【新文速递】2026年5月6日复合材料SCI期刊最新文章

今日更新：Composite Structures 10 篇，Composites Part A: Applied Science and Manufacturing 4 篇，Composites Part B: Engineering 8 篇，Composites Science and Technology 2 篇

Composite Structures

Prediction of debonding behavior of externally bonded FRP-to-concrete joints with bond defects: An alytical approach and experimental validation

Kaiming Pan, Weifeng Fan, Yanjie Wang, Hongbo Liu, Xinxin Li, Yabin Deng, Zhimin Wu

doi:10.1016/j.compstruct.2026.120417

带粘结缺陷的frp -混凝土外粘接节点的脱粘行为预测：分析方法和实验验证

Externally bonded (EB) fiber-reinforced polymer (FRP) composites are widely used to strengthen reinforced concrete (RC) structures. FRP-to-concrete interfaces often contains randomly distributed bond defects resulting from imperfect installation or service-related deterioration. However, most existing an alytical models fail to consider the effect of such defects. This paper presents a closed-form an alytical model for predicting the debonding performance of defective FRP-to-concrete joints with a single bond defect of arbitrary size and location. The model is formulated using a bilinear cohesive material law (CML) and is validated against experiments from the literature. The predictions agree well with the experimental load-slip responses, with Integral Absolute Error (IAE) values below 15% for the first peak load, the minimum load, the second peak load, and the maximum load. A parametric study quantitatively evaluates the effects of defect size and location on the load-slip response and critical load values. It is found that increasing the defect ratio from 10% to 30% causes over a 50% drop in the minimum load and more than a 15% loss in maximum load. Defects near the loaded end cause unstable responses with significant load drops, whereas those near the free end result in more stable debonding propagation and enhanced load-bearing performance.

外粘接（EB）纤维增强聚合物（FRP）复合材料被广泛用于加固钢筋混凝土（RC）结构。frp -混凝土界面通常包含随机分布的粘结缺陷，这是由于安装不完善或与使用有关的劣化造成的。然而，大多数现有的分析模型都没有考虑到这些缺陷的影响。本文提出了一种预测具有任意大小和位置的单一粘结缺陷的frp -混凝土缺陷节点脱粘性能的封闭分析模型。该模型是使用双线性内聚物质定律（CML）制定的，并根据文献中的实验进行了验证。预测结果与试验荷载-滑移响应吻合较好，第一次峰值荷载、最小荷载、第二次峰值荷载和最大荷载的积分绝对误差（IAE）均小于15%。一项参数化研究定量评估了缺陷尺寸和位置对载荷-滑移响应和临界载荷值的影响。研究发现，将缺陷率从10%增加到30%，会导致最小载荷下降50%以上，最大载荷损失15%以上。靠近加载端的缺陷会导致不稳定的响应，载荷下降明显，而靠近自由端的缺陷会导致更稳定的脱粘传播和增强的承载性能。

Physics-guided prediction of dimensionless parameters for circular sandwich panels under blast loading

Zihao Song, Puxin Ge, Zhiyang Li, Hairen Wang, Jianyin Lei, Zhifang Liu, Zhiyong Wang, Zhihua Wang

doi:10.1016/j.compstruct.2026.120420

爆炸荷载作用下圆形夹层板无量纲参数的物理导向预测

Lightweight sandwich panels are widely used for blast protection. Predicting their behavior is difficult because it requires capturing interactions between impulse transfer, material properties, and panel geometry. This study presents a physics-guided machine learning approach to find key dimensionless parameters that control the blast response of sandwich panels. The approach trains neural networks in a low-dimensional space defined by the Buckingham π theorem to create dynamic response surfaces. Then, the Active Subspace Method (AS M) is used to an alyze gradients of the response surfaces and identify dominant low-dimensional subspaces. Shapley Additive Explanations (SHAP) show how each parameter contributes. Applications demonstrate that this method works for problems of different complexity. For simple homogeneous-core panels, the one-dimensional dominant subspace matches classical theory, confirming that the method can reveal physical principles. For complex gradient-core panels, it shows that the high-dimensional parameter system can also be reduced. The identified dimensionless parameters help build simple semi-empirical equations that work well in practice. Overall, this modeling approach balances clarity and accuracy and offers a reliable way to quickly design complex protective structures using data.

轻质夹层板广泛用于防爆防护。预测它们的行为是很困难的，因为它需要捕捉脉冲传递、材料特性和面板几何形状之间的相互作用。本研究提出了一种物理指导的机器学习方法来寻找控制夹层板爆炸响应的关键无量纲参数。该方法在由Buckingham π定理定义的低维空间中训练神经网络以创建动态响应面。然后，利用主动子空间法（AS M）分析响应面梯度，识别低维优势子空间；Shapley加性解释（SHAP）显示了每个参数是如何起作用的。应用表明，该方法适用于不同复杂程度的问题。对于简单的均匀芯板，一维优势子空间符合经典理论，证实了该方法可以揭示物理原理。对于复杂的梯度芯板，也可以简化高维参数系统。确定的无量纲参数有助于建立简单的半经验方程，在实践中很好地工作。总的来说，这种建模方法平衡了清晰度和准确性，并提供了一种可靠的方法来使用数据快速设计复杂的保护结构。

A reduced-order Maxwell viscoelastic constitutive model for efficient numerical simulation of thermoset composite curing

Yang Liu, Honghua Zhao, Kai Jian, Yaqi Yang, Hongyang Cui, Ruixiang Bai

doi:10.1016/j.compstruct.2026.120419

热固性复合材料固化有效数值模拟的降阶Maxwell粘弹性本构模型

Prediction of curing-induced deformation depends on the constitutive model. A nine-branch generalised Maxwell viscoelastic model is accurate but computationally expensive, whereas the CHILE model is more efficient but less accurate. In this work, AS4/3501–6 composite served as the reference material and a Prony-spectrum order-reduction method was established for finite-element implementation. The nine-branch spectrum was treated as a “teacher”; representative branches were obtained by k-medoids (PAM) clustering along the logarithmic relaxation-time axis, and a reduced “student” spectrum was constructed within a unified shift-factor framework α T , X . Relative error Rel L 2 was adopted as the training metric, wall-clock time defined the computational cost, and an error-efficiency Pareto front was used to select the optimal number of branches. A three-dimensional finite-element model of a U-shaped component was used to compare curing residual stresses and deformations predicted by the nine- and four-branch spectra under different histories and lay-ups. The four-branch spectrum lay near the Pareto knee ( Rel L 2 ≈4.6%), reduced wall-clock time from 60 to 43 min (∼28%) and decreased the Newton iterations per increment from 1.69 to 1.48 without cutback, while maintaining agreement in residual stress fields and curing spring-in. Errors were more sensitive to dwell time than to heating rate.

固化变形的预测依赖于本构模型。九分支广义麦克斯韦粘弹性模型是准确的，但计算成本高，而智利模型效率更高，但精度较低。本文以AS4/3501-6复合材料为基准材料，建立了prony谱降阶法进行有限元实现。九支谱被当作“老师”；通过k- mediids （PAM）沿对数松弛-时间轴聚类获得代表性分支，并在统一的移位因子框架α T， X内构建简化的“学生”谱。采用相对误差rell2作为训练度量，壁钟时间定义计算代价，采用误差效率Pareto前选择最优分支数。采用u形构件三维有限元模型，比较了不同历史和铺层下九支和四支谱预测的固化残余应力和变形。四分支谱位于帕累托膝附近（Rel l2≈4.6%），将壁钟时间从60分钟减少到43分钟（~ 28%），并将牛顿迭代次数从1.69次减少到1.48次而不减少，同时在残余应力场和处理弹簧时保持一致。误差对停留时间比对升温速率更敏感。

Dynamic constitutive modeling and compressive mechanical behavior of needle-punched C/C composites

Fangchao Jia, Chenwei Shan, Wengang Liu, Ziwen Xia, Yifeng Xiong, Ming Luo

doi:10.1016/j.compstruct.2026.120418

针 刺C/C复合材料动态本构建模及压缩力学行为

Needle-punched C/C composites exhibit pronounced strain rate sensitivity during service. However, due to their highly complex internal structure, significant anisotropy is observed. The dynamic mechanical behavior of this material currently lacks adequate explanation and an alysis. This study investigates the dynamic mechanical response under various conditions by conducting compression impact tests on needle-punched C/C composites at different strain rates and along different orientations. The results indicate that under high kinetic energy, crack nucleation and propagation occur within the material. Compression along different orientations alters crack propagation paths, leading to distinct failure modes: along the needling direction, failure primarily manifests as fiber shear fracture and interfacial debonding; perpendicular to it, failure is dominated by interfacial delamination and matrix cracking. Simultaneously, a certain level of kinetic energy exhibits an inhibitory effect on interfacial debonding, enhancing the load-bearing capacity to some extent. Finally, a dynamic constitutive model incorporating the strain rate effects on elastic modulus and peak strength is established. Validation through numerical simulations confirms that the model effectively describes the dynamic mechanical behavior of needle-punched C/C composites.

