Development and validation of a methodology for in-situ detection and model-based evaluation of manufacturing defects in electron beam powder bed fusion

增材制造工艺（例如电子束粉末床熔合（PBF-EB））的一个显著优点是可以在没有工具的情况下制造形状复杂的部件。然而，由于位置相关的工艺条件，在制造的部件中会出现局部缺陷，目前只能通过复杂的无损检测来检测。同时，缺陷对形状复杂部件性能的影响很大程度上取决于其特征（位置、尺寸、形状、方向），而目前对此的评价标准还不够。为了克服这些限制，该项目的目的是开发和验证一种方法，使电子光学（ELO）原位成像的手段，可靠地检测和表征PBF EB中的电子诱导缺陷。此外，它们对局部力学性能的定量影响应基于模型进行评估。这将为评估单个缺陷提供可靠的标准，并消除对组件进行额外非破坏性测试的需要。项目第一阶段的目标之一是开发一种算法，用于从增材制造过程的ELO信号数据中定性检测错误连接，以便在生产过程中可以可靠地检测到这些错误连接。随后将通过X射线计算机断层扫描检测缺陷的特征，并与所研究的试样的力学参数一起用于验证和进一步开发基于缺陷和微观结构的模型。研究将使用钛铝化物（TiAl）的材料系统进行，由于其低缺陷容限和线性弹性材料行为，该材料系统对局部缺陷特别敏感。通过使用已针对PBF-EB进行优化且迄今为止很少进行表征的合金BMBF 3，还将生成新的材料数据，这对于这种有前途的材料的适用性是必要的。在这种情况下，在这两个研究地点进行的广泛分析将使人们能够深入了解工艺参数、微观结构和缺陷形成以及近应用条件下的机械性能之间的关系，并将研究结果转移到其他增材制造合金中。为此，所有测试结果，包括工艺参数、微观结构和缺陷分析、机械性能（准静态和循环材料行为）和断口分析，都将合并到每个试样的数字孪生中，并用于综合分析。

A significant advantage of additive manufacturing processes such as electron beam powder bed fusion (PBF-EB) is the possibility of manufacturing complex-shaped components without tools. However, due to the location-dependent process conditions, local defects occur in the manufactured components, which can currently only be detected by complex, non-destructive testing. At the same time, the influence of a defect on the performance of a complexly shaped component is strongly dependent on its characteristics (position, size, shape, orientation), for which only insufficient evaluation criteria currently exist. To overcome these limitations, this project aims to develop and validate a methodology that allows fracture-inducing defects in PBF EB to be reliably detected and characterized by means of electron optical (ELO) in situ imaging. Additionally, their quantitative influence on the local mechanical performance should be model-based evaluated. This should provide a solid criterion for evaluating individual defects and eliminate the need for additional non-destructive testing of the components. One goal of the first project phase is the development of an algorithm for the qualitative detection of misconnections from ELO signal data of the additive manufacturing process so that these can already be reliably detected during production. The characteristics of the defects will be detected subsequently by means of X-ray computed tomography and, together with the mechanical parameters of the investigated test specimens, will be used for validation and further development of defect- and microstructure-based models. The investigation will be carried out using the material system of titanium aluminides (TiAl), which is particularly sensitive to local defects due to their low defect tolerance and linear-elastic material behavior. By using the alloy BMBF3, which has been optimized for PBF-EB and has rarely been characterized so far, new material data will also be generated, which are necessary for the applicability of this promising material. In this context, extensive analyses at both research locations will enable an in-depth understanding of the relationships between process parameters, microstructure and defect formation as well as mechanical properties under near-application conditions and enable the transfer of the research results to other additively manufactured alloys. For this purpose, all test results, including process parameters, microstructure and defect analyses, mechanical properties (quasi-static and cyclic material behavior), and fractographic analyses, are combined in a digital twin of each specimen and used for the comprehensive analyses.