To solve the false detection and missed detection problems caused by various types and sizes of defects in the detection of steel surface defects,similar defects and background features,and similarities between differ...To solve the false detection and missed detection problems caused by various types and sizes of defects in the detection of steel surface defects,similar defects and background features,and similarities between different defects,this paper proposes a lightweight detection model named multiscale edge and squeeze-and-excitation attention detection network(MSESE),which is built upon the You Only Look Once version 11 nano(YOLOv11n).To address the difficulty of locating defect edges,we first propose an edge enhancement module(EEM),apply it to the process of multiscale feature extraction,and then propose a multiscale edge enhancement module(MSEEM).By obtaining defect features from different scales and enhancing their edge contours,the module uses the dual-domain selection mechanism to effectively focus on the important areas in the image to ensure that the feature images have richer information and clearer contour features.By fusing the squeeze-and-excitation attention mechanism with the EEM,we obtain a lighter module that can enhance the representation of edge features,which is named the edge enhancement module with squeeze-and-excitation attention(EEMSE).This module was subsequently integrated into the detection head.The enhanced detection head achieves improved edge feature enhancement with reduced computational overhead,while effectively adjusting channel-wise importance and further refining feature representation.Experiments on the NEU-DET dataset show that,compared with the original YOLOv11n,the improved model achieves improvements of 4.1%and 2.2%in terms of mAP@0.5 and mAP@0.5:0.95,respectively,and the GFLOPs value decreases from the original value of 6.4 to 6.2.Furthermore,when compared to current mainstream models,Mamba-YOLOT and RTDETR-R34,our method achieves superior performance with 6.5%and 8.9%higher mAP@0.5,respectively,while maintaining a more compact parameter footprint.These results collectively validate the effectiveness and efficiency of our proposed approach.展开更多
High-resolution remote sensing images(HRSIs)are now an essential data source for gathering surface information due to advancements in remote sensing data capture technologies.However,their significant scale changes an...High-resolution remote sensing images(HRSIs)are now an essential data source for gathering surface information due to advancements in remote sensing data capture technologies.However,their significant scale changes and wealth of spatial details pose challenges for semantic segmentation.While convolutional neural networks(CNNs)excel at capturing local features,they are limited in modeling long-range dependencies.Conversely,transformers utilize multihead self-attention to integrate global context effectively,but this approach often incurs a high computational cost.This paper proposes a global-local multiscale context network(GLMCNet)to extract both global and local multiscale contextual information from HRSIs.A detail-enhanced filtering module(DEFM)is proposed at the end of the encoder to refine the encoder outputs further,thereby enhancing the key details extracted by the encoder and effectively suppressing redundant information.In addition,a global-local multiscale transformer block(GLMTB)is proposed in the decoding stage to enable the modeling of rich multiscale global and local information.We also design a stair fusion mechanism to transmit deep semantic information from deep to shallow layers progressively.Finally,we propose the semantic awareness enhancement module(SAEM),which further enhances the representation of multiscale semantic features through spatial attention and covariance channel attention.Extensive ablation analyses and comparative experiments were conducted to evaluate the performance of the proposed method.Specifically,our method achieved a mean Intersection over Union(mIoU)of 86.89%on the ISPRS Potsdam dataset and 84.34%on the ISPRS Vaihingen dataset,outperforming existing models such as ABCNet and BANet.展开更多
Acidic electrochemical CO_(2) reduction(CO_(2) RR)mitigates CO_(2) loss and energy inefficiencies but suffers from limited selectivity.Insufficient understanding of the interfacial microenvironment and cation specific...Acidic electrochemical CO_(2) reduction(CO_(2) RR)mitigates CO_(2) loss and energy inefficiencies but suffers from limited selectivity.Insufficient understanding of the interfacial microenvironment and cation specificity hinders the development of efficient interfacial design methods.Here,we integrate ab initio-derived reaction kinetics with mass transfer modeling into a multiscale framework that reproduces the bell-shaped Faradaic efficiency profile inaccessible to the Butler-Volmer equations.Our results emphasize the role of hydrogen bonding in CO_(2) activation and reveal a potential-dependent shift in the rate-determining steps.We also demonstrate that cations inhibit competing hydrogen evolution by strengthening the interfacial electric field and disrupting the hydrogen-bond network.However,their accumulation near the outer Helmholtz plane induces strong steric effects,impeding CO_(2) supply.Furthermore,the parametric analysis highlights the critical role of strategies such as pressurization and pore-confined electrolyte control in overcoming interfacial CO_(2) transport limitations,enhancing selectivity,and broadening the operating potential window.This work advances a multiscale perspective on interfacial mass transfer and cation effects,establishing a unified framework for reaction interface design in acidic CO_(2) RR.展开更多
Accurately estimating depth from underwater monocular images is essential for the target tracking task of unmanned underwater vehicles.This work proposes a method based on the Lpg-Lap Unet architecture.First,the Unet ...Accurately estimating depth from underwater monocular images is essential for the target tracking task of unmanned underwater vehicles.This work proposes a method based on the Lpg-Lap Unet architecture.First,the Unet architecture integrates Laplacian pyramid depth residuals and Sobel operators to improve the boundary details in depth images,which may suffer from the feature loss caused by upsampling and the blurriness of underwater images.Multiscale local planar guidance layers then fully exploit the intermediate depth features,and a comprehensive loss function ensures robustness and accuracy.Experimental results on benchmarks demonstrate the effectiveness of Lpg-Lap Unet and its superior performance over state-of-the-art models.An underwater target tracking system is then designed to further validate its real-time capabilities in the AirSim simulation platform.展开更多
Traditional data-driven fault diagnosis methods depend on expert experience to manually extract effective fault features of signals,which has certain limitations.Conversely,deep learning techniques have gained promine...Traditional data-driven fault diagnosis methods depend on expert experience to manually extract effective fault features of signals,which has certain limitations.Conversely,deep learning techniques have gained prominence as a central focus of research in the field of fault diagnosis by strong fault feature extraction ability and end-to-end fault diagnosis efficiency.Recently,utilizing the respective advantages of convolution neural network(CNN)and Transformer in local and global feature extraction,research on cooperating the two have demonstrated promise in the field of fault diagnosis.However,the cross-channel convolution mechanism in CNN and the self-attention calculations in Transformer contribute to excessive complexity in the cooperative model.This complexity results in high computational costs and limited industrial applicability.To tackle the above challenges,this paper proposes a lightweight CNN-Transformer named as SEFormer for rotating machinery fault diagnosis.First,a separable multiscale depthwise convolution block is designed to extract and integrate multiscale feature information from different channel dimensions of vibration signals.Then,an efficient self-attention block is developed to capture critical fine-grained features of the signal from a global perspective.Finally,experimental results on the planetary gearbox dataset and themotor roller bearing dataset prove that the proposed framework can balance the advantages of robustness,generalization and lightweight compared to recent state-of-the-art fault diagnosis models based on CNN and Transformer.This study presents a feasible strategy for developing a lightweight rotating machinery fault diagnosis framework aimed at economical deployment.展开更多
