The Development of Data-Driven Algorithms and Models for Monitoring Void Transport in Liquid Composite Molding Using a 3D-Printed Porous Media

In Liquid Composite Molding (LCM), the high variability present in reinforcement properties such as permeability creates additional challenges during the injection process, such as void formation. Although improved injection strategy designs can mitigate the formation of defects, these processes can benefit from real-time process monitoring and control to adapt the injection conditions when needed. In this study, a machine vision algorithm is proposed, with the objective of detecting and tracking both fluid flow and bubbles in an LCM setup. In this preliminary design, 3D-printed porous geometries are used to mimic the architecture of textile reinforcements. The results confirm the applicability of the proposed approach, as the detection and tracking of the objects of interest is possible, without the need to incur in elaborate experimental preparations, such as coloring the fluid to increase contrast, or complex lighting conditions. Additionally, the proposed approach allowed for the formulation of a new dimensionless number to characterize bubble transport efficiency, offering a quantitative metric for evaluating void transport dynamics. This research underscores the potential of data-driven approaches in addressing manufacturing challenges in LCM by reducing the overall process monitoring complexity, as well as using the acquired reliable data to develop robust, data-driven models that offer new understanding of process dynamics and contribute to improving manufacturing efficiency.