The effective stress principle is a fundamental methodology for investigating fluid-solid coupling behavior in rocks. Accurately determining the effective stress coefficient (b) is a key parameter in effective stress analysis and is crucial for understanding rock deformation and failure mechanisms. This paper systematically reviews both domestic and international research progress regarding testing methods, characterization models, and influencing factors of the effective stress coefficient. A comprehensive comparison of isotropic testing methods, anisotropic testing methods, fractured rock testing methods, and tight rock testing methods is presented, highlighting their distinct applicability: (1) the drainage method exhibits high accuracy across all porosity ranges and can validate other methods; (2) the stiffness matrix method effectively captures the anisotropic effective stress coefficient; (3) for fractured and low-permeability rocks, the equivalent effective stress coefficient method and the cross-plotting method are recommended. Existing characterization models, including the critical porosity model, effective medium theory, and the BISQ model, are evaluated using experimental data. Results indicate that the Biot-squirt (BISQ) model and the isoframe model achieve high precision suitable for accurate calculations; however, their parameter requirements and computational complexity limit practical applications. The critical porosity model, which requires only two parameters while maintaining satisfactory accuracy, is recommended for engineering applications. Key influencing factors include intrinsic factors (such as pore structure, mineral composition, and pore fluids) and external conditions (such as temperature and stress). Intrinsic factors dominate fundamental behavior, while external factors induce microstructural changes that affect the effective stress coefficient. Future research directions should focus on: (1) real-time measurement of the effective stress coefficient under dynamic disturbances; (2) mechanisms of effective stress coefficient evolution under ultra-high temperature and pressure conditions; and (3) testing methods for engineering-scale rock.

To address the issue of potential secondary damage to the original rock mass caused by increased grouting pressure in traditional grouting methods, a negative-pressure-assisted grouting technique for reinforcing micro-fractures is proposed. This method, building upon conventional grouting practices, involves applying negative pressure to micro-fractures or distant ends, thereby increasing the pressure differential between the grouting end and the fracture end without raising the grouting pressure itself. This approach accelerates the grout flow rate and enhances the efficiency of grouting reinforcement. The research established planar and two-dimensional models and constructed a negative-pressure-assisted grouting experimental system. Additionally, numerical simulations were conducted using COMSOL Multiphysics software to investigate the mechanisms of grout diffusion and the dynamic response characteristics of internal pressure by comparing experimental results with simulation data. The findings indicate that applying negative pressure at specific locations results in a pressure reduction at those sites, thereby creating an enhanced pressure differential between the grouting end and the negative-pressure application end. This drives the rapid and efficient movement of grout towards the negatively pressured area, significantly accelerating the filling speed of grout within complex fracture networks and effectively expanding the filling range. Furthermore, it was discovered that maintaining a ratio of grouting pressure to negative pressure between 1:1 and 3:1 can effectively enlarge the grout diffusion range during injection. This strategy broadens the diffusion range of the grout throughout the grouting process, thereby further improving the effectiveness and quality of grouting. The study of negative-pressure-assisted technology in enhancing grout diffusion effectiveness will provide novel ideas and methods for grouting reinforcement in fractured rock masses.

Roof-guided hydraulic fracturing technology presents significant potential for regional gas drainage in outburst-prone coal seams. To investigate the dynamic cross-interface propagation behavior of fractures, we conducted hydraulic fracturing experiments on “roof rock-coal” composite specimens using a self-developed true triaxial fracturing system equipped with visual observation capabilities. Real-time fracture propagation was monitored through an observation window, while digital image correlation (DIC) was employed to analyze the evolution of the strain field during cross-interface fracture extension. Key findings include: (1) Significant heterogeneous deformation was observed during fracture propagation, with distinct zones of tensile and compressive strain localization flanking the fracture. Fractures predominantly developed within the tensile strain localization zones, with propagation paths closely aligned with the morphology of these zones. (2) The influence ranges of the tensile and compressive strain localization zones measured 532–706 times and 459–481 times the fracture width, respectively. During cross-interface propagation, strain fields exhibited marked disparities across the rock-coal interface, with strain localization zones extending into the adjacent medium as fractures advanced. (3) Propagation rates were notably lower in the roof strata, where fracture-tip tensile strain localization zones were short and wide. Conversely, higher propagation rates were observed in lower-strength coal seams, characterized by long and narrow tensile strain localization zones at fracture tips. (4) Fracture path deviation occurred due to shear strain generation at fracture tips induced by coal-rock heterogeneity. Propagation paths consistently followed the direction of maximum shear strain intensity, resulting in a tensile-shear (Mode I-II) hybrid fracture morphology.