Increased exposure to campus green spaces can make a positive contribution to the healthy development of students.However,understanding of the current supply of campus green space(CGS)and its drivers at different educ...Increased exposure to campus green spaces can make a positive contribution to the healthy development of students.However,understanding of the current supply of campus green space(CGS)and its drivers at different education stages is still limited.A new framework was established to evaluate the spatial heterogeneity and its influencing factors across all education stages(kindergarten,primary school,middle school,college)in 1100 schools at the urban scale of Xi’an,China.The research results show that:1)CGS is lower in the Baqiao district and higher in the Yanta and Xincheng districts of Xi’an City.‘Green wealthy schools are mainly concentrated in the Weiyang,Chang’an and Yanta districts.2)CGS of these schools in descending order is college(31.40%)>kindergarten(18.32%)>middle school(13.56%)>primary school(10.70%).3)Colleges have the most recreation sites(n(number)=2),the best education levels(11.93 yr),and the lowest housing prices(1.18×10^(4) yuan(RMB)/m^(2));middle schools have the highest public expenditures(3.97×10^(9) yuan/yr);primary schools have the highest CGS accessibility(travel time gap(TTG)=31.33).4)Multiscale Geographically Weighted Regression model and Spearman’s test prove that recreation sites have a significant positive impact on college green spaces(0.28–0.35),and education level has a significant positive impact on kindergarten green spaces(0.16–0.24).This research framework provides important insights for the assessment of school greening initiatives aimed at fostering healthier learning environments for future generations.展开更多
Metal Additive Manufacturing(MAM) technology has become an important means of rapid prototyping precision manufacturing of special high dynamic heterogeneous complex parts. In response to the micromechanical defects s...Metal Additive Manufacturing(MAM) technology has become an important means of rapid prototyping precision manufacturing of special high dynamic heterogeneous complex parts. In response to the micromechanical defects such as porosity issues, significant deformation, surface cracks, and challenging control of surface morphology encountered during the selective laser melting(SLM) additive manufacturing(AM) process of specialized Micro Electromechanical System(MEMS) components, multiparameter optimization and micro powder melt pool/macro-scale mechanical properties control simulation of specialized components are conducted. The optimal parameters obtained through highprecision preparation and machining of components and static/high dynamic verification are: laser power of 110 W, laser speed of 600 mm/s, laser diameter of 75 μm, and scanning spacing of 50 μm. The density of the subordinate components under this reference can reach 99.15%, the surface hardness can reach 51.9 HRA, the yield strength can reach 550 MPa, the maximum machining error of the components is 4.73%, and the average surface roughness is 0.45 μm. Through dynamic hammering and high dynamic firing verification, SLM components meet the requirements for overload resistance. The results have proven that MEM technology can provide a new means for the processing of MEMS components applied in high dynamic environments. The parameters obtained in the conclusion can provide a design basis for the additive preparation of MEMS components.展开更多
Based on waveform fitting,full waveform inversion(FWI)is an important inversion method with the ability to reconstruct multi-parameter models in high precision.However,the strong nonlinear equation used in FWI present...Based on waveform fitting,full waveform inversion(FWI)is an important inversion method with the ability to reconstruct multi-parameter models in high precision.However,the strong nonlinear equation used in FWI presents the following challenges,such as low convergence efficiency,high dependence on the initial model,and the energy imbalance in deep region of the inverted model.To solve these inherent problems,we develop a timedomain elastic FWI method based on gradient preconditioning with the following details:(1)the limited memory Broyden Fletcher Goldfarb Shanno method with faster convergence is adopted to im-prove the inversion stability;(2)a multi-scaled inversion strategy is used to alleviate the nonlinear inversion instead of falling into the local minimum;(3)in addition,the pseudo-Hessian preconditioned illumination operator is involved for preconditioning the parameter gradients to improve the illumination equilibrium degree of deep structures.Based on the programming implementation of the new method,a deep depression model with five diffractors is used for testing.Compared with the conventional elastic FWI method,the technique proposed by this study has better effectiveness and accuracy on the inversion effect and con-vergence,respectively.展开更多
[Objective]This study aims to develop a thermodynamically consistent phase-field framework for modeling the initiation and evolution of discontinuous structures in geomaterials.[Methods]Our model introduces crack driv...[Objective]This study aims to develop a thermodynamically consistent phase-field framework for modeling the initiation and evolution of discontinuous structures in geomaterials.[Methods]Our model introduces crack driving forces derived from the volumetric-deviatoric strain decomposition strategy,incorporating distinct tension,compression,and shear degradation mechanisms.Inertia effects capture compaction-band formation driven by wave-like disturbances,grain crushing,and frictional rearrangement.A monolithic algorithm ensures numerical stability and rapid convergence.[Results]The framework reproduces tensile,shear,mixed tensile-shear,and compressive-shear failures using the Benzeggagh-Kenane criterion.Validation against benchmark simulations-including uniaxial compression of rock-like and triaxial compression of V-notched sandstone specimens-demonstrates accurate predictions of crack initiation stress,localization orientation,and energy dissipation.[Conclusions]The framework provides a unified and robust numerical tool for analyzing the spatiotemporal evolution of strain localization and fracture in geomaterials.[Significance]By linking microscale fracture dynamics with macroscale failure within a thermodynamically consistent scheme,this study advances predictive modeling of rock stability,slope failure,and subsurface energy systems,contributing to safer and more sustainable geotechnical practice.展开更多
Magneto-electro-elastic(MEE)materials are widely utilized across various fields due to their multi-field coupling effects.Consequently,investigating the coupling behavior of MEE composite materials is of significant i...Magneto-electro-elastic(MEE)materials are widely utilized across various fields due to their multi-field coupling effects.Consequently,investigating the coupling behavior of MEE composite materials is of significant importance.The traditional finite element method(FEM)remains one of the primary approaches for addressing such issues.However,the application of FEM typically necessitates the use of a fine finite element mesh to accurately capture the heterogeneous properties of the materials and meet the required computational precision,which inevitably leads to a reduction in computational efficiency.To enhance the computational accuracy and efficiency of the FEM for heterogeneous multi-field coupling problems,this study presents the coupling magneto-electro-elastic multiscale finite element method(CM-MsFEM)for heterogeneous MEE structures.Unlike the conventional multiscale FEM(MsFEM),the proposed algorithm simultaneously constructs displacement,electric,and magnetic potential multiscale basis functions to address the heterogeneity of the corresponding parameters.The macroscale formulation of CM-MsFEM was derived,and the macroscale/microscale responses of the problems were obtained through up/downscaling calculations.Evaluation using numerical examples analyzing the transient behavior of heterogeneous MEE structures demonstrated that the proposed method outperforms traditional FEM in terms of both accuracy and computational efficiency,making it an appropriate choice for numerically modeling the dynamics of heterogeneous MEE structures.展开更多