针 刺C/C复合材料在使用过程中表现出明显的应变率敏感性。然而，由于其内部结构高度复杂，观察到显著的各向异性。目前对这种材料的动态力学行为缺乏充分的解释和分析。通过不同应变速率、不同取向的针 刺C/C复合材料压缩冲击试验，研究了不同条件下的动态力学响应。结果表明：在高动能作用下，材料内部出现裂纹形核和扩展；沿不同方向的压缩改变了裂纹扩展路径，导致不同的破坏模式：沿针 刺方向，破坏主要表现为纤维剪切断裂和界面脱粘；在垂直方向上，破坏以界面分层和基体开裂为主。同时，一定的动能对界面脱粘有抑制作用，在一定程度上提高了材料的承载能力。最后，建立了考虑应变速率对弹性模量和峰值强度影响的动态本构模型。通过数值模拟验证，该模型有效地描述了针 刺C/C复合材料的动态力学行为。

Reliability-based design optimization of composite wind turbine blade based on global sensitivity a nalysis and a two-level adaptive Kriging

Jia Wei, Yuan-Zhuo Ma, Wei-Dong Liu, Bo-Feng Xu, Zhen-Zhou Zhao, Hong-Shuang Li

doi:10.1016/j.compstruct.2026.120416

基于全局灵敏度分析和两级自适应Kriging的复合材料风力机叶片可靠性优化设计

The optimization design of large composite wind turbine blades is of great significance to the performance, reliability and cost of wind power systems. Given the complex loads and multi-source uncertainties experienced by the blades during operation, reliability-based design optimization (RBDO) has become a key approach to improving structural reliability. This paper proposes an efficient two-level coupled adaptive Kriging-based RBDO method for the composite wind turbine blades. First, a parametric finite element modeling and an alysis of the composite wind turbine blade is coded using the ANSYS Parametric Design Language (APDL), facilitating a corresponding fully parametric RBDO framework. Secondly, the reliability-oriented global sensitivity a nalysis is performed on the initial design variables to identify key design variables and reduce dimensionality to address the high-dimensional complexity caused by random parameters. Finally, a two-level adaptive Kriging based RBDO method is proposed, where the inner loop Kriging asymptotically depict the limit state functions with the help of importance boundary sampling while the outer-loop one directly approximates the objective function and probabilistic constraints base on the complete constrained expected improvement criterion. Taking a 10 MW wind turbine blade as an example, the proposed method is validated for its performance.

大型复合材料风力机叶片的优化设计对风力发电系统的性能、可靠性和成本具有重要意义。考虑到叶片在运行过程中承受的复杂载荷和多源不确定性，基于可靠性的设计优化（RBDO）已成为提高结构可靠性的关键方法。针对复合材料风力机叶片，提出了一种高效的两级耦合自适应kriging RBDO方法。首先，利用ANSYS参数化设计语言（APDL）对复合材料风力发电机叶片进行参数化有限元建模和分析，实现相应的全参数化RBDO框架。其次，对初始设计变量进行面向可靠性的全局灵敏度分析，识别关键设计变量并降维，解决随机参数带来的高维复杂性问题；最后，提出了一种基于两级自适应Kriging的RBDO方法，其中内环Kriging借助于重要边界采样渐近描述极限状态函数，而外环Kriging则基于完全约束期望改进准则直接逼近目标函数和概率约束。以10 MW风力机叶片为例，验证了该方法的性能。

Critical buckling an alysis and multi-objective optimal design of hygrothermally stable laminates with extension-twist and bend-twist couplings

Da Cui, Muyao Miao, Yuanzhi Xiong, Daokui Li

doi:10.1016/j.compstruct.2026.120413

具有伸扭和弯扭耦合的湿热稳定层压板的临界屈曲分析及多目标优化设计

Compared with A S B t D S laminates (laminates with extension-twist coupling), asymmetric composites with both extension-twist and bend-twist couplings show significant potential for improving mechanical properties such as coupled characteristics and buckling performance. However, existing studies indicate that introducing bend-twist coupling tends to degrade the buckling performance of laminates. To address this, this study presents a buckling an alysis modelling approach for asymmetric A S B t D F laminates (laminates with both bend-twist and extension-twist couplings), enabling the precise semi‑an alytical determination of their critical buckling loads using the high‑mode Rayleigh‑Ritz method. A multi-objective optimization algorithm, GA-SQP-WC (Genetic Algorithm-Sequential Quadratic Program-Weight Coefficient), is proposed. Under hygrothermal stability constraints derived an alytically, the algorithm achieves multi‑objective design optimization for asymmetric laminates by incorporating both buckling performance and coupling. Results show that semi-a nalytical critical buckling loads closely match numerical simulations. Compared with A S B t D S laminates, A S B t D F laminates achieve significant improvements in both critical buckling load and extension-twist coupling: for 7-ply laminates, the maximum increases reach 71.0% and 17.4%, respectively. Robustness an alysis validates the effectiveness of the layup designs, while numerical simulations and buckling tests confirm the hygrothermal stability, coupling effects, and buckling load accuracy. The proposed method can be extended to the buckling a nalysis of asymmetric laminates with more types of couplings.

与A - S - B -t - D - S复合材料（具有拉伸-扭转耦合的复合材料）相比，具有拉伸-扭转和弯曲-扭转耦合的非对称复合材料具有显著的力学性能，如耦合特性和屈曲性能。然而，已有研究表明，引入弯扭耦合往往会降低层合板的屈曲性能。为了解决这一问题，本研究提出了一种非对称a - S - B -t - D - F层压板（具有弯曲-扭转和拉伸-扭转耦合的层压板）的屈曲分析建模方法，能够使用高模态瑞利-里兹方法精确地确定其临界屈曲载荷。提出了GA-SQP-WC（遗传算法-序列二次规划-权系数）多目标优化算法。该算法在解析导出的湿热稳定性约束条件下，兼顾屈曲性能和耦合性能，实现了非对称层合板的多目标优化设计。结果表明，半解析临界屈曲载荷与数值模拟结果吻合较好。与A S B t D S复合材料相比，A S B t D F复合材料在临界屈曲载荷和伸扭耦合方面均有显著改善，7层复合材料的临界屈曲载荷和伸扭耦合性能最大分别提高了71.0%和17.4%。鲁棒性分析验证了铺层设计的有效性，而数值模拟和屈曲试验证实了湿热稳定性、耦合效应和屈曲载荷的准确性。所提出的方法可以推广到具有更多类型耦合的非对称层合板的屈曲分析。

Saccharum-spontaneum stem based lightweight load bearing composite structures

Deepak Kumar Maurya, Chandra Shekhar Upadhyay

doi:10.1016/j.compstruct.2026.120410

基于糖-自发茎的轻质承重复合结构

Saccharum spontaneum (cane grass) is an exciting structural material, because the full stem can be used effectively. The outer epidermis region consists of cellulose-rich long fibers - strips of which can be directly used as building-blocks for load-bearing facesheets and laminae of unidirectional composite laminates. Concurrently, the cellular parenchyma core is an excellent energy-absorber. The current study presents an array of natural structural composites made of cane-grass - for both load-bearing and energy-absorbing functions. In order to characterize the material rigorously, detailed mechanical testing of strips obtained from cane grass stem (processed using heating and pressure loading) is carried out. It is shown that heating to 12 0 ∘ C improves the mechanical properties substantially. The specific modulus and specific strength of cane strip ranges from 45 − 72 M P a / k g / m 3 and 0 . 36 − 0 . 44 M P a / k g / m 3 , respectively. This puts cane strips in the bracket of best natural fibers. The study also presents several potential applications, including high-strength cane-strip based laminates and facesheets for sandwich panels, which can be used as load-bearing natural bricks, logs, table tops and wall panels. Another exciting application is the use of the lightweight core to make highly-efficient energy absorbing panels. This is substantiated by low velocity impact (LVI) tests at 5 J , 10 J and 15 J , which reveal excellent energy dissipation and extremely localised damage. No delamination, matrix cracking or debonding of facesheet was observed. This makes cane grass a versatile candidate for varied structural applications.

