首页/文章/ 详情

【新文速递】2026年5月6日固体力学SCI期刊最新文章

5月前浏览196

今日更新:Journal of the Mechanics and Physics of Solids 1 篇,Mechanics of Materials 2 篇,Thin-Walled Structures 3 篇

Journal of the Mechanics and Physics of Solids

Infimal Representation of Landau Potentials

Christophe Denoual, Nicolas Bruzy

doi:10.1016/j.jmps.2026.106653

朗道势的有限表示

A free energy potential for an arbitrary set of stable or metastable states is defined in a large strain formulation to model displacive transitions such as martensitic transformation or twinning. This potential is based on the use of a graph of stable states (vertices of the graph) and continuous transformations (segments of the graph). This graph is built knowing, for each stable or metastable state, the point group rotations of the lattice, the elasticity tensor, and the transformation (represented as a stretch tensor) describing the continuous evolution between stable states. The potential is based on an infimal convolution of a quadratic potential representing the elastic response and a transformational potential containing the information about the possible succession of states and their energies. Thanks to its expression as an infimal operation, this potential is a function of the total and plastic strain only, and therefore defines the displacive transformation as a non-dissipative process. A particular attention is paid to the representation of elasticity for stable states (including stiffening under compression), and the definition of a non-convex potential for the transition states. Despite the inherent complexity and adaptability of the method, a closed-form expression for stress, characterized as the derivative of free energy, is being proposed. We detail through two illustrative examples the construction of potentials and examine the influence of plasticity and the convexity of the potential on the hysteresis observed in the body-centered cubic (bcc) to hexagonal close-packed (hcp) martensitic transition in iron subjected to pressure.

在大应变公式中定义了任意一组稳定或亚稳态的自由能势,以模拟马氏体相变或孪晶等位移转变。这种潜力是基于稳定状态(图的顶点)和连续转换(图的段)的图的使用。对于每一个稳定或亚稳定状态,我们都知道晶格的点群旋转、弹性张量和描述稳定状态之间连续演化的变换(用拉伸张量表示)。势是基于表示弹性响应的二次势和包含可能的状态序列及其能量信息的转换势的无限卷积。由于它的表达式是一个无穷大的操作,这个势是一个函数的总应变和塑性应变,因此定义位移转换为一个非耗散过程。特别注意稳定状态(包括压缩下的硬化)的弹性表示,以及过渡状态的非凸势的定义。尽管该方法具有固有的复杂性和适应性,但仍提出了一种以自由能导数为特征的应力封闭表达式。我们通过两个例子详细说明了势的构造,并研究了在压力作用下铁的体心立方(bcc)到六方密堆积(hcp)马氏体转变过程中所观察到的磁滞现象的塑性和势的凹凸性对磁滞的影响。


Mechanics of Materials

Delaying matrix damage in composite laminates by fibre pre-stressing: A micromechanical numerical study

Pere Maimí, Lucas de Sousa Rodrigues, Laura Carreras, Daniel Sánchez-Rodríguez, Josep Costa

doi:10.1016/j.mechmat.2026.105712

纤维预应力延缓复合材料层合板基体损伤的细观力学数值研究

Composite laminates offer high specific stiffness and strength, but their performance is often limited by matrix cracking, which acts as a precursor to delamination and provides pathways for moisture ingress and environmental degradation. Fibre prestressing can generate beneficial residual stresses in the matrix, delaying the onset of such cracks. However, the magnitude of this effect and the mechanis ms by which prestressing alters the local matrix stress state remain difficult to quantify using experimental observations alone. This work presents a numerical methodology to quantify the influence of prestressing on the matrix stress field in unidirectional plies and multidirectional laminates. A combination of representative volume elements and laminate theory is used to characterise hydrostatic pressure, von Mises stress and stress triaxiality, which are key indicators of cavitation-driven matrix failure. Illustrative results obtained for representative cross-ply laminates show that prestress levels well below the fibre strength can increase the longitudinal load required to reach comparable hydrostatic pressure levels by one order of magnitude.