The acoustic wave propagation in gas-saturated double-porosity materials composed of a microporous matrix and mesopores with arrays of plate-type resonators is investigated.A macroscopic description,established with t...The acoustic wave propagation in gas-saturated double-porosity materials composed of a microporous matrix and mesopores with arrays of plate-type resonators is investigated.A macroscopic description,established with the two-scale asymptotic homogenization method,evidences the combined effect of inner resonances on the acoustic properties of the respective effective visco-thermal fluid.One type of resonance originates from strong pore-scale fluid-structure interaction,while the other one arises from pressure diffusion.These phenomena respectively cause weakly and highly damped resonances,which are activated by internal momentum or mass sources,and can largely influence,depending on the material's morphology,either the effective fluid's dynamic density,compressibility,or both.We introduce semi-analytical models to illustrate the key effective properties of the studied multiscale metamaterials.The results provide insights for the bottom-up design of multiscale acoustic metamaterials with exotic behaviors,such as the negative,very slow,or supersonic phase velocity,as well as sub-wavelength bandgaps.展开更多
In contemporary society,rapid and accurate optical cable fault detection is of paramount importance for ensuring the stability and reliability of optical networks.The emergence of novel faults in optical networks has ...In contemporary society,rapid and accurate optical cable fault detection is of paramount importance for ensuring the stability and reliability of optical networks.The emergence of novel faults in optical networks has introduced new challenges,significantly compromising their normal operation.Machine learning has emerged as a highly promising approach.Consequently,it is imperative to develop an automated and reliable algorithm that utilizes telemetry data acquired from Optical Time-Domain Reflectometers(OTDR)to enable real-time fault detection and diagnosis in optical fibers.In this paper,we introduce a multi-scale Convolutional Neural Network–Bidirectional Long Short-Term Memory(CNN-BiLSTM)deep learning model for accurate optical fiber fault detection.The proposed multi-scale CNN-BiLSTM comprises three variants:the Independent Multi-scale CNN-BiLSTM(IMC-BiLSTM),the Combined Multi-scale CNN-BiLSTM(CMC-BiLSTM),and the Shared Multi-scale CNN-BiLSTM(SMC-BiLSTM).These models employ convolutional kernels of varying sizes to extract spatial features from time-series data,while leveraging BiLSTM to enhance the capture of global event characteristics.Experiments were conducted using the publicly available OTDR_data dataset,and comparisons with existing methods demonstrate the effectiveness of our approach.The results show that(i)IMC-BiLSTM,CMC-BiLSTM,and SMC-BiLSTM achieve F1-scores of 97.37%,97.25%,and 97.1%,(ii)respectively,with accuracy of 97.36%,97.23%,and 97.12%.These performances surpass those of traditional techniques.展开更多
Physiological and pathological processes such as embryonic development and tumor progression involve complicated interplay of mechanical,chemical,and biological factors cross a wealth of spatial and temporal scales.In...Physiological and pathological processes such as embryonic development and tumor progression involve complicated interplay of mechanical,chemical,and biological factors cross a wealth of spatial and temporal scales.In this paper,we review some recent advances in the field of mechano-chemo-biological coupling theories in biological tissues and cells,and their applications in cancer,immunological,and other diseases.Key issues in the mechano-chemo-biological modeling of specific dynamic processes of cells and tissues are discussed.A mechano-chemo-biological growth theory is introduced,which interrogates the mechanical,chemical,and biological coupling mechanisms underpinning the growth,remodeling and degradation of tissues such as tumors.The mechano-chemo-biological instabilities of cells and tissues are systematically analyzed,with particular attention to those induced by coupled mechano-chemo-biological mechanisms.Furthermore,we provide a mechano-chemo-biological multiscale computational framework to investigate the dynamic processes of cells and tissues,for example,the migration and metastasis of cancer cells.Besides,we discuss some recent theoretical and experimental findings in the mechano-chemo-biological dynamics of collective cells.Finally,perspectives and clinical applications of the mechanochemo-biological theories of cells and tissues are proposed.展开更多
Layered rocks(LR)exhibit inherent anisotropic stiffness and strength induced by oriented rough weakness planes,along with stress induced anisotropy and friction related plastic deformation occurs during loading.Furthe...Layered rocks(LR)exhibit inherent anisotropic stiffness and strength induced by oriented rough weakness planes,along with stress induced anisotropy and friction related plastic deformation occurs during loading.Furthermore,microcracks located in intact rock matrix(IRM)of LR are also critically important for friction and damage dissipation processes.In this paper,we first present a novel multiscale friction-damage(MFD)model using a two-step Mori-Tanaka homogenization scheme,with the aim of describing the multiscale friction-damage mechanics in LR.Physically,the initiation and propagation of flaws at different scales(i.e.microcracks and weakness planes)induced damage,and the plastic deformation is closely associated with frictional sliding along these flaws.In the thermodynamics framework,the macroscopic stress-strain relations,the local driving forces respectively conjuncted with flaws propagation and plastic deformation are derived.An analytical macroscopic strength criterion is subsequently deduced,which takes into account the variation of inclination angle and confining pressure.Notably,the failure mechanisms of IRM shearing and weakness planes sliding are inherent included in the criterion.As an original contribution,a new multisurface semi-implicit return mapping algorithm(MSRM)is developed to integrate the proposed MFD model.The robustness of MSRM algorithm is assessed by numerical tests with different loading steps sizes and convergence conditions.Finally,the effectiveness of the MFD model is confirmed using data from experiments under conventional triaxial compression,all main features of mechanical behaviors of LR are well captured by the proposed model,including initial anisotropy,stress-induced anisotropy and strain hardening/softening.展开更多
Alkaline electrolytic hydrogen production has emerged as one of the most practical methods for industrial-scale hydrogen production.However,the initial hydrolysis dissociation in alkaline media impedes the hydrogen ev...Alkaline electrolytic hydrogen production has emerged as one of the most practical methods for industrial-scale hydrogen production.However,the initial hydrolysis dissociation in alkaline media impedes the hydrogen evolution reaction(HER)kinetics of commercial catalysts.To overcome this limitation,this study focuses on the development of a highly efficient electrocatalyst for alkaline HER.Ni-based intermetallic compounds exhibit remarkable catalytic activity for HER,with the NiMo alloy being among the most active catalysts in alkaline environments.Here,we designed and fabricated self-supported multiscale porous NiZn/NiMo intermetallic compounds on a metal foam substrate using a versatile dealloying method.The resulting electrode exhibits excellent HER activity,achieving an overpotential of just 204 mV at 1000 mA/cm^(2),and dem-onstrates robust long-term catalytic stability,maintaining performance at 100 mA/cm^(2) for 400 h in an alkaline electrolyte.Thesefindings underscore the potential of nanosized intermetallic compounds fabricated via a dealloying approach to deliver exceptional catalytic performance for alkaline water electrolysis.展开更多
