Compressed air energy storage lined rock caverns are subject to various complex failure modes during operation, particularly under high-pressure conditions, where tensile and shear failures are prone to occur, posing serious threats to structural stability and operational safety. Focusing on the coupling characteristics of failure mechanisms, an analytical model of the surrounding rock failure zone was developed based on the full stress path from excavation to operation. The influences of rock mass quality, burial depth, tensile strength, and in-situ stress anisotropy on the evolution of failure zones were systematically analyzed. The results indicate that the failure mode during operation is significantly affected by the initial state of the surrounding rock after excavation: for high-quality rock masses that remain elastic after excavation, both high-pressure tensile and shear failures may occur during operation; whereas for lower-quality rock masses where shear failure zones have already developed after excavation, only further shear zone expansion is observed during operation. When the excavation-induced failure zone is smaller than the high-pressure failure zone, increasing burial depth effectively suppresses the expansion of both tensile and shear failure zones. The occurrence of tensile failure under high pressure significantly increases the overall degree of surrounding rock failure. However, relatively low tensile strength can inhibit its initiation. The critical tensile strength required for suppression decreases with increasing burial depth. In-situ stress anisotropy generates direction-dependent. As anisotropy increases, the critical internal pressure for high-pressure failure decreases, whereas the tensile strength required to suppress tensile failure increases.

The cavern constructed in the layered inter-bedded salt rock contains significant insoluble sediment, whose heat storage and utilization potential remains underexplored. To investigate the feasibility of integrating compressed gas energy storage in salt caverns with heat storage in the sediment, a thermo-hydro-mechanical numerical model was developed using Comsol Multiphysics, accounting for the porous medium characteristics of the sediment layer. Firstly, the thermal response of the cavern was analyzed under varying sediment contents during conventional compressed gas storage. The specific heat capacity of the sediment was found to mitigate the thermal impact of hot air on the cavern wall. Temperature fluctuations in the surrounding rock decreased when the sediment layer’s vertical thickness exceeded 30 m. During the gas production stage, temperature fluctuations were restricted to below 0.5 ℃, while during the gas injection stage, they were less than 1 ℃. Subsequently, a dual-cavern model with connected channels was constructed to study the temperature field changes under short-term heat storage, respectively. Simulations showed that temperature fluctuations at the outflow interface during operation ranged from 45 ℃ to 55 ℃, which is 66% lower than at the injection interface, ranging from 30 ℃ to 60 ℃. After three operation cycles, sediment temperature was observed to vary periodically between 49 ℃ and 53 ℃, validating the feasibility of using sediment as an underground heat storage module for compressed air energy storage systems. Results demonstrate that the sediment’s heat capacity contributes to cave wall stability and offers a reference for developing a comprehensive system for utilizing compressed air and heat energy storage in sediment-filled salt caverns.

The surrounding rock mass serves as the primary load-bearing structure in underground CAES caverns, making its stress and deformation behaviour during the charging and discharging process critically significant. Nevertheless, the theoretical framework for elastoplastic deformation of the surrounding rock mass under long-term cyclic expansion dynamic loading remains poorly understood, and the stress paths of the whole process from excavation to cyclic charging and discharging operations are not clear. Accordingly, based on the stress characteristics of the CAES caverns, an analytical solution for elastoplastic deformation of surrounding rock mass throughout the excavation and operational phases is proposed in this study. The reliability of the proposed solution is verified by comparing with the numerical simulation results of commercial software FLAC3D. Furthermore, a parametric sensitivity analysis is conducted using a fixed computational scheme. The analysis evaluates how geological conditions and operational parameters influence the mechanical response of the surrounding rock mass. The main conclusions are as follows: (1) The mechanism that surrounding rock mass will not continue to expand outward is clarified. Under the set working conditions, surrounding rock mass mainly exhibits plastic cumulative deformation inward, resolving the long-standing issue of CAES caverns expansion that has plagued engineering practice. (2) The evolution of the stress path in the surrounding rock mass from excavation through operation is revealed, and the stress path remains between the high- and low-pressure yield lines throughout this period. (3) A method for determining the elastic operating pressure range of the surrounding rock mass of the CAES caverns is developed. This range depends on cohesion and internal friction angle. Operating within this range ensures the surrounding rock mass remains in an elastic state without plastic deformation, thereby addressing a theoretical gap in CAES caverns operating pressure range theory. (4) It is clear that the deformation of surrounding rock mass is most significantly influenced by the minimum and maximum gas storage pressures, while the frequency of charging-discharging also plays an important role. These findings provide theoretical support for the design and construction of CAES power plants.