复合材料层压板具有很高的比刚度和强度,但其性能通常受到基体开裂的限制,这是分层的前兆,并为水分进入和环境退化提供了途径。纤维预应力可以在基体中产生有益的残余应力,延缓此类裂缝的发生。然而,这种影响的大小和预应力改变局部基质应力状态的机制仍然难以单独使用实验观察来量化。本文提出了一种量化预应力对单向层合板和多向层合板基体应力场影响的数值方法。结合代表性体积单元和层压理论来表征静水压力、冯米塞斯应力和应力三轴性,这是空化驱动的基质破坏的关键指标。对代表性的交叉层压板获得的说明性结果表明,远低于纤维强度的预应力水平可以使达到可比静水压力水平所需的纵向载荷增加一个数量级。


Hybrid ANN/FEM-Based inverse identification procedure using the Levenberg–Marquardt method: Application to X100CrMoV5 steel alloy

Najem A., Altmeyer G., Coudray Z., Duchosal A.

doi:10.1016/j.mechmat.2026.105702

基于Levenberg-Marquardt方法的混合ANN/ fem逆识别程序:在X100CrMoV5钢合金中的应用

This study investigates the mechanical response of the X100CrMoV5 steel alloy under severe thermochemical conditions and aims to identify its constitutive parameters through an inverse procedure. This approach integrates experimental force–displacement curves, finite element numerical simulations, and artificial neural network predictions to enhance both accuracy and computational efficiency. The experimental tests are conducted using a thermo-mechanical Gleeble machine under dynamic conditions at ambient and elevated temperature. A finite element model is developed to replicate the experimental test, enabling a detailed comparison with the experimental data. Additionally, an artificial neural network, trained on numerical simulation results, is employed to predict the outcomes of the material tests, significantly accelerating the identification process. The inverse identification procedure is formulated as an optimization problem and solved using the deterministic Levenberg–Marquardt method, minimizing the cost between experimental and numerical results. The proposed hybrid identification strategy demonstrates the effectiveness of coupling numerical simulations with artificial neural networks, particularly Long Short-Term Memory (LSTM) networks, to enhance accuracy and significantly reduce computational costs. The findings of this study highlight the potential of AI-driven approaches in improving material characterization and identification.

本研究研究了X100CrMoV5钢合金在严酷热化学条件下的力学响应,旨在通过反求程序确定其本构参数。该方法将实验力-位移曲线、有限元数值模拟和人工神经网络预测相结合,提高了精度和计算效率。利用热机械式Gleeble试验机在常温和高温动态条件下进行了试验。建立了一个有限元模型来复 制实验测试,以便与实验数据进行详细的比较。此外,利用数值模拟结果训练的人工神经网络对材料试验结果进行预测,大大加快了识别过程。将反辨识过程表述为优化问题,并采用确定性Levenberg-Marquardt方法求解,使实验结果与数值结果之间的代价最小。所提出的混合识别策略证明了将数值模拟与人工神经网络,特别是长短期记忆(LSTM)网络相结合的有效性,可以提高准确性并显着降低计算成本。这项研究的发现突出了人工智能驱动的方法在改善材料表征和鉴定方面的潜力。


Thin-Walled Structures

Coupled dynamics and scaling relations of thin flexible wings with multiple distributed feathers

Guojun Li, Lanlan Wang, Weitao Jiang, Hongzhong Liu

doi:10.1016/j.tws.2026.115055

多分布羽薄柔性翼的耦合动力学及尺度关系

This study investigates the coupled aeroelastic dynamics and performance enhancement of thin flexible wings equipped with multiple distributed feathers, motivated by bio-inspired design of agile micro-air and unmanned aerial vehicles. A multi-body aeroelastic framework is developed by coupling an incompressible Navier–Stokes solver with large-eddy simulation in an arbitrary Lagrangian–Eulerian frame and a flexible multi-body structural model. The aeroelastic solver is applied to examine the aerodynamic performance of thin flexible wings without feathers, with a single leading-edge feather and with multiple overlapping feathers over a range of angles of attack, feather preset angles and feather lengths at R e = 2500. The results show that multi-feather configurations reorganize leading-edge vortex dynamics, yielding higher lift and lift-to-drag ratios than membrane-only wings. This is particularly evident in pre-stall and transitional regimes, where the design also keeps drag penalties moderate and reduces force fluctuations. A shallow positive preset angle combined with an intermediate feather length establishes a soft chordwise slot that sustains a gently convecting leading-edge vortex, accelerates pressure recovery and improves aerodynamic efficiency. When feathers become too long or too steep, blockage effects re-inflate drag and degrade performance. A new empirical scaling relation incorporating bounded feather coverage and preset-angle modifiers is proposed to predict the maximum membrane deformation under a Weber-number backbone. These findings provide quantitative guidelines for designing feathered morphing membrane wings with tunable lift, efficiency and stability across a wide angle-of-attack envelope.