Cryptocurrency is a remarkable financial innovation that has affected the financial system in fundamental ways.Its increasingly complex interactions with the conventional financial market make precisely forecasting it...Cryptocurrency is a remarkable financial innovation that has affected the financial system in fundamental ways.Its increasingly complex interactions with the conventional financial market make precisely forecasting its volatility increasingly challenging.To this end,we propose a novel framework based on the evolving multiscale graph neural network(EMGNN).Specifically,we embed a graph that depicts the interactions between the cryptocurrency and conventional financial markets into the predictive process.Furthermore,we employ hierarchical evolving graph structure learners to model the dynamic and scale-specific interactions.We also evaluate our framework’s robustness and discuss its interpretability by extracting the learned graph structure.The empirical results show that(i)cryptocurrency volatility is not isolated from the conventional market,and the embedded graph can provide effective information for prediction;(ii)the EMGNN-based forecasting framework generally yields outstanding and robust performance in terms of multiple volatility estimators,cryptocurrency samples,forecasting horizons,and evaluation criteria;and(iii)the graph structure in the predictive process varies over time and scales and is well captured by our framework.Overall,our work provides new insights into risk management for market participants and into policy formulation for authorities.展开更多
The successful development of shale oil and gas reservoirs is the biggest technological revolution in the oil and gas industry.Its key technologies are horizontal well drilling and fracturing,which are based on unders...The successful development of shale oil and gas reservoirs is the biggest technological revolution in the oil and gas industry.Its key technologies are horizontal well drilling and fracturing,which are based on understanding the mechanical properties of reservoir rocks.Therefore,it is critical to obtain the reservoir mechanical parameters quickly,efficiently,and inexpensively.In this study,shale samples were collected from three basins in Southwest China,and the elastic modulus of shale in the indentation depth range of 0-5000 nm was obtained by nanoindentation experiments.Experimental results showed that different indentation depths had different physical characteristics.The shallower depths had the mechanical properties of single minerals,while the deeper depths had the mechanical properties of a multi-mineral composite.The difference between the two represented the cementation strength between the mineral particles.The error between the calculation results of the existing equivalent medium theoretical model and experimental data reached 324%.In this study,a weak cementation model was adopted,and three parameters obtained by nanoindentation experiments were considered:the soft component volume content,intergranular cementation strength,and mineral particle size.This solved the problem of assuming rather than calculating the values of some parameters in the existing model and realized the prediction of the macroscopic mechanical parameters of shale.The calculation error was reduced to less than 20%,and the test method and calculation model can be popularized and applied in engineering.展开更多
The suprachiasmatic nucleus in the hypothalamus is the master circadian clock in mammals,coordinating physiological processes with the 24-hour day–night cycle.Comprising various cell types,the suprachiasmatic nucleus...The suprachiasmatic nucleus in the hypothalamus is the master circadian clock in mammals,coordinating physiological processes with the 24-hour day–night cycle.Comprising various cell types,the suprachiasmatic nucleus(SCN)integrates environmental signals to maintain complex and robust circadian rhythms.Understanding the complexity and synchrony within SCN neurons is essential for effective circadian clock function.Synchrony involves coordinated neuronal firing for robust rhythms,while complexity reflects diverse activity patterns and interactions,indicating adaptability.Interestingly,the SCN retains circadian rhythms in vitro,demonstrating intrinsic rhythmicity.This study introduces the multiscale structural complexity method to analyze changes in SCN neuronal activity and complexity at macro and micro levels,based on Bagrov et al.’s approach.By examining structural complexity and local complexities across scales,we aim to understand how tetrodotoxin,a neurotoxin that inhibits action potentials,affects SCN neurons.Our method captures critical scales in neuronal interactions that traditional methods may overlook.Validation with the Goodwin model confirms the reliability of our observations.By integrating experimental data with theoretical models,this study provides new insights into the effects of tetrodotoxin(TTX)on neuronal complexities,contributing to the understanding of circadian rhythms.展开更多
The multiscale computational method with asymptotic analysis and reduced-order homogenization(ROH)gives a practical numerical solution for engineering problems,especially composite materials.Under the ROH framework,a ...The multiscale computational method with asymptotic analysis and reduced-order homogenization(ROH)gives a practical numerical solution for engineering problems,especially composite materials.Under the ROH framework,a partition-based unitcell structure at the mesoscale is utilized to give a mechanical state at the macro-scale quadrature point with pre-evaluated influence functions.In the past,the“1-phase,1-partition”rule was usually adopted in numerical analysis,where one constituent phase at the mesoscale formed one partition.The numerical cost then is significantly reduced by introducing an assumption that the mechanical responses are the same all the time at the same constituent,while it also introduces numerical inaccuracy.This study proposes a new partitioning method for fibrous unitcells under a reduced-order homogenization methodology.In this method,the fiber phase remains 1 partition,but the matrix phase is divided into 2 partitions,which refers to the“12”partitioning scheme.Analytical elastic influence+functions are derived by introducing the elastic strain energy equivalence(Hill-Mandel condition).This research also obtains the analytical eigenstrain influence functions by alleviating the so-called“inclusion-locking”phenomenon.In addition,a numerical approach to minimize the error of strain energy density is introduced to determine the partitioning of the matrix phase.Several numerical examples are presented to compare the differences among direct numerical simulation(DNS),“11”,and“12”partitioning schemes.The numerical simulations show improved++numerical accuracy by the“12”partitioning scheme.展开更多
Existing imaging techniques cannot simultaneously achieve high resolution and a wide field of view,and manual multi-mineral segmentation in shale lacks precision.To address these limitations,we propose a comprehensive...Existing imaging techniques cannot simultaneously achieve high resolution and a wide field of view,and manual multi-mineral segmentation in shale lacks precision.To address these limitations,we propose a comprehensive framework based on generative adversarial network(GAN)for characterizing pore structure properties of shale,which incorporates image augmentation,super-resolution reconstruction,and multi-mineral auto-segmentation.Using real 2D and 3D shale images,the framework was assessed through correlation function,entropy,porosity,pore size distribution,and permeability.The application results show that this framework enables the enhancement of 3D low-resolution digital cores by a scale factor of 8,without paired shale images,effectively reconstructing the unresolved fine-scale pores under a low resolution,rather than merely denoising,deblurring,and edge clarification.The trained GAN-based segmentation model effectively improves manual multi-mineral segmentation results,resulting in a strong resemblance to real samples in terms of pore size distribution and permeability.This framework significantly improves the characterization of complex shale microstructures and can be expanded to other heterogeneous porous media,such as carbonate,coal,and tight sandstone reservoirs.展开更多
基金funded by Ministry of Education Humanities and Social Science Research Project,grant number 23YJAZH034The Postgraduate Research and Practice Innovation Program of Jiangsu Province,grant number SJCX25_17National Computer Basic Education Research Project in Higher Education Institutions,grant number 2024-AFCEC-056,2024-AFCEC-057.
