In order to eliminate metallurgical defects and improve mechanical properties of the builds, the interlayer friction stir processing (FSP) technology was employed to assist in wire-arc additive manufacturing (WAAM) high-strength 2319 aluminium alloy, and comprehensive analysis was investigated on the microstructural evolution and mechanical properties. The grains at the bottom of the builds grew, and the proportion of recrystallized grains increased due to cyclic thermal influence, resulting in a more isotropic microstructure. By applying interlayer FSP to WAAM, the coarse columnar grains in the top regions were refined by 95%, and proportion of high angle grain boundaries (HABs) increased by 24.92 times, effectively reducing the dislocation density. During friction stir processing, the precipitation phase was fragmented, thereby providing a pinning effect that hinders dislocation accumulation. With an increase in plastic deformation, the dislocation density increased, triggering dynamic recrystallisation. Additionally, the mechanical properties of the builds prepared using hybrid method were assessed. The as-deposited builds exhibited an average ultimate tensile strength (UTS) and elongation (EL) of 225 MPa and 7.2%, respectively. By contrast, the tensile properties of stirring builds in the stable region were increased, with an average ultimate tensile strength (UTS) and elongation (EL) of 248.5 MPa and 12.7% respectively. In summary, this work provides practical guidelines for optimizing the additive manufacturing quality of high-strength aluminium alloys.

To elucidate the origin of mechanical anisotropy in cylindrical Mg alloy components containing long-periodic stacking-ordered (LPSO) phases and to provide guidance for component-level microstructure design, this study systematically investigated the combined effects of texture and oriented LPSO phases on the three-dimensional mechanical anisotropy of a cylindrical Mg-Gd-Y-Zn-Zr components fabricated by back-extrusion. The results of uniaxial tensile tests show that there are significant differences in three-dimensional mechanical properties of cylindrical components, and the plastic anisotropy is more significant compared to the strength anisotropy. By comparing Schmid factor (SF) distribution and slip mode of primary-texture and secondary-texture, it is revealed that primary-texture grains have a greater influence on strength anisotropy than secondary-texture grains. The larger SF difference between primary-texture grains and secondary-texture grains leads to significant gradient slip, uncoordinated strain and dislocation accumulation, which results in stronger dislocation strengthening, strain strengthening, and strength anisotropy. The fracture of blocky LPSO phases leads to straight cleavage planes, and the matrix is prone to the formation of numerous tearing ridges. The oriented blocky LPSO phase affects the strength anisotropy through load-bearing strengthening effect, and affects the plastic anisotropy by changing microcrack propagation mode and fracture mechanism. Overall, the oriented blocky LPSO phase has a more significant effect on mechanical anisotropy than the texture.

The selective recovery of rhenium (Re) from molybdenum (Mo)-rich hydrometallurgical solutions is a persistent challenge, hindered by the chemical similarity of the two elements and the poor performance of conventional extraction systems. This study elucidates a highly efficient separation strategy using the tertiary amine extractant N235, uniquely enabled by a synergistic phosphoric-sulfuric mixed-acid medium. We reveal a novel dual-pathway separation mechanism termed “acidity regulation and ion competition.” Specifically, sulfuric acid modulates the solution acidity to promote the dissociation of extractable molybdate anions (PMo₁₂O₄₀³⁻) into non-extractable MoO₂²⁺. Concurrently, phosphoric acid introduces a high concentration of competitive phosphate anions that saturate the extractant's active sites, effectively suppressing the co-extraction of any residual anionic Mo species. This synergistic effect, corroborated by FTIR and DFT calculations, leads to exceptional selectivity. Under optimized conditions, a single-stage Re extraction of 98.6% was achieved with only 1.4% Mo co-extraction, yielding an outstanding separation factor (βRe/Mo) of 4183. Furthermore, the loaded Re was efficiently stripped (>98.2%) using 1 mol/L aqueous ammonia, and the organic system demonstrated excellent stability, maintaining over 96.3% extraction efficiency after six cycles. This work presents not only a robust and industrially viable process but also a new mechanistic paradigm for separating chemically analogous metals by synergistically manipulating solution speciation and competitive equilibria.

