Process-induced defects are the leading cause of discrepancies between as-designed and as-manufactured additive manufacturing (AM) product behavior. Especially for metal lattices, the variations in the printed geometry cannot be neglected. Therefore, the evaluation of the influence of microstructural variability on their mechanical behavior is crucial for the quality assessment of the produced structures. Commonly, the as-manufactured geometry can be obtained by computed tomography (CT). However, to incorporate all process-induced defects into the numerical analysis is often computationally demanding. Thus, commonly this task is limited to a predefined set of considered variations, such as strut size or strut diameter. In this work, a CT-based binary random field is proposed to generate statistically equivalent geometries of periodic metal lattices. The proposed random field model in combination with the Finite Cell Method (FCM), an immersed boundary method, allows to efficiently evaluate the influence of the underlying microstructure on the variability of the mechanical behavior of AM products. Numerical analysis of two lattices manufactured at different scales shows an excellent agreement with experimental data. Furthermore, it provides a unique insight into the effects of the process on the occurring geometrical variations and final mechanical behavior.

工艺引起的缺陷是设计和制造的增材制造 (AM) 产品行为之间差异的主要原因。特别是对于金属晶格，印刷几何形状的变化不可忽视。因此，评估微观结构变异性对其力学行为的影响对于所生产结构的质量评估至关重要。通常，可以通过计算机断层扫描 (CT) 获得制造时的几何形状。然而，将所有过程引起的缺陷纳入数值分析通常需要计算量。因此，通常这个任务仅限于一组预定义的考虑变化，例如 strut尺寸或支柱直径。在这项工作中，提出了基于 CT 的二进制随机场来生成统计上等效的周期性金属晶格几何。所提出的随机场模型与有限单元法 (FCM)（一种浸入式边界方法）相结合，可以有效地评估底层微观结构对 AM 产品机械行为可变性的影响。以不同尺度制造的两种晶格的数值分析显示与实验数据非常吻合。此外，它提供了对过程对发生的几何变化和最终机械行为的影响的独特见解。