甘蔗草（Saccharum spontanum）是一种令人兴奋的结构材料，因为它的全茎可以有效地利用。外表皮区域由富含纤维素的长纤维组成，其中的长条可以直接用作承重面板和单向复合层压板的构件。同时，细胞薄壁核心是一个很好的能量吸收体。目前的研究展示了一系列由藤草制成的天然结构复合材料，具有承重和吸收能量的功能。为了严格地表征材料，对从甘蔗草茎（使用加热和压力加载处理）中获得的条进行了详细的机械测试。结果表明，加热到120°C时，机械性能大大改善。甘蔗条的比模量和比强度分别为45 ~ 72 M mpa / k g / m3和0。36−0。分别为44mpa / kg / m3。这使甘蔗条成为最好的天然纤维。该研究还提出了几种潜在的应用，包括高强度的基于藤条的层压板和夹心板的面板，它们可以用作承重的天然砖、原木、桌面和墙板。另一个令人兴奋的应用是使用轻质核心制造高效的能量吸收板。在5 J、10 J和15 J的低速冲击（LVI）试验中证实了这一点，这些试验显示出良好的能量耗散和极局部的损伤。未观察到面板的分层、基体开裂或脱粘。这使得甘蔗草成为各种结构应用的通用候选人。

Reinforced Thermoplastic Pipes (RTPs): A review of design, modeling, structural health monitoring and future outlook

S. Ezzeldin, M. Elkhodbia, F. Salamat-Zadeh, A. AlFantazi, I. Barsoum

doi:10.1016/j.compstruct.2026.120405

增强热塑性管（RTPs）的设计、建模、结构健康监测及未来展望

Reinforced Thermoplastic Pipes (RTPs) are steadily emerging as a competitive alternative to steel pipelines, combining corrosion resistance, reduced weight, rapid installation, and resistance to scale build-up due to its s mooth interior surface with the potential for lower lifecycle costs. These attributes make RTPs increasingly relevant to industries seeking reliable and efficient solutions for energy transport, water distribution, and chemical processing. This review provides a comprehensive overview of RTP systems, addressing their material constituents, structural design, service performance, and the main failure mechanis ms encountered under operational and environmental loads. Particular focus is placed on the contribution of computational modeling in advancing the design and qualification of RTPs. Constitutive descriptions of thermoplastic liners and covers, multi-scale simulations of non-metallic reinforcements including glass, carbon and aramid fibres and tapes, as well as metallic cord, wire, and strip reinforcements, and interface formulations using cohesive zone and friction-based approaches are examined. In parallel, Structural Health Monitoring (SHM) is highlighted as a key enabler for long-term reliability. Techniques ranging from radiographic and ultrasonic inspection to embedded piezoelectric, optical, and electromagnetic sensors are discussed for their role in enabling early fault detection and integration into integrity management systems. By connecting advances in modeling, SHM, and experimental validation, this review underscores how RTP technology is positioned to deliver both technical and economic value. It identifies current research directions and outlines pathways for improved qualification, digital integration, and demonstration of durability in challenging service environments-factors that are critical for broader industry adoption.

增强热塑性塑料管道（rtp）正稳步成为钢管道的竞争替代品，其具有耐腐蚀性、减轻重量、快速安装、耐水垢积聚（由于其光滑的内部表面）以及降低生命周期成本的潜力。这些特性使得rtp与寻求能源运输、水分配和化学加工等可靠高效解决方案的行业越来越相关。这篇综述提供了RTP系统的全面概述，阐述了它们的材料成分、结构设计、服务性能以及在运行和环境载荷下遇到的主要失效机制。特别重点放在计算建模在推进rtp的设计和资格的贡献。研究了热塑性衬垫和覆盖物的本构描述，非金属增强材料的多尺度模拟，包括玻璃、碳和芳纶纤维和胶带，以及金属绳、线和带增强材料，以及使用内聚区和基于摩擦的方法的界面配方。同时，结构健康监测（SHM）被强调为长期可靠性的关键促成因素。从射线照相和超声波检测到嵌入式压电、光学和电磁传感器，讨论了它们在早期故障检测和集成到完整性管理系统中的作用。通过将建模、SHM和实验验证方面的进展联系起来，本综述强调了RTP技术是如何定位于提供技术和经济价值的。它确定了当前的研究方向，并概述了改进资质、数字集成和在具有挑战性的服务环境中展示耐久性的途径——这些因素对更广泛的行业采用至关重要。

Machine learning-enabled inverse design of bimaterial thermoelastic lattice metamaterials

Xiang-Long Peng, Bai-Xiang Xu

doi:10.1016/j.compstruct.2026.120398

基于机器学习的双材料热弹性晶格超材料的逆向设计

In this work, we develop a general framework combining machine learning and high-throughput simulations for the inverse design of bimaterial thermoelastic lattice metamaterials with unconventional and tailorable effective properties. As a representative case study, the inverse design of a hybrid-honeycomb structure exhibiting negative Poisson’s ratios and negative thermal expansion coefficients is considered. A large dataset is generated through computational homogenization of structures with varying geometrical parameters and base material properties. A forward ML model is first trained to predict the effective thermoelastic properties for a given structure design. Subsequently, inverse ML models are developed to suggest design variables for prescribed target properties. To address different practical scenarios, six inverse models with different combinations of target properties and design variables are proposed. The trained forward model is integrated into the loss functions of inverse models and is further employed to generate additional datasets for cases with fixed materials. These ML models with demonstrated good performance can be applied to efficiently solve inverse design tasks involved in the practical application of the thermoelastic structure as structural or functional components in aerospace systems and temperature-adaptive structures. The methodology can be extended to the inverse design of other thermoelastic metamaterials.

在这项工作中，我们开发了一个结合机器学习和高通量模拟的通用框架，用于具有非常规和可定制有效性能的双材料热弹性晶格超材料的反设计。以负泊松比和负热膨胀系数的混合蜂窝结构为例进行了反设计。通过对具有不同几何参数和基材性质的结构进行计算均匀化，生成一个大型数据集。首先训练前向ML模型来预测给定结构设计的有效热弹性特性。随后，开发了逆ML模型，以建议指定目标属性的设计变量。针对不同的实际情况，提出了目标属性和设计变量不同组合的6种逆模型。将训练好的正演模型集成到逆模型的损失函数中，并进一步用于生成固定材料情况下的附加数据集。这些机器学习模型具有良好的性能，可以有效地解决热弹性结构在航空航天系统和温度自适应结构中作为结构或功能部件的实际应用中的反设计任务。该方法可推广到其他热弹性超材料的反设计中。

Modeling delamination and transverse cracking in laminated beams using an efficient layerwise decohesion approach

Sreenu Hari, Sankha Aditya, Bishwajit Kar, Pratyush Kumar, Arun R. Srinivasa, Samit Roy, J.N. Reddy

doi:10.1016/j.compstruct.2026.120383

用有效分层解聚方法模拟层合梁的分层和横向开裂

Thin laminated composite structures under mechanical loading exhibit coupled interlaminar delamination and intralaminar transverse cracking, which govern progressive failure and residual capacity. Capturing the interaction between these damage modes typically requires separate numerical frameworks and fine through-thickness discretization, resulting in high computational cost. This work presents a computationally efficient, thermodynamically consistent layerwise decohesion model for simulating coupled delamination and transverse cracking in laminated composite beams. Fracture is treated as an intra-element phenomenon, with multiple potential crack directions at each material point. A rate-independent decohesion model governs crack-plane degradation through internal damage variables derived from the Helmholtz free energy, ensuring irreversible evolution and non-negative dissipation. Tensile-shear mixed-mode fracture is modeled using the Benzeggagh–Kenane criterion, and crack opening/closure is captured via s mooth kinematic ramp functions. Model validation uses three-point bending experiments on carbon/epoxy laminates with varying pre-embedded delamination lengths. A single calibrated parameter set accurately predicts delamination-dominated, transverse-crack-dominated, and coupled responses across all configurations. Comparisons with Abaqus simulations using cohesive-zone and Hashin damage models show comparable accuracy while reducing computational cost by over two orders of magnitude. The framework enables rapid, robust an alysis of laminated composites, providing a practical tool for structural assess ment and design optimization.

