Compressed air energy storage(CAES) in underground lined rock caverns(LRC), with its advantages of long power generation time, large scale, short construction period, flexible site selection, low project cost, long operation period, and environmental friendliness, has demonstrated strong vitality in the field of new energy storage, and will significantly promote the construction of new power systems and the high-quality development of renewable energy. Unlike the operational characteristics of traditional underground spaces, the underground lined rock caverns storing compressed air not only have to withstand alternating high internal pressure expansion during the inflation and deflation process, but also experience significant temperature changes. This article focuses on underground lined rock caverns, elaborates on its working principle and the design concept of flexible sealing structure; in view of the load characteristics such as high pressure, alternating stress, and temperature changes, it systematically analyzes the research progress in related theories and analysis methods of underground lined rock caverns, mainly including the temperature and pressure response in the cavern and the heat transfer characteristics of the sealing structure, the stress path and mechanical response of surrounding rock, the cracking and control standards of reinforced concrete lining, the stability and safe burial depth of overlying rock mass, the sealing layer and the sealing plug, etc., and provides an outlook on the development trends of the basic principles and analysis methods for underground lined rock caverns.

To investigate the energy evolution and damage mechanisms of gangue cemented backfill (GCB) with varying water content, we prepared dry, natural, and saturated GCB samples (with water contents of 0%, 10%, and 25%, respectively). Using the stress-strain curves obtained from uniaxial compression tests, we analyzed the energy evolution laws of the GCB samples, proposed an energy strengthening/weakening coefficient, revealed the damage evolution mechanisms, and established a segmented damage constitutive model for the total stress-strain curve. The results show that the stress-strain curves of GCB samples in dry and natural states exhibit distinct “four-stage” characteristics, whereas GCB samples in a saturated state display only “three-stage” characteristics. The elastic modulus and peak strength of GCB samples decrease exponentially from the dry to the saturated state. The weakening mechanism of backfill strength due to water content can be categorized into physical, chemical, and structural effects. The drying effect strengthens the energy of the backfill, primarily through the enhancement of dissipated energy. The saturation effect weakens the energy of the backfill, predominantly through the reduction of elastic strain energy and total strain energy. During the transition from dry to saturated state, the failure modes of GCB samples mainly undergoes a transformation from tensile failure, tensile-shear mixed failure and V-shaped shear failure, accompanied by block detachment. The deformation and failure of GCB samples with different water contents follow the same type of damage evolution process. However, increased water content promotes the development of backfill damage. The established segmented damage constitutive model of GCB samples with different water contents accurately describes the load damage and failure processes of GCB samples. The research results can provide theoretical reference for further studies on the mechanical behavior and stability of gangue cemented backfill.

Red mudstone is highly sensitive to water content variations. Lime treatment is recommended when using red mudstone as subgrade fill material. The mechanical properties of lime-treated red mudstone fill material (LRMF) degrade due to wetting-drying (WD) cycles caused by seasonal environmental effects. A series of WD cycle tests, unconfined compression tests, and bender element tests were conducted to investigate the degradation of strength and small strain stiffness of LRMF. Combining with the successive water-dripping scanning electron microscope (SEM) tests, the microstructure disturbance of LRMF after WD cycles was examined. Swelling of specimens on both the wet and dry sides was observed during low-amplitude WD cycles. For high-amplitude WD cycles, swelling on the wet side was also observed. On the dry side, initial volume shrinkage was recorded, followed by swelling in successive cycles, even though the water content was significantly lower than the initial state. Swelling results in the degradation of strength and small strain stiffness. Volumetric shrinkage increased strength, but small strain stiffness was still reduced due to crack propagation. A unified model is proposed to identify the degradation of strength and volumetric strain, while the small strain stiffness for dry specimens under large-amplitude WD cycles is significantly below the degradation line. The degradation rate of small strain stiffness is significantly higher than that of strength. After water exposure, the LRMF generally retains its initial microstructure. However, loosened aggregates, slaking, and crack propagation are clearly seen in water-exposed specimens. Degradation of the mechanical properties of LRMF can be attributed to damage to the soil fabric.

