Numerical instabilities like ray flickering and non-conservation of energy fre-quently occur at coplanar contact interfaces in multi-body physical simu-lations due to unaligned boundary geometries. We present CoplanarMesh,an open-source Python framework designed to resolve overlapping planarboundaries via a boundary-aligned topological remeshing pipeline. By com-bining region-growing planar clustering, loop extraction, collinear simplifica-tion, and constrained planar retriangulation governed by a unidirectional de-pendency tree, the software guarantees exact spatial alignment of shared in-terface facets across 3D meshes without cyclic dependencies. CoplanarMeshintegrates seamlessly into Python and PyTorch scientific workflows as a reli-able geometric pre-processor.

Fracturing phenomena driven by temperature variations present substantial challenges across geotechnical engineering applications. Accurate computational representation of such behavior demands robust numerical architectures that can reliably capture discontinuous crack evolution within materials characterized by pronounced directional dependencies. This paper introduces a novel approach that integrates dual-horizon non-ordinary state-based peridynamics (DH-NOSBPD) with a variational damage model for simulating thermomechanical fracture in anisotropic media. The proposed framework overcomes the limitations of conventional peridynamic methods in representing anisotropic thermal and mechanical coupling while eliminating numerical instabilities inherent in bond-breaking criteria. A staggered coupling strategy is employed to synchronize thermal and mechanical field updates, incorporating anisotropic constitutive relationships for both heat conduction and stress-strain behavior. The variational damage model introduces a history-dependent scalar damage field derived from strain energy density, thereby circumventing the spurious energy release and mesh dependence associated with abrupt bond deletion. This approach yields physically consistent crack evolution. Numerical examples validate the framework's accuracy in anisotropic heat transfer, mechanical deformation, and complex fracture patterns under combined thermo-mechanical loading. The model demonstrates superior stability and predictive capability compared to conventional bond-breaking approaches.

Freckles are a typical grain defect originating from the thermal-solutal-fluid flow within the mushy region, which can significantly increase the scrap rate of single-crystal blades. Segregation channels are a macroscopic morphology of freckle chains, typically manifested as solute-enriched eutectic phases. In this work, anisotropic permeability was incorporated into a volume-averaged model to investigate chimney behaviours and the evolution of channel segregation. We notice that the magnitude of permeability and its derivative with respect to the liquid phase volume fraction (fl) in high fl regions (fl > 0.9) determine the formation of segregation channels. In addition, the criteria for local solute redistribution (LSR), dendritic remelting, and mushy zone instability were compared to predict the formation location of segregation channels. The results indicate that LSR and remelting criteria are not necessary conditions for the formation of a channel. The mechanism of channel formation can be attributed to the instability of the mushy region. Subsequently, a novel Rayleigh number model that accounts for the amplification of the mushy instability was proposed. The study further emphasizes the synergistic effect of permeability and the characteristic length of the mushy zone on the accuracy of the Rayleigh number criterion. Using the model, the local Rayleigh number of the entire casting was calculated, and the geometric effects of freckles were determined. Finally, the sensitivity of the geometric structure and local cooling conditions of castings to the formation of segregation channels was emphasized. This study elucidates the synergistic mechanism of thermal-solutal flow and dendrite growth during the directional solidification process and provides a theoretical basis for the preparation of high-quality single-crystal blades.

The environments of high temperature and high ground stress in the process of underground coal gasification and geothermal energy development will affect the seepage characteristics of rock strata. Studying the seepage and damage characteristics of rock under temperature-stress coupling conditions is of great significance in deep underground engineering applications. Through the dynamic monitoring test system of hydraulic wetting range of coal rock under true triaxial stress, the true triaxial stress-seepage test of sandstone treated at different temperatures (25-1000 degrees C) was carried out. Based on the currently reported damage constitutive models for rock, a statistical temperature-stress-seepage damage constitutive model for sandstone considering temperature effect and gas seepage was established. The results show that the permeability of sandstone undergoes a descending stage, a constant stage, and a sharply rising stage with the increase of stress at all test temperatures. The permeability eigenvalue of sandstone fluctuates under the influence of temperature, showing a trend of decreasing, increasing, decreasing, and then increasing with the increasing of temperature. The turning point of the permeability changes with stress is affected by temperature. The higher the temperature, the smaller the initial damage stress and the more obvious the damage to sandstone. The initial fracture stress at 600 degrees C is the highest, and that at 1000 degrees C is the lowest. The permeability K of sandstone and the damage variable D show an () () exponential function relationship K = A1 expD+A2 expD+B with the correlation coefficient of deter lambda 1 lambda 2 mination (R2) greater than 0.95, indicating a certain correlation between them.

