Ageing treatment is a simple yet effective method for enhancing the mechanical performance of magnesium (Mg) alloys. This study investigates the precipitation behaviour, precipitation-free zone (PFZ) evolution, and strengthening mechanisms of Mg–7Gd–2Nd–0.5Zr (wt.%) alloy under different ageing temperatures and times. The as-cast microstructure was found to consist mainly of α-Mg matrix, Mg₅ (Gd, Nd) phase, and a small amount of cubic GdH₂ phase. After solution treatment, the eutectic Mg₅ (Gd, Nd) phase was dissolved into the matrix, while cubic GdH₂ remained. Ageing treatment promoted the formation of fine β′ precipitates, which served as the primary strengthening agent. The key finding is that lower-temperature ageing (200 °C, 96 h) maximised the retention of β′ phases, leading to optimal strength with a yield strength (YS) of 193.3 MPa, ultimate tensile strength (UTS) of 347.8 MPa, and elongation (EL) of 2.7%. In contrast, higher-temperature ageing (250 °C, 3 h) accelerated the β′ → β₁ transformation, reduced the density of effective precipitates, and resulted in a lower YS of 185 MPa and UTS of 313.3 MPa. However, this condition concurrently produced the narrowest PFZ (110 nm), leading to an enhanced EL of 5.5%. These findings provide practical guidance for optimising ageing treatments to enhance the high-performance potential of Mg alloys in lightweight structural applications.

Alloy corrosion affects the service life and performance of materials and poses serious safety and economic risks. This study employed methods such as scanning electron microscopy, potentiodynamic polarization, and current efficiency tests to investigate the effects of Mg and Sn elements on the microstructure, electrochemical performance, and corrosive behavior of Al-5.5Zn-0.03In-0.03Ga sacrificial anodes. The results show that the grain size of Al-Zn-In-Ga alloy decreases from 366.77 µm to 261.14 µm with the addition of 1 wt.% Mg and further decreases to 192.34 µm with the addition of 0.05 wt.% Sn. Moreover, the second phase evolves from Zn/In-rich phase to the MgZn₂ phase with the addition of Mg, and further transforms into the Mg₉Sn₅ phase with the addition of Sn. The corrosion process begins with pitting and intergranular corrosion, which gradually spread to the interior of the grains. The addition of Mg and Sn in the Al-Zn-In-Ga alloy inhibits second phase and grain shedding during discharging. The Al-5.5Zn-0.03In-0.03Ga-1Mg-0.05Sn sacrificial anode exhibits the most superior discharge performance, achieving an actual discharge capacity of 2604.28 A·h/kg and a current efficiency of 91.7%.

The corrosion kinetics mechanism of HRB400 steel in alkaline and neutral environments was elucidated through advanced characterization and computational methods. Results reveal that corrosion products formed in 2% Na₂SO₄ impede ionic transport, thereby retarding the corrosion process. At 4% concentration, corrosion products rupture the surface protective film, accelerating corrosion. In simulated sulfate corrosion solutions, the competitive adsorption of OH⁻ ions induces a 0.24 eV increase in surface potential, indicating that hydroxide participation effectively mitigates erosion by highly corrosive ions. The molecular dynamics (MD) simulations were conducted in the NVT ensemble (constant number of particles, volume and temperature) at a temperature of 298.0 K and constant volume. The density functional theory (DFT) calculations were performed using the generalized gradient approximation (GGA) with the Perdew-Wang (PW91) functional with a plane-wave basis set. The charge transfer rate of SO₄²⁻ was found to be 1.24 times that of Cl⁻ in neutral solution and 1.55 times greater in alkaline conditions.

