Fundamentals of process design for dimensionally accurate roll forming of asymmetrical profile geometries

滚压成形是冷轧型材的连续制造工艺，在汽车和家具行业中也有应用。与不连续弯曲方法（如模具弯曲）相反，金属板材是在三维成形过程中成形的。除了截面弯曲外，还会发生与工艺相关的变形，从而导致型材缺陷。特别是非对称型材经常遭受扭转和多维弯曲。目前，在工艺设计中使用有限元模拟是检测轮廓缺陷的唯一途径。由于计算时间较长，限制了有限元模拟的应用。此外，工艺参数与轮廓缺陷之间的相互作用尚未完全了解，因此工艺设计主要是经验性的，由此产生的轮廓缺陷必须在耗时的校正过程中进行补偿。该项目旨在预测、分析和减少不对称帽型辊压成形后的缺陷。不对称的帽型最适合，因为它可以表示具有多个弯曲区域的复杂轮廓几何形状。在项目中，流程设计将从两个方面进行改进。一方面，建立了不需要有限元模拟的非对称帽型扭转预测的解析模型，作为数值优化的初始值。另一方面，对非对称帽型辊压成形进行了数值和实验研究，以表征和分析与工艺相关的附加变形、由此产生的型材缺陷以及辊压成形过程中工艺参数的影响。在此基础上，推导出了有效的应对措施，并进行了数值研究。在此基础上，利用优化算法结合有限元模拟对最有效的对策进行了优化。因此，针对轮廓内应变的平衡状态优化了对策的参数。最后，对对策和优化算法进行了实验验证。通过加深对复杂轮廓几何形状的轮廓缺陷产生的理解，开发有效的应对措施，以及开发在设计早期预测扭转的分析模型，该项目有助于更准确和有效的工艺设计，旨在显着缩短校正过程并提高生产率。

Roll forming is a continuous manufacturing process for cold rolled profiles, which sees use besides others in the automotive and furniture industry. In contrast to discontinuous bending methods, like die bending, the sheet metal is formed in a three-dimensional forming process. Besides the bending of the cross section additional process-related deformations happen which lead to profile defects. Especially asymmetrical profiles are always subjected to torsion and multi-dimensional bowing.Nowadays, the use of Finite-Element-Simulations during the process design is the only way to detect the profile defects. Due to the time intensive calculations the application of FEM-simulations is limited. Furthermore, the interactions between process parameters and profile defects are not fully understood, so the process design is mainly empirical and the resulting profile defects have to be compensated in time-consuming correction processes.The project aims at the prediction, analysis and reduction of the defects after roll forming of an asymmetrical hat profile. The asymmetrical hat profile is suited best, because it can represent complex profile geometries with several bending zones. Within the project the process design will be improved in two ways. On the one side, an analytical model for the prediction of torsion of asymmetrical hat profiles, without the necessity of FEM-simulations, is developed to be used as the initial value for the numerical optimization. On the other side, numerical and experimental investigations of roll forming of asymmetrical hat profiles are conducted to characterize and analyse the process-related additional deformations, the resulting profile defects as well as the influence of process parameters of the roll forming process. Based on the results, effective counter measures can be deduced and numerically investigated. Following the investigation, the counter measure found most effective, is optimized by an optimization algorithm coupled with the FEM-simulations. Thereby, the parameters of the counter measures are optimized with regard to a balanced state of strains within the profile. Finally, the counter measure and the optimization algorithm are validated experimentally.By deepening the understanding of the creation of profile defects of complex profile geometries, the developing of effective counter measures and the development of an analytical model to predict torsion in the early design stage, this project contributes to an more accurate and efficient process design with the aim to significantly shorten correction processes and to increase the production rate.