A composite sealing layer structure of “bentonite-thin steel plate” has been constructed in the lined rock caverns (LRC) for compressed air energy storage (CAES). In this sealing layer, bentonite absorbs water and expands, thereby squeezing the thin steel plate. The thin steel plate generates an elastic reaction force under pressure, which in turn restrains the expansion of bentonite. Therefore, the swelling behavior of bentonite falls under the category of expansion under constant stiffness constraint. To simulate this behavior, a swelling apparatus with comprehensive functions, convenient operation and low cost was developed. The instrument mainly consists of the instrument base, sample chamber, piston, constant stiffness constraint reaction frame, water injection controller, data acquisition box, and so on. With granular bentonite as the research object, verification tests were conducted. The findings indicate that the instrument is capable of accurately replicating the swelling behavior of bentonite under the constant stiffness constraint. Compared with the results obtained by the staged loading equilibrium method, the swelling pressure of the former is generally higher than that of the latter at the same swelling ratio. Moreover, the results of the former are more consistent with the real situation. By changing the constraint combination of the reaction frame, the instrument can also be used to simulate swelling behaviors under constant volume constraint and zero-loading constraint. Meanwhile, in conjunction with the specially designed high- pressure loading system (with a maximum loading capacity of 20 MPa), the loading and unloading tests of bentonite after expansion can be carried out to investigate the deformation behavior of bentonite during the charging and discharging processes of the CAES. The preliminary demonstration of the rationality of the “bentonite-thin steel plate” composite sealing layer structure fully underscores the reliability and scalability of the instrument, highlighting its functional advantage of “one machine serving multiple purposes”

As a key branch of emerging energy storage technologies, underground compressed air energy storage (CAES) is gaining increasing attention for its advantages in large-scale capacity, long-duration storage, and environmental sustainability, making it a crucial support for new power systems. However, underground CAES projects often face challenges such as complex geological conditions, significant multi-physical field coupling, and frequent injection-production cycles, where traditional methods show clear limitations in modeling accuracy and operational efficiency. In recent years, artificial intelligence (AI) technologies, with their powerful nonlinear modeling and data-driven capabilities, have offered novel approaches to intelligent site selection, structural prediction, system operation, and risk warning in underground CAES. This paper employs bibliometric analysis and knowledge mapping techniques to systematically review the current state of AI applications in underground CAES, covering typical scenarios such as site selection and geological modeling, intelligent cavern construction, stability prediction, injection-production optimization, multiphysical coupling modeling, and safety monitoring. The findings reveal that research in this field remains in its early stages, lacking a comprehensive and systematic framework. Based on existing studies, this paper proposes several key directions for future development, including physics-informed modeling, multi-source data integration, and the construction of intelligent engineering platforms, aiming to provide theoretical insights and technical references for advancing the intelligent development of underground CAES and supporting the realization of China’s dual carbon goals.

To better understand the in-service performance of the lining structural system in artificial underground gas storage facilities under repeated high-pressure cyclic loading, we conducted numerical simulations and physical model tests. The analyses accounted for long-term damage-induced deformation of the surrounding rock mass and examined the evolution of stress and deformation in an adaptable support system during cyclic pressurization. This study reveals the deformation and mechanical response characteristics of the lining structure during cyclic pressurization and depressurization in artificial underground gas storage, and verifies the feasibility of constructing underground gas storage caverns with thin steel linings in medium-hardness rock layers such as limestone and sandstone. The results indicate that the proposed adaptable support system design—comprising a 6 mm thin steel lining, wave-arch structures, rubber filling, and a secondary lining can achieve coordinated deformation with the surrounding rock and effectively transfer internal pressure under coupled cyclic loading. This system demonstrates the capacity to absorb high internal pressure and maintain sealing integrity. Under high internal pressure cyclic loading, the stress-strain time-history curves of the sealing steel lining and the circumferential reinforcement respond synergistically with the pressure curve, showing a stepwise increasing trend. Within the 0-6 MPa range, the strain of the steel lining increases rapidly with significant variation. Between 6 MPa and 10 MPa, the strain growth rate gradually decreases as the pressure rises. With an increasing number of cycles, the strain of the steel lining increases to varying degrees, but none exceed the yield strain of 2‰. Moreover, the stress in the outer-ring reinforcement of the secondary lining is consistently higher than that in the inner-ring reinforcement. Therefore, in future designs of underground gas storage, an asymmetric reinforcement design with larger-diameter outer-ring reinforcement could be adopted to enhance the overall load-bearing capacity of the lining. Additionally, the developed 15 MPa-level integrated water-air cyclic loading and unloading test and monitoring system for sealing structures in gas storage caverns will serve as a powerful tool for further analysis of artificial gas storage lining structures. The findings of this study provide fundamental experimental data and valuable references for subsequent key technologies in the construction of compressed air energy storage facilities in medium-hard rock formations.