Reducing thermal conductance is an effective way to improve thermoelectric performance, which remains a challenge due to the restrictive relationship between conductance and electron thermal conductance. Based on density functional theory (DFT) and first-principles calculations combined with Nonequilibrium Green’s function method, thermoelectric properties of armchair phosphorene nanoribbons (APNRs) with divacancy (DV) defects at different positions have been investigated. The results show that defect introduction can effectively capture phonon transport behavior thereby reducing phonon thermal conductance. Thermoelectric properties of phosphorene nanodevices can be further enhanced by introducing multiple edge DV defects to construct multiple defect sequences. Meanwhile, APNRs have unique central electron transport properties, and introduction of edge defects can significantly reduce thermal conductance to synergistically regulate thermoelectric conversion performance without destroying electron transport channel of the devices. Based on phonon local resonance effect, side branching structure induced by spaced-defect sequence makes heat dissipation behavior surged, which further superimposes the effect on reducing thermal conductance of the nanodevices. Finally, a double-edged spaced-defect sequence is constructed with a theoretically predicted ZT value of 0.94, which is more than two times compared to intact nanoribbon system. The above theoretical studies provide profound theoretical support for the development of high-performance thermoelectric nanodevices.

Magnesium alloys, among the most promising biomaterials for orthopedic applications, face challenges with post-implantation infection. Copper offers potent antibacterial activity while exhibiting low biotoxicity at appropriate concentrations. This study investigated the incorporation of copper oxide nanoparticles into micro-arc oxidation (MAO) electrolytes to develop a coating combining enhanced antibacterial performance with improved corrosion resistance for Mg alloys. We systematically examined the influence of Cu content on the microstructure, corrosion resistance, antibacterial efficacy, cytotoxicity, and osteogenic properties of the coated Mg alloy samples. Electrochemical tests demonstrated that MAO coatings incorporating 1 g/L and 3 g/L CuO significantly enhanced corrosion resistance, and the corrosion rates were reduced to 0.16 mm/y and 0.38 mm/y, respectively. In immersion tests, the lowest corrosion rate of 0.31 mm/y was recorded for the 1 g/L CuO coating, which represents a 40% reduction compared to the 0 g/L CuO coating. However, further increases in CuO concentration degraded the coating’s protective properties. Antibacterial assays revealed excellent efficacy against both Staphylococcus aureus (S. aureus) and Escherichia coli (E. coli) for coatings containing ⩾3 g/L CuO. In vivo animal testing indicated that the 3 g/L CuO MAO coating promoted optimal osteogenesis, with substantial new bone formation observed after 4 weeks in vivo. Based on the comprehensive in vitro and in vivo results, the MAO coating modified with 3 g/L CuO exhibited the greatest potential for orthopedic implant applications, offering a balanced combination of corrosion resistance, antibacterial activity, biocompatibility, and osteogenic capability.

Graphitic carbon nitride (g-C₃N₄) has been widely applied in advanced oxidation processes based on persulfate (PS) for photocatalytic degradation aqueous pollutants, yet it still suffers from limitations such as weak redox capability, low electrical conductivity and severe charge recombination. In this study, via building a confined environment, the doped-C and nitrogen vacancy (Nv) were simultaneously introduced in g-C₃N₄ through one-step calcination. Compared to CN-M derived from melamine, the urea-derived CN-U exhibits higher concentrations of doped-C and Nv, which leads to the different band structures. The valence band (VB) and conduction band (CB) of CN-M shift more positively than those for CN-U with ΔE_VB and ΔE_CB being 0.31 and 0.36 eV, respectively. As a result, a Z-type g-C₃N₄/g-C₃N₄ homojunction (CN-UM) derived from the mixture of urea and melamine was constructed with the minimum resistance, the lowest charge recombination rate and the high redox capacity retained. The tetracycline degradation efficiency and degradation rate constant by CN-UM coupling with PS reach 99% and 0.08989 min⁻¹, respectively, after irradiation for 60 min, along with the excellent cycling stability. The active species h⁺, ·O₂⁻, ·OH and