Zinc anode slime (ZAS), a metallurgical byproduct generated during zinc electrowinning, contains Mn, Pb, Fe, and trace amounts of Ag. Hydrodynamic instabilities in the electrolyte circulation system cause ZAS particles to remain suspended, thereby compromising electrodeposition kinetics and the purity of cathodic zinc. Consequently, periodic removal and valorization of ZAS are necessary. This study proposes a novel hydrometallurgical route for recovering high-purity Mn₂O₃ from ZAS. The process integrates sulfuric acid leaching with thiourea (TU) reduction, followed by sequential purification, precipitation, and calcination. Under optimized conditions (ZAS:TU mass ratio of 10:3, 200 g/L H₂SO₄, solid-to-liquid ratio of 1:5, temperature of 60°C, duration of 2 hours, and 100-mesh ZAS particle size), a leaching efficiency of 91.39% for Mn is achieved. Subsequent precipitation recovers 96.21% of Mn as Mn(OH)₂, resulting in Mn₂O₃ after calcination. Kinetic analysis indicates that the reaction follows the unreacted shrinking core model, with mixed control by interfacial reactions and ion diffusion (E_a=35.87 kJ/mol). This work elucidates the reaction mechanisms and establishes a viable route for the valorization of industrial metallurgical residues.

To address the growing issue of the greenhouse effect, CO₂ capture has emerged as a key area of research. In this paper, zeolite imidazolium salt framework-8 (ZIF-8) was synthesized via a one-step method involving the linkage and in situ growth of graphitic phase carbon nitride (g-C₃N₄) on iron trioxide (Fe₂O₃). Subsequently, the material was amine-functionalised, yielding the PEI-Fe-C-ZIF-8 adsorbent, which exhibited dual endo-external recognition capability. The adsorbent exhibited a high CO₂ adsorption capacity, which can rapidly adsorb CO₂ at room temperature and low pressure. The ideal adsorbed solution theory (IAST) selectivity for CO₂/N₂ (15/85) reached 963.76 at 298 K, 0–100 kPa. Molecular simulations revealed that the material trades partial specific surface area for an increased number of adsorption sites. The synergistic effect of multiple sites facilitated the separation of CO₂/N₂ by the material.

Platinum diselenide (PtSe₂) exhibits a distinctive thickness-modulated metal-to-semiconductor transition, which makes it suitable for diverse applications in nanoelectronics and optoelectronics. This study systematically investigates the spatiotemporal dynamics of photoexcited carrier relaxation in PtSe₂. Through temperature-dependent ultrafast spectroscopy, two distinct low-frequency acoustic phonons (AP) were identified in multilayer PtSe₂. Transient absorption microscopy (TAM) measurements reveal thickness-dependent relaxation dynamics and distinct carrier diffusion behavior in multilayer PtSe₂. These findings indicate superior carrier transport properties, with a measured mobility of 394.1±38.5 cm²·V⁻¹·s⁻¹ for multilayers. Temperature-dependent ultrafast dynamics, acquired using a custom-built cryogenic pump-probe system, reveal two coherent AP modes, ω₁ and ω₂, with central frequencies of 1.27 and 0.17 THz, respectively. The higher frequency ω₁ mode corresponds to shear mode with a nominal electron-phonon coupling constant λ_ω1 = 2.22. To our knowledge, this study represents a pioneering investigation that employs low-temperature ultrafast spectroscopy to elucidate the AP mode and strong electron-phonon coupling in semimetallic PtSe₂ systems. These findings offer fundamental insights into the semimetallic nature of PtSe₂ and establish a basis for designing ultrafast photonic devices that harness its distinct optoelectronic response.

