Bahoz Abbas

AutoHD—Automated Handling and Draping of Reinforcing Textiles

In almost all industrial sectors handling processes are automated through the use of robotic systems. However, in the manufacture of fiber-reinforced structures with complex geometries, the handling of dry, pre-impregnated semi-finished textiles is still performed mainly manually resulting in long processing times, low reproducibility and high manufacturing costs. A previous AiF research project “AutoPreforms” aimed at the automation of the entire production process of components with uniaxial curvature. The scope of this AiF research project “AutoHD” is to fully automate the draping and handling process of complex, three-dimensional fiber composite structures with high degrees of deformation and multiaxial curvature (e.g. car wings). Based on a draping simulation wrinkles can already be recognized during the draping process and counteracted by the developed mechanical structure. This is achieved by the utilization of the bending stiffness of textile semi-finished products, a flexible end-effector and a built-in optical quality assurance process. In this paper the main aspects of preforming processes are described revealing the challenges of the project. With examples of currently existing systems, the objective and innovative contribution of the project are described. The paper serves as initial presentation of the project and its solution approaches.

In-Line Quality Control System for the Industrial Production of Multiaxial Non-crimp Fabrics

A new approach is presented in which existing systems for error detection are used to build up a closed loop control for quality optimization in the production process of NCF. Only with this online system, productivity and product quality can be increased. The fibre material is screened for fibre disorientation and gaps and deviations in areal weight at different positions alongside the production line using state of the art equipment for error detection. Additionally, the machine-parameters are monitored and the yarn tension of both reinforcement fibres and stitching yarn are measured continuously. An automated and self-learning control is developed that automatically adjusts the settings of the production machine. By this, a minimum of errors in the textile is achieved. In addition, the quantitative description generated in the process is used to develop an error map for each batch of material produced. This error map allows for production waste to be minimized, as the produced material can be automatically sorted into different quality grades, meeting the demands of different customers. Furthermore, the material cutting can be adjusted that only low quality areas are removed. By this, only high quality material will be used for the component part production. An entire roll of non-crimp fabric would not be declared as waste. The quality control is realized by an automatic control device. By a pressure roll in front of the warp knitting unit the rovings are homogenized.