The Cu-Sb-Te system has attracted keen interest because of the existence of Cu₂Te, Cu₄Te₃, Sb₂Te₃, etc., which show extraordinary properties in thermoelectricity. This work establishes the first complete experimental liquidus projection of the Cu-Sb-Te system based on the microstructural and compositional analysis of a series of key as-cast alloys. Key findings include: 12 primary solidification fields: (Cu), (Sb), (Te), Sb₂Te₃, CuTe, Cu₄Te₃, ζ/ε-Cu₂Te, γ-(Sb, Te), δ-(Sb, Te), β-Cu₃Sb, η-Cu₂Sb and ternary τ-Cu₂Sb_xTe_1−x; a Cu-rich liquid immiscibility dome; 13 invariant reactions including 3 eutectic reactions (E type), 8 transition reactions (U type), 1 peritectic reaction (P type), and a dual-liquid invariant reaction. The projection provides a diagram foundation for designing Cu-Te thermoelectrics.

The synergistic interaction between carbon materials and metals has been widely utilized in industrial selective hydrocarbon oxidation. In this study, a pronounced synergistic effect between g-C₃N₄-M and MnO₂ nanoparticles was observed, which significantly promoted the oxidation of cumene, facilitated the decomposition of cumene hydroperoxide (CHP), and enhanced the selectivity toward 2-phenyl-2-propanol (PP). Specifically, the 30-MnO₂/g-C₃N₄-M catalyst achieved a cumene conversion of 73.83% with a PP selectivity of 69.61%, representing the most efficient performance among all investigated catalysts. The structural characteristics of the synthesized catalysts were systematically analyzed by SEM, TEM, XRD, FTIR, and XPS. Furthermore, gas chromatography confirmed PP as the predominant reaction product. Mechanistic investigations revealed that the decomposition of CHP constitutes a crucial step in the cumene oxidation process. The facile synthesis and low cost of g-C₃N₄-M catalysts offer significant advantages for industrialscale aromatic hydrocarbon oxidation.

The occurrence of rockburst disasters is closely associated with spatial structures. Taking the complex geological setting of steeply inclined extra-thick coal seam groups as the research background, this study focuses on three key factors: mining depth, rock pillar thickness, and coal seam characteristics. Through an integrated approach combining mechanical analysis, numerical simulation, and microseismic monitoring, the mechanisms of rockburst under complex geological conditions were elucidated. The evolution of stress was characterized, the patterns of microseismic activity were clarified, the rockburst hazard was assessed, and zonal prevention and control measures were proposed and their effectiveness evaluated. Results show that: (1) The rock pillar acts as the primary disaster-inducing structure. The bending deformation and energy accumulation effects of the rock pillar are positively correlated with mining depth and inversely correlated with rock pillar thickness. The superposition of static and dynamic loads triggers rockburst events, where static loads store energy and dynamic loads initiate bursts. The greater the mining depth and the narrower the rock pillar thickness, the higher the rockburst risk. (2) The rock pillar stress exceeds that of the roof, and the roof stress exceeds that of the floor. The degree of stress concentration is proportional to mining depth and inversely proportional to rock-pillar thickness. With increasing mining depth, both the intensity and spatial extent of rockburst hazards increase. The rock pillar and surrounding strata stresses during B3+6 coal seam extraction are significantly higher than those of the B1+2 seam, and the mining of B3+6 leads to stress relief in B1+2. (3) In regions where the rock pillar is narrower, microseismic events occur more frequently and release higher energy, exhibiting stronger spatiotemporal activity and resulting in a higher probability and intensity of rockburst occurrence. Compared with the mining of the B1+2 coal seam, extraction of the B3+6 seam produces more frequent and energetic microseismic events with enhanced spatiotemporal activity, thereby increasing the likelihood of rockburst. (4) Zonal prevention and control measures were proposed and implemented at the +425 level. Microseismic monitoring indicated that, following the implementation of these measures, the energy of microseismic events significantly decreased while their frequency notably increased. Borehole observations revealed extensive fracture development within the rock pillars, reducing their structural integrity. No rockburst occurred during the extraction of the B3+6 coal seam at the +425 level. Comprehensive assessment demonstrated that, after the implementation of the measures, the coal–rock mass stored less energy, dissipated energy more efficiently, and released energy in a controlled and distributed manner, thereby substantially reducing the risk of rockburst. These findings provide a scientific basis for safe mining in other mines with similar geological conditions.