在机械载荷作用下，薄层复合材料结构表现为层间分层和层间横向开裂的耦合作用，控制着结构的渐进破坏和残余承载力。捕获这些损伤模式之间的相互作用通常需要单独的数值框架和精细的穿透厚度离散化，从而导致高计算成本。本文提出了一种计算效率高、热力学一致的分层脱黏模型，用于模拟层合复合材料梁的耦合脱黏和横向开裂。断裂被视为一种单元内现象，在每个材料点上都有多个潜在的裂纹方向。一个速率无关的退聚模型通过由亥姆霍兹自由能导出的内部损伤变量来控制裂纹面退化，确保不可逆演化和非负耗散。拉伸-剪切混合模式断裂采用Benzeggagh-Kenane准则建模，裂纹张开/闭合通过平滑运动斜坡函数捕获。模型验证使用三点弯曲实验的碳/环氧层压板具有不同的预埋分层长度。一个单一的校准参数集可以准确地预测所有配置中的分层主导、横向裂缝主导和耦合响应。与使用黏结区和哈希损伤模型的Abaqus模拟进行比较，显示出相当的准确性，同时将计算成本降低了两个数量级以上。该框架能够快速、可靠地分析层压复合材料，为结构评估和设计优化提供实用工具。

Composites Part A: Applied Science and Manufacturing

Engineering spatial distributions of montmorillonite for enhanced low-temperature toughness of polypropylene random copolymer nanocomposites

Yuhao Guo, Ran Xiao, Zhicheng Bai, Yilan Ye, Yonggang Shangguan

doi:10.1016/j.compositesa.2026.109903

蒙脱土的工程空间分布对增强聚丙烯无规共聚物纳米复合材料低温韧性的影响

Polypropylene random copolymer/ethylene-propylene-diene monomer (PPR/EPDM) nanocomposites with distinct spatial distributions of organically modified montmorillonite (OMMT) were prepared by tailoring melt-processing sequences and compatibilizer types. The engineered distributions of OMMT within the rubber phase, at the PPR/EPDM interface, or within the PPR matrix dictates the low-temperature impact performance of the nanocomposites. OMMT localized within the rubber phase maximized low-temperature toughness, achieving a notched impact strength of 112 kJ/m2 at −10 °C and 72 kJ/m2 at −30 °C, and lowering the brittle-ductile transition temperature to −40 °C. In contrast, OMMT at the interface had a minimal effect on low-temperature toughness, while OMMT within the matrix reduced impact resistance but increased stiffness. Rheological and dynamic mechanical an alyses revealed that rubber-phase-localized OMMT enhances internal friction and energy dissipation, delaying crack initiation. Interfacial OMMT primarily stabilizes the interface and transfers stress but does not significantly promote bulk energy dissipation. Matrix-localized OMMT restricts molecular mobility, increasing storage modulus (G′) but reducing impact toughness. These findings establish structure–property-mechanis m relationships and provide principles for designing polypropylene-based nanocomposites with superior low-temperature impact resistance.

通过调整熔体加工顺序和增容剂类型，制备了具有不同空间分布的有机改性蒙脱土（OMMT）的聚丙烯无规共聚物/乙丙二烯单体（PPR/EPDM）纳米复合材料。OMMT在橡胶相、PPR/EPDM界面或PPR基体中的工程分布决定了纳米复合材料的低温冲击性能。位于橡胶相的OMMT使低温韧性最大化，在- 10°C和- 30°C时的缺口冲击强度分别达到112 kJ/m2和72 kJ/m2，并将脆性-韧性转变温度降至- 40°C。相比之下，界面处的OMMT对低温韧性的影响最小，而基体内的OMMT降低了抗冲击性，但增加了刚度。流变学和动态力学分析表明，橡胶相局部化的OMMT增强了内摩擦和能量耗散，延缓了裂纹萌生。界面OMMT主要起到稳定界面和传递应力的作用，但不显著促进体能耗散。基体局域化的OMMT限制了分子迁移率，增加了存储模量（G '），但降低了冲击韧性。这些发现建立了结构-性能-机理关系，为设计具有优异低温抗冲击性能的聚丙烯基纳米复合材料提供了理论依据。

Polymer systems for liquid silicon infiltrated ceramic matrix composites

Felix Wich, Martin Demleitner, Farnaz Ebrahimi, Nico Langhof, Gaetano La Delfa, Stefan Ellinger, Holger Ruckdäschel, Stefan Schafföner

doi:10.1016/j.compositesa.2026.109867

液态硅渗透陶瓷基复合材料的聚合物体系

This work investigated the suitability of different commercially available polymers for manufacturing ceramic matrix composites (CMC) based on the liquid silicon infiltration (LSI) process. Carbon fiber reinforced composites based on thermoplastic PEEK, phenolic novolak resin and three polyaddition curing cyanate ester resins were produced and characterized using DMA, TGA and dilatometry. The composite samples were pyrolyzed and infiltrated using liquid silicon to obtain the desired CMC. The manufactured CMC were characterized through light-microscopy, three-point bending tests, SEM an alysis and X-ray diffraction with phase composition an alysis. Interactions between the properties in the polymer and ceramic state were revealed. The pyrolysis mechanis m and char yield clearly determine the microstructure, the delamination tendency during pyrolysis and the phase composition in the ceramic state and ultimately define the mechanical properties of the CMC. It was shown that high char yield cyanate ester resins are well suited for the LSI process. Damage-tolerant C/C-SiC ceramics with a bending strength of more than 130MPa were obtained for all resin systems. Overall, cyanate ester resin based CMC displayed a unique combination of favorable processing properties such as low-viscosity, polyaddition cure, high char yield and exceptional fiber-protection qualities and present a promising resin system for future CMC development.

