Designing Composites with Target Effective Young’s Modulus using Reinforcement Learning

Designing Composites with Target Effective Young’s Modulus using Reinforcement Learning
复制标题

使用强化学习设计具有目标有效杨氏模量的复合材料

DOI:
10.1145/3485114.3485123
复制
发表时间:
2021
期刊:
ACM Symposium on Computational Fabrication
影响因子:
--
通讯作者:
Whiting, Emily
Whiting, Emily
中科院分区:
--
文献类型:
--
作者:
Gongora, Aldair E.;Mysore, Siddharth;Li, Beichen;Shou, Wan;Matusik, Wojciech;Morgan, Elise F.;Brown, Keith A.;Whiting, Emily

文献摘要

参考文献

被引文献

相似文献

增材制造的进步已经实现了以前无法实现的材料和结构的设计和制造。特别是,复合材料和结构的设计空间大大扩展,由此产生的尺寸和复杂性对传统的设计方法提出了挑战,例如蛮力探索和一次一因素(OFAT)探索,以找到最佳或定制的设计。为了应对这一挑战,监督机器学习方法已经出现,使用策划的训练数据对设计空间进行建模;然而,训练数据的选择通常由用户决定。在这项工作中,我们开发并利用了一个基于强化学习(RL)的框架来设计复合材料结构,从而避免了对用户选择的训练数据的需求。对于一个由柔软和柔顺的组成材料块组成的5 × 5复合设计空间,我们发现使用这种方法,模型可以使用由225种设计可能性组成的总设计空间的2.78%进行训练。此外,开发的基于RL的框架能够以超过90%的成功率找到设计。这种方法的成功激励未来的学习框架利用RL来设计复合材料和其他材料系统。
Advancements in additive manufacturing have enabled design and fabrication of materials and structures not previously realizable. In particular, the design space of composite materials and structures has vastly expanded, and the resulting size and complexity has challenged traditional design methodologies, such as brute force exploration and one factor at a time (OFAT) exploration, to find optimum or tailored designs. To address this challenge, supervised machine learning approaches have emerged to model the design space using curated training data; however, the selection of the training data is often determined by the user. In this work, we develop and utilize a Reinforcement learning (RL)-based framework for the design of composite structures which avoids the need for user-selected training data. For a 5 × 5 composite design space comprised of soft and compliant blocks of constituent material, we find that using this approach, the model can be trained using 2.78% of the total design space consists of 225 design possibilities. Additionally, the developed RL-based framework is capable of finding designs at a success rate exceeding 90%. The success of this approach motivates future learning frameworks to utilize RL for the design of composites and other material systems.
DOI: 10.1016/j.isci.2021.102262
发表时间: 2021-04-23
期刊: iScience
影响因子: 5.8
作者:
Gongora AE;Snapp KL;Whiting E;Riley P;Reyes KG;Morgan EF;Brown KA
通讯作者: Brown KA
PuzzleFlex:具有松动接头的链条的运动学
DOI: 10.1109/icra40945.2020.9196854
发表时间: 2020
期刊: International Conference on Robotics and Automation
影响因子: --
作者:
Lensgraf, Samuel;Itani, Karim;Zhang, Yinan;Sun, Zezhou;Wu, Yijia;Li, Alberto Quattrini;Zhu, Bo;Whiting, Emily;Wang, Weifu;Balkcom, Devin
通讯作者: Balkcom, Devin
用于雕塑稳定性分析和保护的增强现实
DOI: 10.2312/gch.20201298
发表时间: 2020
期刊: Eurographics Workshop on Graphics and Cultural Heritage
影响因子: --
作者:
Henneman, Dennis;Li, Yichen;Ochsendorf, John;Betke, Margrit;Whiting, Emily
通讯作者: Whiting, Emily
DOI: 10.1126/sciadv.abb6987
发表时间: 2020-09-01
期刊: SCIENCE ADVANCES
影响因子: 13.6
作者:
Leinen, Philipp;Esders, Malte;Tautz, F. Stefan
通讯作者: Tautz, F. Stefan
FOLDLINGS:交互式弹出卡设计的可视化工具
DOI: --
发表时间: 2015
期刊:
影响因子: --
作者:
Marissa Allen
通讯作者: Marissa Allen