今日更新:Composite Structures 1 篇,Composites Part A: Applied Science and Manufacturing 1 篇,Composites Part B: Engineering 1 篇,Composites Science and Technology 1 篇
Stochastic interpretable machine learning based multiscale modeling in thermal conductivity of Polymeric Graphene-enhanced composites
Liu Bokai, Lu Weizhuo, Olofsson Thomas, Zhuang Xiaoying, Rabczuk Timon
doi:10.1016/j.compstruct.2023.117601
基于随机可解释机器学习的聚合物石墨烯增强复合材料导热性多尺度建模
We introduce an interpretable stochastic integrated machine learning based multiscale approach for the prediction of the macroscopic thermal conductivity in Polymeric graphene-enhanced composites (PGECs). This method encompasses the propagation of uncertain input parameters from the meso to macro scale, implemented through a foundational bottom-up multi-scale framework. In this context, Representative Volume Elements in Finite Element Modeling (RVE-FEM) are employed to derive the homogenized thermal conductivity. Besides, we employ two sets of techniques: Regression-tree-based methods (Random Forest and Gradient Boosting Machine) and Neural networks-based approaches (Artificial Neural Networks and Deep Neural Networks). To ascertain the relative influence of factors on output estimations, the SHapley Additive exPlanations (SHAP) algorithm is integrated. This interpretable machine learning methodology demonstrates strong alignment with published experimental data. It holds promise as an efficient and versatile tool for designing new composite materials tailored to applications involving thermal management.
我们介绍了一种基于随机综合机器学习的多尺度可解释方法,用于预测聚合物石墨烯增强复合材料(PGEC)的宏观热导率。该方法通过自下而上的多尺度基础框架,将不确定的输入参数从中观尺度传播到宏观尺度。在此背景下,我们采用了有限元建模中的代表性体积元素(RVE-FEM)来推导均质化热导率。此外,我们还采用了两套技术:基于回归树的方法(随机森林和梯度提升机)和基于神经网络的方法(人工神经网络和深度神经网络)。为了确定各因素对输出估计的相对影响,还集成了 SHapley Additive exPlanations(SHAP)算法。这种可解释的机器学习方法与已公布的实验数据非常吻合。它有望成为设计新型复合材料的高效多功能工具,为涉及热管理的应用量身定制。
Electromagnetic interference shielding composites and the foams with gradient structure obtained by selective distribution of MWCNTs into hard domains of thermoplastic polyurethane
Wang Xiaohan, Zou Fangfang, Zhao Yishen, Li Guangxian, Liao Xia
doi:10.1016/j.compositesa.2023.107861
电磁干扰屏蔽复合材料,以及通过在热塑性聚氨酯硬域中选择性分布 MWCNT 而获得的梯度结构泡沫
In this paper, multilayer thermoplastic polyurethane (TPU)/multiwalled carbon nanotubes (MWCNTs) electromagnetic interference shielding composite foams with gradient structure was prepared. The gradient distribution of effective concentration of filler and cell size were realized by selective distribution of MWCNTs into hard domains of TPU, which improved interlayer interface polarization of electromagnetic waves and impedance matching between the material and the air. The average electromagnetic interference shielding efficiency (EMI SE) of TPU/MWCNTs composites with gradient structure is 1.2 times larger than that of homogeneous composites. After foaming, the average EMI SE of the gradient foams was higher than that of the homogeneous foams, with maximum average EMI SE of 35.4 dB. This work is the first time to correlate the interaction of fillers with the soft domains and hard domains of TPU and EMI shielding performance, providing a feasible method for designing lightweight composites with low filler and better EMI shielding performance.
本文制备了具有梯度结构的多层热塑性聚氨酯(TPU)/多壁碳纳米管(MWCNTs)电磁干扰屏蔽复合泡沫。通过将 MWCNTs 选择性地分布在 TPU 的硬域中,实现了填料有效浓度和单元尺寸的梯度分布,从而改善了电磁波的层间界面极化以及材料与空气之间的阻抗匹配。梯度结构 TPU/MWCNTs 复合材料的平均电磁干扰屏蔽效率(EMI SE)是均质复合材料的 1.2 倍。发泡后,梯度泡沫的平均 EMI SE 高于均质泡沫,最大平均 EMI SE 为 35.4 dB。这项研究首次将填料与热塑性聚氨酯软域和硬域的相互作用与 EMI 屏蔽性能联系起来,为设计低填料、EMI 屏蔽性能更好的轻质复合材料提供了可行的方法。
Multifunctional basalt fiber polymer composites enabled by carbon nanotubes and graphene
Balaji K.V., Shirvanimoghaddam Kamyar, Naebe Minoo
doi:10.1016/j.compositesb.2023.111070
利用碳纳米管和石墨烯实现玄武岩纤维聚合物复合材料的多功能性
Basalt fiber (BF) is an eco-friendly fiber that can mitigate environmental footprint by enabling lightweight composite systems and components. When combined with carbonaceous structures such as carbon nanotubes (CNT) and graphene, these fibers can form multi-scale composites with remarkable potential for creating smart composites with added functionalities. In this paper, we review various fiber treatment methods used for nanomaterials, which include chemical vapor deposition (CVD), electrophoretic deposition, sizing, dipping, and chemical grafting. Compared to matrix modification, fiber treatment methods are more efficient in facilitating better load transfer between the matrix resin and fiber reinforcement through a nanomaterial bridge. Custom sizing with nanomaterials and CVD processes without a catalyst have been found the most effective methods for immobilizing nanomaterials onto the fibers. This advancement sets the stage for a new generation of sustainable and functional polymer composites that can support a circular economy.
