材料科学
有限元法
失真(音乐)
残余应力
格子(音乐)
变形(气象学)
结构工程
机械
复合材料
声学
物理
工程类
放大器
光电子学
CMOS芯片
作者
Lisa Wiesent,Felix Stocker,Aida Nonn
出处
期刊:Materialia
[Elsevier BV]
日期:2023-04-26
卷期号:28: 101774-101774
被引量:1
标识
DOI:10.1016/j.mtla.2023.101774
摘要
Maintaining dimensional accuracy is a major challenge of laser powder bed fusion (L-PBF) preventing its application for more complex and filigree L-PBF structures in industrial practice. Previous studies have shown that residual stresses and distortion of benchmark L-PBF components may be predicted by sequential thermo-mechanical analyses. However, the reliability of these analyses for more complex structures must be critically questioned, as comprehensive validation and sensitivity analyses are scarce. In this paper, we present a calibrated and validated low-fidelity sequential thermo-mechanical finite element analysis (FEA) of a tubular L-PBF lattice structure, i.e., an aortic stent, where pronounced local deformation is expected. As a first step, the finite element model was extensively calibrated using experimental data to ensure reproducibility of the simulation results. Thereupon, geometric features critical to the distortion of L-PBF lattice structures and measures to compensate for the distortion, such as inversion of the distorted L-PBF structure, were investigated. It was found that the distortion of the L-PBF lattice structures can be reduced, but not completely prevented, by increasing the strut angles, increasing the strut thickness, and decreasing the transition radius in the area of merging struts. FEA-based inversion of the numerically predicted deformed structure minimized distortion, resulting in the L-PBF aortic stent approximating the intended CAD geometry even with a small strut thickness. This work shows that low-fidelity sequential thermo-mechanical FEA can be used not only for the analysis and deformation compensation of reference structures, but also for the analysis of more complex filigree structures with pronounced local deformation.
科研通智能强力驱动
Strongly Powered by AbleSci AI