicted by using a hydro-chemo-mechanical constitutive model incorporating chemical factors. This theoretical framework provides a valuable reference for geotechnical engineering in the presence of chemical corrosion.

The ongoing global energy shift from fossil fuels to renewable sources highlights the importance of underground hydrogen storage (UHS) as a sustainable mechanism to counterbalance the seasonal inconsistencies of renewable energy. Comparable to geological CO2 sequestration, UHS necessitates precise subsurface imaging and a thorough understanding of hydrogen behavior within geological formations, including its location and migration patterns. Since direct subsurface measurement is unattainable, there is a pronounced need for robust subsurface characterization and monitoring techniques. This research focuses on UHS feasibility in enduring storage scenarios, employing seismic monitoring strategies adapted from CO2 storage practices. Specifically, we integrate Continuous Active Source Seismic Monitoring (CASSM) and Time-Lapse Full Waveform Inversion (TLFWI) to scrutinize UHS. The study employs synthetic crosshole surveys that initially model the coexistence of hydrogen and water within fractured reservoirs, aiming to pinpoint hydrogen-related velocity alterations in acoustic waveforms. Subsequently, we utilize a time-lapse synthetic model of hydrogen injection to emulate CASSM observations. Utilizing TLFWI in conjunction with White's model, we successfully process the CASSM data to discern and quantify temporal variations in velocity and saturation, which effectively maps the spatial and temporal distribution of mobile hydrogen. The results affirm the capabilities of TLFWI and CASSM as proficient monitoring systems for UHS. This study aspires to offer a dependable methodology for the administration of large-scale and long-term geological hydrogen storage.

Cryoablation, as an important method for tumor treatment, possesses the advantages of safety, efficiency, and minimally invasive characteristics. However, the phenomenon of intracranial pressure fluctuation caused by brain tumor cryoablation has not received sufficient attention. Investigating the mechanism behind the interaction between intracranial temperature and pressure may contribute to addressing this issue. Considering the coupling effect of temperature and confining pressure, the evolution equation and constitutive model of tumor mechanical damage after freezing were constructed to evaluate the influence of freezing on mechanical properties and damage law of biological tissues in a low temperature environment. Based on the Lemaitre's strain-equivalence principle, the microelement strength of tumor ice body under the coupling of low temperature and confining pressure is assumed to follow a Weibull distribution. The thermal and mechanical damage variables are introduced using continuous damage mechanics and statistical theory, and the microelement failure is in accordance with the SMP criterion. Adopts the numerical simulation of intracranial glioma cryoablation COMSOL temperature - mechanical coupling response mechanism, by numerical simulation and theoretical derivation method to obtain the required constitutive model parameters, the theory of stress-strain curve compared with simulated curve, accord well with those of two kinds of curve. The results show that the established damage statistical constitutive model can accurately reflect the stress-strain characteristics of biological tissues after freezing, and verify the rationality and reliability of the model and its parameter determination method.

Under the influence of external factors such as high temperature, chemical corrosion, and loading-unloading cycles, micro-cracks and pores may develop within rock structures. Continuum damage mechanics (CDM) characterizes these micro-defects arising within rocks, investigating the physico-mechanical behaviors of rocks in damaged states. This study, grounded in the theoretical framework of CDM and utilizing the Mohr-Coulomb failure criterion, integrates different probability distributions and damage coupling methods to develop three statistical damage models. The investigation primarily explores the effects of strength criteria, probability functions, and damage coupling mechanisms on the statistical damage models. In addressing the complex interplay between rock micro-defects and external stressors such as high temperatures and chemical corrosion, this study leverages the principles of CDM within the Mohr-Coulomb criterion framework. It advances the field by developing three innovative statistical damage models, enriched by diverse probability distributions and damage coupling methods. The research delineates the substantial impact of strength criteria, probability functions, and damage coupling mechanisms on the behavior of these models. Notably, it achieves a synthesis of theoretical constructs and experimental validation, demonstrating the models' capacity to reflect the nuanced behaviors of damaged rock accurately. Furthermore, this investigation contributes a methodological blueprint for statistical damage model construction, underscoring the critical selection of strength criteria and probability distributions. These efforts fortify the theoretical underpinnings necessary for the enhanced engineering application of statistics-driven damage analysis.