Accurately identifying the landslide area is the key to disaster prevention and reduction. However, the regional characteristics, suddenness and complexity of landslide accident make it difficult to obtain large-scale and highquality images. The few-shot conditions significantly affects the recognition abilities of the currently available neural networks. To solve this problem, we propose a new model for identifying landslide areas under few-shot conditions. The model first extracts information in multiple dimensions, such as color, space, and frequency, through a multi-channel feature decomposition process. An efficient hybrid channel-space attention mechanism (EHCS attention) is then designed and introduced to enhance the ability of the model to capture the key features. The generalization ability of the model is then enhanced by data augmentation and L2 regularization, and the hyperparameters are tuned by Bayesian optimization. Finally, a post-processing method is integrated to generate accurate landslide area identification results. The performance of the model is evaluated with the indicators including Dice similarity coefficient (DSC), volumetric overlap error (VOE) and relative volume difference (RVD) and compared with those of the original U-Net architecture and region growing image segmentation method. The DSC, VOE and RVD of the model reach 0.93, 0.14 and 0.07, respectively, significantly better than those of the baseline methods. The advantages of the model proposed in our study can be summarized as follows. First, the multi-channel feature decomposition and EHCS attention mechanism effectively improve the ability of the model to capture key features. Second, the low parameter demands and simple structure of the EHCS attention mechanism make it adaptable to the multi-layer features learning needs. In addition, under the conditions of few samples and high noises, the model shows high robustness and stability through the pre-processing, preliminary identification and post-processing steps.

An improved delayed detached-eddy simulation (IDDES) methodology is used to investigate the effect of the fineness ratio on flow characteristics of supersonic opposing jets over a blunt body. The complex shock/vortex interactions mechanisms between the opposing jet and supersonic free stream are systematically examined across various fineness ratios. Numerical results indicate that an increase in the fineness ratio reduces the pressure in the recirculation zone, which increases the jet pressure ratio. This induces a transition of the flow field from the Long Penetration Mode (LPM) to the Short Penetration Mode (SPM) characterized by a stable shock wave structure and a smaller recirculation zone. Specifically, at larger fineness ratios, the flow field demonstrates typical SPM features, maintaining effective drag reduction while significantly improving flow stability. In contrast, smaller fineness ratios achieve better drag reduction but induce strong oscillations. Further analysis of the oscillation mechanism reveals that the instability of the long penetration mode at small fineness ratios originates from Kelvin-Helmholtz(K-H) instability in the shear layer. The vortex structures generated by this instability impinge on the wall and break up during downstream development, resulting in periodic pressure disturbances that propagate within the subsonic zone. As the fineness ratio increases, the supersonic region in SPM mode effectively obstructs the propagation path of pressure disturbances, making the flow field transition from asymmetric oscillation to a symmetric stable structure.

The injection of CO2 and/or N2 into coal seams represents an effective strategy for enhancing coalbed methane (CBM) recovery while achieving carbon sequestration. Understanding the competitive adsorption dynamics and flow kinetics among these gases is essential for optimizing engineering efficiency. In this work, well-designed experiments were conducted to quantify the competitive adsorption characteristics and flow dynamics of binary and ternary CO2/CH4/N2 mixtures in coal seams. Key parameters including concentration, flow rate, adsorption amount, effective diffusion coefficient, length of adsorption zone, and separation factor were evaluated. A novel concept, i.e., the competitive adsorption coefficient (CAC) is proposed to quantitatively analyze the competitive adsorption characteristics of binary and ternary gas mixtures in coal beds, enabling temporal quantification of competitive intensity. Due to differences in adsorption capacity, the outlet flow rate exhibits a “clumping” phenomenon. The CAC values for the ternary mixture follow N2 (0.909 × 10−4 Scm3·g/s2) > CH4 > CO2 (0.132 × 10−4 Scm3·g/s2), reflecting N2's high mobility and weak adsorption versus CO2's deep penetration and prolonged retention. The separation factor correlates positively with adsorption capacity, with CO2/CH4 reaching 6.95, higher than that (0.76) of N2/CH4. Competitive displacement in enhanced coalbed methane recovery is not an instantaneous equilibrium process but a time-dependent, diffusion-controlled transport phenomenon. The Bohart-Adams model further indicates the adsorption rate hierarchy: kS2-CO2 > kS1-CO2 > kT1-CO2. These microscopic kinetic insights provide a theoretical basis for designing efficient CBM recovery and secure CO2 storage strategies.