Hemihydrate phosphogypsum-based filling cementitious materials (HCM) have been widely applied in mine goaf treatment. However, the rapid setting of HCM slurry significantly limits its applicability. This study examined the impacts of citric acid (CA), sucrose (SU), and protein retarder (PR) on the setting time and strength of HCM through the determination of setting time and uniaxial compressive strength tests. The discussion addressed the factors influencing the retardation of three retarders. Hydration heat test, scanning electron microscope, mercury intrusion porosimetry and other test methods were used to analyze the retarding mechanism in alkaline environment from the perspectives of hydration and microstructure. The results indicated that the SU was the optimal choice for HCM, resulting in an extension of the final setting time by 5.45 h, and an increase in 3-day strength by 0.46 MPa with the addition of 0.05% SU. The increase in strength associated with the retarding effect was closely related to a greater enhancement in water reduction than the deterioration of retardation. The former indicated excess free water seeped out due to diminished adsorption force between particles, reducing the pores between hydration products. The latter occurred because the retarder modified the morphology of the hydration products to reduce the interlocking structures. These findings will ensure that HCM slurry is suitable for various application scenarios without compromising its strength and increasing costs.

Selective flotation of niobite is crucial for the efficient utilization of niobium resources. In this study, di(2-ethylhexyl) phosphate (DEHPA) was systematically investigated as a selective collector for niobite flotation. Microflotation experiments demonstrated that DEHPA exhibits strong collecting ability toward niobite in a weakly acidic to neutral pH range (pH 5.0–8.0), resulting in high flotation recovery. Artificially mixed minerals flotation further confirmed that DEHPA enables efficient separation of niobite from calcite and quartz at pH 6 with an initial dosage of 1 × 10⁻⁴ mol/L, indicating its excellent selectivity. Zeta potential measurements revealed a pronounced surface charge shift of niobite after DEHPA adsorption, suggesting specific interfacial interaction. Fourier transform infrared spectroscopy (FTIR) and X-ray photoelectron spectroscopy (XPS) analyses indicated that DEHPA chemically adsorbs onto the niobite surface through chemical reaction between its O atom(s) of the P(=O)-O⁻ groups and Nb(V) or Fe(II/III) species on niobite surface, generating the Nb(V) or Fe(II/III)-O-P bonds, thereby enhancing mineral hydrophobicity. These results elucidate the selective adsorption mechanism of DEHPA on niobite surface and provide theoretical guidance for the development of efficient phosphate collectors for niobium mineral flotation.

In cold regions, water within the joints and fractures of rock masses freezes into ice, significantly influencing the transmission and attenuation of blasting stress waves, and thereby affecting rock blasting outcomes such as crack propagation and fragmentation. However, the effect of ice-filled joints on the blasting response of frozen rock masses has rarely been studied. This research examines blast-induced damage and fragmentation in such rock masses using theoretical analysis and numerical simulation. Theoretically, the transmission coefficient of blasting stress waves decreases with lower ice temperatures and greater joint thickness, but increases as the joint dip angle rises from 45° to 90°. A numerical model of ice-filled joints was developed to further analyze their impact on blasting damage and fragmentation. Comparative results show that frozen rock masses with ice-filled joints experience significantly more damage and fragmentation than those with non-filled joints. Moreover, blasting-induced damage and fragmentation vary considerably with the conditions of the ice-filled joints, including thickness, angle, length, and ice temperature. These findings offer important insights for optimizing precision blasting in ice-jointed frozen rock masses.