The comprehensive analysis of the constitutive relationship of the shear behavior of rock joints is crucial for predicting geologic hazards and ensuring the stability of underground engineering. This study investigates the full-process shear stress-strain relationship of sawtooth joints, with a particular focus on the residual stage, which has been largely overlooked in previous research. Through numerical simulations, the shear stress-strain curves of sawtooth joints under various sawtooth inclination angles and normal stress conditions are analyzed. The study reveals that the complete stress-strain curve consists of four distinct stages: shearing compaction, elastic, softening, and residual stages, with the residual stage exhibiting cyclic oscillations under certain conditions. A novel constitutive model is proposed to accurately describe the entire shear process, especially the residual stage, by integrating the effects of sawtooth inclination and normal stress. The model’s validity is confirmed through comparisons with experimental data from the literature. The development of a comprehensive constitutive model that fits the entire shear stress-strain curve, providing a more accurate representation of the shear behavior of sawtooth joints. Simultaneously, the derivation of functional relationships between peak strength, residual strength, and sawtooth inclination angle, offering new insights into the shear mechanics of rock joints.

The discrimination of rockburst proneness is crucial for ensuring the safe construction of deep underground rock engineering. This study aims to propose a novel criterion for predicting rockburst proneness based on rock crack damage strength (σ_CD) and the load/unload response ratio (LURR) theory. Uniaxial compression (UC) tests and uniaxial compression graded load-unload (UCGL) tests were conducted on 13 distinct rocks. The consistency between σ_CD points and LURR start-rise points was analyzed, and an improved acoustic emission (AE) Kneedle algorithm was developed to quantitatively calculate σ_CD based on AE data. A novel lag time ratio index (T_CD^AE)was defined as the ratio of the loading time interval between σ_CD and peak strength. The fitting degree between σ_CD points and LURR start-rise points reached 0.983, demonstrating strong consistency. The improved AE method for determining σ_CD showed a correlation of 0.993 with the classical crack volume strain method, confirming its effectiveness. The proposed TCDAE-based criterion significantly reduced discrimination time by at least 50% while maintaining high accuracy. A novel rockburst proneness criterion based on TCDAE was established, offering improved efficiency and reliability compared to existing methods.

Under the dynamic-static coupling of explosive shock and high in-situ stress, the failure mechanisms of rocks become more complex. Therefore, this paper establishes a numerical model for blasting in deep fractured rock masses, simulating the evolution process of explosion-induced initiation and propagation of rock fractures in deep environments. The effects of confining pressure, lateral pressure coefficient, and crack inclination angle on rock damage are studied. By analyzing the displacement trend lines, the fracture failure modes are identified, revealing a new co-directional tensile failure mode. Furthermore, the fracture propagation mechanism is investigated from the perspective of dynamic stress intensity factors. The study reveals that the explosive-induced stress waves undergo scattering, reflection, and transmission when passing through fractures, resulting in a complex stress distribution around the fractures. To achieve optimal blasting outcomes in deep rock masses, it is advisable to orient boreholes in alignment with the direction of the maximum stress.

Anti-dip slopes represent a prevalent slope type in mining engineering. To systematically investigate the influence of various factors on the stability of anti-dip slopes under overburden pressure, this study conducted uniaxial compression tests on micro shrinkage layered anti-dip slope specimens. The analysis was performed on the macroscopic failure mode, stress-strain curves, and displacement fitting curves of the specimens with particular emphasis on examining how different factors affect the failure characteristics and mechanical properties of specimens. The study demonstrated that the interlayer thickness, interlayer angle, and anchorage conditions have different effects on the stability of anti-dip slope, and the bolt support anchorage significantly altered both the failure mode and mechanical properties of slope specimens. NPR bolt exhibited superior performance that enhanced axial load-bearing capacity by 6.12%. Furthermore, it effectively controlled horizontal and vertical displacements by 20.2% and 5.4%, respectively. These findings confirmed that NPR bolts could substantially improve the stability of anti-dip slopes through their exceptional mechanical properties and displacement control capabilities.

