Using simulation to accelerate autonomous experimentation: A case study using mechanics.
Using simulation to accelerate autonomous experimentation: A case study using mechanics.
复制标题
DOI：
10.1016/j.isci.2021.102262
复制
发表时间：
2021-04-23
期刊：
影响因子：
5.8
通讯作者：
Brown KA
中科院分区：
文献类型：
--
作者：
Gongora AE;Snapp KL;Whiting E;Riley P;Reyes KG;Morgan EF;Brown KA
自主实验（AE）通过将自动化和机器学习结合到以顺序的方式智能和快速地进行实验来加速研究。虽然AE系统最需要研究无法在分析或计算上预测的属性，但原则上即使是不完善的预测也可能有用。在这里，我们调查了来自模拟的不完美数据是否可以使用有关添加性生产结构的机制的案例研究加速AE。最初，我们研究了弹性，这是一种通过有限元分析（FEA）进行良好预测的属性，发现FEA可用于构建贝叶斯先验和实验数据，并可以使用差异建模来整合实验，以减少所需的实验数量10 -折叠。接下来，我们研究韧性，这是FEA不受欢迎的属性，发现FEA仍然可以通过转换实验数据和指导实验选择来改善学习。这些结果强调了通过转移学习改善AE的多种方法。
模拟和自主实验合并以加速研究
精确模拟的弹性在少10 x实验中学习了
模拟相关特性，即产量力，加速学习韧性
使用转移学习将模拟引入实验学习循环
机械性能；材料科学中的计算方法；材料科学中的模拟
Autonomous experimentation (AE) accelerates research by combining automation and machine learning to perform experiments intelligently and rapidly in a sequential fashion. While AE systems are most needed to study properties that cannot be predicted analytically or computationally, even imperfect predictions can in principle be useful. Here, we investigate whether imperfect data from simulation can accelerate AE using a case study on the mechanics of additively manufactured structures. Initially, we study resilience, a property that is well-predicted by finite element analysis (FEA), and find that FEA can be used to build a Bayesian prior and experimental data can be integrated using discrepancy modeling to reduce the number of needed experiments ten-fold. Next, we study toughness, a property not well-predicted by FEA and find that FEA can still improve learning by transforming experimental data and guiding experiment selection. These results highlight multiple ways that simulation can improve AE through transfer learning. Simulation and autonomous experimentation were combined to accelerate research Resilience, which was accurately simulated, was learned in 10 x fewer experiments Simulating related properties, i.e. yield force, accelerated learning toughness Simulation was introduced to an experimental learning loop using transfer learning Mechanical Property; Computational Method in Materials Science; Simulation in Materials Science
登录
查看更多内容
影响因子：
29.4
作者：
Epps, Robert W.;Bowen, Michael S.;Abolhasani, Milad
通讯作者：
Abolhasani, Milad
影响因子：
56.9
作者：
Coley, Connor W.;Thomas, Dale A., III;Jensen, Klavs F.
通讯作者：
Jensen, Klavs F.
影响因子：
16.6
作者：
Porwol, Luzian;Kowalski, Daniel J.;Cronin, Leroy
通讯作者：
Cronin, Leroy
影响因子：
2.9
作者：
Li, Jiagen;Lu, Yihua;Zhu, Xi
通讯作者：
Zhu, Xi
影响因子：
4.6
作者：
Bowden, Gregory D.;Pichler, Bernd J.;Maurer, Andreas
通讯作者：
Maurer, Andreas