Development, Implementation and Evaluation of a Prototype System for Data-Driven Optimization of a Preforming Process

Abstract Modern production of fiber reinforced composites via the preforming process is widely used in industry. A common way to create dry, semi-finished products forming or draping a textile into three-dimensional component geometry. The punch and die often for resin transfer molding (RTM) composite manufacturing. Due major influence step on later mechanical performance component, detailed knowledge architecture beneficial. To enable in-situ monitoring specific deformation woven fabric, novel kind single-use two-dimensional strain sensors has already been developed characterized. We show that by using industrial communication standards, data from various sources can be consolidated an edge computer improve process. this end, we hardware firmware device reads out printed transfers IO-Link. In addition, collects programmable logic controller capable connecting further IO-Link sensors. Our demonstrator intended as proof concept monitoring, data-driven analysis improvement will developed. propose machine learning-based analytics approach detecting defects increasing quality during Forming tests with double-dome benchmark geometry carbon fabric which suitable industry have carried validate our prototype system.