Accurate identification of crack types in rock masses is critical for understanding damage mechanisms and ensuring the structural safety of rock engineering. This study presents a novel unsupervised classification framework based on Gaussian mixture modeling (GMM) for distinguishing acoustic emission (AE) signatures associated with different fracture modes in sandstone samples that contain prefabricated fissures at varying inclination angles. The frequency-domain characteristics of the AE signals were extracted using fast Fourier transform (FFT), while the RA–AF (rise time/amplitude versus average frequency) parameter space was employed to characterize the crack mechanisms. To increase classification accuracy and model robustness, the Bayesian information criterion (BIC) was introduced to determine the optimal number of Gaussian components. Experimental results from uniaxial compression tests reveal that fissure inclination significantly affects crack evolution behavior: low-angle fissures favor shear and hybrid cracks, whereas high-angle fissures cause tensile failure. The proposed GMM-based method effectively identifies tensile, shear, and hybrid cracks with increased objectivity and accuracy, outperforming traditional empirical RA–AF thresholding techniques. This research provides a reliable and generalizable approach for AE signal classification, which presents theoretical insights and practical support for real-time monitoring, early warning, and structural health assessment in fractured rock masses.

Constructing salt caverns in deep formations poses significant challenges due to high geostresses, pronounced creep behavior, and particularly intense pressure fluctuations. This study first conducted mechanical experiments to investigate the long-term creep behavior of salt rock, and to examine the differences in its mechanical response under cyclic loading compared with traditional triaxial loading. The results revealed a confining-pressure-dependent nonlinear creep behavior as well as a degradation mechanism induced by cyclic loading. A numerical model was then developed that incorporate the nonlinear creep law with periodic parameter weakening. Comparative analyses of cavern dilatancy under cyclic versus constant pressure conditions were conducted, validating the necessity of integrating the periodic weakening mechanism into the numerical model. Results indicate that cyclic loading enhances the plastic deformation capacity while lowering its dilatancy threshold; For cyclic gas pressure (CGP) mode, a minimum operational pressure of 9.6 MPa is infeasible due to excessive sidewall convergence and extensive spalling risk zones, with 12 MPa recommended as the lower limit; The constant brine pressure (CBP) mode exhibits superior performance in controlling deformation and damage; For constant gas pressure (GP) mode, a constant pressure of 19.2 MPa results in no significant dilatancy damage zones in salt layer; Critically, neglecting the dynamic weakening of parameters induced by cyclic loading leads to substantial underestimation of long-term deformation, by 20.2% in this study, primarily accumulated during the unloading (gas production) phase. The findings are expected to provide valuable insights into deep salt caverns with high-pressure fluctuations.

In order to explore the ejection failure mechanism and energy variation law of the in-focused energy rock burst and the concentrated stress rock burst. Based on the true triaxial disturbance unloading rock test system, high-stress unloading and different second principal stress loading tests were conducted. The mechanical properties and AE characteristics of coal samples under high-stress unloading and varying secondary principal stress loading conditions have been systematically analyzed. The area of the loading and unloading curve calculates the energy of the loaded coal sample. The elastic energy and dissipation energy ratio, the pre-peak energy and post-peak energy ratio, and the dissipation energy and elastic energy ratio are used to characterize the energy accumulation, dissipation, and release behavior. The energy mutation mechanism of coal samples under different loading and unloading conditions is revealed. The results indicate that the ejection failure characteristics exhibit the characteristics of tension-shear composite failure, in which tension failure is an inevitable occurrence during the ejection failure process. Unloading is more sensitive to energy accumulation ejection failure, and the ejection failure phenomenon tends to become more evident as the unloading degree deepens. The second principal stress has a great influence on the ejection failure of stress concentration rock burst, and the intensity of ejection failure increases with the increase of the second principal stress. The energy variation law of coal samples under high stress unloading and different second principal stress loading conditions is similar. With the increase in coal sample strength, the kinetic energy, impact tendency, and ejection failure probability of coal sample ejection fragments will also increase. The more fully damaged after unloading, the more prone to ejection. The evolution characteristics of AE have three stages, such as rising period, quiet period and destruction, in which the local ejection phenomenon appears in the destruction stage. The research method is certain rationality for the analysis of the energy evolution mechanism of coal rock. The research results provide an experimental basis for the ejection failure of rock burst. To provide support for targeted classified control measures.