基于仿生设计的敏捷微型飞行器和无人机,研究了多分布羽毛薄柔性机翼的耦合气动弹性动力学和性能增强问题。将不可压缩Navier-Stokes求解器与任意拉格朗日-欧拉框架的大涡模拟和柔性多体结构模型相结合,建立了多体气动弹性框架。应用气动弹性求解器对无羽毛、单前缘羽毛和多前缘羽毛在不同迎角、羽毛预设角和羽毛长度范围内的气动性能进行了研究。结果表明,多羽毛结构重组了前缘涡动力学,产生了比单膜机翼更高的升力和升阻比。这一点在失速前和过渡阶段尤为明显,在这些阶段,设计还能保持适度的阻力损失,减少力波动。一个较浅的正预设角与中间羽毛长度相结合,建立了一个柔软的弦向槽,维持一个温和的前缘对流涡流,加速压力恢复,提高气动效率。当羽毛变得太长或太陡时,堵塞效应会重新增加阻力并降低性能。提出了一种结合有界羽毛覆盖度和预设角度修正因子的经验标度关系式,用于预测韦伯数骨架下膜的最大变形。这些发现为设计具有可调升力、效率和稳定性的大迎角羽毛变形膜翼提供了定量指导。


LamiNetDef: A novel physics-informed U-Net++ framework for predicting deflection shapes of laminated composite plates

Quang-Viet Vu, Dai-Nhan Le, Nguyen Pham Dinh, George Papazafeiropoulos, Sawekchai Tangaramvong

doi:10.1016/j.tws.2026.115041

LamiNetDef:一种新的基于物理的U-Net++框架,用于预测层压复合材料板的挠度形状

This study introduces LamiNetDef, a novel physics-informed deep learning framework developed to accurately predict the mid-surface deflection patterns of laminated composite plates subjected to uniformly distributed loads. A comprehensive dataset is generated through an automated finite element modeling pipeline, integrating Abaqus and MATLAB via the Abaqus2Matlab toolbox. The model parameters encompass geometric dimensions, stacking sequence, ply orientations, and boundary conditions, sampled using a Latin Hypercube Sampling scheme to efficiently explore the design variable space. The finite element nodal displacements are subsequently normalized and processed to form structured data for training. The proposed LamiNetDef architecture, built upon a modified U-Net++ network, incorporates a Boundary Condition Deflection Form that encodes physical boundary effects as a two-dimensional spatial input to better inform the encoder in learning physically consistent deformation modes. In parallel, design parameters including geometric and material descriptors are processed by Kolmogorov–Arnold Networks to derive adaptive conditioning parameters for the Feature-wise Linear Modulation layers. This mechanis m enables dynamic coupling between physics and features, enriching the model’s representational capacity. The network is trained under a physics-informed loss function that integrates both magnitude and gradient terms, ensuring s moothness and physical consistency in the predicted deformation fields. The LamiNetDef framework achieves outstanding predictive performance, with R 2 = 0.951 and MSSIM = 0.987 on independent test data, demonstrating excellent agreement with finite element results. Furthermore, the framework is deployed as a cloud-based interactive platform via Hugging Face, enabling rapid and reliable deflection prediction directly from user-defined design parameters.

本研究引入了LamiNetDef,这是一种新颖的物理信息深度学习框架,用于准确预测受均匀分布载荷作用的层压复合材料板的中表面挠度模式。通过自动化有限元建模管道生成综合数据集,通过abaquus2matlab工具箱集成Abaqus和MATLAB。模型参数包括几何尺寸、堆叠顺序、层向和边界条件,使用拉丁超立方采样方案进行采样,以有效地探索设计变量空间。随后将有限元节点位移归一化处理,形成结构化数据进行训练。LamiNetDef架构基于改进的U-Net++网络,包含一个边界条件偏转形式,该形式将物理边界效应编码为二维空间输入,以便更好地告知编码器学习物理一致的变形模式。同时,包括几何和材料描述符在内的设计参数由Kolmogorov-Arnold网络处理,以获得面向特征的线性调制层的自适应调节参数。这种机制使物理和特征之间的动态耦合成为可能,从而丰富了模型的表示能力。该网络在一个物理信息损失函数下进行训练,该函数集成了幅度和梯度项,确保了预测变形场的平滑性和物理一致性。LamiNetDef框架具有出色的预测性能,独立试验数据的R 2 = 0.951, MSSIM = 0.987,与有限元结果吻合良好。此外,该框架通过hug Face作为基于云的交互平台部署,可以直接从用户定义的设计参数中快速可靠地预测挠度。