本文研究了基于液态硅渗透（LSI）工艺制造陶瓷基复合材料（CMC）的不同商用聚合物的适用性。以热塑性PEEK、酚醛novolak树脂和三种聚加成固化氰酸酯树脂为基体制备了碳纤维增强复合材料，并用DMA、TGA和膨胀法对其进行了表征。用液态硅对复合样品进行热解和渗透，得到所需的CMC。通过光学显微镜、三点弯曲试验、扫描电镜分析和x射线相组成分析对制备的CMC进行了表征。揭示了聚合物性质与陶瓷态之间的相互作用。热解机理和炭产率清楚地决定了CMC的微观结构、热解过程中的分层倾向和陶瓷状态下的相组成，最终决定了CMC的力学性能。结果表明，高炭收率的氰酸酯树脂非常适合于大规模集成电路工艺。所有树脂体系均获得了弯曲强度超过130MPa的耐损伤C/C- sic陶瓷。综上所述，氰酸酯树脂基CMC具有低粘度、多加成固化、高炭收率和优异的纤维保护性能，是未来CMC发展的一种很有前途的树脂体系。

Impact of fabrication dynamics and coating interaction on interfacial integrity and distributed strain sensing in 3D-Printed composites

Ahmad Shawki Charkieh, Sergei Mikhailov, Efstratios Polyzos, Yifan Zhu, Sidney Goossens, Francis Berghmans, Lincy Pyl

doi:10.1016/j.compositesa.2026.109900

制备动力学和涂层相互作用对3d打印复合材料界面完整性和分布应变传感的影响

This study investigates the integration of distributed optical sensing arrays into Fused Deposition Modeling composites to enable embedded strain monitoring. The integration of a distributed optical fiber network within pris matic composite coupons enables internal strain mapping beyond the capabilities of surface-based pointwise or full-field optical strain measurement techniques. The fabrication process utilized a precise stop-and-go embedding technique within optimized diamond-shaped channels. In-situ infrared thermography identified a “heat cone” effect, where thermal consolidation in the coupon core facilitated superior sensor encapsulation compared to the rapid cooling at the periphery. This was verified via Micro-Computed Tomography, which revealed that central sensing sections achieved significantly higher interfacial contact percentages than edge sensing sections. Comparative an alysis of sensor coatings showed that acrylate and bare optical fibers achieved effective mechanical coupling, whereas thermally stable polyimide coatings exhibited poor wettability and stick–slip failure under load. Mechanical testing confirmed these findings by demonstrating that the sensing array could detect fabrication-induced bond asymmetries and accurately map the location and orientation of an oblique structural anomaly. These results establish a validated framework for advanced structural health monitoring, enabling precise strain detection and internal defect mapping in complex additive manufacturing architectures.

本研究将分布式光学传感阵列集成到熔融沉积建模复合材料中，以实现嵌入式应变监测。在棱镜复合材料片内集成分布式光纤网络，使内部应变映射超越了基于表面的点向或全场光学应变测量技术的能力。制造过程在优化的菱形通道中使用了精确的走走停停嵌入技术。原位红外热成像发现了一种“热锥”效应，与外围的快速冷却相比，联片核心的热固结有助于更好的传感器封装。通过微计算机断层扫描验证了这一点，结果显示，中心传感切片的界面接触百分比明显高于边缘传感切片。对传感器涂层的对比分析表明，丙烯酸酯和裸光纤实现了有效的机械耦合，而热稳定的聚酰亚胺涂层在负载下表现出较差的润湿性和粘滑破坏。机械测试证实了这些发现，表明传感阵列可以检测到制造引起的键不对称，并准确地绘制出倾斜结构异常的位置和方向。这些结果为先进的结构健康监测建立了一个经过验证的框架，可以在复杂的增材制造体系结构中实现精确的应变检测和内部缺陷映射。

Design of 3D printed cellulose-protein scaffolds for tissue engineering

Iraia Osquila, Hegoa Craamer, Ainhoa Irastorza, Koro de la Caba, Pedro Guerrero

doi:10.1016/j.compositesa.2026.109899

组织工程用3D打印纤维素蛋白支架的设计

Inks based on soy protein and gelatin were developed in this study. Cellulose, incorporated as a filler into these formulations, was dissolved in a deep eutectic solvent (DES) in different concentrations (0, 10, 15 and 20 wt%). The inks exhibited shear-thinning behavior and solid-like properties, enabling easy extrusion and maintaining shape fidelity after 3D printing. FTIR an alysis indicated interactions between proteins and cellulose and morphological studies confirmed good integration of cellulose within the protein matrix, leading to the improvement of the mechanical performance of cellulose-protein scaffolds. The incorporation of cellulose also improved water uptake capacity, without compromising integrity. Finally, biocompatibility assays demonstrated cell viability, confirming the potential of the developed scaffolds for tissue engineering.

本研究开发了基于大豆蛋白和明胶的油墨。将纤维素作为填料加入到这些配方中，以不同的浓度（0,10,15和20% wt%）溶解在深共晶溶剂（DES）中。油墨表现出剪切变薄行为和类似固体的特性，使3D打印后易于挤压和保持形状保真度。FTIR分析表明蛋白质与纤维素之间存在相互作用，形态学研究证实纤维素在蛋白质基质内具有良好的整合，从而提高了纤维素-蛋白质支架的机械性能。纤维素的掺入也提高了吸水能力，而不损害完整性。最后，生物相容性试验证明了细胞活力，证实了所开发的支架在组织工程中的潜力。

Composites Part B: Engineering

Dual-resonance hybrid composite metastructures for broadband acoustic absorption under high hydrostatic pressure

Guoqing Sun, Xiyue An, Changliang Lai, Hongbo Tao, Hualin Fan

doi:10.1016/j.composites b.2026.113755

高静水压力下宽带声吸收的双共振杂化复合元结构

The progressive increase in the depth of submarine operations, coupled with the heightened need for acoustic stealth, has imposed more stringent demands on underwater sound-absorbing materials. Nevertheless, conventional sound-absorbing materials frequently undergo deformation or structural failure when subjected to the extreme pressures encountered in deep-water environments, leading to a deterioration in their sound absorption efficacy. This study introduces a novel composite metastructure featuring a hybrid configuration of dual-resonance units, designed to achieve broadband sound absorption in deep-water environments. The proposed structure encapsulates the local-resonance and cavity-resonance acoustic units within a sandwich configuration, consisting of a carbon fiber orthogonal grid core and buoyant material facesheets. Additionally, each acoustic unit type is combined with a specifically selected matrix material. The research findings indicate that both the local resonance structure and the cavity structure exhibit specific underwater sound absorption capabilities. The sound absorption performance of the composite hybrid structure is further improved as a result of the coupling effect, surpassing the simple combined performance of the two individual structures. The incorporation of carbon fiber grids enhanced the mechanical load-bearsing capacity of the acoustic material. Experimental results confirmed that the hybrid structure sustains broadband sound absorption efficacy under hydrostatic pressures ranging from 0.5 MPa to 6.0 MPa. This design presents a feasible approach for broadband sound absorption in deep-water environments and holds significant potential for applications in submarine acoustic stealth and underwater acoustic engineering.

潜艇作业深度的不断增加，加上声隐身需求的提高，对水下吸声材料提出了更严格的要求。然而，传统吸声材料在深水环境中受到极端压力时，往往会发生变形或结构破坏，导致吸声效果下降。本研究介绍了一种新型复合元结构，具有双共振单元的混合配置，旨在实现深水环境中的宽带吸声。提出的结构将局部共振和腔共振声学单元封装在三明治结构中，由碳纤维正交网格核心和浮力材料面板组成。此外，每种声学单元类型都与特定选择的基质材料相结合。研究结果表明，局部共振结构和腔体结构均表现出特定的水声吸收能力。由于耦合作用，复合杂化结构的吸声性能进一步提高，超越了两个单独结构的简单组合性能。碳纤维网格的加入增强了声学材料的机械承载能力。实验结果证实，在0.5 MPa ~ 6.0 MPa的静水压力范围内，混合结构仍能保持宽带吸声效果。该设计为深水环境下的宽带吸声提供了一种可行的方法，在潜艇声隐身和水声工程中具有重要的应用潜力。

Simulation-Guided Process Development for Thermoforming of Aligned Discontinuous Fibre Composites

Burak Ogun Yavuz, Ian Hamerton, Marco L. Longana, Jonathan P.-H. Belnoue

doi:10.1016/j.composites b.2026.113754

对准不连续纤维复合材料热成形的模拟导向工艺开发

Aligned discontinuous fibre-reinforced composites (ADFRCs) combines near-continuous fibre mechanical properties with the formability and recyclability needed for sustainable manufacturing of composites. Structural properties have been demonstrated at coupon scale, and recent studies have shown that ADFRCs can be formed into complex geometries. However, these forming trials remain largely empirical. The interplay between matrix viscoelasticity, fibre overlap evolution, and the requirement to form below the melting temperature to preserve fibre alignment creates a narrow process window that is difficult to navigate through trial-and-error alone. This study presents a forming simulation framework to contribute process development. ADFRCs comprising 3 mm carbon fibres in a bio-based poly(L-lactic acid) matrix were manufactured via the HiPerDiF process. Viscoelastic constitutive models implemented within Abaqus/Explicit predict fibre sliding, overlap evolution, and forming feasibility, where the tensile constitutive model is experimentally validated; shear and bending models are supported by sensitivity an alysis demonstrating their secondary role under the conditions studied. Parametric development identified that a forming temperature of 140 °C, a rate of 375 mm min-1, and corner-constrained boundaries yield defect-free laminates. This was confirmed experimentally in a single forming trial without iterative optimisation. The framework demonstrates that simulation-guided process development can reduce dependence on empirical trial-and-error by narrowing the process window before physical trials.