Retropropulsion is a deceleration technique that directs thrust along the flight path and serves as the driving technology for vertical rocket recovery and reusability. The flow structure and scales of the retropropulsion flow field are governed by the complex interaction between the jet and the incoming freestream. To investigate this interaction in detail, high-fidelity numerical simulations of a single-nozzle retropropulsion configuration were employed in this work. The effects of thrust coefficients, freestream Mach numbers, and specific jet heat ratios on the flow field structure were systematically examined. Through quantitative analysis of the flow structure, the influence of these parameters was revealed on key flow features, such as the shape and location of shocks and the penetration depth of the jet. The results show that the thrust coefficient is the dominant parameter governing the flow structure. The axial position and characteristic scale of typical flow structures in the interaction region scale linearly with the square root of the thrust coefficient. Further analysis indicates that under hypersonic conditions, the flow structure is insensitive to the freestream Mach number. In contrast, the specific heat ratio of the jet gas affects the penetration and expansion behavior of the jet. Subsequently, the physical model based on the momentum conservation principle and the empirical relation are used to explain the mechanisms by which these parameters affect the flow field structure, thereby enhancing the understanding of retropropulsion flow fields.

The microstructure of crystalline rock comprises mineral aggregates and randomly distributed microcracks that strongly govern macroscopic mechanical behavior. However, existing numerical approaches struggle to simultaneously capture finite-width microcrack closure, mineralogical heterogeneity, and realistic compressive-to-tensile strength ratios. This study develops a three-dimensional heterogeneous rock model within a finite-discrete element method (FDEM) framework that explicitly represents the polycrystalline mineral structure and embeds microcracks with prescribed intensity and finite aperture. Systematic uniaxial compression simulations on granite show that increasing microcrack width and intensity increases both crack-closure strain and crack-initiation strain, while decreasing crack-closure stress and crack-initiation stress. The crack-closure stage becomes more pronounced with increasing microcrack intensity and width, but via distinct mechanisms: higher intensity lowers the initial tangent modulus, whereas greater width extends the crack-closure strain range. Increasing microcrack intensity and width reduces elastic modulus and uniaxial compressive strength, with intensity exerting the stronger influence. As intensity increases, the failure mode transitions from localized shear fracture to diffuse fragmentation. Based on these parametric analyses, we establish an efficient calibration procedure for FDEM micromechanical parameters that incorporates microcrack characteristics. The calibrated model shows excellent agreement with laboratory measurements (relative errors < 5%), reproducing the nonlinear compaction stage and granite's high compressive-to-tensile strength ratio. Applications to thermo-mechanical and hydro-mechanical coupling demonstrate that the model captures temperature-induced strength degradation and stress-controlled hydraulic fracture propagation in microcracked granite. This work provides a physically consistent framework for modeling the nonlinear compaction behavior and strength characteristics of crystalline rocks.