文摘To solve the false detection and missed detection problems caused by various types and sizes of defects in the detection of steel surface defects,similar defects and background features,and similarities between different defects,this paper proposes a lightweight detection model named multiscale edge and squeeze-and-excitation attention detection network(MSESE),which is built upon the You Only Look Once version 11 nano(YOLOv11n).To address the difficulty of locating defect edges,we first propose an edge enhancement module(EEM),apply it to the process of multiscale feature extraction,and then propose a multiscale edge enhancement module(MSEEM).By obtaining defect features from different scales and enhancing their edge contours,the module uses the dual-domain selection mechanism to effectively focus on the important areas in the image to ensure that the feature images have richer information and clearer contour features.By fusing the squeeze-and-excitation attention mechanism with the EEM,we obtain a lighter module that can enhance the representation of edge features,which is named the edge enhancement module with squeeze-and-excitation attention(EEMSE).This module was subsequently integrated into the detection head.The enhanced detection head achieves improved edge feature enhancement with reduced computational overhead,while effectively adjusting channel-wise importance and further refining feature representation.Experiments on the NEU-DET dataset show that,compared with the original YOLOv11n,the improved model achieves improvements of 4.1%and 2.2%in terms of mAP@0.5 and mAP@0.5:0.95,respectively,and the GFLOPs value decreases from the original value of 6.4 to 6.2.Furthermore,when compared to current mainstream models,Mamba-YOLOT and RTDETR-R34,our method achieves superior performance with 6.5%and 8.9%higher mAP@0.5,respectively,while maintaining a more compact parameter footprint.These results collectively validate the effectiveness and efficiency of our proposed approach.
基金provided by the Science Research Project of Hebei Education Department under grant No.BJK2024115.
文摘High-resolution remote sensing images(HRSIs)are now an essential data source for gathering surface information due to advancements in remote sensing data capture technologies.However,their significant scale changes and wealth of spatial details pose challenges for semantic segmentation.While convolutional neural networks(CNNs)excel at capturing local features,they are limited in modeling long-range dependencies.Conversely,transformers utilize multihead self-attention to integrate global context effectively,but this approach often incurs a high computational cost.This paper proposes a global-local multiscale context network(GLMCNet)to extract both global and local multiscale contextual information from HRSIs.A detail-enhanced filtering module(DEFM)is proposed at the end of the encoder to refine the encoder outputs further,thereby enhancing the key details extracted by the encoder and effectively suppressing redundant information.In addition,a global-local multiscale transformer block(GLMTB)is proposed in the decoding stage to enable the modeling of rich multiscale global and local information.We also design a stair fusion mechanism to transmit deep semantic information from deep to shallow layers progressively.Finally,we propose the semantic awareness enhancement module(SAEM),which further enhances the representation of multiscale semantic features through spatial attention and covariance channel attention.Extensive ablation analyses and comparative experiments were conducted to evaluate the performance of the proposed method.Specifically,our method achieved a mean Intersection over Union(mIoU)of 86.89%on the ISPRS Potsdam dataset and 84.34%on the ISPRS Vaihingen dataset,outperforming existing models such as ABCNet and BANet.
基金supported by the National Natural Science Foundation of China(52394202 and 52476056)key project of the Joint Fund for Innovation and Development of Chongqing Natural Science Foundation(CSTB2022NSCQ-LZX0013)+1 种基金the Innovative Research Group Project of the National Natural Science Foundation of China(52021004)the Natural Science Foundation of Chongqing,China(CSTB2024NSCQ-MSX0915).
文摘Acidic electrochemical CO_(2) reduction(CO_(2) RR)mitigates CO_(2) loss and energy inefficiencies but suffers from limited selectivity.Insufficient understanding of the interfacial microenvironment and cation specificity hinders the development of efficient interfacial design methods.Here,we integrate ab initio-derived reaction kinetics with mass transfer modeling into a multiscale framework that reproduces the bell-shaped Faradaic efficiency profile inaccessible to the Butler-Volmer equations.Our results emphasize the role of hydrogen bonding in CO_(2) activation and reveal a potential-dependent shift in the rate-determining steps.We also demonstrate that cations inhibit competing hydrogen evolution by strengthening the interfacial electric field and disrupting the hydrogen-bond network.However,their accumulation near the outer Helmholtz plane induces strong steric effects,impeding CO_(2) supply.Furthermore,the parametric analysis highlights the critical role of strategies such as pressurization and pore-confined electrolyte control in overcoming interfacial CO_(2) transport limitations,enhancing selectivity,and broadening the operating potential window.This work advances a multiscale perspective on interfacial mass transfer and cation effects,establishing a unified framework for reaction interface design in acidic CO_(2) RR.
基金partially supported by the Natural Science Foundation of Shandong Province,China(No.ZR2023ME009)the National Natural Science Foundation of China(No.51909252)。
文摘Accurately estimating depth from underwater monocular images is essential for the target tracking task of unmanned underwater vehicles.This work proposes a method based on the Lpg-Lap Unet architecture.First,the Unet architecture integrates Laplacian pyramid depth residuals and Sobel operators to improve the boundary details in depth images,which may suffer from the feature loss caused by upsampling and the blurriness of underwater images.Multiscale local planar guidance layers then fully exploit the intermediate depth features,and a comprehensive loss function ensures robustness and accuracy.Experimental results on benchmarks demonstrate the effectiveness of Lpg-Lap Unet and its superior performance over state-of-the-art models.An underwater target tracking system is then designed to further validate its real-time capabilities in the AirSim simulation platform.
基金supported by the National Natural Science Foundation of China(No.52277055).
文摘Traditional data-driven fault diagnosis methods depend on expert experience to manually extract effective fault features of signals,which has certain limitations.Conversely,deep learning techniques have gained prominence as a central focus of research in the field of fault diagnosis by strong fault feature extraction ability and end-to-end fault diagnosis efficiency.Recently,utilizing the respective advantages of convolution neural network(CNN)and Transformer in local and global feature extraction,research on cooperating the two have demonstrated promise in the field of fault diagnosis.However,the cross-channel convolution mechanism in CNN and the self-attention calculations in Transformer contribute to excessive complexity in the cooperative model.This complexity results in high computational costs and limited industrial applicability.To tackle the above challenges,this paper proposes a lightweight CNN-Transformer named as SEFormer for rotating machinery fault diagnosis.First,a separable multiscale depthwise convolution block is designed to extract and integrate multiscale feature information from different channel dimensions of vibration signals.Then,an efficient self-attention block is developed to capture critical fine-grained features of the signal from a global perspective.Finally,experimental results on the planetary gearbox dataset and themotor roller bearing dataset prove that the proposed framework can balance the advantages of robustness,generalization and lightweight compared to recent state-of-the-art fault diagnosis models based on CNN and Transformer.This study presents a feasible strategy for developing a lightweight rotating machinery fault diagnosis framework aimed at economical deployment.