With the advancement of underground engineering to greater depths and the continuous increase in mining intensity, the prevention and control of dynamic disasters, such as rockbursts and coal bumps, have become critical safety issues necessitating urgent solutions. To address the deficiencies of conventional support structures in energy absorption efficiency, deformation capacity, and impact resistance, a steel pipe shrinkable energy-absorbing cable, based on the principle of plastic deformation of steel pipes, was developed. Systematic comparative drop-weight impact tests were conducted using a self-designed impact-shear testing device, which included conventional cables as well as 160 kN and 350 kN energy-absorbing cables. Full-scale anchored rock mass specimens were utilized to investigate the effects of impact energy and frequency on the impact-shear resistance characteristics. The test results demonstrate that: (1) The rock mass reinforced with the 160 kN energy-absorbing cable was able to withstand an impact energy of 103 kJ without failure, as its constant resistor effectively absorbed energy through plastic deformation; (2) Compared to conventional cables, the energy- absorbing cables exhibited longer impact durations and smaller fluctuation ranges of impact force; (3) Under multiple consecutive impacts, the constant resistor maintained stable yielding deformation and constant resistance. The steel pipe shrinkable energy-absorbing cable exhibits excellent impact-shear resistance characteristics, providing reliable technical support for dynamic disaster prevention in deep engineering projects.

 The shear behavior of rock joints is often considered a “black box” problem, as current experimental methods are inadequate for directly capturing the evolution of surface morphology and contact characteristics. In this study, transparent joint specimens exhibiting natural surface morphology and mechanical properties akin to marlstone are fabricated using high-transparency Veroclear material and advanced 3D printing technology. Visualized direct shear tests are subsequently conducted under varying normal stresses and shear rates. The results reveal that: (1) in the initial shear stage, contact is predominantly concentrated at the center of the joint surface. As normal stress increases, the contact area gradually expands, radiating outward from the center toward the edges. Compared to the initial contact area, the newly formed contact area during shearing is relatively small and primarily located near the edges, with its distribution varying across different stages of the shear process. (2) The shear rate significantly influences the contact behavior of the joint asperities. At lower shear rates, the extended contact duration between asperities facilitates greater compaction and an increase in the contact area. Conversely, at higher shear rates, the contact time is reduced, and the contact mode between asperities transitions from slip-dominated to localized transient contact. (3) An analysis of the variation in contact angles of asperities corresponding to both the initial and newly formed contact areas indicates that the initial contact angle is closely linked to the peak dilation angle. Based on this relationship, a method for quantifying the initial contact angle is proposed to estimate the shear strength of rock joints. These conclusions provide valuable insights for a deeper understanding of the shear mechanism of rock joints.

To address the challenge posed by agglomerated drilling cuttings under water-rich conditions, which impede the accurate characterization of the true stress state in coal seams, this study conducted true triaxial drilling experiments on coal samples with varying moisture contents. Acoustic emission (AE) signals recorded during drilling were denoised, and key features were extracted and spatially localized. The main findings are as follows: (1) the combined application of Variation mode decomposition-independent component analysis (VMD-ICA) and Fast-Akaike information criterion (Fast-AIC) algorithms effectively suppresses noise and extracts key AE features; (2) the extreme Gradient Boosting (XGBoost) algorithm accurately identifies damage locations in coal under different moisture conditions; (3) damage ellipsoids fitted using the least squares method effectively represent the spatial distribution of damage under true triaxial loading; and (4) based on the equivalent cuttings principle, a quantitative model reconstructs the relationship between cuttings volume and coal failure mechanisms, including rupture, damage, and borehole deformation, revealing the coupling mechanism between moisture content and cuttings generation. This study provides a theoretical foundation for evaluating rockburst hazards in water-rich coal seams through the drilling cuttings method, particularly relevant to mines in the Inner Mongolia—Shaanxi region.