Compatibility an alysis between two-phase local/nonlocal and classical theories for nanobeams in s mooth and non-s mooth fields

Z.W. Song, S.K. Lai, Y.T. Zhang, C.W. Lim

doi:10.1016/j.tws.2026.115031

纳米梁在光滑场和非光滑场中两相局部/非局部与经典理论的相容性分析

In general, it is assumed that both Eringen’s two-phase local/nonlocal theory and the stress-driven two-phase local/nonlocal theory reduce to the classical local theory as the nonlocal length-scale parameter l 0 approaches zero, despite the lack of a rigorous theoretical proof for this fundamental issue. In addition, the relationship between the local phase volume fraction β and l 0 remains unclear. This work demonstrates that the constitutive boundary and interface conditions present in two-phase local/nonlocal theories do not automatically vanish or become satisfied in the classical local theory as l 0 tends to zero. Achieving full compatibility between nonlocal and local theories requires that these conditions are also satisfied within the classical framework. Furthermore, it is concluded that β is associated with l 0 , and the stress-driven nonlocal theory is not fully compatible with the classical local theory. These findings are validated through illustrative examples to highlight the compatibility conditions between nonlocal and local theories and to elucidate the relationship between β and l 0 .

一般认为,当非局部长度尺度参数l0趋近于零时,Eringen的两相局部/非局部理论和应力驱动的两相局部/非局部理论均可归结为经典局部理论,尽管这一基本问题缺乏严格的理论证明。此外,局部相体积分数β与l0之间的关系尚不清楚。本研究表明,两相局域/非局域理论中存在的本构边界和界面条件在经典局域理论中不会随着l0趋于零而自动消失或满足。要实现非局部理论和局部理论之间的完全兼容,需要在经典框架内也满足这些条件。进一步得出β与l0相关的结论,应力驱动的非局部理论与经典局部理论不完全相容。通过举例验证了这些发现,突出了非局部理论和局部理论之间的相容条件,并阐明了β和l0之间的关系。




来源:复合材料力学仿真Composites FEM
ACTMechanicalInspireAbaqusDeform复合材料化学理论材料仿生试验人工智能无人机管道
著作权归作者所有,欢迎分享,未经许可,不得转载
首次发布时间:2026-05-13
最近编辑:5月前
Tansu
签名征集中
获赞 22粉丝 7文章 1031课程 0
点赞
收藏
作者推荐