Understanding the evolution of damage, fracturing, and permeability in shale under stress-permeability coupling is crucial for safe and efficient shale gas extraction. In this study, a coupled stress-permeability numerical model was developed to describe the fracturing behavior and permeability evolution of shale under varying bedding angles and stress conditions. The model incorporates mesoscale heterogeneity, anisotropy, and a local material degradation law to capture progressive failure. Validation against experimental data confirmed its reliability. The interlaminar strength ratio (b) was introduced to investigate the damage and failure processes of layered shale, as well as permeability evolution, under uniaxial and triaxial compression. Results indicate significant variations in failure modes and mechanical properties across bedding configurations, with permeability increasing as microcracks propagate until failure. The model was also extended to simulate hydraulic fracturing, showing that the coefficient of lateral stress (λ = σ₃/σ₁) strongly affects fracture propagation. When λ > 0.83, fractures propagate in a markedly more complex and random manner; when λ < 0.67, they predominantly align with the maximum principal stress. Five types of hydraulic fractures were identified, highlighting that trans-layer and bifurcated fractures are essential for complex fracture networks. These findings aid hydraulic fracturing design and shale gas recovery.

To investigate the fault damage process and slip mechanism when the longwall working face encounters a fault, the slip characteristics and the evolution of the stress environment around the fault during the longwall–fault distance decrease was analyzed through numerical simulation. Based on these numerical results, biaxial loading–unloading tests were performed on fault specimens to examine their damage process, shear stress evolution, and energy release characteristics under different stress states. The results show that the decrease in the lateral pressure coefficient, caused by horizontal stress reduction and vertical stress increase, is the main factor inducing fault slip. The progressive failure process of the fault involves evolution from the upper roof fault distant from the coal seam to the fault adjacent to the coal seam. The fault near the coal seam slips later, but the seismic moment and seismic energy are large. Biaxial loading-unloading test results indicate that during the stick-slip instability stage, the specimen undergoes a sudden drop in shear stress and releases energy, accompanied by tensile fracture and shear slip in the fault zone. The accuracy of the numerical simulations is verified from an experimental perspective and generalized to an engineering context.

Discovering and revealing the stress path effect and load failure mechanism of surrounding rock in steeply dipping pitching oblique stope is the basis for realizing safe and efficient mining of such coal seams. Based on the study of the evolution law of roof caving and filling in pitching oblique stope, the three-dimensional physical simulation experiment, numerical calculation, theoretical analysis and field monitoring are used to reveal the load and instability mechanism of roof rock mass structure in different areas under the non-equilibrium constraint of gangue, and the temporal and spatial evolution law of overburden mining stress in different positions along the strike and tendency of working face is deeply analyzed. The results show that there are obvious regional characteristics of roof fracture and migration in steeply dipping pitching oblique stope. The lower fracture lags behind the middle and upper parts, and the effective support area of filling gangue is about 2 / 5 of the length of working face. The basic roof stress evolves from wedge to inverted trapezoid. The roof strata are damaged by tension and compression in the middle and upper regions, and by compression in the lower region. With the advancement of the working face, the magnitude and direction of the three-dimensional mining stress of the overlying rock have changed significantly. The asymmetric deflection transfer of the stress path leads to the failure and instability of the rock layer inside the bearing arch, and the bearing arch increases first and then tends to be stable. The peak abutment pressure of surrounding rock has a cumulative effect. The stress concentration in the lower area has a lag. The peak abutment pressure in the upper area is smaller than that in the true inclined working face, and the stability of the roof is enhanced. The peak values of the bending moment in the upper, middle and lower regions of the inclination are 483.71, 306.95 and 230.26 KN·m, respectively, and the peak values are located at 2–3 m in front of the coal wall. The relationship between the ‘support-surrounding rock’ of the stope presents significant regional characteristics. The research results have important guiding significance for the stability control of surrounding rock and safe and efficient mining of scarce high-quality coal seams such as steeply dipping coal seams.

