Wolfson School of Mechanical, Electrical and Manufacturing Engineering, Loughborough University, Loughborough LE11 3TU, UK.
The Nonwovens Institute, North Carolina State University, 1010 Main Campus Dr, Raleigh, NC 27606, USA.
Micron. 2022 Sep;160:103321. doi: 10.1016/j.micron.2022.103321. Epub 2022 Jun 27.
Quantitative analysis of fibre orientation in a random fibrous network (RFN) is important to understand their microstructure, properties and performance. 2D fibre orientation distribution presents an in-plane fibre orientation without any information on fibre orientation in thickness direction. This research introduces a fully parametric algorithm for computing 3D fibre orientation as thickness is important for high-density or thick fibrous networks. The algorithm is tested for 3 major classes of nonwoven fabrics called low- (L), medium- (M) and high-density (H) ones. H fabric density is 6-8 times larger than the L fabric density. M fabric density (traditional intermediate fabric density) is 3-4 times larger than the L fabric density. Voxel models of experimental nonwoven webs were generated by an X-ray micro-CT (µCT) system and evaluated with the algorithm. Statistical results showed that a fraction of fibres orientated along the thickness direction increases as fibre density grows. To validate the accuracy of findings, deterministic voxelated virtual fibrous structures, created using mathematical functions were used. This novel algorithm is able to produce a 3D orientation distribution function (ODF) for any RFN including, models of nonwovens produced with various manufacturing parameters, experimentally verified and validated with X-ray µCT. Also, it can compute 2D ODFs of various types of RFNs to evaluate 2D behaviour of fibrous structures. The obtained results are useful for applications in many fields including finite element analysis, computational fluid dynamics, additive manufacturing, etc.
定量分析随机纤维网络（RFN）中的纤维取向对于理解其微观结构、性质和性能非常重要。二维纤维取向分布呈现出平面内纤维取向，而没有任何关于厚度方向纤维取向的信息。本研究介绍了一种完全参数化的算法，用于计算 3D 纤维取向，因为厚度对于高密度或厚纤维网络很重要。该算法经过了低（L）、中（M）和高密度（H）三种非织造布的测试。H 织物的密度比 L 织物的密度大 6-8 倍。M 织物的密度（传统的中间织物密度）比 L 织物的密度大 3-4 倍。实验性非织造织物的体素模型是通过 X 射线微计算机断层扫描（µCT）系统生成的，并使用该算法进行了评估。统计结果表明，随着纤维密度的增加，沿厚度方向取向的纤维分数增加。为了验证研究结果的准确性，使用了使用数学函数创建的确定性体素化虚拟纤维结构。该新算法能够为任何 RFN 生成 3D 取向分布函数（ODF），包括使用各种制造参数生产的非织造布模型，并通过 X 射线 µCT 进行了实验验证和验证。此外，它还可以计算各种类型 RFN 的 2D ODF，以评估纤维结构的 2D 行为。所得结果可用于许多领域的应用，包括有限元分析、计算流体动力学、增材制造等。