In practical engineering, soil strength displays characteristics of spatial heterogeneity and anisotropy. Neglecting these characteristics complicates reliably evaluations of slope stability. Therefore, this study conducts an in-depth analysis of slope stability considering the spatial heterogeneity and anisotropy of soil strength. First, improvements were made to the existing spatial heterogeneity model and the original Casagrande anisotropy model to enhance their universality and practicality. Next, the spatial heterogeneity and anisotropy of soil strength were coupled and incorporated into the Mohr-Coulomb (M-C) strength criterion using an improved tensile-shear mode. Subsequently, within the framework of the limit equilibrium (LE) theory, a calculation mode of slip surface stress was employed to replace the conventional assumption mode of inter-slice force. This was achieved by constructing slip surface stress functions and introducing the concept of the local factor of safety for the slip surface, along with stress constraint conditions at the ends of the slip surface. This approach integrates the combined mechanisms of tension-shear and compression-shear, as well as the progressive failure modes of slopes. Finally, based on the overall mechanical equilibrium conditions satisfied by the sliding body, a rigorous LE solution for slope stability was established, accounting for the characteristics of the spatial heterogeneity and anisotropy in soil strength. Through comparative analysis of specific examples, the feasibility and effectiveness of the proposed method were validated. Additionally, this research can also be applied to thoroughly elucidate the slope failure mechanism influenced by the spatial heterogeneity and anisotropy of soil strength.

Cyclic disturbances during the excavation of underground powerhouses in large hydropower stations can damage the surrounding rock mass, compromising its mechanical properties. To reveal the failure characteristics and damage process of deep-buried granite in the Yebatan underground powerhouse after cyclic disturbance, this study conducted pre-peak constant amplitude cyclic disturbance tests on granite samples with a confining pressure of 30 MPa. The tests used cyclic stresses of 12.5%–17.5% (low stress), 10.0%–37.5% (medium stress), and 62.5%–72.5% (high stress) of their peak strengths. Subsequently, three types of graded unloading confining pressure tests were conducted, followed by an analysis of the evolution of characteristic parameters and the construction of damage variables for deep-buried granite under pre-peak cyclic disturbances. Under different graded unloading confining pressure paths, the deformation of rock specimens is primarily concentrated during the reduction of confining pressure in the final three stages, and their macroscopic failure mode is compression-shear failure. The elastic modulus shows a downward trend, while the Poisson’s ratio exhibits an upward trend; the maximum dilatancy angle remains largely consistent. After constant amplitude cyclic disturbance under high axial pressure, the rock specimen shows the highest cumulative strain energy in the graded unloading confining pressure test, indicating an increased tendency towards brittle failure. Furthermore, the strain energy of rock specimens under graded unloading confining pressure exceeds that under conventional triaxial compression. This serves as a threshold to identify if rock specimens under graded unloading confining pressure transition into the accelerated damage rupture stage. Using the lateral strain response method and crack volumetric strain method, the closure stress, crack initiation stress, and damage stress of deep-buried granite are found to be 17.9%, 42.7%, and 73.8% of the peak strength, respectively. Based on these insights, a damage variable derived from crack volumetric strain is formulated, allowing the delineation of four distinct stages in the damage process of deep hard rock. The research results provide experimental and theoretical support for analyzing the failure process of high-stress granite.

Based on the simplified Bishop method, formulas for calculating slope seismic stability are derived, which include the pseudo-static method (PSM), classical pseudo-dynamic method (PDM) and modified pseudo-dynamic method (MPDM). Among these, MPDM treats soil as Kelvin-Voigt material, considering its viscoelastic and amplification characteristics. MPDM is then applied to slope conditions. Subsequently, Morlet wavelet analysis is used to determine the spectral distribution characteristics of seismic waves, with the El Centro wave as the research object. A method for evaluating the seismic stability of slopes is proposed, taking into account the time-frequency characteristics of ground motion by identifying the predominant frequency at the moment of peak ground acceleration (PGA). The effectiveness of this method is validated through illustrative examples. The results show that slope seismic stability can be assessed rapidly and accurately using the pseudo-dynamic method, which requires fewer parameters and provides clear interpretations, closely resembling the results of the time-history analysis method. These research findings establish a theoretical basis for the seismic risk assessment of slopes, the analysis and prediction of the scope of seismic landslide disasters and earthquake emergency response.

Establishing a constitutive model that reflects the local bonding breakage process has always been a core task in soil mechanics and is crucial for solving engineering stability issues. Based on thermodynamic principles and breakage mechanics, this paper proposes a macro-micro thermodynamic constitutive model. This model quantitatively describes the thermodynamic behavior of local bonding breakage and the non-uniform distribution of stress-strain at the microscale. It improves the prediction accuracy of the model for deformation characteristics, which is similar to the Cambridge model in mathematical form. Firstly, based on the law of conservation of thermodynamic energy, the mathematical expression of structural breakage work during compression deformation was determined. It was found that the dissipated energy of breakage can be mainly divided into two parts: the frictional effect between bonded elements and frictional elements, and the irreversible transformation from bonded elements to frictional elements. Furthermore, a macro-micro constitutive model framework considering the thermodynamic behavior of local bonding breakage was established. Secondly, based on the constitutive framework and the deformation mechanism of loess (frictional, bonded, and damaged), the expressions for free energy, dissipated energy, and damage dissipated energy were determined. The damage yield function and elastic-plastic constitutive model considering the evolution laws of volume breakage and shear breakage were derived. Finally, the established model was used to predict the experimental data of other scholars, and its rationality and simulation advantages were verified through comparison. This model aligns better with thermodynamic principles, and its parameters are easy to determine.