Coal seam water injection is a key technology for source control of hazards such as dust and gas. The complex pore-fracture structure within the coal mass directly determines the effectiveness of water injection, making it essential to clarify the distribution of coal’s microstructure and the underlying water seepage mechanisms. Among various techniques for characterizing coal structure, digital imaging technologies including Scanning Electron Microscopy (SEM), Focused Ion Beam-Scanning Electron Microscopy (FIB-SEM), and X-ray Computed Tomography (CT) are widely used for analyzing coal pores and fractures due to their capabilities in visualization and quantification. This review systematically analyzes the principles, observational scales, and image processing procedures of these digital imaging technologies, discusses their effectiveness in characterizing coal pore-fracture structures and analyzing water seepage characteristics, and further prospects their future development directions in the field of coal and rock seepage. SEM, FIB-SEM, and CT collectively form a core technical system for characterizing coal pore-fracture structures from the nanometer to micrometer scale. SEM is primarily used for surface morphological analysis in two dimensions, FIB-SEM enables non-destructive three-dimensional detection of nanopores, and CT is suitable for the three-dimensional visualization of micrometer-scale pores and fractures. The Representative Elementary Volume (REV) concept effectively addresses the scale correlation between the heterogeneity of microscopic pores and fractures in coal digital images and macroscopic engineering parameters, with the REV determined collaboratively based on multiple structural parameters being particularly important. Digital images require processing steps such as denoising, threshold segmentation, and three-dimensional reconstruction to effectively observe the pore-fracture structure. In the segmentation process, the MP-Otsu method and the deep learning approach based on U-Net demonstrate significant advantages. Seepage simulation based on three-dimensional reconstruction models from digital images is an effective pathway to reveal the intrinsic mechanisms linking pore-fracture structure and water seepage behavior. Research using controlled methods, such as the digital image artificial fracture technique, indicates that pore-fracture shape is a key factor influencing permeability, with significant differences in the contribution of different shapes. The influence of fracture aperture on seepage exhibits a nonlinear segmented characteristic and produces a coupling effect with surface roughness. Connected pores and fractures are confirmed as the primary channels for water migration, significantly enhancing the understanding of the true state of water distribution within the coal mass. Currently, digital imaging technology still faces key challenges in coal seam water injection seepage research, including the insufficient applicability of existing pore-fracture characterization methods to organic rocks like coal, difficulty in defining the permeability REV, and the contradiction between image resolution and field of view. Future research should focus on developing new in-situ dynamic imaging technologies, establishing multi-scale coupled seepage prediction models, and vigorously advancing multi-scale image fusion and super-resolution reconstruction algorithms based on artificial intelligence to promote the precise and efficient development of coal seam water injection technology.

Hydraulic fracturing in fluid-saturated porous media constitutes a challenging multiphysics problem involving the coupled evolution of fluid flow, solid deformation, and fracture propagation. This work presents a novel computational framework that integrates dual-horizon non-ordinary state-based peridynamics (DH-NOSBPD) with variational damage mechanics and Biot’s poroelasticity theory for simulating hydraulic fracture propagation in brittle porous materials. The dual-horizon formulation permits spatially varying discretization while rigorously preserving linear and angular momentum balance, thereby overcoming a fundamental constraint of conventional peridynamic methods. A dynamically consistent energy functional is constructed that couples mechanical deformation, pore pressure diffusion, and damage evolution, with governing equations derived systematically through variational principles. By employing an energetically motivated damage variable, the proposed approach obviates phenomenological bond-breaking criteria, enabling autonomous crack nucleation, propagation, branching, and coalescence without recourse to explicit fracture tracking algorithms. A stabilized formulation is adopted to suppress zero-energy modes while preserving computational efficiency. The resulting coupled system is solved via an adaptive staggered scheme that combines forward Euler integration for fluid diffusion with adaptive dynamic relaxation for quasi-static mechanical equilibrium. The framework is validated against analytical solutions and benchmark problems, including Terzaghi’s consolidation, five-spot well flow, pressure-driven fracture initiation, and interaction with natural discontinuities, demonstrating its accuracy, robustness, and capability to model complex hydraulic fracturing scenarios.

The research on gas diffusion mechanisms in coal matrix is of great significance to utilize coal gas resource and prevent dynamic disasters. However, the gas transport behavior has significant scale effect, which has great impact on the applicability of gas diffusion models and related research is not yet clear. In this paper, the mercury intrusion porosimetry, N2 adsorption, and CO2 adsorption experiment were carried out to quantitatively study the pore structure of coal particle. Then, the gas diffusion experiment was conducted to obtain the gas desorption-diffusion and adsorption-diffusion characteristics under different pressure. Finally, the applicability differences of gas diffusion models in characterizing gas migration behavior at coal particle scale and coal seam scale were discussed. Results show that the characterization of gas migration in coal exhibits significant scale effects, which is closely related to the coal structure. For millimeter scale coal particles, the bidisperse model, time-based model, and pressure-based model all have good characterization effects. For the on-site scale, the time-based model representation effect is poor due to the neglection of spatial evolution characteristic of gas diffusion coefficient, and the bidisperse model has too many parameters that are difficult to determine. The pressure-based model we proposed contains only two parameters that need to be determined and considers the dynamic evolution characteristics of diffusion coefficient on both temporal and spatial dimensions, which can better characterize the gas diffusion behavior. Through the pressure-based model we have established, the clear and observable gas pressure is correlated with the virtual and unobservable equivalent transport channel resistance, achieving precise description of the gas diffusion coefficient evolution. This research provides an insight into the gas diffusion models and has theoretical guidance significance for gas extraction practice.