As an innovative energy technology that deeply integrates CO₂ geological sequestration with compressed gas energy storage, compressed CO₂ storage in saline aquifer achieves dual benefits by using CO₂ as a circulating working fluid, simultaneously enabling large-scale carbon sequestration and grid-level energy storage. This technology aligns well with China's strategic development needs for energy transition and the “dual carbon” goals, offering broad application prospects. However, current research on compressed CO₂ energy storage in saline aquifer primarily focuses on optimizing ground systems, with limited attention given to the underground factors that ultimately determine the success or failure of the entire project. Therefore, this paper reviews the key geomechanical issues related to compressed CO₂ energy storage in saline aquifer. First, it summarizes state of the art in compressed gas energy storage technologies worldwide, and introduces the basic principles and operational characteristics of compressed CO₂ energy storage in saline aquifer systems. Second, it summarizes the thermodynamic properties of CO₂, highlighting the advantages of CO₂ as a working fluid compared to air. Third, it addresses key geomechanical challenges facing geological storage systems, detailing the multiphase coupling mechanisms between CO₂, brine, and rock during system construction and operation. It examines the potential stability and integrity issues arising from these interactions, including formation and surface deformation, sand production, caprock fatigue and fracturing, fault reactivation and induced seismicity, sealing mechanisms of caprock/faults, and leakage pathways. Finally, it summarizes the shortcomings and challenges in current research on compressed CO₂ energy storage in saline aquifer and proposes several research recommendations.

Air temperature and pressure in underground caverns used for compressed air energy storage fluctuate throughout the charging and discharging cycles. These fluctuations alter the thermal field and induce structural deformation in the sealing layer, lining, and surrounding rock. Accurate prediction of stress and deformation in the sealing system during inflation and deflation is essential for reliable cavern design. Incorporating the thermal and physical properties of the materials in each layer of the sealing structure can enhance the accuracy of stress and deformation calculations for the sealing structure and also facilitate the analysis of the cavern’s energy storage capabilities. By integrating the air state analysis algorithm of the cavern with the numerical algorithm for multi-layer material heat conduction on the cavern wall, the temperature distribution across the multi-layered sealed structure can be ascertained, and the stress and strain at each point of the structure can be further determined. By accounting for the energy loss from thermal conduction through the cavern walls, and based on the concept of air exergy within the cavern, the energy recovery rate of the underground cavern during multiple inflation and deflation cycles can be quantified. Studies have shown that caverns constructed in surrounding rock with good thermal conductivity results in rapid diffusion of air heat within the caverns, leading to a more uniform temperature distribution across each sealing layer across the walls, a smaller deformation gradient of the sealing structure, and better adaptability to extreme charging and discharging conditions. However, this rapid heat diffusion can decrease energy-recovery efficiency. In low-conductivity rock, the opposite trend is expected. Therefore, engineering design should account for the differences in thermal and physical properties of sealed structures, as theses can substantially affect cavern operation.

Due to the inevitable dry shrinkage of concrete, there will be construction gaps between the steel lining and concrete lining of the compressed gas energy storage artificial chamber, which brings great risks to the stability and tightness of the artificial chamber. In order to study the influence of construction gaps on the force of the steel lining of the compressed gas energy storage artificial chamber, a theoretical analysis method was proposed to simplify the concrete lining-surrounding rock double-layer cylinder model into a single-cylinder model with equivalent characterization parameters. An analytical model is developed to characterize the stress distribution in steel linings before and after gap closure. Its effectiveness is verified by numerical simulation. Based on the equivalent theoretical model, the sensitivity analysis of different parameters such as air pressure, elastic modulus of surrounding rock, cohesion of surrounding rock, friction angle of surrounding rock and burial depth is carried out. Analytical solutions for the stress in the steel lining under varying parameter conditions were obtained for both before and after the closure of the construction gap. The hoop-prestress ratio in the steel lining under various parameter settings was determined prior to the closure of the construction gap. The results indicate that the width of the construction gap is the key factor influencing the steel lining’s prestress. A larger gap leads to a greater proportion of hoop prestress and a higher hoop stress after the gap closes during gas injection. Furthermore, the internal air pressure, the elastic modulus of the surrounding rock, its cohesion, and the burial depth of the cavern significantly influence the mechanical response of the steel lining. By contrast, the friction angle of the surrounding rock exerts only a minor effect. During construction, the gap width must be strictly controlled. Furthermore, priority should be given to sites exhibiting a high elastic modulus of the surrounding rock, high rock cohesion, and greater burial depth during the site selection phase. The maximum gas storage pressure must be maintained within the permissible limit. Together, these measures underpin the gas storage facility’s long-term operational safety.

Underground lined rock caverns (LRCs) used for compressed air energy storage are highly susceptible to cracking and the development of leakage pathways under cyclic high-pressure loading. Therefore, sealing performance and crack control are critical design challenges. This study employs the finite-discrete element method (FDEM) to develop an integrated rock-lining-sealing layer model, to systematically compare flat steel plate and wave-arch liners, as well as to analyze the influence of preset joint design parameters on crack evolution and sealing behavior. The results indicate that flat steel plate liners exhibit elevated stress levels, numerous cracks, and poor structural integrity, while preset joints provide only limited mitigation. In contrast, wave-arch liners significantly reduce peak stress and redistribute cracking, concentrating damage beneath arches, thereby lowering crack density, albeit with locally larger crack widths. A combined wave-arch and preset joint design further uniform crack propagation and alleviates liner stress. Increasing the number of wave arches and preset joints improves stress uniformity and crack control. However, the maximum crack width exhibits a “decrease-increase” trend, accompanied by a transition in liner stress mode from tension-dominated to bending-shear dominated behavior. When joints are placed at arch bottoms, cracks develop uniformly along the preset paths; when combined with waterproofing and drainage measures, leakage risks can be effectively managed. Overall, the wave-arch and preset joint composite design offers notable advantages in guiding crack development, releasing strain, and enhancing sealing reliability, providing a robust reference for coordinated optimization of sealing and lining design in underground LRCs.