The macroscopic properties of soil are primarily influenced by its microstructure and pore characteristics. Understanding the microscopic evolution of soil under external conditions like freeze-thaw is crucial for geotechnical studies. Turfy soil, a seasonally frozen and unique type of soil, exhibits high compressibility and low strength due to its high humus and plant fiber content. Therefore, this study focused on turfy soil to investigate its microstructures, pore characteristics and the effects of freeze-thaw using methods such as geotechnical tests, nuclear magnetic resonance (NMR), X-ray computed tomography (CT) and scanning electron microscopy (SEM). Utilizing geotechnical and NMR theories, micro-image segmentation was performed on CT slices and SEM images to identify air and water-storage pores. Combined with microscopic images and compositional analysis, the microstructure of turfy soil reveals that the organic matter component forms a matrix capable of containing and conducting water. The pore size distribution of turfy soil after freeze-thaw shows an increased proportion of mesopores and a significant increase in the number of pores. Consequently, quantitative characterization of microscopic parameters indicates enhanced pore connectivity and reduced pore shape complexity in turfy soil before and after freeze-thaw, thereby enhancing permeability. Verification of theoretical calculations for unsaturated soil shows that the NMR method effectively measures the permeability of freezing and thawing soil. The research findings can serve as a basis for studies on soil with high organic matter and fiber content and can be applied as a parameter basis for engineering construction in regions with turfy soil.

The explosion replacement method is an effective treatment for weak soil layers exceeding 15 m in thickness. To investigate the effect of explosive deposition depth on treatment efficacy, model tests involving silt blasting were conducted, and vane shear tests were performed using an automatic cross shear apparatus. The effects of explosive deposition depth on the undrained shear strength and water content of silt were examined. Additionally, the soil microstructure before and after blasting at various depths was observed using a scanning electron microscope. The results indicated that blasting disrupted the soil structure around the explosion point, increasing the void ratio and causing a sudden drop in undrained shear strength, thereby forming a blasting disturbance zone. The soil outside the disturbed zone was compacted, resulting in increased undrained shear strength and forming a blast-compacted zone. After blasting, soil water content decreased; the farther from the explosion center, the more significant the decrease. When the explosive deposition depth was 0.3 times the soil layer thickness, the reduction in undrained shear strength was most significant at about 27%, and the disturbed zone was largest at approximately 19.2d (where d is the explosive diameter), making the explosion most effective. Therefore, there exists an optimal deposition depth of the explosive, which makes the best explosion treatment effect and the largest range of disturbed soil. And, the range of blasting disturbance can be determined by undrained shear strength. The results of the study can provide technical support and reference for the design of explosive deposition depth in related projects.

The consolidation characteristics of dredged slurry improved by vacuum preloading significantly deviate from conventional patterns due to its unique properties. However, there is a lack of large strain consolidation models that consider the development of soil columns during vacuum preloading, especially regarding the variation of soil clogging characteristics over time and spatial distribution. To address this issue, this study developed a large strain consolidation model for slurries improved by the vacuum preloading, incorporating the increasing radius of soil columns. The model integrates factors such as the variation of permeability with radius, time-dependent discharge capacity, soil self-weight, nonlinear compressibility and nonlinear permeability. The governing equations using void ratio as the control variable, were solved by the finite difference method to calculate consolidation behavior considering soil column variations. The proposed model was validated by comparing its predicted results with those from an existing analytical model. Based on this, a parametric study analyzed the influences of radial permeability distribution, soil column formation time and drain discharge capacity. The results indicate the necessity of considering soil column radius variation over time and its permeability relative to the distance from the prefabricated vertical drain in the computational model due to their significant influence. Ignoring the soil column development process underestimates both consolidation speed and degree, with errors exceeding 10%. Assuming the soil column’s permeability coefficient remains constant at different position along the radius also underestimates consolidation speed and degree, with the errors increasing as permeability decreases. The consolidation differences under various soil column radius growth times are substantial and increase as the soil column radius increases and the permeability coefficient decreases.