The wetting of coal seams during water injection can be effectively reflected by changes in ultrasonic P-wave velocity. To investigate the ultrasonic P-wave velocity of water-bearing coal with different structures, this study examines both raw coal and briquettes. A 3D coal body model was first reconstructed using computed tomography technology. Ultrasonic measurements were then performed on wet raw coal samples to analyze the effects of water on the wave velocity of coal with varying structures. Relationships between P-wave velocity, wet ratio, porosity, and fractal dimension were explored. Additionally, ultrasonic tests were conducted on briquette samples prepared under different conditions to evaluate the influence of macroscopic cracks on ultrasonic characteristics under wetting. The results show that dry Class A samples, characterized by fewer pores, lower porosity, and smaller fractal dimensions, exhibit high P-wave velocities, but their velocities increase slowly after wetting, indicating a limited wetting effect. Conversely, Class B samples, with higher microporosity and larger fractal dimensions, which are more sensitive to water, display lower initial velocities but significant increases after wetting. This sensitivity can be attributed to the higher microporosity and greater tortuosity of micropore walls in Class B samples. The wave velocity increment is positively correlated with porosity and fractal dimension. For briquettes, a distinct bedding effect on P-wave velocity is observed. The influence of crack thickness on wave velocity increases with the degree of wetting, while crack size has the opposite effect. Under dry conditions, crack inclination is positively correlated with wave velocity; however, this correlation diminishes as wetting increases. These findings provide theoretical guidance for evaluating coal seam wetting levels using ultrasonic detection techniques.

Currently, the immersed boundary method (IBM) is widely used in fluid–structure interaction simulations. However, due to the fact that the mesh used in IBM is often fixed, when the motion or deformation range of the object is large, it causes significant waste of computing resources. We combine a fluid simulation method based on IBM and adaptive mesh refinement with a structural solver to form an efficient computational framework suitable for three-dimensional fluid–structure interaction. Three complex fluid–structure interaction cases are tested, including a flapping flag in a free stream, an inverted flag in a free stream, and an elastic wing for hovering motion. The results show that the framework significantly reduces the number of fluid mesh cells compared to fixed mesh cases (with a reduction of more than 60%) while meeting accuracy requirements. Finally, this computational framework was used to simulate the flapping of flexible wings in a butterfly model. Flexible wings are composed of skeletons and membranes. The results indicate that leading-edge vortices are generated on the inner and outer sides of the flexible wing during the downstroke, which increases the lift during the downstroke. However, during the upstroke, the leading-edge vortices and trailing edge vortices of the flexible wing are stronger, thereby increasing the thrust during the upstroke.

The growing application of physics-informed neural networks (PINNs) for solving parametric partial differential equations (PDEs) in fluid dynamics has demonstrated their potential for modeling complex multiscale flows; however, conventional PINNs often exhibit spectral bias and slow, unstable convergence, limiting accuracy in boundary layers and wakes. This research presents novel physics-informed feature decomposition in residual dense block neural networks (PI-RDB-NN), which embeds physical constraints directly into the network architecture rather than relying solely on soft constraints. PI-RDB-NN uses hierarchical residual dense blocks for multi-scale feature extraction, allocates feature channels to velocity and pressure in a 2:1 ratio consistent with two-dimensional incompressible Navier–Stokes physics, and enforces mass conservation via a learnable divergence-aware projection applied at the feature level. The model is evaluated on National Advisory Committee for Aeronautics (NACA) 0012 airfoil flow at Reynolds numbers (Re)=5000 and Re=1000 using a hybrid loss combining PDE residuals, boundary conditions, and sparse computational fluid dynamics (CFD) data. PI-RDB-NN reduces PDE residual and divergence error by 91.2% and 71.7% vs traditional PINNs (Re=5000) and by 85.5% and 85.4% vs a physics-informed Deep Operator Network (DeepONet) baseline (Re=1000). These physics consistency gains improve aerodynamic force predictions and CFD agreement, confirmed by velocity, wake, and pressure coefficient (Cp) distributions. Consistent accuracy across both Reynolds regimes supports the framework's generality, with three-dimensional and unsteady extensions identified as future work.