To reveal the influence of moisture content on the mechanical properties and damage of gangue cemented backfill, uniaxial compression tests were carried out on gangue cemented backfill with four different moisture contents (dry, natural, immersed, and saturated). The influence of moisture content on the characteristic parameters, energy evolution, distribution characteristics and peak point energy index of gangue cemented backfill was considered. The unit characteristic change rate and unit energy change rate were proposed to describe the degree of variation in the characteristic parameters and peak point energy indicators with moisture content. From the perspective of energy dissipation, a damage constitutive model considering the initial compaction closure and post-peak failure stages was established, and the model was revised to compensate for the shortcomings of the current damage constitutive model research. The stress – strain curve of the gangue-cemented backfill before reaching the saturated state exhibited a typical four-stage characteristic, whereas the gangue-cemented backfill in the saturated state lost the initial compaction closure stage. The characteristic parameters of the cemented gangue backfill decreased in the form of a quadratic function with the increase of moisture content. The unit characteristic change rate gradually decreased with the increase of moisture content, and the decreased amplitude gradually increased, indicating that the order of the influence of moisture state on the unit characteristic change rate of gangue cemented backfill was: drying effect < immersion effect < saturation effect. The energy evolution law of gangue cemented backfill with different moisture contents was consistent with the stage characteristics of the stress – strain curve. Before the saturated state, the elastic energy ratio curves of the gangue-cemented backfill all showed a trend of first increasing and then decreasing, whereas the dissipation energy ratio curves showed a trend of first decreasing and then increasing. The order of influence of the moisture state on the unit energy change rate of the total strain energy and elastic energy of the gangue-cemented backfill was as follows: drying effect < immersion effect < saturation effect, while the order of influence on the unit energy change rate of the dissipated energy was as follows: saturation effect < immersion effect < drying effect. With the increase in the moisture content of the gangue-cemented backfill, the failure mode of the backfill mainly underwent a transformation from tensile failure to tensile-shear mixed failure and then to “V” -shaped shear failure. The modified damage constitutive model based on energy dissipation considering the initial compaction closure stage and the post-peak failure stage has a high consistency with the test curve, and the energy dissipation curve and damage evolution curve of the cemented gangue backfill with different moisture content were also consistent. When the dissipation energy curve reached the threshold, the damage evolution curve also reached the threshold. These results provide a theoretical basis for studying the long-term stability of gangue-cemented backfill with different moisture contents in the goaf.

Rock shear failure-induced instability, characterized by pronounced nonlinearity and abrupt transitions, frequently leads to severe geological hazards in deep resource extraction and rock engineering structures. This study uses the critical phase transition theory and catastrophe theory to investigate the acoustic emission (AE) characteristics associated with the transition from stable crack propagation to dynamic shear instability. Real-time AE monitoring was performed during direct shear tests on sandstone to analyze AE responses during the evolution from microcracking to through-going fracture. The results show that during the instability stage, AE energy release, event count, and amplitude increased markedly, whereas fluctuations in AE parameter values decreased, indicating enhanced crack interactions and a sudden shift in failure mode during nonlinear instability. The b-value derived from the maximum likelihood method exhibited a significant decline, reflecting the rapid development of large fractures and the onset of instability. The variance and autocorrelation coefficient of AE energy and count exhibited a sharp increase immediately before instability. As the normal stress increased, the multifractal spectrum width (Δα) of AE energy and count gradually decreased, suggesting that high-energy AE events increasingly dominated destabilization. A novel early warning model based on swallowtail catastrophe theory was developed to overcome the limitations of conventional instability warning methods. This model accurately captures the nonlinear evolution of AE parameters and provides high predictive accuracy and engineering applicability. It is superior to existing models that use the b-value, variance, and autocorrelation coefficient as damage precursors because they exhibit only slight fluctuations before instability.

In the complex stress environment of deep underground mines, blasting disturbances are more likely to cause damage to shaft linings. To investigate the scattering of P-waves and dynamic response within a shaft lining in an inhomogeneous in-situ stress field, a theoretical model for the critical peak particle velocity (PPV) of shaft linings under the combined action of non-uniform horizontal stress and incident P-waves was established based on multiple-angle method and wave function expansion method. The theory was validated using numerical simulation. The influence of key engineering parameters on the critical PPV of the shaft lining was quantitatively analyzed. Results indicate that the critical PPV of the shaft lining decreases with increasing elastic modulus of the lining and rock, as well as increasing rock density. When the shaft lining is relatively thin, the critical PPV decreases significantly. The stress non-uniformity coefficient (β) shows a significant negative correlation with the critical PPV of the shaft lining. The relationship between the in-situ stress magnitude and the critical PPV of the shaft lining changes from positive to negative as β increases.