基金Under the auspices of Natural Science Basic Research Plan in Shaanxi Province of China(No.2024JC-YBMS-196)。
文摘Increased exposure to campus green spaces can make a positive contribution to the healthy development of students.However,understanding of the current supply of campus green space(CGS)and its drivers at different education stages is still limited.A new framework was established to evaluate the spatial heterogeneity and its influencing factors across all education stages(kindergarten,primary school,middle school,college)in 1100 schools at the urban scale of Xi’an,China.The research results show that:1)CGS is lower in the Baqiao district and higher in the Yanta and Xincheng districts of Xi’an City.‘Green wealthy schools are mainly concentrated in the Weiyang,Chang’an and Yanta districts.2)CGS of these schools in descending order is college(31.40%)>kindergarten(18.32%)>middle school(13.56%)>primary school(10.70%).3)Colleges have the most recreation sites(n(number)=2),the best education levels(11.93 yr),and the lowest housing prices(1.18×10^(4) yuan(RMB)/m^(2));middle schools have the highest public expenditures(3.97×10^(9) yuan/yr);primary schools have the highest CGS accessibility(travel time gap(TTG)=31.33).4)Multiscale Geographically Weighted Regression model and Spearman’s test prove that recreation sites have a significant positive impact on college green spaces(0.28–0.35),and education level has a significant positive impact on kindergarten green spaces(0.16–0.24).This research framework provides important insights for the assessment of school greening initiatives aimed at fostering healthier learning environments for future generations.
基金funded by the National Natural Science Foundation of China Youth Fund(Grant No.62304022)Science and Technology on Electromechanical Dynamic Control Laboratory(China,Grant No.6142601012304)the 2022e2024 China Association for Science and Technology Innovation Integration Association Youth Talent Support Project(Grant No.2022QNRC001).
文摘Metal Additive Manufacturing(MAM) technology has become an important means of rapid prototyping precision manufacturing of special high dynamic heterogeneous complex parts. In response to the micromechanical defects such as porosity issues, significant deformation, surface cracks, and challenging control of surface morphology encountered during the selective laser melting(SLM) additive manufacturing(AM) process of specialized Micro Electromechanical System(MEMS) components, multiparameter optimization and micro powder melt pool/macro-scale mechanical properties control simulation of specialized components are conducted. The optimal parameters obtained through highprecision preparation and machining of components and static/high dynamic verification are: laser power of 110 W, laser speed of 600 mm/s, laser diameter of 75 μm, and scanning spacing of 50 μm. The density of the subordinate components under this reference can reach 99.15%, the surface hardness can reach 51.9 HRA, the yield strength can reach 550 MPa, the maximum machining error of the components is 4.73%, and the average surface roughness is 0.45 μm. Through dynamic hammering and high dynamic firing verification, SLM components meet the requirements for overload resistance. The results have proven that MEM technology can provide a new means for the processing of MEMS components applied in high dynamic environments. The parameters obtained in the conclusion can provide a design basis for the additive preparation of MEMS components.
基金supported by the Marine S&T Fund of Shandong Province for Pilot National Laboratory for Marine Science and Technology(Qingdao)(Grant No.2021QNLM020001)the National Key R&D Program of China(Grant No.2019YFC0605503C)+2 种基金the Major Scientific and Technological Projects of China National Petroleum Corporation(CNPC)(Grant No.ZD2019-183-003)the National Outstanding Youth Science Foundation(Grant No.41922028)the National Innovation Group Project(Grant No.41821002).
文摘Based on waveform fitting,full waveform inversion(FWI)is an important inversion method with the ability to reconstruct multi-parameter models in high precision.However,the strong nonlinear equation used in FWI presents the following challenges,such as low convergence efficiency,high dependence on the initial model,and the energy imbalance in deep region of the inverted model.To solve these inherent problems,we develop a timedomain elastic FWI method based on gradient preconditioning with the following details:(1)the limited memory Broyden Fletcher Goldfarb Shanno method with faster convergence is adopted to im-prove the inversion stability;(2)a multi-scaled inversion strategy is used to alleviate the nonlinear inversion instead of falling into the local minimum;(3)in addition,the pseudo-Hessian preconditioned illumination operator is involved for preconditioning the parameter gradients to improve the illumination equilibrium degree of deep structures.Based on the programming implementation of the new method,a deep depression model with five diffractors is used for testing.Compared with the conventional elastic FWI method,the technique proposed by this study has better effectiveness and accuracy on the inversion effect and con-vergence,respectively.
文摘[Objective]This study aims to develop a thermodynamically consistent phase-field framework for modeling the initiation and evolution of discontinuous structures in geomaterials.[Methods]Our model introduces crack driving forces derived from the volumetric-deviatoric strain decomposition strategy,incorporating distinct tension,compression,and shear degradation mechanisms.Inertia effects capture compaction-band formation driven by wave-like disturbances,grain crushing,and frictional rearrangement.A monolithic algorithm ensures numerical stability and rapid convergence.[Results]The framework reproduces tensile,shear,mixed tensile-shear,and compressive-shear failures using the Benzeggagh-Kenane criterion.Validation against benchmark simulations-including uniaxial compression of rock-like and triaxial compression of V-notched sandstone specimens-demonstrates accurate predictions of crack initiation stress,localization orientation,and energy dissipation.[Conclusions]The framework provides a unified and robust numerical tool for analyzing the spatiotemporal evolution of strain localization and fracture in geomaterials.[Significance]By linking microscale fracture dynamics with macroscale failure within a thermodynamically consistent scheme,this study advances predictive modeling of rock stability,slope failure,and subsurface energy systems,contributing to safer and more sustainable geotechnical practice.
基金supported by the National Natural Science Foundation of China(Grant Nos.42102346,42172301).
文摘Magneto-electro-elastic(MEE)materials are widely utilized across various fields due to their multi-field coupling effects.Consequently,investigating the coupling behavior of MEE composite materials is of significant importance.The traditional finite element method(FEM)remains one of the primary approaches for addressing such issues.However,the application of FEM typically necessitates the use of a fine finite element mesh to accurately capture the heterogeneous properties of the materials and meet the required computational precision,which inevitably leads to a reduction in computational efficiency.To enhance the computational accuracy and efficiency of the FEM for heterogeneous multi-field coupling problems,this study presents the coupling magneto-electro-elastic multiscale finite element method(CM-MsFEM)for heterogeneous MEE structures.Unlike the conventional multiscale FEM(MsFEM),the proposed algorithm simultaneously constructs displacement,electric,and magnetic potential multiscale basis functions to address the heterogeneity of the corresponding parameters.The macroscale formulation of CM-MsFEM was derived,and the macroscale/microscale responses of the problems were obtained through up/downscaling calculations.Evaluation using numerical examples analyzing the transient behavior of heterogeneous MEE structures demonstrated that the proposed method outperforms traditional FEM in terms of both accuracy and computational efficiency,making it an appropriate choice for numerically modeling the dynamics of heterogeneous MEE structures.