定向不连续纤维增强复合材料（ADFRCs）将近乎连续的纤维力学性能与复合材料可持续制造所需的可成形性和可回收性相结合。ADFRCs的结构特性已经在券面尺度上得到了证明，最近的研究表明，ADFRCs可以形成复杂的几何形状。然而，这些成形试验在很大程度上仍然是经验性的。基体粘弹性、纤维重叠演变以及在熔化温度以下形成以保持纤维排列的要求之间的相互作用造成了一个狭窄的工艺窗口，很难通过单独的试错来导航。本研究提出了一个成形模拟框架，以促进工艺开发。ADFRCs由3mm碳纤维在生物基聚乳酸（l -乳酸）基质中通过HiPerDiF工艺制造。在Abaqus/Explicit中实现的粘弹性本构模型预测纤维滑动、重叠演变和成型可行性，其中拉伸本构模型经过实验验证；剪切和弯曲模型得到敏感性分析的支持，表明它们在研究条件下的次要作用。参数化开发表明，在成形温度为140°C、速度为375 mm min-1和角约束边界的条件下，可以得到无缺陷的层压板。这在没有迭代优化的单次成形试验中得到了实验证实。该框架表明，通过缩小物理试验前的过程窗口，仿真指导的过程开发可以减少对经验试错的依赖。

Effect of Anchor Fan-Radius Ratio on the Compressive Behavior and Damage Evolution of CFRP-Anchored RC Columns

Zhengxie Zhang, Yonglai Zheng, Tanbo Pan, Xin Lan, Chenyu Hou

doi:10.1016/j.composites b.2026.113753

锚扇半径比对cfrp锚固混凝土柱压缩性能及损伤演化的影响

This study evaluates how the anchor fan-length–to–column radius ratio governs the mechanical response and damage evolution of CFRP-anchored reinforced concrete (RC) short columns. Uniaxial compression tests were conducted with acoustic emission (AE) monitoring to track failure progression and cracking mechanis ms. Unstrengthened and wrap-only specimens were dominated by corner stress concentrations, leading to longitudinal splitting or abrupt brittle rupture of the CFRP jacket. As the fan-radius ratio increased, the failure mode shifted from localized damage to a cooperative, compression-dominated response over the full section. At low ratios (0.27–0.53), drilling-induced cross-section weakening partially offset the strengthening benefit. In contrast, at a ratio of 0.80 the ultimate capacity reached 1334.5 kN, representing a 49.6% increase over the control, and the ductility index rose to 1.438. Stress–strain responses and circumferential strain distributions indicate that a large fan-radius ratio homogenizes interfacial strains and stabilizes the confined core concrete in an equivalent triaxial compression state. AE results reveal a three-stage evolution (initiation–propagation–failure); for high ratios, ringing counts, cumulative energy, HI and Sr were markedly delayed, and peak severity was ∼one-tenth of the control. GMM-based RA–AF an alysis further suggests a shift from shear-dominated cracking to coupled shear–tension energy dissipation. These findings highlight the critical role of a rational fan-radius ratio in maximizing strengthening efficiency and improving failure controllability.

研究了锚扇长柱半径比对cfrp锚固短柱力学响应和损伤演化的影响。采用声发射（AE）监测进行单轴压缩试验，以跟踪破坏进展和开裂机制。未加固和仅包覆试样以拐角应力集中为主，导致CFRP护套纵向劈裂或突然脆性断裂。随着扇形半径比的增加，破坏模式从局部破坏转变为整个截面的协同、压缩主导响应。在较低比率（0.27-0.53）时，钻井引起的截面弱化部分抵消了加固效益。当比例为0.80时，极限承载力达到1334.5 kN，比对照提高49.6%，延性指数提高到1.438。应力应变响应和周向应变分布表明，较大的扇形半径比使界面应变均匀化，使承压核心混凝土处于等效三轴压缩状态。声发射结果显示了三个阶段的演化过程（起始-传播-失效）；对于高比率，振铃计数、累积能量、HI和Sr明显延迟，峰值严重程度为对照组的十分之一。基于gmm的RA-AF分析进一步表明，从剪切为主的开裂向剪切-张拉耦合耗能转变。这些发现强调了合理的风扇半径比在最大限度地提高效率和改善故障可控性方面的关键作用。

Mechanical Property Evaluation of SiO2/Al2O3 Hybrid Fiber Porous Materials: Sintering-Informed Microstructural Modeling and Experimental Validation

Yuxin Li, Hong Ye, Xiaoliang Ma, Xuefeng Mu, Kaili Yin, Yuanpeng Fang, Liping Shi, Yesheng Zhong, Xiaodong He

doi:10.1016/j.composites b.2026.113752

SiO2/Al2O3杂化纤维多孔材料力学性能评价：烧结通知微观结构建模和实验验证

The macroscopic properties of ceramic fiber porous materials sintered from random fibers are influenced by complex couplings of microscopic factors. For hybrid ceramic fiber porous materials, the lack of effective simulations accounting for fiber interactions makes performance prediction a challenge. In this study, a finite element micro-model with stochastically distributed fibers was established to simulate the compressive behavior of SiO2/Al2O3 hybrid ceramic fiber porous materials. In addition to conventional approaches, a novel inter-fiber bonding model accounting for sintering degree was developed based on fiber surface diffusion simulation. Verified against experimental data, the developed model demonstrates predictive fidelity in simulating compressive behavior, reducing the average strength deviation to -2.29%. Simulation results demonstrate that the macroscopic strength can be enhanced by 14.4% through fully sintering and elevating the sintering degree effectively suppresses fiber buckling. Parameter coupling an alysis indicates that porosity and fiber ratio both negatively correlate with compressive strength but exert opposing influences on density. This divergence creates a design trade-off. Ultimately, a comprehensive evaluation framework for the mechanical performance of composite ceramic fiber porous materials was proposed, which can be extended to support the on-demand design and fabrication of hybrid ceramic fiber materials.

随机纤维烧结陶瓷纤维多孔材料的宏观性能受到微观因素的复杂耦合影响。对于杂化陶瓷纤维多孔材料，由于缺乏有效的纤维相互作用模拟，使得性能预测成为一个挑战。本文建立了具有随机分布纤维的有限元微观模型，模拟了SiO2/Al2O3杂化陶瓷纤维多孔材料的压缩行为。在传统方法的基础上，基于纤维表面扩散模拟，建立了考虑烧结程度的纤维间键合模型。通过实验数据验证，该模型在模拟压缩行为方面具有良好的预测保真度，将平均强度偏差降低至-2.29%。模拟结果表明，充分烧结可使宏观强度提高14.4%，提高烧结程度可有效抑制纤维屈曲。参数耦合分析表明，孔隙率和纤维比与抗压强度呈负相关，而对密度的影响相反。这种分歧造成了设计上的取舍。最后，提出了复合陶瓷纤维多孔材料力学性能的综合评价框架，该框架可扩展到支持混合陶瓷纤维材料的按需设计和制造。

Interfacial Porosity Modulates Fracture of Brick-and-Mortar Multi-Metal Composites in Additive Manufacturing

Zifan Wang, Radim Kocich, Miaomiao Zou, Darren Batey, Matthew Spink, Lenka Kunčická

doi:10.1016/j.composites b.2026.113751

增材制造中界面孔隙度对砖混复合材料断裂的调节作用

Multi-material structures have shown great versatility in wide applications. However, additive manufacturing of multi-metal mechanical composite structures is challenging. Beyond this, a comprehensive and multi-scale understanding of the fracture mechanis ms in such structures has not been sufficiently elucidated. In this study, we exploited synchrotron phase contrast X-ray computed tomography and synchrotron X-ray ptychographic tomography to achieve in situ, continuous observation of the fracturing process in large-scale brick-and-mortar multi-metal composite structures, resolving phenomena spanning from the micro- to nano- scale. Findings suggest that nano-pores prevailingly exist in additively manufactured metals, and interfacial porosity as a transitional geometry between different materials can retard the crack growth and improve fracture toughness. This multi-scale study directly informs the designing, manufacturing, and testing of multi-metal composite structures.