【新文速递】2026年5月9日复合材料SCI期刊最新文章

今日更新:Composite Structures 1 篇,Composites Part B: Engineering 4 篇,Composites Science and Technology 1 篇Composite StructuresInterfacial strengthening and dynamic mechanical properties of CF/PEEK composites with a water-based PAAs/Montmorillonite nanocomposite sizing agentYang Liu, Kaibao Wang, Zhilong Chang, Huirong Ledoi:10.1016/j.compstruct.2026.120434水基PAAs/蒙脱土纳米复合施胶剂对CF/PEEK复合材料界面强化及动态力学性能的影响This article conducted an investigation into the interfacial strengthening of carbon fiber-reinforced PEEK (CF/PEEK) through the modification of CF with a water-based PAAs/Montmorillonite (MMT) nanocomposite sizing agent. The surface roughness of CF exhibited a significant growth with the augmentation of MMT content. Compared with desized CF (e.g., DCF), the surface energy of PAAs/MMT-sized CF (e.g., PMCF1) increased by 503.95 %. The flexural strength, modulus, and interlaminar shear strength (ILSS) of PMCF1/PEEK increased by 61.24 %, 19.87 %, and 12.95 %, respectively. The interfacial strengthening can be explained by two factors: 1) increasing roughness and surface energy of CF due to MMT, and 2) enhanced interaction and molecular chains entanglement due to PI. Thinner CF fabrics contributed to a further enhancement of the mechanical properties of the composites. The flexural strength, modulus, and ILSS of TPMCF0.3/PEEK reached 800.82 MPa, 95.03 GPa, and 62.98 MPa, respectively. This is due to the higher CF volume fraction and resin impregnation of the thin CF fabric system. The flexural strength achieved in this study have an outstanding performance among related studies. The interface modification method described in this paper holds promise to enhance the strength of the CF/PEEK.采用水基PAAs/蒙脱土(MMT)纳米复合施胶剂对CF进行改性,研究了碳纤维增强PEEK (CF/PEEK)的界面强化。随着MMT含量的增加,CF的表面粗糙度显著增大。与浆料CF(如DCF)相比,PAAs/ mmt浆料CF(如PMCF1)的表面能提高了503.95%。PMCF1/PEEK的抗弯强度、模量和层间剪切强度(ILSS)分别提高了61.24%、19.87%和12.95%。界面强化可以用两个因素来解释:1)MMT增加了CF的粗糙度和表面能;2)PI增强了相互作用和分子链纠缠。更薄的CF织物有助于进一步提高复合材料的机械性能。TPMCF0.3/PEEK的抗弯强度、模量和ILSS分别达到800.82 MPa、95.03 GPa和62.98 MPa。这是由于较高的CF体积分数和薄CF织物体系的树脂浸渍。本研究获得的抗弯强度在相关研究中表现突出。本文所描述的界面改性方法有望提高CF/PEEK的强度。Composites Part B: EngineeringElectronic-Structure-Guided Ta/O Co-Doping Engineering Bulk and Interphase in Argyrodite Electrolytes for Solid-State Li Metal BatteriesJie Yang, Siwu Li, Ziling Jiang, Kunpeng Tang, Lei Shi, Long Chen, Xiaobo Chen, Chen Liu, Chuang Yudoi:10.1016/j.composites b.2026.113779固态锂金属电池用银汞石电解质的电子结构引导Ta/O共掺杂工程体与界面Argyrodite-based sulfide electrolytes are promising solid electrolytes for all-solid-state lithium metal batteries owing to their high ionic conductivity and favorable mechanical properties. However, poor air stability and severe interfacial reactivity with Li metal, which trigger continuous side reactions and dendrite growth, remain major obstacles to their practical application. Herein, we design a series of Ta/O co-doped chlorine-rich argyrodite electrolytes with the nominal composition Abstract to simultaneously regulate the PS4 3− tetrahedral framework and the Li|solid-state electrolyte (SSE) interphase. Rietveld refinement and density functional theory calculations reveal that Ta preferentially occupies P sites while O substitutes S in PS4 3− tetrahedra, leading to lattice expansion and a modified Li+ migration landscape. Although the room-temperature ionic conductivity decreases moderately with increasing Ta/O content, the optimized composition Li5.5P0.95Ta0.05S4.375O0.125Cl1.5 exhibits a substantially enhanced critical current density of 3.0 mA