基金Project supported by the Chilean National Agency for Research and Development(ANID)through Grants ANID FONDECYT Regular(Nos.1211310 and 1250496)ANID Anillo de Tecnologia(No.ACT240015)the Polish National Science Centre(NCN)through Grant Agreement(No.2021/41/B/ST8/04492)。
文摘The acoustic wave propagation in gas-saturated double-porosity materials composed of a microporous matrix and mesopores with arrays of plate-type resonators is investigated.A macroscopic description,established with the two-scale asymptotic homogenization method,evidences the combined effect of inner resonances on the acoustic properties of the respective effective visco-thermal fluid.One type of resonance originates from strong pore-scale fluid-structure interaction,while the other one arises from pressure diffusion.These phenomena respectively cause weakly and highly damped resonances,which are activated by internal momentum or mass sources,and can largely influence,depending on the material's morphology,either the effective fluid's dynamic density,compressibility,or both.We introduce semi-analytical models to illustrate the key effective properties of the studied multiscale metamaterials.The results provide insights for the bottom-up design of multiscale acoustic metamaterials with exotic behaviors,such as the negative,very slow,or supersonic phase velocity,as well as sub-wavelength bandgaps.
基金supported in part by the Guangxi Science and Technology Department Key Research and Development Project(Grant No.23026149)in part by the Guangxi Key Research and Development Plan Project(Grant No.AB24010073).
文摘In contemporary society,rapid and accurate optical cable fault detection is of paramount importance for ensuring the stability and reliability of optical networks.The emergence of novel faults in optical networks has introduced new challenges,significantly compromising their normal operation.Machine learning has emerged as a highly promising approach.Consequently,it is imperative to develop an automated and reliable algorithm that utilizes telemetry data acquired from Optical Time-Domain Reflectometers(OTDR)to enable real-time fault detection and diagnosis in optical fibers.In this paper,we introduce a multi-scale Convolutional Neural Network–Bidirectional Long Short-Term Memory(CNN-BiLSTM)deep learning model for accurate optical fiber fault detection.The proposed multi-scale CNN-BiLSTM comprises three variants:the Independent Multi-scale CNN-BiLSTM(IMC-BiLSTM),the Combined Multi-scale CNN-BiLSTM(CMC-BiLSTM),and the Shared Multi-scale CNN-BiLSTM(SMC-BiLSTM).These models employ convolutional kernels of varying sizes to extract spatial features from time-series data,while leveraging BiLSTM to enhance the capture of global event characteristics.Experiments were conducted using the publicly available OTDR_data dataset,and comparisons with existing methods demonstrate the effectiveness of our approach.The results show that(i)IMC-BiLSTM,CMC-BiLSTM,and SMC-BiLSTM achieve F1-scores of 97.37%,97.25%,and 97.1%,(ii)respectively,with accuracy of 97.36%,97.23%,and 97.12%.These performances surpass those of traditional techniques.
基金supported by the National Natural Science Foundation of China(Grant Nos.12032014,T2488101,and 12325209)。
文摘Physiological and pathological processes such as embryonic development and tumor progression involve complicated interplay of mechanical,chemical,and biological factors cross a wealth of spatial and temporal scales.In this paper,we review some recent advances in the field of mechano-chemo-biological coupling theories in biological tissues and cells,and their applications in cancer,immunological,and other diseases.Key issues in the mechano-chemo-biological modeling of specific dynamic processes of cells and tissues are discussed.A mechano-chemo-biological growth theory is introduced,which interrogates the mechanical,chemical,and biological coupling mechanisms underpinning the growth,remodeling and degradation of tissues such as tumors.The mechano-chemo-biological instabilities of cells and tissues are systematically analyzed,with particular attention to those induced by coupled mechano-chemo-biological mechanisms.Furthermore,we provide a mechano-chemo-biological multiscale computational framework to investigate the dynamic processes of cells and tissues,for example,the migration and metastasis of cancer cells.Besides,we discuss some recent theoretical and experimental findings in the mechano-chemo-biological dynamics of collective cells.Finally,perspectives and clinical applications of the mechanochemo-biological theories of cells and tissues are proposed.
基金jointly supported by Science and Technology Projects in Guangzhou(Grant No.SL2023A04J01079)Zhejiang ProvincialWater Conservancy Science and Technology Plan Project(Grant No.RC2405)Thematic Five of the Second Scientific Expedition of Qinghai-Tibet Plateau(Grant No.2019QZKK0905).
文摘Layered rocks(LR)exhibit inherent anisotropic stiffness and strength induced by oriented rough weakness planes,along with stress induced anisotropy and friction related plastic deformation occurs during loading.Furthermore,microcracks located in intact rock matrix(IRM)of LR are also critically important for friction and damage dissipation processes.In this paper,we first present a novel multiscale friction-damage(MFD)model using a two-step Mori-Tanaka homogenization scheme,with the aim of describing the multiscale friction-damage mechanics in LR.Physically,the initiation and propagation of flaws at different scales(i.e.microcracks and weakness planes)induced damage,and the plastic deformation is closely associated with frictional sliding along these flaws.In the thermodynamics framework,the macroscopic stress-strain relations,the local driving forces respectively conjuncted with flaws propagation and plastic deformation are derived.An analytical macroscopic strength criterion is subsequently deduced,which takes into account the variation of inclination angle and confining pressure.Notably,the failure mechanisms of IRM shearing and weakness planes sliding are inherent included in the criterion.As an original contribution,a new multisurface semi-implicit return mapping algorithm(MSRM)is developed to integrate the proposed MFD model.The robustness of MSRM algorithm is assessed by numerical tests with different loading steps sizes and convergence conditions.Finally,the effectiveness of the MFD model is confirmed using data from experiments under conventional triaxial compression,all main features of mechanical behaviors of LR are well captured by the proposed model,including initial anisotropy,stress-induced anisotropy and strain hardening/softening.
基金Taishan Scholar Project of Shandong Province(No.tsqn202306226)Natural Science Foundation of Shandong Prov-ince(No.ZR2023ME155)+1 种基金The project of“20 Items of University”of Jinan(No.202228046)Luzhou Municipal Science and Technol-ogy Plan Project(Nos.2024JYJ016 and 2024JYJ018).
文摘Alkaline electrolytic hydrogen production has emerged as one of the most practical methods for industrial-scale hydrogen production.However,the initial hydrolysis dissociation in alkaline media impedes the hydrogen evolution reaction(HER)kinetics of commercial catalysts.To overcome this limitation,this study focuses on the development of a highly efficient electrocatalyst for alkaline HER.Ni-based intermetallic compounds exhibit remarkable catalytic activity for HER,with the NiMo alloy being among the most active catalysts in alkaline environments.Here,we designed and fabricated self-supported multiscale porous NiZn/NiMo intermetallic compounds on a metal foam substrate using a versatile dealloying method.The resulting electrode exhibits excellent HER activity,achieving an overpotential of just 204 mV at 1000 mA/cm^(2),and dem-onstrates robust long-term catalytic stability,maintaining performance at 100 mA/cm^(2) for 400 h in an alkaline electrolyte.Thesefindings underscore the potential of nanosized intermetallic compounds fabricated via a dealloying approach to deliver exceptional catalytic performance for alkaline water electrolysis.