多材料结构在广泛的应用中显示出极大的通用性。然而，多金属机械复合材料结构的增材制造具有挑战性。除此之外，对此类构造中断裂机制的全面和多尺度理解尚未得到充分阐明。在这项研究中，我们利用同步加速器相衬x射线计算机断层扫描和同步加速器x射线断层扫描技术，实现了对大型砖瓦复合材料结构中压裂过程的现场连续观察，解决了从微观到纳米尺度的现象。结果表明，增材制造金属中普遍存在纳米孔隙，界面孔隙作为不同材料之间的过渡几何形态，可以延缓裂纹扩展，提高断裂韧性。这种多尺度的研究直接指导了多金属复合材料结构的设计、制造和测试。

Thermo-forging CF/PPS rivet for joining CF/PEEK composite and TC4 titanium alloy structures

Xiangfei Kong, Jun Lin, Cong Shao, Dong Quan, Yanjin Guan, Liang Chen, Cunsheng Zhang, Guilong Wang, Guoqun Zhao

doi:10.1016/j.composites b.2026.113750

CF/PPS热锻铆钉，用于连接CF/PEEK复合材料和TC4钛合金结构

Owing to the recyclability, light weight and high strength, thermoplastic composites are increasingly applied in aircraft and automotive fields, and joining composite skins and lightweight alloy frames by riveting process is of great importance. To overcome the problems like weight increase and potential corrosion existed in metallic rivets, the present study proposes thermo-forging carbon fiber reinforced polyphenylene sulfide (CF/PPS) thermoplastic composite rivet to join CF reinforced polyetheretherketone (PEEK) composite and metal structures, where CF/PEEK may not be melted in the riveting process since the melting point of CF/PPS is relatively low. A thermoforming platform with controllable heating and cooling rates is developed based on electric-hydraulic servo press. During thermal deformation, resin percolation and fiber buckling occur, leading to an improvement in forming quality due to an elevated temperature. Excellent mechanical performance is demonstrated by the joints fabricated at high temperatures and low cooling rates during riveting, by which increased crystallinity and effective resin encapsulation of the fibers are achieved. The specific strengths of CF/PPS joints in tensile shear and pull-through tests are 227.60% and 151.66% higher than those connected by Ti45Nb rivets. The fracture mode of the joint is rivet failure, with multiple damage forms observed at the fracture surface, such as fiber breakage and interfacial debonding. Compared to joints connected by composite fasteners reported in the literature, the present joint exhibits superior strength, demonstrating that the proposed thermo-forging and riveting process using relatively low-melting-point composite rivet is of significant potential for joining thermoplastic composite and lightweight alloy structures in aerospace industry.

热塑性复合材料具有可循环利用、重量轻、强度高等优点，在飞机和汽车领域的应用越来越广泛，通过铆接工艺将复合材料蒙皮与轻质合金车架连接起来具有重要意义。为了克服金属铆钉存在的重量增加和潜在腐蚀等问题，本研究提出了热锻碳纤维增强聚苯硫醚（CF/PPS）热塑性复合铆钉，用于连接CF增强聚醚醚酮（PEEK）复合材料与金属结构，由于CF/PPS的熔点较低，在铆接过程中CF/PEEK可能不会熔化。在电液伺服压力机的基础上，研制了一种加热和冷却速率可控的热成形平台。在热变形过程中，树脂渗透和纤维屈曲发生，导致成形质量的改善，由于温度升高。在高温和低冷却速率下制造的接头具有优异的机械性能，从而提高了纤维的结晶度和有效的树脂包封性。CF/PPS接头在拉伸剪切和拉穿试验中的比强度分别比Ti45Nb铆钉连接的接头高227.60%和151.66%。接头断裂方式为铆钉破坏，断口处存在多种损伤形式，如纤维断裂、界面脱粘等。与文献中报道的复合材料紧固件连接的接头相比，本接头具有优越的强度，表明采用相对低熔点复合材料铆钉的热锻铆接工艺在航空航天工业中连接热塑性复合材料和轻质合金结构具有重要的潜力。

Composite CrN/HVOF Coatings for Superior High-Temperature Wear Resistance

Lei Shao, Ru Liu, Wensheng Li, Yanru Sun, Xuedan Song, Xiaojie Song, Canming Wang, Qiang Song, Lunlin Shang, Sheng Guo

doi:10.1016/j.composites b.2026.113748

复合CrN/HVOF涂层具有优异的高温耐磨性

CrN coatings possess excellent hardness but suffer from rapid degradation under high-temperature sliding, largely due to limited toughness and substrate softening. In this work, we introduce a composite coating architecture, consisting of a CrN top layer supported by NiCrBSi or Stellite12 high-velocity oxy-fuel (HVOF) interlayers, designed to improve high-temperature load-bearing capacity and wear performance. Reciprocating sliding tests up to 600 °C demonstrate that the composite systems exhibit substantially reduced wear rates compared with monolithic CrN, enabled primarily by the mechanical support of the HVOF interlayers, which effectively suppress substrate plasticity and crack formation. Cross-sectional TEM of worn regions reveals that CrN accommodates high-temperature deformation through a sequence of dislocation activities, partial dislocation slip generating stacking faults and nanotwins, and stress-assisted amorphization at grain boundaries. While these mechanis ms provide additional strain accommodation, they are not sufficient to prevent severe damage in monolithic CrN. In contrast, the composite coating maintains structural integrity and shows a transition to predominantly mild oxidative wear, highlighting the critical role of the interlayer in stabilizing the CrN film under thermomechanical loading. This study demonstrates that engineering a load-bearing composite architecture provides an effective pathway to achieve high-temperature wear resistance in nitride ceramic coatings, while offering new insight into the high-temperature deformation behaviors of CrN.

CrN涂层具有优异的硬度，但在高温滑动下会迅速降解，这主要是由于有限的韧性和基体软化。在这项工作中，我们介绍了一种复合涂层结构，由NiCrBSi或Stellite12高速氧燃料（HVOF）夹层支撑的CrN顶层组成，旨在提高高温承载能力和磨损性能。高达600°C的往复滑动测试表明，与单片CrN相比，复合材料系统的磨损率大大降低，这主要得益于HVOF夹层的机械支撑，它有效地抑制了基体的塑性和裂纹的形成。磨损区域的透射电镜显示，CrN通过一系列位错活动、部分位错滑移产生层错和纳米孪晶以及晶界应力辅助非晶化来适应高温变形。虽然这些机制提供了额外的应变调节，但它们不足以防止单片CrN的严重损伤。相比之下，复合涂层保持了结构的完整性，并呈现出以轻度氧化磨损为主的转变，突出了中间层在热机械载荷下稳定CrN膜的关键作用。该研究表明，设计承载复合材料结构为氮化陶瓷涂层提供了实现高温耐磨性的有效途径，同时为CrN的高温变形行为提供了新的见解。

Functionally graded composites with triply periodic minimal surface fabricated by micro-laser powder bed fusion for superior mechanical response

Dien Hu, Jianying Wang, Jun-Yuan Zheng, M.W. Fu

doi:10.1016/j.composites b.2026.113747

采用微激光粉末床熔合技术制备了具有三周期最小表面的功能梯度复合材料

Composite triply periodic minimal surface (TPMS) structures offer a promising approach to tailor material properties by combining the advantages of constituent materials. In this research, a novel composite SS316L-AlCoCrFeNi2.1 functionally graded (FG) TPMS structure was developed by micro laser powder bed fusion (μLPBF). A dual structural design enables the integration of the inherent high strength of the low-relative-density AlCoCrFeNi2.1 part with the high relative density-induced excellent load-bearing capacity of the SS316L section. During compression along the building direction, the deformation behavior of the composite FG structures was transformed from layer-by-layer to a coordinated deformation during the stress plateau stage, characterized by a relatively stable stress level as strain increases, undergoing strengthening, stress plateau, and strengthening stages at a high stress level. The participation of the AlCoCrFeNi2.1 section also induces an enhanced mechanical response during the transversal compression. As a result, the developed structure processes enhanced mechanical properties, including plateau stress and energy absorption capacity, compared to the pure SS316L counterparts. The developed composite structure demonstrates a novel cost-effective strategy for achieving excellent mechanical response, showcasing significant potential for high-performance metamaterial applications.