The expansive performance of in-situ expansive soil is significantly influenced by its natural characteristics such as water content and structural properties. Effectively and accurately determining the in-situ expansion potential of expansive soil is a key challenge in engineering. To evaluate the expansiveness of in-situ expansive soil, artificial and natural expansive soils were selected as research subjects. Low-frequency electrical tests and expansion ratio experiments were conducted to establish an evaluation model for the non-loaded expansion ratio of expansive soil based on low-frequency electrical indicators. The research results indicate that the expansion ratio without load decreases as water content increases at a constant dry density. The soil’s expansiveness is significantly influenced by the initial water content and pore structure characteristics. The final expansion ratio is correlated with the water absorption characteristics of soil mineral particles. Additionally, changes in water content and dry density can significantly impact the resistivity index. The strong sensitivity between resistivity index and soil water content, along with the characterization of water absorption capacity in expanded soil minerals, suggests that the resistivity index is a valuable tool for assessing the expansion behavior of in situ expanded soil. Based on this foundation, resistivity, dry density and water content were normalized. A comprehensive indicator Q, which combines normalized water and electricity data, was introduced. Furthermore, an evaluation model for the non-loaded expansion ratio of in-situ expansive soil was developed. Through validation with existing experimental data, the proposed electrical evaluation model for the expansion ratio of in-situ expansive soil exhibits direct and efficient attributes, offering a dependable assessment of the expansive potential of in-situ expansive soil.

Resonance occurs when the natural frequency of an offshore wind turbine matches its rotational or blade passing frequency, potentially causing severe structural damage. Existing research on the resonance frequency characteristics of offshore wind turbines has mainly focused on elastic analysis, neglecting the nonlinear dynamic interaction between the foundation and soil. Based on the dynamic Winkler foundation model, the hyperbolic soil resistance around the pile-lateral displacement (p-y) backbone curve was used to consider the stiffness nonlinear of the pile-soil system. The shear strain-dependence of hysteretic damping was considered for soil energy dissipation. A simplified nonlinear frequency domain analysis method for calculating the resonance frequency of monopile-supported offshore wind turbines was proposed. The validity of the method was confirmed through comparisons with model test results and field measurements from the Lely (A2) offshore wind turbine. A parametric study was conducted to investigate the influence of sand density, pile length and pile diameter on the nonlinear resonance frequency of offshore wind turbines. The results show that the resonance frequency of the offshore wind turbine system decreases with increasing loading amplitude. Additionally, the influence of soil nonlinearity on the resonance frequency for systems is more obvious when the sand is looser, the pile length is shorter, or the pile diameter is smaller.

The deformation and failure of deep tunnels in a high in-situ stress field are driven by energy, which results from the dissipation and release of energy exceeding the storage limit in the surrounding rock. Existing theoretical studies mainly focus on calculating the elastic strain energy of the surrounding rock during the elastic unloading of deep circular tunnels. Few authors have conducted theoretical research on energy calculation and evolution in deep circular tunnels under elastic-plastic unloading. Firstly, this study clarifies the definitions of elastic strain energy, dissipated energy, and released energy based on the energy evolution of rocks. Subsequently, utilizing elastic-plasticity theory and Hoek-Brown strength criteria, the radius increment method is proposed to calculate the unloading stress path of the surrounding rock, taking into account strain-softening behavior. Finally, an energy calculation method for deep circular rock tunnels based on the unloading stress path is established, mathematically proving the energy conservation relationship of the surrounding rock. The research results indicate that energy is input from the remote surrounding rock after tunnel excavation. During the elastic deformation phase, input energy is all converted into elastic strain energy, with higher elastic strain energy density closer to the tunnel wall. In the plastic deformation stage, the total elastic strain energy continues to increase with ongoing energy input. However, the elastic strain energy density near the tunnel reaches the energy storage limit of the surrounding rock, leading to a plastic zone. Subsequently, the energy storage limit in the plastic zone decreases due to strain softening behavior, resulting in an exponential increase in the dissipated and released energy of the surrounding rock. For deep rocks with high peak strength, energy release dominates during plastic deformation, leading to rockbursts caused by the energy release mechanism. Rocks with low peak strength exhibit plastic deformation primarily driven by energy dissipation, leading to the initiation of tunnel squeezing deformation.

Indoor measurements of the soil-water characteristic curve (SWCC) often require a long equilibrium time, yet the actual soil changes may not meet this requirement. When the time scale is small, the SWCC may not reach equilibrium. If the equation for unsaturated soil is still established using the soil-water characteristic curve under equilibrium conditions, errors may occur. Therefore, the dynamic effect of the soil-water characteristic curve under non-equilibrium conditions was studied. Based on existing theories, we derived the relationship between the parameters of the soil-water characteristic curve and the rate of saturation change, and subsequently established a dynamic capillary hysteresis model using the derived dynamic parameters. We developed a self-designed SWCC rapid measurement device and conducted soil-water characteristic curve measurements for coarse sand and fine sand under varying rates of saturation change. The experimental research revealed that: 1) the parameters of the soil-water characteristic curve exhibit significant dynamic effects; 2) the air entry value and residual saturation are not constant, but vary depending on the rate of saturation change; and 3) the model was validated using the experimental results, and the prediction outcomes of the dynamic model closely matched the experimental results, demonstrating the model's rationality. This study provides a more practical theoretical basis for solving problems related to deformation, strength, and seepage in unsaturated soils.