The seepage mechanism of fractured rock masses provides a basis for deep geothermal development, tunnel engineering, and nuclear waste disposal. This study developed a coupled thermal, hydraulic, and mechanical (THM) dual-medium model based on thermodynamic theory. The model derives constitutive relationships for saturated porous and fractured media, characterizing interactions through mass and heat exchange. Model accuracy was validated against analytical solutions for one-dimensional thermoelastic consolidation. Simulations reveal seepage evolution patterns: fracture flow velocity initially depends on aperture distribution, while pore flow follows pressure gradient; at steady-state, fracture flow is controlled by overall pressure gradient, and pore flow is controlled by fluid exchange. This transformation occurs as the fracture apertures adjust from thermo-mechanical coupling, where high-temperature injection causes thermal expansion and closure stress. The analysis shows that the mass flux creates preferential flow patterns, lateral stress causes pathway shifts, and temperature gradients lead to fracture-dominated patterns. A sensitivity analysis indicates that the fracture conductivity depends on the mechanical parameters, whereas temperature propagation relates to the convection intensity. This study explains the seepage transformation under multifield coupling for rock mass control and stability assessment.

A thorough investigation into the mechanical properties and damage characteristics of water-saturated coal samples, subsequent to adsorption-desorption under a range of gas pressures, is vital for informing and guiding engineering applications. The TMC uniaxial loading system and acoustic emission (AE) technology were used to study the effects of adsorption-desorption under different gas pressures on the physical-mechanical parameters, failure, and AE of water-saturated coal samples during splitting tests. A new damage constitutive model was established based on AE parameters. The damage evolution of water-saturated coal samples was quantitatively analyzed after adsorption-desorption under different gas pressures, revealing the damage and deterioration mechanisms in water-saturated samples during Brazilian splitting tests. Gas pressure reduced the tensile strength and splitting modulus of coal samples. The damage evolution rate of coal samples underwent an abrupt change at a critical moment, which could be qualitatively evaluated through the peak ringing counts and cumulative energy. The peak cumulative energy of coal samples gradually decreased with the increased gas pressure. The failure mode exhibited significant macroscopic fractures and severe fragmentation. Precursor to destabilization in saturated coal samples was characterized by a synchronous, sharp increase—and in some cases, an order-of-magnitude rise—in cumulative AE energy and cumulative ringing counts. b gradually increased, indicating reduced high-energy events within the sample and a progressive enhancement in the complexity of the microcrack network. According to damage mechanics theory, the damage variable derived from cumulative ringing counts could characterize tensile damage in saturated coal samples. Additionally, a continuous damage model was proposed, using AE ringing counts as a variable. The model described the damage of water-saturated coal samples during deformation failure after adsorption-desorption under different gas pressures. The findings provide a reference for the safe and efficient extraction of pressure-relief gas in high-gas, soft coal seams through hydraulic fracturing.

Microseismic monitoring and signal recognition constitute critical technologies for accurately assessing rockburst risks and ensuring the safe construction of underground rock engineering. This study developed a “surface + underground” microseismic intelligent monitoring system to evaluate dynamic disaster risks during construction at the Beishan High-Level Radioactive Waste Geological Disposal Laboratory in China. The data sets of four typical one-dimensional time-domain microseismic signals of rock fracture, blasting, TBM tunneling and drilling are constructed, and the BO-CNN-LSTM model is model was developed to identify and classify these signals. Based on the classification results, the typical time-frequency domain characteristics of the four types of signals are analyzed. The classification results of BO-CNN-LSTM, CNN, LSTM and CNN-LSTM model show that the recognition accuracy of the four models is 98.2 %, 86.7 %, 67.7 % and 92 % respectively. Among all types, the four models demonstrate the highest effectiveness in identifying rock fracture and borehole signals. The findings further confirm that the BO-CNN-LSTM model efficiently recognizes and extracts microseismic signal features, demonstrating superior performance and stability in the classification task. Finally, the study suggests several future research directions, particularly in the areas of raw signal denoising, and the automation and interpretability of feature extraction.