When subjected to continuous dry-wet cycles, rock slopes experience significant disturbances due to water-rock interactions, which lead to dissolution and weathering that can trigger landslide disasters. To investigate the microstructural deterioration and dissolution mechanisms of rock under these conditions, basalt was chosen as the research subject. Experiments were conducted under dry-wet cycles in solutions with varying pH values of 2, 4, 7, 9, and 11. Uniaxial compressive tests, scanning electron microscopy (SEM), and other analytical methods were employed to assess the uniaxial compressive strength and microstructural deterioration of basalt across different pH values and dry-wet cycle conditions. The migration behaviors of key chemical elements within the solutions were analyzed by evaluating their chemical compositions. Based on the dissolution kinetics of basalt, the study explored the microstructural deterioration process and the dissolution mechanisms of water-rock reactions in both non-equilibrium and equilibrium states. The results indicated that, across the various solutions, an increase in dry-wet cycles resulted in a decrease in the uniaxial compressive strength of the rocks and an increase in the proportion of internal microscopic defects. The water-rock reactions in both acidic and alkaline environments transitioned from a non-equilibrium state to an equilibrium state, while those in neutral conditions remained in a non-equilibrium state throughout. Furthermore, the reaction processes in non-equilibrium states across different pH solutions exhibited similar characteristics. The sequence of chemical element release during non-equilibrium water-rock reactions was observed as follows: Ca, Si, and Al, with the concentration of Si increasing as Al substitution intensified. The deterioration of the basalt microstructure was primarily influenced by the release of Si; greater Si release correlated with more severe dissolution of the rock’s internal skeleton. Overall, basalt exhibited the most significant deterioration effects in acidic environments, followed by neutral and alkaline conditions. These findings provide a scientific basis for evaluating the stability of basalt slopes and for understanding the developmental processes of slope-related disasters.

 To investigate the combined effects of strain rate and confining pressure on the nonlinear mechanical behavior and energy evolution characteristics of sandstone, a series of constant and variable strain rate tests were conducted across various strain rates (10－5, 10－4, 10－3, 10－2 s－1) and confining pressures (0, 20, 40, 60 MPa). The influence of strain rate and confining pressure on the stress-strain curve, characteristic strength, failure mode, energy evolution, and brittleness index was analyzed. The results indicate that the characteristic strength increases nonlinearly with confining pressure and linearly with the logarithm of strain rate, demonstrating that these effects are independent of one another. The modified Hoek-Brown strength criterion effectively captures the nonlinear evolution of characteristic strength under varying strain rates and confining pressures. An increase in strain rate intensifies rock damage and fracturing, thereby enhancing brittleness, while confining pressure has the opposite effect. Both the energy absorption and storage capacity of the rock increase with confining pressure and strain rate. Higher confining pressure results in a greater proportion of pre-peak dissipated energy density, which intensifies the nonlinearity of the pre-peak stress-strain curve. Conversely, an increase in strain rate raises the proportion of total dissipated energy density throughout the deformation process. The energy-based brittleness index derived from the dissipated energy evolution curve more accurately reflects the influence of strain rate on brittle-ductile characteristics under different confining pressures. These findings provide robust support for tunnel stability analysis and the optimized design of support structures in deep, high-stress, and high-energy geological environments.

Hot dry rock (HDR) geothermal reservoirs contain abundant natural fractures. Utilizing supercritical carbon dioxide (ScCO2) fracturing can activate and connect these fractures to form a complex fracture network, providing an effective method to enhance reservoir permeability and heat exchange. In-situ high-temperature fracturing experiments on fractured granite using ScCO2 are conducted at temperatures ranging from 100 ℃ to 300 ℃. The injection pressure curves, post-fracturing fracture network morphology, fracture surface characteristics, permeability and convective heat transfer coefficients for both ScCO2 and water fracturing under high-temperature conditions are analyzed. The results indicate that: (1) Compared to water fracturing, ScCO2 fracturing exhibits a slower pressure increase and a lower fracture initiation pressure. ScCO2 fracturing results in more spalling on the fracture surfaces, rendering them rougher. Additionally, ScCO2 fracturing of granite generates multiple branch fractures that connect more effectively with natural fractures. (2) The transient phase change of the fracturing fluid and the convective heat transfer of the fluid due to crack propagation lead to a decrease in fluid temperature at the fracturing zones as the fractures propagate. Notably, the greater the initial temperature of the granite, the more significant the decrease in fluid temperature. (3) The permeability of granite following ScCO2 fracturing is 4.38 to 5.18 times greater than that of water fracturing, with a non-steady seepage phenomenon occurring as the pressure gradient increases, further amplifying the difference in permeability between the two fracturing methods. For both ScCO2- and water-fractured granite, both outlet flow rate and permeability decrease with increasing temperature, with ScCO2 fracturing exhibiting a more pronounced reduction. Specifically, as the temperature rises from 100 ℃ to 300 ℃, the outlet flow rate and permeability of ScCO2-fractured granite decrease by 59.5% and 85.3%, respectively. (5) Although the produced fluid temperature increases with rising granite temperature, the convective heat transfer coefficient decreases due to reduced permeability under higher confining stress. Nevertheless, ScCO2-fractured granite maintains a convective heat transfer coefficient that is 8.68 to 9.68 times greater than that of water fracturing, demonstrating superior thermal exchange performance. This study provides theoretical guidance for fracturing stimulation and efficient geothermal energy extraction in hot dry rock reservoirs