cm−2 (vs. 0.8 mA cm−2 for pristine L5.5PS4.5Cl1.5) and enables stable Li plating/stripping for over 500 h in Li|SSE|Li symmetric cells. Ta/O co-doping also improves the environmental robustness of the electrolyte by increasing its resistance to water adsorption and hydrolysis, thereby enhancing air stability. Interfacial an alysis via X-ray photoelectron spectroscopy depth profiling and morphology characterization demonstrates the formation of a Li–Ta–Sx rich interphase and Li–Ta alloy phases together with inorganic decomposition products, which homogenize Li deposition and effectively suppress dendrite penetration. When paired with a high-Ni ZrO2@NCM955 cathode, the optimized Ta/O co-doped electrolyte delivers a high discharge capacity of 118.4 mAh g−1 after 100 cycles at 1.0 C with 86.9% capacity retention, underscoring the potential of Ta/O engineered chlorine-rich argyrodites for high-energy, long-life sulfide-based all-solid-state lithium metal batteries.银硫化物电解质具有较高的离子电导率和良好的力学性能,是全固态锂金属电池中很有前途的固体电解质。然而,较差的空气稳定性和与锂金属的强烈界面反应性引发了持续的副反应和枝晶生长,仍然是其实际应用的主要障碍。在此,我们设计了一系列的Ta/O共掺杂富氯银镁石电解质,其标称成分为Abstract,以同时调节PS4 3 -四面体框架和Li|固态电解质(SSE)界面相。Rietveld细化和密度泛函数理论计算表明,在PS4 3 -四面体中,Ta优先占据P位,而O取代S,导致晶格扩展和Li+迁移景观的改变。尽管室温离子电导率随着Ta/O含量的增加而适度降低,但优化后的组合物li5.5 p0.95 ta0.05 s4.3750 o0.125 cl1.5的临界电流密度显著提高至3.0 mA cm - 2(原始L5.5PS4.5Cl1.5为0.8 mA cm - 2),并能在Li|SSE|Li对称电池中稳定镀/剥离超过500小时。Ta/O共掺杂还通过增加其对水的吸附和水解的抵抗力,从而提高了电解质的环境稳健性,从而提高了空气稳定性。通过x射线光电子能谱深度剖面和形貌表征对界面进行分析,发现富Li - ta - sx界面相和Li - ta合金相以及无机分解产物的形成均质化了Li沉积,有效抑制了枝晶的渗透。当与高镍ZrO2@NCM955阴极配对时,优化的Ta/O共掺杂电解质在1.0 C下循环100次后可提供118.4 mAh g- 1的高放电容量,容量保留率为86.9%,这突显了Ta/O工程设计的富氯银离子电池在高能、长寿命硫化物基全固态锂金属电池中的潜力。MIL-125(Ti)-derived core-shell TiO2@C composites for tunable and efficient microwave absorptionLiang Zhou, MUHAIMINUL ISLAM SHAKIL, Jingyi He, Xiyue Zhang, Hongbo Wangdoi:10.1016/j.composites b.2026.113764MIL-125(Ti)衍生核壳TiO2@C复合材料可调谐和有效的微波吸收Metal-organic framework (MOF)-derived carbon materials exhibit significant advantages for microwave absorption applications, owing to their high electrical conductivity, structural tunability, and abundant intrinsic defects and heterogeneous interfaces. However, achieving an optimal balance between magnetic and dielectric losses remains challenging for most magnetic metal-based MOFs-derived absorbers. In this study, MIL-125(Ti) was prepared as a template, and TiO2@C-x composites were synthesized via a two-step pyrolysis method. Experimental results demonstrate that the pyrolysis temperature significantly affects the graphitization degree of the carbon matrix and the formation of TiO2 nanoparticles, thereby playing a dominant role in the dielectric and microwave absorption properties of the investigated MIL-125(Ti)-derived core-shell TiO2@C composites. Notably, the TiO2@C-700 composite delivers an effective absorption bandwidth of 3.2 GHz in 9.2-12.4 GHz at a thickness of 2.01 mm. Therefore, the MIL-125(Ti)-derived TiO2@C composites fabricated in this study establish a viable and nonmagnetic strategy for designing tunable, lightweight, and high-performance MOF-derived microwave absorbers.金属有机骨架(MOF)衍生的碳材料由于其高导电性、结构可调性、丰富的本征缺陷和非均相界面,在微波吸收应用中具有显著的优势。然而,实现磁性和介电损耗之间的最佳平衡对于大多数磁性金属基mofs衍生吸收剂来说仍然是一个挑战。本研究以MIL-125(Ti)为模板,采用两步热解法合成TiO2@C-x复合材料。实验结果表明,热解温度显著影响碳基体的石墨化程度和TiO2纳米颗粒的形成,从而对MIL-125(Ti)衍生核壳TiO2@C复合材料的介电和微波吸收性能起主导作用。值得注意的是,TiO2@C-700复合材料在厚度为2.01 mm时,在9.2-12.4 GHz范围内提供了3.2 GHz的有效吸收带宽。因此,本研究中制备的MIL-125(Ti)衍生TiO2@C复合材料为设计可调谐、轻量化和高性能的mof衍生微波吸收器建立了一种可行的非磁性策略。