The cross-sectional shape of an underground gas storage in a compressed air energy storage (CAES) power plant significantly influences its stress state, thereby affecting gas tightness and stability. It is conventionally considered that a circular cross-section yields the most favorable stress distribution. However, under anisotropic in-situ stress conditions, the hoop stress around a circular cavern is uneven and can induce localized lining cracking and gas leakage. To address this issue, this study investigates elliptical cross-section storage caverns using theoretical analysis based on elasticity theory. First, we propose a gas-tightness criterion based on the elastic stress state and derive an analytical solution for the elliptical cross-section’s optimal axial ratio. Second, we establish a stability criterion for lined rock caverns under anisotropic in-situ stress and develop a method to determine the operating pressure range for elliptical cross-sections, comparing it with the circular cross-section. Furthermore, we perform a parameter sensitivity analysis using FLAC3D to evaluate how factors affect the elliptical cross-section’s optimal axial ratio and operating pressure range. Finally, we establish an integrated calculation process to determine the optimal axial ratio and pressure range. Results show that, under anisotropic in-situ stress conditions, an elliptical cross-section designed with an optimal axial ratio can produce uniform hoop stress around the cavern. When an elliptical cross- section gas storage facility is designed with the optimal axis ratio, its pressure operating range is maximized. The lateral pressure coefficient, in-situ stress and internal pressure of gas storage are the main factors affecting the operating pressure range of elliptical section gas storage. These findings provide theoretical guidance for the shape optimization and pressure design of underground gas storage caverns.

The burial depth issue of compressed air energy storage(CAES) caverns significantly influences initial site selection and overall stability assessment, serving as a critical aspect in the design of artificial gas storage caverns. To investigate the uplift failure mode of tunnel-type gas storage reservoirs for CAES, we propose a multi-cavern conical failure model based on the single-cavern limit equilibrium framework. The model accounts for rock-mass friction and cohesion, as well as interactions between adjacent caverns. We introduce parameters describing the equivalent slip surface and its inclination angle ω . By enforcing equality of the inter-block force (E_a) between the central and lateral blocks, we derive a governing equation for ω. Solving this equation yields the safe burial depth. Research findings demonstrate that the calculation results of multi-cavern model, which considers the influence of adjacent caverns, are more reasonable compared to those of the single-cavern model. To further explore the response characteristics of the multi-cavern model to design parameters, we conduct a systematic analysis of the impacts of cohesion, horizontal in-situ stress coefficient, cavern diameter, and cavern spacing on buried depth. The results indicate that burial depth is negatively correlated with cohesion and decreases as cohesion increases. In contrast, burial depth is positively correlated with the horizontal in-situ stress coefficient and increases as this coefficient rises. Burial depth also increases with cavern diameter but decreases as the spacing between adjacent caverns increases. Notably, when the spacing between adjacent caverns exceeds four times the cavern diameter, the multi- cavern effect can be neglected.

To mitigate the impact of renewable energy on the stable operation of power grids, the development of energy storage systems is necessary. Compressed air energy storage, as a large-scale long-duration energy storage system, has attracted widespread attention. In tunnel-type artificial caverns, heat exchange between the gas at the cavern end and the incoming cold airflow is limited during inflation. Consequently, the air temperature increases rapidly during pressurization, producing a pronounced temperature gradient between the inlet and the cavern end. The excessively high temperature affects the durability of the sealing materials at the end of the cavern, which is detrimental to the force-bearing of the sealing structure, necessitating the adoption of necessary temperature control measures. Based on the Bernoulli effect, this paper proposes a scheme to add a temperature control pipe at the central axis position of the cavern. By adjusting the airflow lines within the cavern, the mixing between incoming air and the existing air is enhanced, thereby promoting internal air circulation within the cavern. After optimizing the parameters of the temperature control scheme, the temperature difference within the cavern can be effectively reduced. The results show that the optimized scheme can keep the temperature difference within the underground cavern below 80 ℃, which has high practical value and research and development potential.