Earthquakes play a critical role in the catastrophic destabilization of concealed bedding rock slopes (CBRSs). However, established methods for assessing the stability of these slopes under seismic conditions are lacking. This study proposes a method for determining the pseudo-static coefficient by analyzing extensive results from shaking table model tests of CBRSs, which accounts for the peak acceleration amplification effect with elevation. Based on the shattering-sliding failure mechanisms of CBRSs under seismic loading, a double-slip model along with its kinematically admissible velocity field is developed. Utilizing the upper bound theorem of limit analysis, a stability analysis method for CBRSs subjected to seismic conditions is presented. Additionally, a comparative analysis of slope stability is conducted through two case studies, further validating the proposed method. The findings indicate that under seismic action, these slopes are particularly vulnerable to catastrophic sliding failure, exhibiting nonlinear and segmented characteristics in both the peak acceleration amplification effect and the surface amplification effect. As the amplitudes of horizontal and vertical seismic waves increase, the stability coefficient of the CBRSs decreases linearly, accompanied by a slight reduction in the thickness of the potential sliding mass. Notably, the stability coefficient exhibits a more significant decrease when accounting for the dynamic elevation amplification effect of seismic acceleration. Furthermore, the stability coefficient of CBRSs decreases as a power function in relation to increasing slope angle and height, while the thickness of the potential sliding mass decreases with an increasing slope angle but increases with rising slope height and structural plane strength. A comparative analysis between the proposed method and two case studies from the literature reveals strong agreement in the calculated results, thereby validating the accuracy and reliability of the method. These findings provide valuable insights for the stability evaluation and treatment design of CBRSs under seismic conditions.

This study proposes a refined prediction method that leverages mutual feedback of spatiotemporal modal information from multiple monitoring points. This approach addresses the limitations of traditional step-like landslide displacement prediction methods regarding granularity and timeliness at annual and monthly scales, as well as the insufficient accuracy arising from increased nonlinearity and redundant information interference at the daily scale. Initially, the variational modal decomposition method, optimized using the sparrow search algorithm (SSA-VMD), is employed to minimize the system?s redundant information entropy. The displacement sequence is decomposed into trend term displacements that reflect material properties, along with periodic and random term displacements influenced by environmental factors. Subsequently, Pearson correlation analysis and Granger causality analysis are performed to identify significant statistical associations between the displacement modes and ecological factors, as well as to filter relevant statistical relationships between displacement modes and environmental variables. This is followed by the design of a CNN-Informer deep learning structure. Spatiotemporal information is extracted using convolutional neural networks (CNN), while the sparse attention mechanism and self-attention distillation mechanism of the informer are utilized to mitigate interference from redundant information and dynamically capture long-term dependency relationships between features and displacements. The ultimate outcome is a point-level daily displacement forecast achieved through time-series reconstruction. This study utilizes daily monitoring data from the monitoring points GSCX3, ZGX111, and GSCX5, collected from October 2017 to December 2021, to analyze the step-like Bazimen landslide. The results indicate that: (1) The model achieves RMSE values of 2.61, 3.84, and 3.91 mm for the three monitoring points, demonstrating a 71.94% reduction in RMSE at GSCX3 compared to alternative models, effectively predicting daily step-like displacement evolution patterns; (2) In the presence of noise interference, significant data absence, and limited samples, the maximum RMSE and MAE reach 6.10 mm and 4.17 mm, respectively, showcasing substantial robustness; (3) Sensitivity analysis reveals a 313.79% increase in RMSE for single-point predictions without mutual feedback, underscoring the critical role of multi-monitoring point spatiotemporal information mutual feedback in enhancing displacement predictions for step-like landslides.