玄武岩纤维(BF)是一种生态友好型纤维,可实现复合材料系统和组件的轻量化,从而减少对环境的影响。当与碳纳米管(CNT)和石墨烯等碳质结构相结合时,这些纤维可以形成多尺度复合材料,在创造具有附加功能的智能复合材料方面具有显著的潜力。本文综述了用于纳米材料的各种纤维处理方法,包括化学气相沉积(CVD)、电泳沉积、上浆、浸渍和化学接枝。与基体改性相比,纤维处理方法能更有效地通过纳米材料桥在基体树脂和纤维增强材料之间更好地传递载荷。使用纳米材料定制尺寸和不使用催化剂的 CVD 工艺是将纳米材料固定在纤维上的最有效方法。这一进步为新一代可支持循环经济的可持续功能性聚合物复合材料奠定了基础。
Super-tough, super-elastic, temperature-responsive, and tunable viscoelastic elastomer enabled by embedding nanosized liquid metal droplets
Li Sai, Zhao Hengheng, Liu Minghui, Zeng Xiaofei, Wei Yuan, Zhang Ganggang, Liu Jun, Zhang Liqun
doi:10.1016/j.compscitech.2023.110311
通过嵌入纳米级液态金属液滴实现超强韧性、超强弹性、温度响应性和可调粘弹性弹性体
Liquid metal (LM) based composites are playing irreplaceable roles in many emerging fields such as stretchable and wearable electronics, soft robotics. However, it is still challenging to facilely fabricate the LM-based elastomers with nanosized and well-dispersed LM domains. Herein, the LM droplets filled elastomer nanocomposites (ENCs) with temperature-responsive, super-tough, super-elastic and tunable viscoelastic properties are introduced. By employing the conventional rubber processing method, LM is fragmented into nanoscale droplets and dispersed uniformly in cross-linked natural rubber (NR) without compromising the soft and highly stretchable properties of the matrix. In addition to the remarkable enhancement in tear resistance, the toughness of the resulting composites is strikingly improved as lowering the applied temperature, which is attributed to the phase transition and the simultaneous volume expansion of LM droplets. Surprisingly, for the viscoelasticity, this LM-based ENCs exhibit almost the same dynamic hysteresis with the pure NR system at the service condition of automobile tires, which is remarkably reduced compared to the traditional ENCs filled with rigid nanoparticles. Furthermore, this material also shows a good damping property for noise attenuation in the case of submarine covering. Collectively, this work opens a new avenue for the next generation of high-performance and multifunctional ENCs equipped in low-temperature working conditions.
基于液态金属(LM)的复合材料在许多新兴领域发挥着不可替代的作用,例如可拉伸和可穿戴电子设备、软机器人技术等。然而,如何方便地制造具有纳米尺寸和良好分散的液态金属畴的液态金属基弹性体仍是一项挑战。本文介绍了具有温度响应、超韧性、超弹性和可调粘弹性能的 LM 液滴填充弹性体纳米复合材料(ENCs)。通过采用传统的橡胶加工方法,LM 被破碎成纳米级液滴,并均匀地分散在交联天然橡胶(NR)中,而不会影响基体的柔软和高拉伸性能。除了抗撕裂性显著增强外,随着应用温度的降低,所得复合材料的韧性也显著提高,这归因于 LM 液滴的相变和同时的体积膨胀。令人惊讶的是,在粘弹性方面,这种基于 LM 的 ENC 在汽车轮胎的使用条件下表现出与纯 NR 系统几乎相同的动态滞后,与填充了刚性纳米颗粒的传统 ENC 相比明显减少。此外,这种材料还显示出良好的阻尼特性,可用于海底覆盖层的噪声衰减。总之,这项工作为在低温工作条件下装备下一代高性能、多功能 ENC 开辟了一条新途径。