Outil prédictif pour la conception et la fabrication de préformes textiles pour composites appliqués à l'aérospatiale

该项目的重点是设计用于液相成型（LCM）制造非结构航空复合材料零件的多层纺织品预成形件。具体而言，该项目侧重于设计制造这些预成型（“预成型”）的工业方法。目标是开发一种数字工具，使工业合作伙伴能够快速建立用于给定几何形状预成型的设计，制造步骤和纺织品增强。该工具将在许多场景中指导这些选择，并确定最佳选项。数字化工具必须1)建立简化和加速制造的预成形设计；2)识别制造步骤及其顺序，模式，锚的位置，并通过“胶合板”向技术人员清楚地传达所有内容；3)选择最适合作品几何形状的特定挑战的面料；4)提出一种方案，以确保在较短的时间内预制体的高质量和低可变性-质量将通过在树脂注入制造复合材料之前检查干燥的预制体来评估。任何复杂的零件几何形状都可以简化为一组简单的、参数化的基本几何形状，如平面、单曲率和双曲率曲面以及扭转曲面。该项目的核心策略是通过建立两个独立于复杂几何形状的数据库进行分析，将悬垂模拟和组织表征转移到工业合作伙伴收到的给定复杂几何形状的上游。该策略的核心是在接收要分析的几何形状之前完成大部分模拟和表征工作，而不是之后；数据以前存储在两个数据库中，以一种允许它们快速用于分析任何几何块的形式，由客户提供。合作伙伴在竞争激烈的国际环境中工作：预制件的设计时间、制造时间和质量至关重要，创建报价所需的时间也是如此。因此，数字工具必须快速且易于使用。

The project focuses on the design of multilayer textile preforms used in the manufacture of non-structural aeronautical composite parts by liquid-phase molding (LCM). Specifically, the project focuses on the design of industrial methods of manufacturing these preforms ('preforming'). The goal is to develop a digital tool that allows the industrial partner to quickly establish the design, manufacturing steps and textile reinforcement used for the preform of a given geometry. The tool will guide these choices among many scenarios, and identify the best option. The digital tool must 1) establish a preform design that will simplify and speed up manufacturing; 2) identify the manufacturing steps and their sequence, the patterns, the position of the anchors, and communicate everything to the technicians clearly via a 'plybook'; 3) choose the fabric best suited to the specific challenges posed by the geometry of the piece; and 4) propose a scenario that ensures high quality and low variability of preforms in a shorter time - quality will be assessed by inspecting the preforms while dry before the resin is infused to create a composite material. Any complex part geometry can be reduced to a set of simple, parametric basic geometries such as planes, single and double curvature surfaces, and torsion surfaces. The central element of the proposed strategy for this project is to move draping simulations and tissue characterization upstream of the industrial partner's receipt of a given complex geometry, through the construction of two databases independent of the complex geometry to analyze. The central point of the strategy is to do most of the simulation and characterization work before receiving a geometry to be analyzed - and not after; the data is previously stored in two databases, in a form that allows them to be used quickly for any piece of geometry to be analyzed, provided by a client. The partner works in a very competitive international context: the design time, the manufacturing time and the quality of the preforms are critical, as is the time required to create a quote. So, the digital tool must be fast and easy to use.