As the internal pressure gradually increases, the surrounding rock of a tunnel-type hard rock gas storage cavern will undergo three phases: fracture initiation at the cavern wall, fracture propagation, and eventual cavern failure. Fracture initiation occurs due to shear or tensile failure of the surrounding rock at the cavern wall, and is governed by the rock mass’s shear strength and tensile strength. We employed FLAC2D numerical analysis based on the Mohr-Coulomb criterion to locate fracture initiation points on the cavern wall and the associated initiation pressure, accounting for the initial stress field. Boundary effects become negligible when the lateral boundary spacing is at least 15D and the bottom boundary spacing is at least 10D (where D is cavern diameter). The lateral pressure coefficient k and cavern burial depth H significantly influence the fracture initiation angle α, the characteristic curves of fracture initiation angles exhibit two distinct patterns: exponential curve type and horizontal straight-line type; the fracture initiation pressure P₀ is markedly affected by multiple factors including the lateral pressure coefficient k, burial depth H, shear strength (c, φ)and tensile strength R_t, the characteristic curves of fracture initiation pressure display a piecewise-linear pattern. Using elastic theory analytical methods, this paper derives analytical calculation formulas for both fracture initiation pressure and angle. The analytical solutions for fracture initiation pressure show good agreement with the numerical results. However, because finite boundary conditions of overlying rock mass, the analytical solutions for fracture initiation angle demonstrate some discrepancies with numerical results for k in the range 0.95–1.2.

In the context of the rapid development of new energy technologies, energy storage has emerged as a crucial strategic capability. As a new energy storage technology, compressed air energy storage in aquifers (CAESA) is receiving attention due to its advantages of wide distribution and large scale. It has more significant advantages in carrying out compressed air energy storage in aquifers with high exploration and development levels in oil regions. In CAESA, the sealing performance of the caprock system controls the storage capacity and safety. Fully tapping into the sealing potential of the caprock system requires the use of quantitative evaluation indicators. This article proposes a maximum sealing pressure model (P_max model) to define the sealing capacity of a caprock segment, and provides a method for determining P_max via mathematical modeling. This model can accommodate various sealing mechanisms, and for the first time, it explains the sealing mechanism of the thickness of the caprock layer from a mathematical and physical perspective. This indicator can be used as a single indicator to measure the sealing performance of the caprock layer during the site selection stage. Based on this indicator and in combination with the short-board principle, we propose the safe sealing pressure index P_safe to quantify the sealing performance of the entire caprock system. We also provide the calculation method and a flowchart, which can be applied during the engineering development and design stage. Finally, using the KD642-7 pilot project planned by Sinopec Shengli Oilfield as an example, the model’s validity has been preliminarily demonstrated.

A large strain nonlinear radial consolidation model RVTCS for saturated soil foundation under vacuum preloading combined with heating is developed by using piecewise-linear difference method. This model couples radial heat conduction in a soil layer with large-strain thermo-consolidation. It accounts for thermal stress and thermal expansion during consolidation, self-weight, radial and vertical seepage, time-dependent heat-source temperature, nonlinear relationships between compressibility and permeability, and rebound–recompression behavior. The model can analyze large-strain thermo-consolidation under various combinations of vacuum preloading and heating. The model is verified by large-scale vacuum preloading combined with heating consolidation test, and the RVTCS numerical solution of settlement is in good agreement with the indoor test results. The variation laws of excess pore water pressure, settlement, degree of consolidation, temperature and energy consumption under different combined conditions are deeply studied through the analysis of numerical examples. The combination of vacuum preloading and heating can significantly improve the settlement and consolidation rate of soil layer.

The pullout performance of the anchorage system is a key indicator for assessing the effectiveness of rock mass reinforcement. However, existing pullout tests, though widely used, fail to accurately simulate the stiffness of the actual rock mass, reveal the radial mechanical response, and analyze failure modes. To address these issues, this study designed a combination of aluminum sleeves and rock to simulate a soft rock environment and conducted pullout tests on the anchorage system under the stiffness conditions of a soft rock environment. Results indicate two dominant failure modes in a soft-rock environment: rock-splitting failure and unsplit failure, with the corresponding pullout curves showing post-peak sharp drop and post-peak gradual drop, respectively. A significant correlation exists between radial stress and failure mode: peak radial stress ranges from 4 MPa to 10 MPa for rock-splitting failure and 2 MPa to 4 MPa for unsplit failure. Moreover, the anchorage length has the most significant impact on the peak load, while the anchorage length and grout strength are the main factors affecting the failure mode of the anchorage system.

The performance and mix ratio of industrial residue-cement fluid solidified shield muck were studied for promoting the utilization of urban solid wastes. The shield muck collected in Wuhan city were solidified by cement and industrial residues, including blast furnace slag, carbide slag, phosphogypsum, rice husk ash, fly ash and silica fume. The effects of water-cement ratio, cement-soil ratio, industrial residue type and replacement ratio on sample fluidity, shrinkage deformation and strength were discussed. Results indicate that sample fluidity is determined by the water-cement and cement-soil ratios. As the fly ash replacement ratio increases, sample fluidity continuously increases. With increasing replacement ratios of other industrial residues, sample fluidity gradually decreases. The order of reduction is silica fume > rice husk ash > phosphogypsum > slag > carbide slag. The shrinkage deformation mainly occurs within 3~28 d, and the shrinkage rate is significantly reduced when the industrial residue replaces the cement. Sample strength depends on the water-cement ratio and the cement-soil ratio. A low water-cement ratio enhances long-term sample strength, while a high cement-soil ratio boosts early strength. Blast furnace slag and phosphogypsum in the industrial residue promote long-term strength development. Finally, a design method of fluid solidified soil mix ratio based on the theory of slurry rheology and strength development is proposed, providing reference for subsequent construction.