To investigate the creep mechanical properties of layered rock slopes with varying dip angles following excavation unloading, layered phyllite was selected as the research subject. Multistage loading-unloading triaxial creep tests were conducted to establish a creep model for layered rock masses, taking into account both the anisotropic characteristics associated with different dip angles and the time-dependent strain-softening mechanism. The influence of varying dip angles on the creep mechanical properties, long-term strength, and deformation and failure characteristics of layered phyllite was analyzed. The experimental results indicate that: (1) Layered phyllite experiences three sequential stages during multistage loading-unloading triaxial creep: decelerating creep, steady-state creep, and accelerating creep. Notably, dip angles of 0° and 90° have minimal impact on these three creep stages. In contrast, rock with a dip angle of 60° transitions into the accelerating creep stage more rapidly and exhibits the shortest creep duration. (2) The creep rate is significantly influenced by the dip angle, demonstrating anisotropic characteristics. At the final stress level, the increase in circumferential strain surpasses that in axial strain, indicating a pronounced effect of rock lateral dilatancy. (3) The long-term strength initially decreases and then increases as the dip angle rises, reaching its maximum near 45°–60° and approximately following a “U-shaped” pattern. (4) An elasto-plastic variable body incorporating a dip angle factor was introduced to establish the LSVISC creep model, which effectively reflects the multistage loading-unloading creep characteristics of rock masses with varying dip angles. The experimental data align well with the model curves, accurately characterizing the multistage loading-unloading triaxial creep mechanical properties of phyllite under different dip angles and quantifying the “U-shaped” variation of creep parameters with respect to dip angle. This study provides a theoretical foundation for exploring the creep mechanical properties and long-term stability design of layered rock and soil masses in slope engineering.

This study investigates the compaction mechanism of membrane bag polymer grouting in soil by establishing a two-dimensional simulation method that integrates the modified Cam-clay model with polymer chemical reaction theory and the finite element method. The reliability of this method is validated through comparisons of experimental results and simulated outcomes regarding the size, density, and soil squeezing pressure of pile-shaped solids under various conditions. Based on these comparisons, the time-history variations in volume, density, expansion pressure, and other parameters of the polymer solid are analyzed, along with the spatiotemporal characteristics of soil stress, density, and void ratio during soil compaction. Furthermore, the effects of grouting quantity, preheating temperature, and soil density on expansion and reinforcement outcomes are examined. The results indicate that upon injection into the bag, the polymer slurry reacts rapidly, expanding and compressing the surrounding soil, which enlarges the grouting hole. The stress and density of the soil adjacent to the hole progressively increase, while the void ratio decreases. Additionally, as the distance from the center of the grouting hole increases, the effect of soil compaction gradually diminishes, with soil parameters outside of approximately six times the pile diameter largely returning to their initial state. The final diameter and density of the polymer solid increase with the amount of grouting, thereby enhancing the soil reinforcement effect. An increase in preheating temperature accelerates both the expansion rate of the polymer and the compaction process; however, it does not significantly affect the final reinforcement outcome. Denser soil surrounding the hole results in a smaller final diameter and greater density of the polymer solid. Under varying working conditions, the expansion pressure of the grout is zero in the early stage, experiences approximately linear rapid growth in the middle stage, and subsequently shows a decreasing growth rate that stabilizes in the late stage. The work presented in this paper provides a reference for further research on the mechanisms of soil compaction through membrane bag polymer grouting.

To investigate the effects of initial moisture-density conditions and testing paths on the lateral earth pressure and pore structure of expansive soils during soaking, this paper measures the relationships among void ratio, lateral earth pressure, and vertical stress ( ) of Ningming expansive soil specimens compacted at five different initial moisture-dry density (w， ) conditions under the at-rest condition. The constant volume path (CV), swelling-under-load path (SUL), and swell-consolidation path (SC) were followed during the tests, and the characteristics of the pore structure were also obtained. The results show that: (1) the CV path is a specific case of the SUL path. The and relationships derived from the SC and SUL paths are multi-linear and become consistent when the vertical stress exceeds 400 kPa. Below 400 kPa, the slopes of the relationships from the SUL and SC paths are comparable; (2) the swelling potential of the expansive soil decreases nonlinearly with increasing compaction moisture content w, with a more significant reduction observed in the higher w range. The swelling and consolidation indices are not sensitive to the while the swelling stresses increase significantly with the ; (3) as the increases, the micropores of the SC path and the macropores of the SUL path continually decrease. The micropores of the SUL path and the macropores of the SC path decrease when ( is the initial void ratio before soaking). They become insensitive to when . With , comparing with the SUL path, the macropores of the SC path are easier to compress while the micropores of the SC path are harder to compress, at  = 700 kPa where consolidate domains. The pore structures of the specimens with SC and SUL paths are similar.