The erodibility of clay exhibits significant variability across different influencing factors. The existing research using compaction approach for specimen preparation neglected the non-uniformity in soil specimens and is unsuitable for high plasticity clay. In this study, the saturated preconsolidation approach was used to prepare uniform kaolinite specimens to simulate natural consolidating conditions. The prepared specimens were then analyzed using a hole erosion analyzer, and the surface morphology of the eroded hole was quantified using a 3D scanner. A total of 18 hole erosion tests were conducted under various preconsolidation pressures and erosion directions. The erosion resistances were found to increase with higher prestress, and the variation of critical shear stress across different erosion directions reached 29%. The SEM images reveal a stack-packing microstructure in the consolidated specimens, with a denser clay aggregate packing observed under higher pre-stress conditions. The anisotropic erosion property is properly described by the radial anisotropic coefficient κ_-r  and the roughness anisotropic coefficient κ_pr , and the critical shear stress τ_c  is negatively correlated with κ_-r  , while its correlation with κ_pr is not obvious.

To investigate the impact of cyclic loading on the mechanical damage characteristics and seepage evolution of fractured rock mass, the stress-seepage coupling tests were conducted on sandstone samples containing an inclined rough single fracture under conditions of axial and pore pressures low frequency repeatedly loading and unloading. The results indicate that under stress, rock samples exhibit relative slip along pre-existing fractures, with surface protrusions undergoing wear and shear. These protrusions cause a slight stress drop during loading, leading to flow rate fluctuations during compression. For samples with inclined fractures, strength is mainly determined by slip along pre-existing fractures and pore pressure magnitude. During cyclic loading and unloading under axial compression, fractured rock mass undergoes progressive damage and deterioration with increasing cycles. This reduces the samples’ ability to resist deformation. Additionally, samples exhibit a nonlinear decrease in elastic modulus and a nonlinear increase in Poisson’s ratio with increasing cycles. Pore pressure cycling shortens the decay creep stage and reduces strain rates in the stable creep stage. The flow rate dynamically evolves during cyclic loading and unloading. Initially, the flow rate decreases due to fracture closure under stress. Subsequently, it decreases and then increases with cyclic axial stress. The grayscale threshold segmentation method effectively separates connected and contact zones of fractures. It shows that final flow increases exponentially with increases in fractal dimension D and connection ratio Φ.

The occurrence of rock bursts during underground engineering construction is closely related to the time-delay failure characteristics of rocks. Dynamic disturbances such as excavation blasting and mechanical drilling directly impact the formation process of rock bursts. Time-delay failure tests on sandstone under dynamic disturbances of varying amplitudes and frequencies were conducted to analyze their impact and mechanism on rock failure delays. Research findings indicate: 1) Under different amplitudes (8%–24%) and frequencies (0.2–1.0 Hz) of disturbances, the incubation time for rock sample failure significantly shortens, by 39.49%–98.21% compared to undisturbed conditions. With increasing frequency and amplitude of the disturbance load, the proportion of the accelerated deformation stage duration in the rock sample gradually decreases, leading to an increase in the sudden failure. 2) Increasing amplitude and frequency of disturbance loads weaken the flaky peeling phenomenon during rock sample failure and increase the randomness of crack generation, altering the time-delay failure form and degree of rock samples. 3) The damage variable of rock samples under dynamic disturbance exhibits an inverted S-shaped growth pattern over time. A time-delay cumulative damage model for rocks considering static loads and dynamic disturbances has been developed. 4) During the static loading stage, the rock sample accumulates elastic energy and incurs initial damage. The dynamic disturbance intensifies the damage evolution process, leading to continuous induction of new damage. Increasing amplitude and frequency of the disturbance accelerates rock damage, causing a reduction in sample bearing capacity and ultimately triggering rock burst. These research findings can provide valuable references for analyzing and interpreting time-delayed rock bursts induced by dynamic forces in deep rock mass engineering.