Silicone rubber, widely recognized for its exceptional properties, has encountered significant limitations in traditional manufacturing processes when applied to complex structures such as flexible actuators and soft robots. While additive manufacturing, particularly 3D direct writing printing, has emerged as a transformative technology for creating intricate structures with diverse materials, its application in silicone rubber for soft robotics remains underdeveloped and warrants further exploration. Therefore, to address these challenges, this study proposes 3D-printed silicone rubber for gas-driven soft robots. The extrusion process of silicone rubber was simulated using a flow field model, and its shear-thinning characteristics were verified through rheological testing to ensure that it was suitable for direct ink writing. Furthermore, the influence mechanism of wall thickness, number, length, and input air pressure on the bending deformation of the tentacle was analyzed using finite element simulations. Subsequently, a soft tentacle with a gradient structure was successfully prepared, and a pneumatic control system was built to enable clamping and extraction functions. By using direct ink writing, this study provided a new technical solution for soft robots from material property control to integrated manufacturing of functional structures. These findings are expected to enhance the development of silicone rubber for gas-driven soft robots.

The shear behavior of rock joints under dynamic disturbances is still not well understood, especially when subjected to irregular stress waveforms, which are common in real-world scenarios. In this study, a series of cyclic normal loading/unloading direct shear tests were conducted on rough granite fractures using a laboratory direct shear apparatus. The effects of different normal loading rates, unloading rates, and shear velocities on shear stress, apparent friction coefficient, normal displacement, and shear work were systematically analyzed. The experimental results indicated that as the normal loading and unloading rates increase and the shear velocity decreases, the peak shear stress and shear work decrease. Compared with quasi-static shear strength, dynamic normal stress disturbance may strengthen the dynamic shear strength or weaken it, and the strengthening/weakening degree is controlled by the normal loading/unloading rates and shear velocity. Furthermore, three distinct shear stress variation patterns (linear decay, nonlinear decay, and peak delay) are observed. These findings provide a theoretical basis for evaluating the stability of jointed rock masses under complex dynamic disturbances such as earthquakes, tidal effects, traffic loads, and blasting activities.

In response to the difficulty of controlling the stability of the surrounding rock in deep mining chambers, taking a typical deep coal mine pump chamber group as the engineering background, the deformation development mode, loosening range expansion law, and anchor (cable) bearing state of the chamber group under the influence of construction disturbance were analyzed. A compensation control method for deep chamber group excavation with the core of “stress compensation, grouting reinforcement, and excavation disturbance reduction” was proposed. To further verify the rationality of this method, large-scale geomechanical model tests and numerical simulation tests were conducted. The distribution and evolution of stress and displacement fields in the surrounding rock during the construction process of chamber groups were analyzed. The results show that compared with traditional control methods, using the compensation control method for deep chamber group excavation increased the stress of the shallow surrounding rock by 68.2% and reduced the surface displacement of the surrounding rock by 35.5%. Based on the above research, the field tests were conducted. The monitoring results showed that the maximum deformation of the surrounding rock was 115 mm, and the utilization rate of cable strength was 63.8%, achieving stability control of the deep chamber group.

