Deformation and failure mechanisms of metallic additively manufactured triply periodic minimum surface structures

在工业中，拓扑优化的轻质细胞结构的优势使用通常需要巧妙和灵活的生产技术，例如增材制造（AM）。增材制造允许对几何形状进行优化，例如，相对于减少组件的重量。然而，在微观（单杆）、细观（单胞）和宏观（整个晶格结构）尺度上，由于自然引入的缺陷，有时会降低其机械稳定性和性能。因此，有必要很好地了解这种复杂结构在所有三个长度尺度上的变形和破坏机制。本项目旨在了解金属AM三周期最小表面（TPMS）结构的变形和破坏机制。我们的方法是基于高级特征、数据驱动和基于模型的方法的组合。一方面，我们的目的是利用实验数据（即前x射线和原位x射线计算机断层扫描，XCT）来研究这些结构的几何特征和力学行为。另一方面，这些知识将被推广使用模拟模型来预测未经测试的TPMS结构的力学行为（刚度，承载能力）。先进的贝叶斯技术结合几何缺陷和本构参数的随机场表示，用于确定系统的几何制造缺陷以及模型参数的空间分布，包括不确定性量化以预测模型预测的准确性。原位XCT成像和实验数据分析方法的优化和进一步发展，是直接利用数字体积相关获得的完整变形场的改进模型标定技术应用的前提。造成TMPS结构变形和损伤的机制有几种，一些与原材料（微观结构）有关，一些与制造参数（微观和介观效应，例如制造缺陷）有关，还有一些与几何特征（名义几何形状，壁厚偏差）有关。在项目中，我们将对这些影响进行分析：我们将识别几何缺陷，随后分析材料行为（例如（an-）各向同性）。为了获得现实和全面的见解，将研究两种加载场景：压缩（广泛应用于AM细胞结构）和扭转（新颖）。作为一项最终成就，变形和破坏机制的识别将允许指定最相关的设计和制造参数，有助于机械行为，并通过直接包含制造缺陷的仿真模型直接预测机械性能及其不确定性。

The advantageous use of topologically optimized, light-weight cellular structures in industry often requires cunning and flexible production technologies, such as additive manufacturing (AM). AM allows for an optimization of the geometry e.g. with respect to a reduced weight of the component. However, it sometimes possesses reduced mechanical stability and performance caused by naturally introduces defects at the microscopic (single strut), mesoscopic (unit cell) and macroscopic (whole lattice structure) scale. It is therefore imperative to well understand the deformation and failure mechanisms of such complex structures at all of the three length scales. The proposed project aims at understanding the deformation and failure mechanisms of metallic AM Triply Periodic Minimum Surface (TPMS) structures. Our approach is based on a combination of advanced characterization, data-driven, and model-based methodologies. On the one hand, we aim to investigate the geometric features and mechanical behavior of these structures using experimental data (namely ex- and in-situ X-ray Computed Tomography, XCT). On the other hand, this knowledge will be generalized using simulation models to predict the mechanical behavior (stiffness, load bearing capacity) of untested TPMS structures. Advanced Bayesian techniques combined with a random field representation of geometrical defects and constitutive parameters are used to determine both systematic manufacturing deficiencies in the geometry as well as the spatial distribution of model parameters including an uncertainty quantification to predict the accuracy of the model prediction. The optimization and further development of in-situ XCT imaging and experimental data analysis methods are a precondition for the application of improved model calibration techniques that directly use the complete deformation field obtained from the digital volume correlation. There are a few mechanisms contributing to the deformation and damage of TMPS structures, some linked to the raw material (microstructure), some to the manufacturing parameters (micro- and mesoscopic effects, e.g. manufacturing defects), and some to the geometrical features (nominal geometry, wall thickness deviation). In the project, those effects are analyzed: we will identify geometrical defects and subsequently analyze the material behavior (e.g. (an-)isotropy). To derive realistic and comprehensive insights, two loading scenarios will be studied: compression (widely applied for AM cellular structures) and torsion (novel). As an ultimate achievement, the identification the deformation and failure mechanisms will allow specifying the most relevant design and manufacturing parameter contributing to the mechanical behavior and predicting the mechanical performance and its uncertainty directly by a simulation model directly incorporating manufacturing defects.