Analysis of the impregnation process during injection experiments in fibre-reinforced polymer composites has been carried out via a three-dimensional in-situ imaging method, namely Synchrotron Radiation Computed Laminography (SR-CL). A resin transfer moulding device was designed to perform injection experiments at different flow rates while, at the same time, the evolution of the flow front was inspected by SR-CL. Injections with flow rates of 50 μL/min and 300 μL/min in plain woven E-glass fabrics were carried out, resulting in small capillary number fluid propagation with capillary forces dominant over viscous ones. Intratow and intertow voids were accurately segmented by using a machine learning algorithm that enabled the direct measurement of their time and spatial distribution from the raw XCT volumes. The saturation evolution, void content, shape and spatial distribution quantification are reported. Intratow void formation and transport mechanisms were accurately described, in particular those related with large void splitting including the subsequent flow dragging. Advantages and limitations of the use of SR-CL for the analysis of the impregnation process of dry preforms in liquid moulding are discussed in detail, and in particular, those aspects related with the time acquisition limitations.

通过三维原位成像方法，即同步辐射计算机层析成像 (SR-CL) ，对纤维增强聚合物复合材料注射实验过程中的浸渍过程进行了分析。设计了一种树脂传递模塑装置，用于在不同流速下进行注射实验，同时，SR-CL 检查流动前沿的演变。流速为 50  μ L/min 和 300  μ 的进样L/min 在平织 E 玻璃织物中进行，导致小毛细管数流体传播，毛细管力优于粘性流体。Intratow 和 intertow 空隙是通过使用机器学习算法准确分割的，该算法能够从原始 XCT 体积直接测量它们的时间和空间分布。报告了饱和度演变、空隙含量、形状和空间分布量化。流内空隙的形成和传输机制被准确描述，特别是与大空隙分裂相关的那些，包括随后的流动拖曳。详细讨论了使用 SR-CL 分析液体成型中干燥预成型件的浸渍过程的优点和局限性，特别是与时间获取限制相关的那些方面。