To address the high energy demands in aerospace protective systems, this study proposes a bioinspired nonuniform variable-thickness honeycomb (NVTH) structure derived from human femur morphology, aiming to enhance energy absorption beyond conventional hexagonal honeycombs. A mechanical model integrating the upper bound theorem of plastic mechanics and virtual work principle was developed to analyze NVTH’s structural behavior. Finite element simulations systematically evaluated deformation modes, stress-strain responses, load-bearing capacity, and energy absorption characteristics. Key findings reveal that NVTH achieves 26.26% greater energy absorption than standard hexagonal honeycombs and 25.04% improvement over traditional negative Poisson’s ratio (NPR) configurations. The thickness-gradient design enhances buckling resistance by 23.33% compared to uniform counterparts, while exhibiting NPR properties and multi-stage collapse mechanisms under quasi-static compression. Experimental validation confirms the structure’s synergistic performance enhancements through controlled material redistribution. The proposed methodology demonstrates extensibility to diverse metamaterial architectures, including NPR variants, offering a generalized framework for optimizing energy-absorbing structures in advanced engineering applications. This biomimetic approach bridges anatomical efficiency with engineered material systems, establishing new pathways for lightweight, high-performance protective solutions in aerospace and related fields.

The vibrations of the propulsion shaft systems have a critical impact on the performance and noise control of underwater vehicles. Bearings are essential for the system’s dynamic performance as the support components. Existing studies have primarily focused on the impacts of the individual bearing parameters on system vibrations, while the effects of bearing designation, clearance, and tolerance, and bearing arrangement on the multi-bearing propulsion shaft system dynamics remain unclear. There is a lack of optimization design for multi-bearing parameters in propulsion shaft systems. A comprehensive dynamic model of the multi-bearing propulsion shaft system is developed in this study, which includes key components such as support bearings and the propeller. The effects of different bearings at different positions and bearing parameters on key vibration indicators (such as acceleration and displacement) are revealed through dynamic simulations. Based on the simulation results, a vibration optimization model for a multi-bearing propulsion shaft system is proposed, which can select effective bearing parameters. The optimal bearing parameters of the propulsion shaft system can be obtained through the optimization model. The findings not only provide quantitative criteria for low-vibration design of underwater propulsion systems, but also provide a theoretical reference for modeling and vibration control of complex multi-support rotating machinery.

Tunnel cavities are often excavated in layered rock mass characterized by the pronounced cross-anisotropic behavior. To simulate these kinds of problems, a new cross-anisotropic constitutive model is developed to represent both the stiffness and strength anisotropy of layered rock mass, with the aid of a cross-anisotropic elastic compliance matrix and a proposed generalized anisotropic stress tensor, respectively. By formulating the constitutive model in second-order cone programming (SOCP) format, the SOCP optimized finite element method (FEM-SOCP) is established and applied to tunnel cavity analysis involving cross-anisotropic rock mass. The proposed generalized anisotropic stress tensor with three stress scaling factors endows the FEM-SOCP framework with promising feasibility, flexibility and practical applicability in simulating the cross-anisotropic rock mass. By introducing five independent elastic constants into the cross-anisotropic elastic compliance matrix, the bedding plane spacing can be taken into account in cross-anisotropic rock mass. Based on the analyses of three examples, namely an unsupported tunnel, a cylindrical cavity expansion and a tunnel in layered rock mass, it is found that the implicitly modeling FEM-SOCP can capture the deformation behavior and failure mode of cross-anisotropic rock mass induced by tunnel cavity, being consistent with those simulated by the explicitly modeling discrete element method (DEM).