Enhancing CFRP hole-making quality via ultrasonic vibration-assisted bi-directional drillingChaoqun Wu, Wenjian Huang, Shiyu Cao, Yufei Tangdoi:10.1016/j.composites b.2026.113745超声振动辅助双向钻进提高CFRP成孔质量To address the critical issues of delamination, burr formation, and poor hole wall quality that frequently occur during Carbon Fiber Reinforced Plastics (CFRP) drilling and severely compromise assembly accuracy in aviation, aerospace, marine, and automotive industries, this study proposes a novel ultrasonic vibration-assisted bi-directional drilling (UVABD) method that combines the damage suppression advantages of bi-directional drilling and ultrasonic vibration-assisted drilling for high-quality and high-efficiency CFRP hole-making. Owing to the interlaminar bonding characteristics and anisotropy of CFRP that result in hole exit damage and poor hole wall quality, the positive effect of bi-directional drilling on exit damage suppression is first clarified based on a reverse feeding an alysis, a comparative a nalysis of the critical thrust force and drilling process between conventional drilling and bi-directional drilling is performed considering the internal support effect of bi-directional drilling, and ultrasonic vibration is introduced to further improve the hole wall quality. Experimental results from different CFRP drilling methods indicate that compared to conventional drilling, bi-directional drilling and UVABD can reduce delamination by 62.9% and 72.7%, respectively, while reducing burr formation by 98.6% and 97.9%, respectively. Additionally, the introduction of vertical ultrasonic vibration can reduce hole wall roughness by up to 9.1%. The proposed UVABD method achieves synergistic performance improvement, effectively suppressing drilling exit defects and enhancing hole wall quality simultaneously, which can provide an important reference for the precision and efficient hole-making of CFRP.为了解决碳纤维增强塑料(CFRP)钻孔过程中经常出现的分层、毛刺形成和孔壁质量差的关键问题,并严重影响航空、航天、船舶和汽车行业的装配精度,本研究提出了一种新型超声振动辅助双向钻进(UVABD)方法,该方法结合了双向钻进和超声振动辅助钻进的损伤抑制优势,实现了高质量、高效率的CFRP制孔。由于CFRP的层间粘结特性和各向异性导致了孔出口损伤和孔壁质量差,首先通过反向进给分析阐明了双向钻进对出口损伤抑制的积极作用,考虑双向钻进的内部支撑效应,对常规钻进和双向钻进的临界推力和钻进过程进行了对比分析;并引入超声振动,进一步改善孔壁质量。不同CFRP钻孔方式的实验结果表明,与常规钻孔相比,双向钻孔和UVABD可分别减少62.9%和72.7%的分层,减少毛刺的形成分别为98.6%和97.9%。此外,垂直超声振动的引入可使孔壁粗糙度降低9.1%。所提出的UVABD方法实现了性能的协同提升,在有效抑制钻出口缺陷的同时提高了孔壁质量,可为CFRP的精密高效制孔提供重要参考。Multiscale Interfacial Engineering for the Scalable Fabrication of High-Performance Ag2Se-Coated Carbon Fiber Thermoelectric Yarns for Human Body Heat HarvestingXinxing Zhou, Bo Wu, Kerui Li, Chengyi Hou, Qinghong Zhang, Yaogang Li, Hongzhi Wangdoi:10.1016/j.composites b.2026.113728 人体热采集用高性能镀银碳纤维热电纱的多尺度界面工程Fiber-shaped thermoelectric devices hold great promise for practical applications owing to their flexibility, high integration density, and weavability; however, their deployment is still limited by complex fabrication processes, insufficient material performance, and performance degradation induced by friction and fatigue during service. In this work, we use carbon fibers as the core substrate and implement multiscale interfacial engineering to enable high-performance Ag2Se loading. At the microscopic level, surface roughening and chemical activation of the carbon fibers markedly enhance the loading efficiency of Ag2Se on the fiber surface. At the structural level, mechanically twisting numerous fine carbon filaments into robust interlocked yarns effectively suppresses interfacial wear and structural damage during operation. Based on this strategy, we realize high-performance Ag2Se-coated carbon fiber thermoelectric yarns that exhibit a stable and high room-temperature power factor (> 330 μW m-1 K-2) and retain reliable electrical performance after more than 1000 friction cycles. When woven into TE fabrics, the resulting textiles achieve a power density of ∼0.81 μW cm-2 under human body heat, comparable to state-of-the-art fiber-based TE devices, while simultaneously offering simple processing and excellent scalability.