To explore the effect of hob temperature on the rock-breaking characteristics of full-section tunnel boring machines (TBMs) in sandstone strata, high-temperature furnace heating experiments of sandstone and physical and mechanical experiments at room temperature and high temperatures were conducted to obtain the mechanical properties of sandstone at different temperatures. The mechanical properties at different temperatures were calibrated using PFC3D to obtain micro-mechanical and thermodynamic parameters and establish a rock-breaking model. The orthogonal experiments were used to establish the simulation experiments of rock breaking under different temperatures, confining pressure conditions, knife tip distances, and penetration degrees. The results show that the hob tip force is gradually increasing with an increase in the confining pressure. When below 600 °C, there is little temperature transfer from particle to particle as the temperature increases. At this time, with the two sides of the rock slag flaking, the hob knife tip force is the first to reduce. After 600 °C, with the expansion of the rock extrusion hob, the temperature rises on both sides; at this time, the hob tip force also increased. The hob tip force is minimal at a tip distance of 70 mm and an S/P of 14. As the surrounding pressure increases, the rock-breaking efficiency of the hobber decreases. The highest rock-breaking efficiency is achieved at 25 °C and 600 °C. The rock-breaking efficiency is highest when the knife tip distance is designed to be 70 mm, and when the S/P is 14. The three-dimensional constitutive analysis of rock-breaking particles showed that the increment caused by the hob temperature is mainly distributed in the normal force direction in the surrounding rock without any confining pressure, and the increment caused by the hob temperature exposed to the confining pressure occurs in all directions.

In the field of CO2 capture utilization and storage (CCUS), recent advancements in active-source monitoring have significantly enhanced the capabilities of time-lapse acoustical imaging, facilitating continuous capture of detailed physical parameter images from acoustic signals. Central to these advancements is time-lapse full waveform inversion (TLFWI), which is increasingly recognized for its ability to extract high-resolution images from active-source datasets. However, conventional TLFWI methodologies, which are reliant on gradient optimization, face a significant challenge due to the need for complex, explicit formulation of the physical model gradient relative to the misfit function between observed and predicted data over time. Addressing this limitation, our study introduces automatic differentiation (AD) into the TLFWI process, utilizing deep learning frameworks such as PyTorch to automate gradient calculation using the chain rule. This novel approach, AD-TLFWI, not only streamlines the inversion of time-lapse images for CO2 monitoring but also tackles the issue of local minima commonly encountered in deep learning optimizers. The effectiveness of AD-TLFWI was validated using a realistic model from the Frio-II CO2 injection site, where it successfully produced high-resolution images that demonstrate significant changes in velocity due to CO2 injection. This advancement in TLFWI methodology, underpinned by the integration of AD, represents a pivotal development in active-source monitoring systems, enhancing information extraction capabilities and providing potential solutions to complex multiphysics monitoring challenges.

In order to optimize the efficiency and safety of gas hydrate extraction, it is essential to develop a credible constitutive model for sands containing hydrates. A model incorporating both cementation and damage was constructed to describe the behavior of hydrate-bearing cemented sand. This model is based on the critical state theory and builds upon previous studies. The damage factor Ds is incorporated to consider soil degradation and the reduction in hydrate cementation, as described by plastic shear strain. A computer program was developed to simulate the mechanisms of cementation and damage evolution, as well as the stress-strain curves of hydrate-bearing cemented sand. The results indicate that the model replicates the mechanical behavior of soil cementation and soil deterioration caused by impairment well. By comparing the theoretical curves with the experimental data, the compliance of the model was calculated to be more than 90 percent. The new state-dependent elasto-plastic constitutive model based on cementation and damage of hydrate-bearing cemented sand could provide vital guidance for the construction of deep-buried tunnels, extraction of hydrocarbon compounds, and development of resources.

Knowledge of the shear behavior of landslide soil is very important for understanding landslide dynamics. This study investigated the shear behavior of base soil particle materials using ring shear tests with glass beads. Samples with varying particle sizes and distributions were examined to explore how particle size affects shear behavior at different shear velocities. Analysis of the samples included assessment of shear stress, displacement-shear stress fluctuations, and changes in the shear stress standard deviation. Results showed that with increasing shear displacement, shear stress fluctuations stabilize. Rather large difference particle size differences lead to more uniform particle mixing and a lower shear stress standard deviation. Conversely, when smaller particles predominate, particle mixing is less uniform, resulting in a higher shear stress standard deviation. Additionally, this study discussed the shear dilation and compaction mechanisms of samples under large displacement shear.