The future construction of the lunar base advocates the full use of the lunar material resources for the construction of the lunar surface. In this paper, graphene oxide (GO) and nano-SiO₂ (NS) were added to lunar regolith simulant geopolymer by ball milling and ultrasonic dispersion to prepare lunar regolith simulant geopolymer nanocomposites (LRSGN). The mechanical properties of LRSGN under high temperature curing environment of the moon were studied. The surface crack deformation and strain evolution distribution characteristics were analyzed by digital image correlation (DIC) technology. The microstructure evolution was characterized by X-ray diffraction(XRD), Fourier transform infrared spectrometer(FTIR), mercury intrusion porosimetry (MIP) and scanning electron microscope (SEM). The results show that the addition of GO and NS significantly improve the mechanical properties of LRSGN, and the optimum contents are 0.08% and 1%, respectively. The maximum compressive strength of the 28 d samples mixed with GO and NS in the lunar regolith simulant by ultrasonic dispersion were increased by 93% and 145%, the maximum splitting tensile strength were increased by123% and 219%, and the maximum flexural strength were increased by 64% and 144%, respectively, compared with the control group. The enhancement effect of mechanical properties is much better than that of ball milling dispersion. The proper incorporation of GO and NS will also delay the formation of local strain concentration zone of LRSGN. The maximum principal strain is at the crack of the sample surface and appears in the form of local strain concentration zone. The microstructure results show that GO and NS have nucleation effect, filling effect, size effect and bridging effect, which promote the polymerization reaction in the system and improve the original pore characteristics, thus improving the mechanical properties of the samples. However, with the increasing content of GO and NS, the agglomeration effect of nanomaterials will occur, which is not conducive to the development of LRSGN strength. These findings provide a theoretical basis and experimental data for material selection and structural design of lunar engineering.

The heterogeneous accumulation structure of tailings sand poses a serious threat to the safe and stable operation of tailings ponds. However, analyses of how freeze-thaw cycles affect the mechanical properties of heterogeneous tailings-sand accumulations are limited. In this paper, based on triaxial compression tests and nuclear magnetic resonance tests, we analyze the macroscopic mechanical properties and mesoscopic pore evolution of the heterogeneous accumulation structure of tailings sand under the action of freeze-thaw cycles. Results show that as the number of freeze-thaw cycles rises, the shear strength of specimens with different interface inclination angles decreases significantly, and the magnitude of decrease correlates positively with the inclination angle. Under low confining pressure, the stress-strain curves of specimens with larger inclination angles exhibit strain softening. Under the action of freeze-thaw cycles, the deformation modes of coarse sand, fine sand, and specimens at various inclination angles can be categorized as shear dilation or a combination of shear dilation and contraction, with the failure mode being transverse fracture. Freeze-thaw cycles cause the “migration and accumulation” of pore water in tailings sand specimens. Interface structures of tailings sand accumulations with different particle sizes can significantly shorten the “migration and accumulation” period, presenting more intense phenomena of small pore connectivity and large pore collapse. Confining pressure strengthens tailings sand and can amplify freeze-thaw–induced strength deterioration. The filling of the lower-layer pore spaces by fine particles increases the interfacial strength of the tailings-sand at the interface. The small particles located above the interface act as weak structural planes and are more prone to deterioration. These findings provide both macroscopic and microscopic perspectives on how freeze-thaw cycles degrade the mechanical properties of sand and soil.

Starting from the balance differential equation of pore water in unsaturated soils, this study transcends the traditional empirical model framework predicated on Darcy’s law. The equation of motion for unsaturated soil seepage was derived through rigorous mathematical derivation of the pore water balance differential equation. By introducing the condition that flow resistance is proportional to velocity during laminar seepage, a theoretical formula for permeability coefficient was established, revealing the physical essence of soil-water interaction mechanisms in seepage dynamics. To validate the theoretical model, an improved apparatus for simultaneous measurement of soil-water characteristic curve (SWCC) and permeability coefficient was employed. Multi-step seepage tests were conducted on various soil types, with transient seepage velocity and moisture content parameters obtained synchronously. Experimental results demonstrate that the calculated seepage velocities from the theoretical formula show excellent agreement with measured data across the effective saturation variation range, preliminarily verifying the applicability of the derived equation of motion for unsaturated soil seepage.