纤维形状的热电器件由于其灵活性、高集成密度和可织性而具有很大的实际应用前景;然而,它们的部署仍然受到复杂的制造工艺、材料性能不足以及在使用过程中摩擦和疲劳引起的性能下降的限制。在这项工作中,我们使用碳纤维作为核心衬底,并实施多尺度界面工程,以实现高性能Ag2Se加载。微观层面上,碳纤维表面粗化和化学活化显著提高了Ag2Se在纤维表面的负载效率。在结构层面,将大量的细碳丝机械捻制成坚固的互锁丝,有效地抑制了运行过程中的界面磨损和结构损伤。基于该策略,我们实现了高性能的ag2se涂层碳纤维热电纱线,具有稳定的高室温功率因数(> 330 μW m-1 K-2),并且在超过1000次摩擦循环后仍保持可靠的电气性能。当编织到TE织物中时,所得纺织品在人体热量下的功率密度达到~ 0.81 μW cm-2,与最先进的基于纤维的TE设备相当,同时提供简单的加工和出色的可扩展性。Composites Science and TechnologyPrecision orientation of hydrogen evolution performance of iron-based electrocatalysts empowered by machine learning: From electronic structure regulation to rational constructionYan Hu, Senlin Tian, Yingjie Lidoi:10.1016/j.compscitech.2026.111682 基于机器学习的铁基电催化剂析氢性能的精确定向:从电子结构调控到合理构造The efficient realization of hydrogen economy urgently requires the development of low-cost and high-performance electrocatalytic hydrogen evolution (HER) catalysts. Although iron-based materials are abundant in resources, low-cost, and have adjustable electronic structures, their performance optimization faces challenges such as multivariable nonlinear coupling, difficulty in maximizing active sites and coordinating stability. Traditional trial and error methods and single scale theoretical simulations have limitations in efficiency and accuracy. This review systematically elaborates on how machine learning (ML), as a transformative research paradigm, empowers iron-based HER catalysts from precise electronic structure control to rational macroscopic performance construction through deep mining of hidden patterns in high-throughput computing and experimental data. ML can not only establish fast quantitative prediction models from composition, structure to key electronic descriptors (such as d-band center, ΔG*H), but also reverse engineer new material structures with ideal performance through generative models and optimization algorithms, and intelligently plan their synthesis paths. A detailed overview of the methodological progress and application cases of ML in electronic structure quantification, defect and interface effect an alysis, single atom catalyst design, and stability evaluation was provided, revealing the key role of ML in discovering novel physical descriptors and elucidating complex synergistic mechani s ms. Finally, a closed-loop intelligent development system that integrates physical information modeling, automated experiments, and digital twin technology is proposed, providing a comprehensive theoretical framework and technical roadmap for the precise directional design of the new generation of high-performance and high stability iron-based electrocatalysts.高效实现氢经济迫切需要开发低成本、高性能的电催化析氢催化剂。尽管铁基材料具有资源丰富、成本低、电子结构可调等优点,但其性能优化仍面临多变量非线性耦合、活性位点难以最大化和协调稳定性等挑战。传统的试错法和单尺度理论模拟在效率和准确性上存在局限性。本文系统阐述了机器学习(ML)作为一种变革性的研究范式,如何通过深入挖掘高通量计算和实验数据中的隐藏模式,使铁基HER催化剂从精确的电子结构控制到合理的宏观性能构建。ML不仅可以建立从成分、结构到关键电子描述符(如d波段中心、ΔG*H)的快速定量预测模型,还可以通过生成模型和优化算法对性能理想的新材料结构进行逆向工程,并智能规划其合成路径。详细概述了机器学习在电子结构量化、缺陷和界面效应分析、单原子催化剂设计和稳定性评价等方面的方法进展和应用案例,揭示了机器学习在发现新的物理描述符和阐明复杂协同机制方面的关键作用。最后,提出了集物理信息建模、自动化实验和数字孪生技术于一体的闭环智能开发系统,为新一代高性能高稳定性铁基电催化剂的精准定向设计提供了全面的理论框架和技术路线图。来源:复合材料力学仿真Composites FEM

未登录
还没有评论
课程
培训
服务
行家
VIP会员 学习计划 福利任务
下载APP
联系我们
帮助与反馈