复合三周期最小表面（TPMS）结构通过结合组成材料的优点，提供了一种有前途的方法来定制材料的性能。本研究采用微激光粉末床熔融（μLPBF）法制备了一种新型复合材料SS316L-AlCoCrFeNi2.1功能梯度（FG） TPMS结构。双重结构设计使低相对密度AlCoCrFeNi2.1部件固有的高强度与高相对密度引起的SS316L截面的卓越承载能力相结合。在沿建筑方向的压缩过程中，复合材料FG结构的变形行为由逐层变形转变为应力平台阶段的协调变形，其特征是随着应变的增加，应力水平相对稳定，经历强化、应力平台和高应力水平下的强化阶段。AlCoCrFeNi2.1截面的参与也会在横向压缩期间引起增强的力学响应。因此，与纯SS316L相比，发达的组织工艺提高了力学性能，包括高原应力和能量吸收能力。开发的复合材料结构展示了一种新的经济有效的策略，可以实现优异的机械响应，展示了高性能超材料应用的巨大潜力。

Composites Science and Technology

Mechanis m-guided machine learning framework for predicting long-term aging of bamboo scrimber

Bingbing Li, Chang Wang, Yue Qi, Hongxia Ma, Wenji Yu, Yahui Zhang

doi:10.1016/j.compscitech.2026.111680

基于机器学习的竹材长期老化预测框架

Accurate understanding of the natural aging processes in bamboo scrimber is essential for enhancing its durability and service life, thereby enabling reliable outdoor applications and supporting sustainable construction. Due to its relatively recent development as an engineered biomass material and its continued application in outdoor environments without observable biodegradation, significant limitations exist in accurately understanding the lifetime of bamboo scrimber. This study employs a data-driven multi-scale framework based on machine learning to predict the natural aging process of bamboo scrimber. By introducing environmental factors as inducers, the model reveals the evolution of macroscopic properties from the perspective of microscopic chemical changes. The chemical components in bamboo scrimber along long-term aging process were examined by Fourier-transform infrared spectroscopy (FTIR) and X-ray photoelectron spectroscopy (XPS). Machine learning methods including XGBoost, GBDT, Random Forest, KNN, and SVR were used to establish a relationship between microscopic indicators and macroscopic mechanical properties based on collected data for a decade. Furthermore, an Autoregressive Integrated Moving Average with Exogenous variables (ARIMAX) model is applied for predicting the temporal dynamics of key microscopic features under varying environmental conditions. An improvement in average predictive accuracy (MAPE) exceeding 10% was achieved over conventional single-stage models, demonstrating the superior long-term forecasting capability of the integrated framework on a 10-year natural aging dataset. The mechanis m-guided approach not only ensures accurate prediction of mechanical performance but, through quantitative SHAP a nalysis, identifies the C1, C2, and C1/C2 ratio (derived from XPS) as the most influential indicators.

准确了解竹材的自然老化过程对于提高其耐久性和使用寿命至关重要，从而实现可靠的户外应用并支持可持续建筑。由于其作为一种工程生物质材料的发展相对较晚，并且在室外环境中持续应用而没有明显的生物降解，因此在准确理解竹纤维的寿命方面存在显着的局限性。本研究采用基于机器学习的数据驱动多尺度框架对竹材自然老化过程进行预测。该模型通过引入环境因素作为诱导剂，从微观化学变化的角度揭示宏观性质的演化。采用傅里叶变换红外光谱（FTIR）和x射线光电子能谱（XPS）对竹材长期老化过程中的化学成分进行了分析。利用XGBoost、GBDT、Random Forest、KNN、SVR等机器学习方法，基于近十年收集的数据，建立微观指标与宏观力学性能之间的关系。在此基础上，提出了一种具有外生变量的自回归综合移动平均（ARIMAX）模型，用于预测不同环境条件下关键微观特征的时间动态。与传统的单阶段模型相比，平均预测精度（MAPE）提高了10%以上，表明综合框架在10年自然老化数据集上具有优越的长期预测能力。以机制为导向的方法不仅确保了机械性能的准确预测，而且通过定量的SHAP分析，确定了C1、C2和C1/C2比率（由XPS导出）是最具影响力的指标。

Morphology-Informed Tow-wise Modelling of AFP-Manufactured Pseudo-Woven Laminates

Saral Mittal, Ebrahim Oromiehie, B. Gangadhara Prusty

doi:10.1016/j.compscitech.2026.111685

afp制造伪编织层压板的形态双向建模

The high-precision fibre tow deposition capability of Automated Fibre Placement (AFP) has enabled fabrication of tailored meso-structural composite architectures, termed as Meso Architected Composites (MAC). These engineered configurations involve fibre tow traversal between adjacent plies, forming pseudo-woven geometries that have demonstrated enhanced impact resistance. However, comprehensive characterisation of their in-plane mechanical behaviour remains limited. Furthermore, conventional ply-wise modelling strategies are inadequate for numerically representing these architectures, necessitating a morphology-informed tow-level modelling framework as presented in this work for accurately predicting their mechanical behaviour. This study presents the first systematic characterisation attempt for investigating the internal meso-structural features in three distinct AFP-manufactured MAC configurations with varying inter-tow spacing. The characterised morphological features are subsequently incorporated into an efficient parametric geometric description for tow-wise numerical modelling. Experimental tensile characterisation of the manufactured MAC configurations indicates less than 5% degradation in tensile strength for the low-gap configurations relative to a conventional non-woven laminate. This behaviour is attributed to pronounced tow spreading at tow junctions during consolidation, resulting in minimal fibre undulation angles. The proposed morphology-informed numerical modelling approach accurately captures the tensile response of the MAC configurations, predicting the tensile strength within a 5% margin, while providing consistent insight into the associated damage mechanis ms at the tow-level. This high-fidelity modelling capability not only deepens the understanding of mechanical behaviour of currently characterised MAC configurations but also establishes a foundation for the rapid performance prediction of a wide range of potential configurations enabled by this tow placement strategy.

自动铺放成型（AFP）的高精度纤维束沉积能力使定制的介观结构复合材料体系结构成为可能，称为介观结构复合材料（MAC）。这些工程配置包括相邻层之间的纤维束穿越，形成伪编织几何形状，具有增强的抗冲击性。然而，对其平面内力学行为的全面表征仍然有限。此外，传统的聚类建模策略不足以在数字上表示这些结构，因此需要在本工作中提出的具有形态信息的二级建模框架，以准确预测其力学行为。本研究首次系统地研究了三种不同的afp制造的具有不同束间间距的MAC结构的内部细观结构特征。特征形态特征随后被纳入有效的参数几何描述，用于双向数值模拟。制造的MAC结构的实验拉伸特性表明，相对于传统的无纺布层压板，低间隙结构的拉伸强度下降小于5%。这种行为归因于在固结过程中明显的束扩散，导致纤维波动角最小。所提出的基于形态的数值模拟方法准确捕获了MAC结构的拉伸响应，预测拉伸强度在5%的范围内，同时提供了对二级相关损伤机制的一致见解。这种高保真建模能力不仅加深了对当前表征的MAC配置的机械行为的理解，而且为通过这种拖缆放置策略实现的广泛潜在配置的快速性能预测奠定了基础。