Seismic isolation design typically emphasizes transverse responses of tunnels, with comparatively limited research on longitudinal isolation responses. Previous analytical solutions for isolation response are inapplicable to variable stiffness tunnels. To address research gaps, analytical solutions for longitudinal seismic responses of variable stiffness tunnels with isolation layers are proposed. The solution can be applied to engineering. The mechanical model of isolation layers is developed using the Kelvin model. The variable stiffness tunnel is simplified as two semi-infinite beams embedded in homogeneous and isotropic soil layers. Governing equations are solved using integral transformations and continuity conditions. Analytical expressions are obtained by introducing displacement phase angles to simulate traveling wave effects. The proposed analytical solutions are validated through comparisons with results from existing literature and verified using numerical simulations. Parametric sensitivity analyses are conducted to investigate effects of tunnels with and without an isolation layer, isolation layer thickness and elastic modulus, tunnel stiffness ratio, and wavelength and amplitude of shear waves on seismic responses of variable stiffness tunnels. Changes in stiffness have a more significant effect on internal forces than displacements. Additionally, isolation layer’s thickness and elastic modulus can be optimized through our method to balance structural performance and economic efficiency.

Pressure transients generated by two trains passing each other within an enclosed noise barrier can induce fatigue loads, cause damage to the noise barrier structures, and pose safety risks to high-speed trains. This study numerically investigated the influence of vents on pressure transients when high-speed trains pass each other at 350 km/h within an enclosed noise barrier, focusing on the vent cross-sectional area and number. Numerical simulations were conducted utilizing the Renormalization Group (RNG) k-ε turbulence model with a dynamic mesh method, and these simulations were validated against full-scale experimental results. The results indicated that vents alter the pressure waveform and significantly reduced the peak pressures induced by train intersections in the enclosed noise barrier. For a single vent, the optimal cross-sectional area ratio between the vent and the noise barrier was determined to be 0.24, achieving a 53.3 % reduction in peak-to-peak pressure. Introducing additional vents at the midpoints of the regions [ML/(1+M), (L-L_tr)/2] and [(L+L_tr]/2, L/(1+M)] optimizes the distribution of peak pressures and further mitigates the pressure amplitudes. The vents significantly contribute to the reduction of pressure transients within the enclosed noise barrier, presenting a promising solution for alleviating train-induced aerodynamic pressure in railway enclosed noise barriers.

Ensuring the operational safety of high-speed trains during earthquakes is a core challenge for China’s extensive high-speed rail network. While machine learning (ML) -based seismic response assessment has become a mainstream approach, conventional ML methods suffer from limitations such as heavy training data demands, poor interpretability, and over-reliance on deterministic predictions. This study proposes an interpretable dynamic ensemble learning model integrated with sample augmentation to predict extreme seismic responses of vehicle-track-bridge (VTB) systems. The framework combines Generative Adversarial Networks (GAN) for data generation, the Kepler Optimization Algorithm (KOA)—chosen for its superior convergence speed and optimization performance over classical algorithms—for hyperparameter tuning, and a dynamically weighted ensemble of Long Short-Term Memory (LSTM)-Attention and Support Vector Machine (SVM). A 3D nonlinear VTB model under bidirectional seismic excitation serves as the physical basis, with GAN-based augmentation mitigating data imbalance. Comprehensive validation against traditional ML models confirms significant accuracy gains, marked by reduced Mean Absolute Error (MAE) and coefficient of determination (R²) values consistently exceeding 0.97. SHapley Additive exPlanation (SHAP) analysis identifies key input features affecting wheel-rail interaction parameters, and Gaussian probabilistic interval prediction quantifies predictive uncertainty with adaptive confidence bounds. The findings offer references for seismic prediction and safety risk assessment of high-speed railways.