To reveal the occurrence mechanism of ultra-low friction-typed rockburst, deep coal-rock is used as the study material. Firstly, a self-developed ultra-low friction experimental device was used to reproduce the ultra-low friction-typed rockburst under laboratory conditions. A high-speed camera and an acoustic-emission system were used to capture the dynamic fracture evolution of the coal seam and the associated roadway squeezing induced by unstable ultra-low-friction slip. Based on laboratory tests and engineering examples, we established a model of ultra-low friction-typed rockburst and derived an energy index using an energy-based criterion., and the energy index of the ultra-low friction-typed rockburst is given based on energy criterion. Using the coal–rock interfacial friction force, vertical displacement difference, and fractured-seam length as key indices, we evaluated how stiffness, damping, perturbation amplitude, perturbation frequency, and elastic-modulus ratio affect the ultra-low friction response. The results show that increasing the stiffness or damping coefficient of the structural plane reduces the possibility of ultra-low friction sliding or mitigates its intensity. High disturbance amplitude is more prone to induce ultra-low friction sliding of the coal seam. Compared to other disturbance frequencies, the disturbance frequency between 14 Hz and 16 Hz most likely trigger ultra-low friction sliding. When the coal–rock interfacial friction forces are comparable, a higher coal elastic-modulus ratio more readily induces ultra-low friction sliding in the fractured coal seam. These findings provide a basis for predicting and preventing rockburst hazards.

To address the limitations of traditional microbially induced calcium carbonate precipitation (MICP) technology, including reliance on exogenous bacterial strains, high costs, and poor applicability to fine-grained loess, this study introduces environmentally friendly calcium lignosulfonate into the cement solution instead of the traditional calcium source, and optimizes concentrations of nutrient yeast extract (YE), NH₄Cl, urea-calcium source through single-factor and response surface Box-Behnken design experiments. By targeting the activation of indigenous urease-producing microorganisms for loess solidification, we systematically investigate the solidification mechanisms using bioactivity monitoring, unconfined compressive strength tests, calcium carbonate quantification, scanning electron microscope (SEM), X-ray diffraction (XRD) and high-throughput sequencing. The results show that: the optimized nutrient concentrations are 1.2 g/L for YE, 125 mmol/L for NH₄Cl and 0.8 mol/L for urea-calcium source; the urease activity, pH value, and viable bacterial count in the optimized group peak at 120 hours of reaction, significantly promoting calcium carbonate deposition; compared to the control group (untreated loess), the optimized group exhibits 131.42%, 194.32%, and 734.65% improvements in unconfined compressive strength, secant modulus, and calcium carbonate content, respectively; needle/rod-shaped calcium carbonate crystals formed during the reaction significantly enhance soil strength and compactness through filling-bridging-cementation effects; the relative abundance of Bacillales in the optimized group reaches 93%, with notable changes in microbial community diversity and composition. These findings provide a reference for the engineering application of targeted activation technology in loess regions.

Accurate determination of constitutive relationships for hydrothermal parameters is crucial for multi-field coupling studies. However, the presence of multi-scale influencing factors and highly nonlinear response behaviors makes it difficult for existing models to accurately capture coupling effects, heat transfer pathways, and transport mechanisms. Consequently, the development of reliable and robust parameter models remains a significant challenge. To address these issues, a physics-informed ensemble learning (PIEL) model was introduced, combining a priori physical knowledge with interpretive analysis. Using soil thermal conductivity () as a case study, the PIEL model’s accuracy, robustness, and physical consistency were comprehensively evaluated. Sensitivities, response patterns, and coupling effects in the decision-making process were visualized using Shapley additive explanations (SHAP) and partial dependence plots (PDPs). The results demonstrate that the proposed ensemble learning framework effectively captures the complex nonlinear coupling behavior governing thermal conductivity, improving prediction accuracy by a factor of 3 to 6 compared to traditional models. By incorporating a priori heat-transfer knowledge, the PIEL model effectively prevents physically implausible predictions—a common limitation of purely data-driven approaches—thereby substantially enhancing the physical consistency of model outputs. Among the evaluated methods, the physics-informed extreme gradient boosting (PXGBoost) model optimized via the sparrow search algorithm exhibited the highest predictive accuracy and robustness. Sensitivity analyses and response patterns revealed by SHAP and PDPs are consistent with established heat-transfer theory, further validating the physical interpretability of the PIEL-based decision-making process. Furthermore, the optimal predictors identified through cumulative SHAP values significantly outperform traditional parameter analysis methods, enabling the development of more computationally efficient and accurate simplified models. The PIEL model, integrating physical knowledge, represents a powerful tool for geotechnical parameter prediction and paves the way for advancing AI-enabled hydrothermal simulations.

A trace is a spatial curve formed by the intersection between rock joints and the rock-mass free surface. Its geometry directly reflects the rock-mass structural characteristics. Therefore, the rapid and accurate extraction of trace information is of both theoretical and practical engineering significance. Current trace-extraction methods for spatial point clouds rely primarily on curvature, and rarely incorporate point-cloud color information. Moreover, current trace-extraction methods for spatial point clouds rely primarily on curvature, and rarely incorporate point-cloud color information. To address these limitations, we propose a new trace line extraction method based on dimensionality reduction projection (NTDR). NTDR conformally projects 3D point clouds onto a 2D plane and performs efficient edge detection using the color information in the projected data. It then clusters and connects candidate points by integrating 3D geometric features (e.g., curvature and point-to-point distance), enabling automatic extraction of joint traces. Experimental results show that, for large-scale point clouds, NTDR reduces processing time by 91.07% relative to manual extraction, substantially improving efficiency. The traces extracted by NTDR achieve a 90.42% overlap with manually extracted traces and preserve more local details, indicating improved accuracy and overall performance. NTDR maintains an identification accuracy above 80% with 20% noisy points, indicating robustness to noise. Compared with similar automated methods, NTDR yields better extraction results and more closely matches the observed trace distribution in the rock mass. NTDR can improve the efficiency of geohazard prediction and provide data to support tunnel-support design and engineering safety assessment.