The wing of a butterfly consists of partially overlapping forewing and hindwing, and forewing sweeping can dynamically change the shape of the whole wing. In this work, the effect of forewing sweeping on aerodynamic performance of a butterfly like model is studied using a solver based on immersed boundary method and adaptive mesh. For aerodynamic performance, adding a “forward-backward-forward” sweeping motion to the forewing makes it more suitable for fast cruising flight, and compared to the situation without forewing sweeping, the drag is reduced by 46% and the lift to drag ratio is increased by 45%. On the contrary, adding a “backward-forward-backward” sweeping motion to the forewing increases lift and makes it more suitable for climb flight. For downstroke and middle to late upstroke, the forewing sweeping affects the Leading-Edge Vortex (LEV) through two factors: sweeping velocity and forward sweeping angle, and their effects are coupled. A large forward sweeping velocity can enhance the strength of LEV, while a large forward sweeping angle can weaken it. For early upstroke, the forewing sweeping can affect the wake capture mechanism, sweeping backward can enhance it while sweeping forward can weaken it. The findings in this work provide insight into the design of butterfly like Micro Air Vehicles (MAVs).

In order to investigate the influence of water-gas domain distribution on the relative permeability of water and gas during the injection of hot water into coal, we establish a two-dimensional pore cavity throat model based on fractal theory, and use the water gas dynamic equilibrium equation as the judgment condition for seepage calculation. The water and gas balance control equation of the cavity throat network was derived, and the distribution law of the water-gas domain during coal seam thermal injection and drainage gas production process was clarified. Finally, the relationship between water injection pressure, fractal dimension, temperature and relative permeability was obtained. In the study of constant temperature water injection process, it was found that as the injection pressure increases, the gas chamber is constantly occupied by water, the saturation of water continuously increases, and the relative permeability of water significantly increases. The injection pressure is positively correlated with the relative permeability of water. During the constant pressure heating and drainage process, the increase of free methane leads to an increase in gas pore pressure in the model, overcoming capillary forces to discharge water. The temperature is higher, the water chamber is occupied by the gas chamber, and the distribution of the gas domain is wider. When percolation occurs, the relative permeability of the gas increases sharply, which is positively correlated with temperature and conducive to the discharge of water.

The benefits of coalbed gas (CBG) development are directly affected by the production capacity level. Previous studies of CBG productivity response have mainly focused on the hydrogeochemical characteristics of produced water, and paid less attention to the correlation between microbial communities and productivity. This study focuses on nine CBG wells in the Liulin Block of the eastern Ordos Basin, systematically analyzing the CBG genesis, as well as the relationships between hydrogeochemical parameters, microbial community characteristics and CBG productivity. The results show that the CBG in Liulin Block is a mixture of secondary biogenic gas and thermogenic gas, and the proportion of biogenic methane is 48.0 % to 49.7 %. The total dissolved solids (TDS) content is positively correlated with the average daily production of CBG, and the concentration of HCO3- is negatively correlated with the average daily gas production. The concentration of elemental Sn in the produced water is significantly correlated with average daily gas production. Positive delta 13CDIC values indicate the occurrence of microbial methanogenesis. Analysis of the microbial community shows that the bacterial community exhibits significant functional diversity, with Hydrogenophaga having the highest relative abundance and a positive correlation with average daily production of CBG. The archaeal community is dominated by methanogenic archaea. From the perspective of community diversity, the Operational Taxonomic Units (OTU) number, Chao1 and Ace indexes of the archaeal community show a positive correlation with average daily gas production. This study provides a novel perspective for research on microbial indicators of CBG production.