基金financial support from the National Natural Science Foundation of China(Grant Nos.71971079,72271087,and 71871088)the Major Projects of the National Social Science Foundation of China(Grant No.21ZDA114)+1 种基金the National Social Science Foundation of China(Grant No.19BTJ018)the Hunan Provincial Natural Science Foundation of China(Grant No.21JJ20019).
文摘Cryptocurrency is a remarkable financial innovation that has affected the financial system in fundamental ways.Its increasingly complex interactions with the conventional financial market make precisely forecasting its volatility increasingly challenging.To this end,we propose a novel framework based on the evolving multiscale graph neural network(EMGNN).Specifically,we embed a graph that depicts the interactions between the cryptocurrency and conventional financial markets into the predictive process.Furthermore,we employ hierarchical evolving graph structure learners to model the dynamic and scale-specific interactions.We also evaluate our framework’s robustness and discuss its interpretability by extracting the learned graph structure.The empirical results show that(i)cryptocurrency volatility is not isolated from the conventional market,and the embedded graph can provide effective information for prediction;(ii)the EMGNN-based forecasting framework generally yields outstanding and robust performance in terms of multiple volatility estimators,cryptocurrency samples,forecasting horizons,and evaluation criteria;and(iii)the graph structure in the predictive process varies over time and scales and is well captured by our framework.Overall,our work provides new insights into risk management for market participants and into policy formulation for authorities.
基金supported by the Key R&D Program Project of Xinjiang Province(2024B01013)the National Key Research and Development Program of China(2022YFE0129800).
文摘The successful development of shale oil and gas reservoirs is the biggest technological revolution in the oil and gas industry.Its key technologies are horizontal well drilling and fracturing,which are based on understanding the mechanical properties of reservoir rocks.Therefore,it is critical to obtain the reservoir mechanical parameters quickly,efficiently,and inexpensively.In this study,shale samples were collected from three basins in Southwest China,and the elastic modulus of shale in the indentation depth range of 0-5000 nm was obtained by nanoindentation experiments.Experimental results showed that different indentation depths had different physical characteristics.The shallower depths had the mechanical properties of single minerals,while the deeper depths had the mechanical properties of a multi-mineral composite.The difference between the two represented the cementation strength between the mineral particles.The error between the calculation results of the existing equivalent medium theoretical model and experimental data reached 324%.In this study,a weak cementation model was adopted,and three parameters obtained by nanoindentation experiments were considered:the soft component volume content,intergranular cementation strength,and mineral particle size.This solved the problem of assuming rather than calculating the values of some parameters in the existing model and realized the prediction of the macroscopic mechanical parameters of shale.The calculation error was reduced to less than 20%,and the test method and calculation model can be popularized and applied in engineering.
基金Project supported by the National Natural Science Foundation of China(Grant Nos.12275179,11875042,and 12150410309)the Natural Science Foundation of Shanghai(Grant No.21ZR1443900).
文摘The suprachiasmatic nucleus in the hypothalamus is the master circadian clock in mammals,coordinating physiological processes with the 24-hour day–night cycle.Comprising various cell types,the suprachiasmatic nucleus(SCN)integrates environmental signals to maintain complex and robust circadian rhythms.Understanding the complexity and synchrony within SCN neurons is essential for effective circadian clock function.Synchrony involves coordinated neuronal firing for robust rhythms,while complexity reflects diverse activity patterns and interactions,indicating adaptability.Interestingly,the SCN retains circadian rhythms in vitro,demonstrating intrinsic rhythmicity.This study introduces the multiscale structural complexity method to analyze changes in SCN neuronal activity and complexity at macro and micro levels,based on Bagrov et al.’s approach.By examining structural complexity and local complexities across scales,we aim to understand how tetrodotoxin,a neurotoxin that inhibits action potentials,affects SCN neurons.Our method captures critical scales in neuronal interactions that traditional methods may overlook.Validation with the Goodwin model confirms the reliability of our observations.By integrating experimental data with theoretical models,this study provides new insights into the effects of tetrodotoxin(TTX)on neuronal complexities,contributing to the understanding of circadian rhythms.
基金funded by the National Key R&D Program of China(Grant No.2023YFA1008901)the National Natural Science Foundation of China(Grant Nos.11988102,12172009)“The Fundamental Research Funds for the Central Universities,Peking University”.
文摘The multiscale computational method with asymptotic analysis and reduced-order homogenization(ROH)gives a practical numerical solution for engineering problems,especially composite materials.Under the ROH framework,a partition-based unitcell structure at the mesoscale is utilized to give a mechanical state at the macro-scale quadrature point with pre-evaluated influence functions.In the past,the“1-phase,1-partition”rule was usually adopted in numerical analysis,where one constituent phase at the mesoscale formed one partition.The numerical cost then is significantly reduced by introducing an assumption that the mechanical responses are the same all the time at the same constituent,while it also introduces numerical inaccuracy.This study proposes a new partitioning method for fibrous unitcells under a reduced-order homogenization methodology.In this method,the fiber phase remains 1 partition,but the matrix phase is divided into 2 partitions,which refers to the“12”partitioning scheme.Analytical elastic influence+functions are derived by introducing the elastic strain energy equivalence(Hill-Mandel condition).This research also obtains the analytical eigenstrain influence functions by alleviating the so-called“inclusion-locking”phenomenon.In addition,a numerical approach to minimize the error of strain energy density is introduced to determine the partitioning of the matrix phase.Several numerical examples are presented to compare the differences among direct numerical simulation(DNS),“11”,and“12”partitioning schemes.The numerical simulations show improved++numerical accuracy by the“12”partitioning scheme.
基金Supported by the National Natural Science Foundation of China(U23A20595,52034010,52288101)National Key Research and Development Program of China(2022YFE0203400)+1 种基金Shandong Provincial Natural Science Foundation(ZR2024ZD17)Fundamental Research Funds for the Central Universities(23CX10004A).
文摘Existing imaging techniques cannot simultaneously achieve high resolution and a wide field of view,and manual multi-mineral segmentation in shale lacks precision.To address these limitations,we propose a comprehensive framework based on generative adversarial network(GAN)for characterizing pore structure properties of shale,which incorporates image augmentation,super-resolution reconstruction,and multi-mineral auto-segmentation.Using real 2D and 3D shale images,the framework was assessed through correlation function,entropy,porosity,pore size distribution,and permeability.The application results show that this framework enables the enhancement of 3D low-resolution digital cores by a scale factor of 8,without paired shale images,effectively reconstructing the unresolved fine-scale pores under a low resolution,rather than merely denoising,deblurring,and edge clarification.The trained GAN-based segmentation model effectively improves manual multi-mineral segmentation results,resulting in a strong resemblance to real samples in terms of pore size distribution and permeability.This framework significantly improves the characterization of complex shale microstructures and can be expanded to other heterogeneous porous media,such as carbonate,coal,and tight sandstone reservoirs.
