ABSTRACT Additive manufacturing (AM) unlocks novel industrial possibilities in relation to design optimisation for lightweight structures, e.g. in aerospace applications. However, the inherent geometric complexity of topology-optimised AM components represents a major challenge for conventional non-destructive testing (NDT) methods. Due to its flexibility and high throughput, industrial X-ray micro-computed tomography (XCT) is the most promising NDT method for AM. In this contribution, we investigate topology-optimised engine brackets that were manufactured from AlSi10 Mg using selective laser melting (SLM). We investigate the respective parts and in-process test coupons in a multiscale approach to be able to extract pore size distributions at different spatial resolutions between 105 and 1.25 µm isometric voxel size. At the lowest spatial resolutions, existing pores cannot be segmented. In contrast, decreasing voxel sizes leads to an increase in total porosity up to 1.53%. Defects like pores in load-carrying areas can profoundly influence the component´s mechanical performance; hence, extensive NDT investigations are mandatory to predict the effect of defects in aluminium AM components.

中文翻译：

---

拓扑优化铝部件的增材制造和无损检测

摘要 增材制造 (AM) 开启了与轻型结构设计优化相关的新型工业可能性，例如在航空航天应用中。然而，拓扑优化的 AM 组件固有的几何复杂性代表了传统无损检测 (NDT) 方法的主要挑战。由于其灵活性和高通量，工业 X 射线微计算机断层扫描 (XCT) 是最有前途的 AM 无损检测方法。在这篇文章中，我们研究了使用选择性激光熔化 (SLM) 由 AlSi10 Mg 制造的拓扑优化发动机支架。我们以多尺度方法研究各个零件和过程中的测试试样，以便能够在 105 和 1.25 µm 等距体素尺寸之间的不同空间分辨率下提取孔径分布。在最低空间分辨率下，无法分割现有孔隙。相比之下，减小体素尺寸会导致总孔隙率增加高达 1.53%。承载区域中的孔隙等缺陷会严重影响部件的机械性能；因此，必须进行广泛的 NDT 调查来预测铝 AM 部件中缺陷的影响。