To investigate the coupling mechanism of seepage and heat transfer in rock fractures, a numerical model was developed based on the physics-informed neural networks method. The Navier-Stokes equations governing fluid seepage and the convection-diffusion equation describing heat transfer were embedded as physical constraints into the neural network training process. Additionally, a dynamic feedback mechanism for temperature-dependent kinematic viscosity was introduced to propose a numerical model simulating the coupled effects of fluid seepage and heat transfer. The accuracy of the proposed model was validated against the classical Poiseuille flow heat transfer problem. Furthermore, comparison with finite element method results demonstrated its superior stability in handling problems with irregular geometric boundaries. Finally, the effects of fluid kinematic viscosity, seepage velocity (hydraulic gradient, fracture aperture, wall roughness, etc.), and fracture wall temperature on the coupling mechanism of fluid seepage and heat transfer were investigated. The results indicate that if the effect of temperature on the fluid kinematic viscosity is considered, the maximum velocity in the fracture center increases from 0.53 mm/s to 1.92 mm/s, representing an increase of 262.3%. This velocity difference further alters the temperature distribution and reduces the central fracture temperature from 160.1 ℃ to 110.2 ℃(a 31.2% decrease). As the hydraulic gradient increases from 1 Pa/m to 4 Pa/m, the convective heat flux peak rises significantly, far exceeding the increase in diffusive heat flux, leading to a 42.3% decrease in the core fluid temperature. An increase in fracture aperture enhances fluid velocity, which effectively reduces boundary layer thickness and significantly improves heat transfer efficiency. An increase in the fractal dimension of the fracture wall leads to greater flow resistance, which enhances heat transfer through the fracture channel and results in a higher fluid temperature at the outlet. When the fracture wall temperature increases from 100 ℃ to 200 ℃, the peak fluid kinematic viscosity decreases by 55.7%, while the peak seepage velocity rises by 126.7%, and the core region temperature difference expands by 372.4%.

Microbially induced carbonate precipitation (MICP) shows strong potential for reinforcing uranium tailings dams because it is environmentally friendly and efficient. However, spatial heterogeneity in the treated mass limits engineering applicability. To quantify strength heterogeneity in MICP-treated uranium tailings sand, we first used response surface methodology to evaluate how grouting parameters affect strength and its dispersion. Next, we determined the probability distribution of strength by fitting macroscopic strength data obtained under individual grouting conditions. Finally, we analyzed strength evolution and failure mechanisms using calcium carbonate content and spatial distribution, scanning electron microscopy-energy dispersive spectroscopy (SEM–EDS) observations, and macroscopic fracture characteristics. We propose a random discrete element model based on a Beta distribution and define a cementation state index to quantify the macroscopic reinforcement effect inferred from the fitted strength distribution. The results show that reinforcement is maximized at a cementation-solution concentration of 1 mol/L and pH 7. Higher solution concentrations (≥0.9 mol/L) and lower pH (≤7) significantly enhance reinforcement. The strength distribution is right- skewed. The Weibull (AD =0.773) and log-normal (AD =0.32) distributions fit the data better than the normal distribution (AD = 3.616). Failure occurs as progressive brittle fracture, and the Beta-distribution-based random discrete element model captures the resulting heterogeneous mechanical behavior. These findings provide a theoretical basis for optimizing MICP parameters and guiding engineering applications.

The shear mechanical properties of the soil-rock interface directly affect the stability of the binary structure slope. Among these factors, the joint roughness coefficient (JRC) of rock structural planes is a key parameter governing mechanical properties of the soil-rock interface. We combined laboratory direct shear tests with numerical simulations to investigate how varying JRC levels influence the shear properties of the soil-rock interface. An empirical formula for the shear strength of soil-rock interface considering different JRCs of the rock surface is proposed. The results indicate that: (1) The shear strength of the soil-rock interface increases with the increase of structural plane JRC, however, higher applied normal stress diminishes the JRC effect on enhancing the shear strength of the contact surface. (2) As structural roughness increases, the shear strength exhibits two stages: a rapid growth stage (JRC<13.42), during which the shear-sliding zone at the soil-rock interface fails; and a plateau stage (JRC≥13.42), during which internal shear failure of the soil occurs. Verification shows that the proposed empirical formula for soil-rock interface shear strength, which accounts for the influence of structural surface roughness, yields results that agree well with experimental data. These methods and findings provide a valuable reference for assessing the mechanical properties of soil–rock interfaces.