arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2309.04499cs.LG
收藏

考虑3D设计数据的几何特征与工程性能的加权无监督域适应

Weighted Unsupervised Domain Adaptation Considering Geometry Features and Engineering Performance of 3D Design Data

Cho Chun Shik Graduate School of Mobility, KAIST(KAIST赵春植移动研究生院);Narnia Labs(纳尼亚实验室)

Seungyeon Shin, Namwoo Kang

AI总结:

针对3D设计数据深度学习预测在新域不准确的问题,提出考虑几何特征与工程性能的双加权无监督域适应方法,通过对抗训练提取域不变特征并避免负迁移,有效降低目标风险并替代传统有限元分析。

摘要:

制造业的产品设计过程涉及迭代的设计建模与分析,以实现目标工程性能,但这一迭代过程耗时且计算成本高昂。近期,基于深度学习的工程性能预测模型被提出以加速设计优化。然而,这些模型仅在训练数据上保证预测准确性,应用于新域数据时可能不准确。特别是3D设计数据具有复杂特征,意味着存在具有不同分布的域。因此,由于繁重的数据收集和训练负担,深度学习的应用受到限制。我们提出了一种双加权无监督域适应方法,该方法考虑了3D设计数据的几何特征和工程性能,专门用于基于深度学习的工程性能预测。通过使用假设差异的对抗训练策略,可以提取域不变特征,并利用提取的特征执行多输出回归任务以预测工程性能。特别是,我们提出了一种适用于3D设计数据的源实例加权方法,以避免负迁移。所开发的基于工程结构几何特征和工程性能的双加权策略被整合到训练过程中。所提出的模型在车轮冲击分析问题上进行了测试,用于预测最大von Mises应力的大小及3D道路车轮的对应位置。该机制能够基于加权多源域知识降低无标签目标域的目标风险,并能高效替代传统的有限元分析。

英文摘要:

The product design process in manufacturing involves iterative design modeling and analysis to achieve the target engineering performance, but such an iterative process is time consuming and computationally expensive. Recently, deep learning-based engineering performance prediction models have been proposed to accelerate design optimization. However, they only guarantee predictions on training data and may be inaccurate when applied to new domain data. In particular, 3D design data have complex features, which means domains with various distributions exist. Thus, the utilization of deep learning has limitations due to the heavy data collection and training burdens. We propose a bi-weighted unsupervised domain adaptation approach that considers the geometry features and engineering performance of 3D design data. It is specialized for deep learning-based engineering performance predictions. Domain-invariant features can be extracted through an adversarial training strategy by using hypothesis discrepancy, and a multi-output regression task can be performed with the extracted features to predict the engineering performance. In particular, we present a source instance weighting method suitable for 3D design data to avoid negative transfers. The developed bi-weighting strategy based on the geometry features and engineering performance of engineering structures is incorporated into the training process. The proposed model is tested on a wheel impact analysis problem to predict the magnitude of the maximum von Mises stress and the corresponding location of 3D road wheels. This mechanism can reduce the target risk for unlabeled target domains on the basis of weighted multi-source domain knowledge and can efficiently replace conventional finite element analysis.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