Based on the mass conservation equation, energy conservation equation, solute migration equation, and soil stress-strain equation for unsaturated soil, a mathematical model was established to couple multiple physical fields, specifically the interactions of water, vapor, heat, salt, and mechanics in unsaturated sulfate saline soil. This model further incorporates the effects of water and vapor convection, including both liquid water and water vapor, as well as the latent heat of water vapor diffusion and convective heat transfer on the physical properties of saline soil, thereby enhancing existing models. Subsequently, the impact of periodic temperature variations on heat and mass migration, along with the deformation of sulfate saline soil, was analyzed through numerical modeling. Additionally, the validity of the proposed model was confirmed through experimental testing. The results indicate that when external water and salt supplies are disregarded, heat transfer in shallow saline soil primarily occurs through conductive heat flux and the latent heat flux associated with the water-salt phase transition, followed by convective heat flux from water vapor and latent heat flux from water vapor diffusion. The convective heat flux from liquid water is relatively minor and can be considered negligible. In shallow saline soil, the predominant mode of water migration is through water vapor flux; however, as depth increases, liquid water flux begins to dominate. When ambient temperature rises, salt concentration increases while crystalline salt content decreases in the saline soil, leading to a reduction in soil displacement, which indicates settlement deformation. Conversely, when ambient temperature decreases, the trends in salt concentration, crystalline salt content, and soil displacement reverse, suggesting that the soil undergoes expansion deformation during this period.

To enhance the seismic design of rigid-faced reinforced soil retaining walls for high-speed railways in China, this study integrates shaking table tests with numerical simulations to analyze post-earthquake horizontal displacement, settlement, and structural acceleration responses, with a particular focus on the mechanical behavior of geogrids under seismic action. The key findings are as follows: (1) the analysis utilizing the peak ground acceleration (PGA) amplification factor effectively evaluates the overall dynamic characteristics of the structure. Both composite and monolithic facings exhibit similar deformation patterns, primarily undergoing rotational deformation during seismic events. (2) The mechanical behavior of geogrids under seismic conditions can be characterized by three parameters: elongation, displacement, and strain. The results indicate that pullout failure is more prevalent than tensile failure in these structures. (3) The intensity of seismic activity significantly influences geogrid mechanisms. During low-intensity earthquakes, the potential failure surface resembles a “0.3H” shape, shifting to “0.14H” when connector effects are disregarded. Conversely, during strong earthquakes, the insufficient pullout resistance of upper geogrids results in compensatory effects in the middle geogrids, leading to an “S”-shaped failure surface. (4) The anchored wedge method employed in strong earthquake design tends to underestimate the load transfer role of middle geogrids, resulting in excessively conservative designs. (5) The mechanical mechanisms of geogrids affect the distribution characteristics of surface settlement, which displays a bimodal pattern with peaks occurring near the panel (0.05H) and at the end of the geogrid. These insights provide valuable guidance for the design of reinforced soil retaining walls in high-speed railway projects.

 In the engineering application of the artificial ground freezing method, the effective establishment of a freezing temperature field is influenced by the synergistic effects of multiple parameters. To investigate the synergistic regulatory mechanisms of cryogenic coolant temperature (ranging from －30 ℃ to －60 ℃), coolant flow rates (1–7 m3/h), and freeze-pipe spacing (0.8–1.2 m) on the evolution of the cryogenic freezing temperature field at －50 ℃, a physical model test system was developed for －50 ℃ cryogenic freezing. A single-factor experimental study was conducted on silty clay strata in Shanghai. The results indicated that as the coolant temperature decreased from －30 ℃ to －50 ℃, the freezing wall closure time was reduced by 26.8%, the time required to achieve an equivalent frozen wall thickness decreased by 55.6%, and the duration to reach the equivalent average interface temperature in the frozen walls was compressed by 71.5%. Under operational conditions with a coolant temperature of －50 ℃ and a flow rate of 5 m3/h, reducing the freezing pipe spacing from 1.2 m to 0.8 m resulted in a significant 58.8% reduction in closure time, a 16.7% increase in the maximum thickness of the main surface frozen wall, and a 33.4% decrease in the average interface temperature. Moreover, the coolant flow rate exhibited a critical threshold at 3 m3/h, beyond which the gain in freezing efficiency remained below 4.8% under cryogenic conditions. These results elucidate the evolution mechanism of the temperature field during －50 ℃ cryogenic freezing. The established quantitative system for parameter sensitivity grading offers theoretical support and serves as a foundation for engineering decision-making regarding the multi-parameter collaborative optimization of the cryogenic freezing process and the rapid construction of high-strength freezing curtains.