Accurately identifying precursor information of rockbursts is a key issue in rockburst monitoring and early warning. Engineering practice has shown that rockburst disasters progress from static incubation to dynamic failure, with the unstable stage being critical for rock mass instability and mutation leading to rockburst occurrence. Studying this stage is crucial for understanding the essence of precursor information of rockbursts. The studies on the acoustic emission characteristics of diorite at different unloading rates and the identification of unstable stages are presented. The characteristics of amplitude, ringing count, frequency, etc. are analyzed, and the random forest (RF) model is used to identify the unstable stages of parameters such as amplitude, ringing count, and energy. The research results indicate that the stress-amplitude-ringing count-time relationship shows a quiet period, during which large, medium, and small values appear simultaneously. The initial frequency, reverberation frequency, cumulative initial frequency, and cumulative reverberation frequency increase sharply after the quiet period . These values are widely distributed and can serve as precursor information for rockbursts. This entire stage is unstable. The RF model can effectively identify the unstable stage of acoustic emission with a minimum accuracy of 81.2%. Conducting rock mechanics experiments to calibrate acoustic emission parameters and identifying unstable stages in a specific gold mine using the RF model is crucial for monitoring and early warning of rockburst disasters.

To investigate the mechanical properties of frozen peat soil derived from Dianchi Lake's lacustrine deposits, a low-temperature triaxial shear test was conducted under various influencing factors, utilizing an improved TSZ-2 fully automatic strain control instrument. This study aimed to examine the mechanical behavior of frozen peat soil at different temperatures, confining pressures and moisture levels. Additionally, the binary medium model theory was introduced to analyze the deviatoric stress-strain relationship in frozen soil. The test results indicate that as strain increases, the deviatoric stress-strain curve divides into three stages: linear-elastic, elastic-plastic and stable stages. The volume deformation primarily involves bulk expansion, and the deformation characteristics of frozen peat soil can be explained using a binary medium model. The peak strength of frozen peat soil is positively correlated with confining pressure and moisture content, but negatively correlated with temperature. In the experimental setup, the impact of confining pressure on strength initially rises and then declines, while moisture content exhibits higher sensitivity to strength. Cohesion increases as temperature decreases, and the internal friction angle fluctuates between 20.56° and 24.89°. Based on the simplified binary medium model, the equations suitable for frozen peat soil are constructed and the results are verified with good applicability.

The connectivity ratio significantly influences the mechanical properties and failure mechanisms of rock samples with structural planes of varying inclinations. Consequently, jointed rock samples with connectivity values k of 0, 0.25, 0.50, and 1.00, and dip angles β  of 15°, 30°, 45°, and 60° were manually prepared. Uniaxial and triaxial compression tests at confining pressures of 200, 400, and 600 kPa were conducted to investigate the effects of connectivity and dip angles on the mechanical properties of the rock mass. The results show that: 1) Connectivity significantly reduces the strength of the rock mass; peak strength decreases as connectivity increases. Stress-strain curves exhibit initial strain hardening followed by strain softening when k≤0.50; for k>0.50, the behavior transitions from strain softening at low angles to strain hardening at high angles. 2) Samples with structural planes initially shrink and then expand; overall expansion decreases with increasing connectivity at each dip angle; abrupt changes occur during shrinkage when k=1.00 at dip angles of 30°, 45°, and 60°. 3) Before energy inflection point α , dissipative energy U_d increases with connectivity while elastic strain energy U_e  slightly decreases; after point α , andU_e  both increase sharply, and different connectivity levels exhibit similar energy release rates. 4) The prediction criteria for rock samples with varying connectivity rates are classified. When k < 0.5, β=45°+φ_j /2 , it is suitable for judging the failure angle, where φ_j  is the friction angle of the structural plane; when k ≥0.5, the Jaeger criterion accurately reflects the failure angle of the rock mass.

In the carbonate belt of Tianxingzhou, Wuhan, there are giant thick-bedded arenated dolomites. Currently, there is no relevant experience or suggestion to determine its mechanical parameters. This study explores a method for determining the parameters of the Hoek-Brown (H-B) criterion for arenated dolomite using the linear transformation theory of the H-B and Mohr-Coulomb (M-C) criteria. It synthesizes previous research findings and conducts laboratory experiments. A simplified method was proposed to determine the upper limit of the maximum confining pressure σ_3max  by utilizing the static lateral pressure coefficient k₁. The mechanical parameters of sandy dolomite were estimated, and recommended values were provided based on the proposed method. The findings indicate that the proposed simplified method of σ_3max  based on k1 effectively narrows the range of confining pressure σ₃ and enhances accuracy. Additionally, the geological strength index GSI of sandy dolomite correlates linearly with the unconfined compressive strength σ_ci , leading to an improved table of suggested GSI values. The material parameter m_i is primarily influenced by rock type and fine-grained particle roughness, with m_i=9 in this study, showing small value range and minimal impact on rock mechanical parameters. Furthermore, the relationship between σ_ci and rock mass shear strength is established by converting the H-B and M-C criteria. These results provide valuable insights for applying the H-B criterion, effectively addressing the challenge of determining the mechanical parameters of highly fractured dolomite rocks. This research is significant for determining the mechanical parameters of rock formations similar to those in Wuhan, China.

