.png){kind=icon}
返回
Digital Material Design Using Tensor-Based Error Diffusion for Additive Manufacturing
.png){kind=icon}
OA
AI
DOI：10.1016/j.cad.2019.05.031.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Recent multi-material additive manufacturing (AM) technologies enable the fabrication of an object with accurate deposition of different types of materials. Hence, in addition to geometric shapes, it is possible to use different material compositions to optimize the mechanical properties of a component for given design requirements. However, current AM processes have a limitation on the number of materials that can be deposited during the fabrication process. Due to the constraint, it is critical to optimize the material distribution using the limited base materials; however, an extremely large design space exists in a design domain that is enabled by the AM technologies. In this paper, we introduce a digital material design framework to generate digital material compositions that can be printed and be able to achieve the desired behavior. We take analog material composition as the input and perform the analog-to-digital conversion using an exemplar-based approach based on a pre-computed material library. The patterns in the library are constructed with different combinations of the given base materials, and their mechanical properties are computed using finite element simulation. Accordingly, the design goal of the analog-to-digital conversion is to find material composition in the design domain with matching mechanical properties. A tensor-based error diffusion algorithm has been developed to reduce the approximation error during the conversion effectively. Experimental tests based on the design framework have been performed. The test results demonstrate that our framework can quickly find effective solutions for various multi-material design problems. (C) 2019 Elsevier Ltd. All rights reserved.
Keyword:
Additive manufacturing
Digital material design
Error diffusion
Tensor
Robotic grasper
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
C
IF:
3.1
论文数:
3.1K
被引数:
6.4K