Design Repository Effectiveness for 3D Convolutional Neural Networks: Application to Additive Manufacturing
Design Repository Effectiveness for 3D Convolutional Neural Networks: Application to Additive Manufacturing
复制标题
DOI：
10.1115/1.4044199
复制
发表时间：
2019-09
影响因子：
3.3
通讯作者：
Glen Williams;N. Meisel;T. Simpson;Christopher McComb
中科院分区：
文献类型：
--
作者：
Glen Williams;N. Meisel;T. Simpson;Christopher McComb
机器学习可用于自动化已经存在足够数据的常见或耗时的工程任务。例如，设计存储库可用于训练深度学习算法以评估组件的可制造性;然而，确定设计存储库与机器学习一起使用的适用性的方法并不存在。我们提供了一个初步的调查，以确定这样一种方法，使用“人工”设计存储库，以实验测试的程度，改变属性的数据集影响评估精度和泛化的神经网络训练的数据。在这个实验中，我们使用3D卷积神经网络直接从基于体素的组件几何形状中估计定量制造指标。增材制造（AM）被用作案例研究，因为最近以AM为重点的设计存储库（如GrabCAD和Thingiverse）的增长很容易在线访问。在这项研究中，我们只关注材料挤出，占主导地位的消费者AM过程，并调查三个AM构建指标：（1）零件质量，（2）支持材料质量，（3）构建时间。此外，我们将卷积神经网络的准确性与基线多元线性回归模型的准确性进行了比较。我们的研究结果表明，在标准化程度较低的方向和位置的设计库上进行训练，可以得到更准确的训练神经网络，并且方向相关的指标比方向无关的指标更难估计。此外，对于所有构建指标，卷积神经网络比基线线性回归模型更准确。
Machine learning can be used to automate common or time-consuming engineering tasks for which sufficient data already exist. For instance, design repositories can be used to train deep learning algorithms to assess component manufacturability; however, methods to determine the suitability of a design repository for use with machine learning do not exist. We provide an initial investigation toward identifying such a method using “artificial” design repositories to experimentally test the extent to which altering properties of the dataset impacts the assessment precision and generalizability of neural networks trained on the data. For this experiment, we use a 3D convolutional neural network to estimate quantitative manufacturing metrics directly from voxel-based component geometries. Additive manufacturing (AM) is used as a case study because of the recent growth of AM-focused design repositories such as GrabCAD and Thingiverse that are readily accessible online. In this study, we focus only on material extrusion, the dominant consumer AM process, and investigate three AM build metrics: (1) part mass, (2) support material mass, and (3) build time. Additionally, we compare the convolutional neural network accuracy to that of a baseline multiple linear regression model. Our results suggest that training on design repositories with less standardized orientation and position resulted in more accurate trained neural networks and that orientation-dependent metrics were harder to estimate than orientation-independent metrics. Furthermore, the convolutional neural network was more accurate than the baseline linear regression model for all build metrics.