The ongoing urban subway construction has led to an increase in foundation pits near existing underground structures. However, there remains a lack of systematic research on the soil instability modes in adjacent foundation pit engineering. This study employs numerical analysis to investigate the instability modes of asymmetric / symmetric limited soil in the adjacent foundation pit project near an existing subway station. It proposes applicable conditions under various instability modes and analyzes the impact of internal support quantity, foundation pit depth, and spatial relationship between the foundation pit and the existing subway station on soil stability and slip surface properties. The research shows that: 1) Adjacent foundation pit engineering exhibits three to six cases of symmetric or asymmetric limited soil instability modes. 2) As the number of internal supports increases, the slip body’s range expands, the starting position shifts downward in the retaining structure and the safety factor of limited soil stability gradually rises. 3) The increase of depth in the foundation pit influences with the sliding body range, initially increasing and then decreasing, leading to a gradual decrease in the safety factor of finite soil stability. 4) Analysis of the asymmetric slip surface in adjacent foundation pit projects indicates that as the overburden soil thickness of the existing subway station increases, the approaching range of the slip surface near the station decreases gradually, and the starting position shifts outward. This research provides a theoretical foundation for calculating limited soil earth pressure, designing foundation pit retaining structures, and selecting safety control measures for existing subway stations.

To enhance the tensile performance of pile foundations, an inclined anchor-short pile foundation with an anchor plate was proposed for transmission lines in areas with overlying soil and underlying rock layers in mountainous regions. An indoor model test of a reinforced concrete short pile with three inclined anchor foundation joints was conducted to study the mechanical characteristics and failure mechanisms of these joints under multi-directional tension. The results show that the short pile and anchor bar work synergistically. The tensile crack first occurs at the joint of the anchor rod, then the vertical main crack extends to the pile top, leading to the overall splitting and failure of the specimen. The cracking load of the specimens is approximately 150 kN, the yield load is approximately 1 611 kN, and the ultimate load is approximately 1 845 kN. Before failure, all the anchor bolts yield, while the longitudinal bar and stirrup inside the pile do not yield, indicating stable anchoring performance of the inclined anchor. When the load on the inclined anchor-short pile foundation is small, the anchoring action between the pile and anchor is mainly borne by the bonding action of the straight anchor section of the steel bar. As the load increases, the anchoring effect is mainly borne by the bearing effect of the anchoring plate. The oblique stress between the anchor ribs combines with the corresponding oblique shear stress τ_α  in the short pile, causing the concrete cracks around the anchor bars to deflect and exacerbating the development of concrete cracks to the pile surface. This study provides a reference for the design of composite foundations and the study of deformation characteristics and failure mechanisms.

The short and long-term mechanical properties and deformation characteristics of rocks significantly impact the engineering project’s long-term stability and safety. Traditional constitutive models struggle to uniformly describe the short- and long-term mechanical properties of various rock materials. In contrast, deep learning methods can predict these properties without additional elastic or plastic parameters and constitutive laws. The long short-term memory (LSTM) deep learning algorithm is well-suited for time series data tasks and excels in predicting the short- and long-term mechanical properties of rocks. This study utilized the LSTM algorithm to construct sequence data based on the triaxial compression loading path and stress relaxation over time. A prediction model for the mechanical properties of gray sandstone under conventional triaxial compression and stress relaxation was established. Comparison with experimental data validates the accuracy of the deep learning-based short- and long-term rock mechanical prediction model. To enhance the model’s engineering application value, the LSTM constitutive model was embedded into the finite element (FEM) framework for numerical implementation and applied to gray sandstone simulation. The comparison results show that the LSTM-FEM method better predicts the short- and long-term deformation characteristics of rocks.