Track slab concrete (TSC) and filling layer self-compacting concrete (FLSCC) are the key materials used in the China Railway Track System (CRTS) III slab ballastless track. Understanding the dynamic damage evolution of TSC and FLSCC under load is essential for assessing the stability and safety of the slab tracks. In this study, the damage characteristics of TSC and FLSCC were investigated under uniaxial compression based on the acoustic emission (AE) technique. The results showed that the AE events occurred in the failure process and were the most significant during the yielding stage. The AE analysis revealed that the damage of TSC and FLSCC specimens was predominantly tensile cracks, accounting for 70.33%–83.07%. When the b-value was less than 1.0 and a large amount of energy was released, it indicated the presence of large cracks. Based on the Weibull random distribution and the statistical damage constitutive model, the correlation between damage variables and AE parameters was analyzed. This study indicates that the AE technique is effective for monitoring damage evolution in the concrete materials of CRTS III slab tracks.

The evolution mechanism of the dynamic response and failure mode of moraines under dynamic loading is unclear because of the special structure of mixed coarse and fine particles and sharp edges. This work established a novel fatigue parameter (a) and a fatigue damage model for a moraine on the basis of its strain development mechanism and energy dissipation observed through cyclic loading experiments. The strain mode, dynamic strength characteristics and fatigue damage development process of the moraine were investigated. The fatigue parameters reflect the speed of fatigue damage, and the fatigue damage model describes the development process and characteristics of fatigue damage. The results revealed that the moraine exhibited a strain hardening mode and fatigue damage under low cyclic loading, and the fatigue life was related to the particle size and water content. The dynamic strength is positively correlated with the coarse particle size and negatively correlated with the water content. Furthermore, the fatigue parameters are positively correlated with the dynamic stress and negatively correlated with the particle size. The fatigue parameters converge when the water content is 8%. The fatigue parameters are influenced by the particles, water content and stress. Under different dynamic stresses, water contents and particle sizes, the moraine exhibited three fatigue damage development modes, including convex, linear and concave fatigue damage. Moreover, the development of convex damage was rapid in the early stage, and concave damage was rapid in the late stage. At a low fatigue parameter (a<1.43), the moraine exhibited convex fatigue damage, whereas the moraine exhibited concave fatigue damage at high fatigue parameters (a>1.9). This work can provide a reference for disaster assessment and prevention of moraine slopes in complex environments.

The elastic modulus of rock mass is a fundamental parameter for the surrounding rock stability analysis and the support scheme design. The traditional testing methods are mainly conducted through indoor experiments, which require further research for in-situ testing of rock mass elastic modulus. This article conducts multi type rock mass digital drilling experiments based on the intelligent rotary cutting testing system for rock masses. The response law of drilling parameters to elastic modulus has been clarified. And a rock rotational ratio energy that integrates four types of drilling parameters is proposed. The rock elastic modulus prediction models (RD-Ei models) are established. The experimental results show that the average testing errors of the model based on drilling pressure, drilling torque, and rotational ratio energy are 21.04%, 18.84%, and 6.44%, respectively. On this basis, the intelligent drilling explore system of geology is used to carry out rock drilling experiments. The identification of rock interfaces and testing of elastic modulus can be achieved. This study lays a theoretical foundation for real-time quantitative measurement of the surrounding rock elastic modulus on site.

Deformation of existing tunnels induced by adjacent deep excavation is a key construction concern. This paper constructed a Layer-wise locally adaptive activation functions physics-informed neural networks (LAAF-PINN) model, driven by physical laws of a two-stage theoretical model, to predict the deformation response of an existing tunnel to deep excavation. The precision of the solution is improved by an enhanced training on poorly convergent regions in the basis of initial model training. The proposed LAAF-PINN model does not require differential processing as used for traditional differential algorithms to outputs continuous longitudinal deformation response, and moreover, the model can accurately predict bending moment value without prior data training. Parametric analysis show that using the Swish adaptive activation function and learning rate decay strategy can reduce the loss value by at least 10 times compared to other strategies. Furthermore, a local enhancement training can effectively mitigate local convergence issues and enhance the prediction accuracy, which means the range of loss value of the physical law differential equations in the region with poor convergence was reduced about 25 times. The proposed method, verified by field measurements, shows the feasibility of intelligent real time deformation prediction for deep excavation in the proximity to existing tunnels.

Utilizing solid waste resources and lowering backfill costs are made possible by the preparation of cementitious materials as cement substitutes using magnesium slag (MS) and blast furnace slag (BFS). Uniaxial compression tests were carried out on MS-BFS-based backfill with different MS contents (20%, 30%, 40%, and 50%) and curing ages (3, 7, and 28 d) to investigate their effects on the mechanical properties and energy evolution characteristics of the MS-BFS-based backfill. The coupled effects of curing age and MS content on the compressive strength and elastic modulus of the MS-BFS-based backfill are discussed. The energy damage evolution characteristics, energy distribution characteristics, and energy indexes at the peak stress point of the MS-BFS-based backfill were examined, and an energy damage constitutive model was constructed based on energy dissipation. The results show that with increasing curing age, the brittleness of the MS-BFS-based backfill specimen itself is gradually enhanced. With increasing MS content, the post-peak brittle deformation capacity of the MS-BFS-based backfill at all curing ages is enhanced, while post-peak plasticity diminishes. A moderate amount of MS (30%) improves the strength properties of the backfill and provides similar enhancement at all curing ages. At 28 days, the strength and elastic modulus of the backfill with 30% MS content can reach 7.677 MPa and 1317.063 MPa, respectively. The established two-factor coupling function can better represent the coupled effect of curing age and MS content on the mechanical parameters and energy indexes of the MS-BFS-based backfill. After introducing the pre-peak compaction coefficient, the damage constitutive model based on energy dissipation effectively characterizes the stress-strain behavior of the MS-BFS-based backfill. The findings can provide support for the application and stability analysis of MS-BFS-based backfill.

Tunnel offset and deformation in active fault zones under complex geological conditions present significant challenges to the operation of slab track structures. This study develops models for various forms of double-block slab track structures in active fault zones, incorporating concrete plastic damage theory and a cohesive zone model to analyze layered deformation, interlayer bond failure, and damage evolution under lateral deformation. The analysis reveals that when lateral deformation loads are applied to the midsection of the track structure, in addition to deformation in the bottom layer, pronounced abrupt rail displacements occur due to the vertical discontinuity of the multilayer structure. The magnitude of structural deformation is directly proportional to the lateral deformation amplitude, while the displacement gradient is inversely proportional to the length of lateral deformation. Under lateral deformation, significant bond failure occurs at the edges of the track slab, with the degree of debonding closely correlated with the length of the track slab elements, where double-block slab track elements exhibit the lowest bonding strength. As the deformation amplitude increases, structural damage progressively intensifies, beginning with the formation of diagonal cracks and primarily concentrated in the middle of the track slab.

Autonomous driving systems impose stringent requirements on the real-time performance and computational efficiency of visual perception tasks, particularly under complex and diverse adverse weather conditions. To address these challenges, a highly robust object detection method called Adverse-Det is proposed, targeting multiple harsh weather scenarios. This model introduces the visual state-space modeling module and the frequency-aware feature fusion module to achieve dual enhancement in global spatial structure modeling and local detail recovery, effectively mitigating the performance degradation caused by image quality deterioration in challenging environments such as rain, fog, sandstorms, and snow. Experimental results on the public DAWN dataset demonstrate that Adverse-Det achieves high detection accuracy across various scenes and weather conditions. Compared with baseline models, Adverse-Det improves the mean Average Precision at intersection over union thresholds from 0.5 to 0.95 (mAP_50:95) by an average of 18.7%, achieving an mAP_50:95 of 0.455 under snowy conditions. In addition, on the self-constructed real-world rainy weather driving dataset Rain-Drive, Adverse-Det achieves a 4.56% improvement in mAP_50:95. These results fully verify the effectiveness and strong generalization capability of the proposed method in complex real-world weather environments, providing solid technical support for the safe and reliable operation of autonomous driving systems under adverse weather conditions.