Based on the fluctuation theory of elastic medium, the fluctuation equation of thermoviscoelastic medium is established by using the Kelvin-Voigt viscoelastic model, the equation of motion of viscoelastic medium and the generalized thermoelasticity theory. This approach considers soil viscosity and thermal effects. Then the dispersive characteristic equation of the body wave in the thermoviscoelastic medium is derived by introducing the displacement potential function of the solid-phase medium. Numerical calculations were performed to analyze the influence of thermophysical parameters, such as thermal expansion coefficient, medium temperature and relaxation time on the wave velocity and attenuation coefficient of thermoelastic waves. The results indicate significant differences in the wave velocity and attenuation coefficient of thermal elastomer waves under the three theoretical models: elastic theory, thermoelastic theory, and thermoviscoelastic theory. For each 0.5×10⁻³ s increase in relaxation time, the P-wave speed and attenuation coefficient increased by up to 5.18% and 34.67%, respectively; the S-wave speed and attenuation coefficient increased by up to 9.27% and 34.6%, respectively; and the T-wave speed and attenuation coefficient decreased by up to 2.18% and 2.24%, respectively. As frequency increases, the wave velocity and attenuation coefficient of each thermoelastic wave gradually increase. An increase in medium temperature results in higher P-wave and T-wave speeds and a higher P-wave attenuation coefficient. Specifically, for every 20 K increase in temperature, the P-wave speed and attenuation coefficient increase by approximately 3% and 2%, respectively. However, temperature changes do not affect the S-wave propagation characteristics and the T-wave attenuation coefficient. An increase in the thermal expansion coefficient leads to an increase in P-wave velocity and a decrease in T-wave velocity, significantly affecting the attenuation coefficients of both P-wave and T-wave. Additionally, heat flux and the phase delay time of the temperature gradient significantly influence the wave velocity and attenuation coefficient of the T-wave.

The rock-concrete interface represents a crucial weakness in engineering structures, substantially impacting their overall structural integrity and stability. To accurately capture the natural roughness of the rock-concrete interface, we developed numerical models of rock-concrete composite Brazilian disk specimens, incorporating randomly generated rough interfaces using the cohesive zone model (CZM). The validity of our method was confirmed through Brazilian splitting tests conducted at various loading angles. Additionally, we investigated the impact of interface roughness and loading angle on the peak load and failure modes of the specimens. The results reveal three typical failure patterns under different loading angles: interface debonding, composite failure, and tensile cracking across the interface of both materials. The mechanical behavior of the specimens is significantly influenced by the loading angle below 70°, whereas its impact becomes negligible above this threshold. The effect of interface roughness on the specimens varies with the loading angle. Specifically, within the range of 15° to 65°, an increase in interface roughness significantly enhances the peak load, improving the bearing capacity of the rock-concrete structure. The failure pattern of the specimens is dictated by the differences in stress states at the interface. A rough interface, however, enhances the bonding and interlocking effect between concrete and rock, influencing the failure pattern. These findings offer deeper insights into the failure mechanisms at the rock-concrete interface and provide valuable implications for engineering applications.

The trial-and-error approach for calibrating the meso-parameters of the parallel bond model is cumbersome and time-consuming, and it fails to quantitatively assess the correlation between cracks from numerical simulations and laboratory tests. The meso-parameter ranges of the parallel bond model over the past decade were summarized, and the crack fractal dimensions post-failure in both numerical simulations and laboratory tests were calculated using the box-counting method. Based on this, a neural network model was developed using four macroscopic parameters, including elastic modulus, Poisson’s ratio, peak strength and crack fractal dimension, as inputs, and six mesoscopic parameters such as bond effective modulus, ratio of normal stiffness to shear stiffness, cohesion, friction angle, tensile strength and friction coefficient, as outputs. The calibration effects of the parallel bond model with and without considering crack fractal dimension were compared. The research indicates that: 1) The developed neural network model exhibits a good convergence rate, prediction accuracy, and generalization ability, with an error of approximately 3.34% between the test set output and the expected values. 2) Incorporating crack fractal dimension into the neural network model results in errors of less than 3.00% for macroscopic parameters like elastic modulus, peak strength, and Poisson’s ratio between numerical and laboratory tests, outperforming the calibration results without considering crack fractal dimension. 3) This approach quantitatively ensures the consistency of crack irregularity between numerical and laboratory tests. The calibrated results can partially correct the existing neural network model’s calibration, offering new insights for enhancing the calibration of the meso-parameters of the parallel bond model.

To address the issue of excessive noise in high-speed railway vibration signals acquired from the surface of shallowly buried tunnels, a high-order local maximum synchrosqueezing transform (HLMSST) method is introduced for noise reduction in high-speed railway vibration signals traversing shallow tunnels. Firstly, a high-order modulation operator is introduced to calculate the instantaneous frequency for high-order estimation. Subsequently, a HLMSST is developed by integrating the local maximum synchrosqueezing transform. Its anti-noise performance is validated through numerical simulation using a multicomponent noise signal. Next, the vibration signal from high-speed railway is analyzed using correlation time-frequency analysis to confirm the noise reduction and applicability of the HLMSST. Finally, the time-frequency and attenuation characteristics of high-speed railway vibration signals are analyzed using the correlation time-frequency analysis method. The study can provide a time-frequency analysis method that offers high spectral accuracy and effective noise reduction for vibration signals from high-speed railways in tunnel environments. It further establishes a theoretical foundation for investigating the environmental impacts of such vibrations.