Design representation for performance evaluation of 3D shapes in structure-aware generative design
Design representation for performance evaluation of 3D shapes in structure-aware generative design
复制标题
DOI：
10.1017/dsj.2023.25
复制
发表时间：
2023-09
期刊：
影响因子：
2.4
通讯作者：
Xingang Li;Charles Xie;Z. Sha
中科院分区：
文献类型：
--
作者：
Xingang Li;Charles Xie;Z. Sha
摘要 数据驱动生成设计（DDGD）方法利用深度神经网络根据现有数据创建新颖的设计。结构感知 DDGD 方法可以处理复杂的几何形状，并将单独的组件自动组装到系统中，在促进创意设计方面显示出前景。然而，确定适当的矢量化设计表示 (VDR) 来评估从结构感知 DDGD 模型生成的 3D 形状在很大程度上仍未得到探索。为此，我们对使用来自两个来源的 VDR 预测 3D 形状的工程性能的替代模型的性能进行了比较分析：编码结构和几何信息的结构感知 DDGD 模型的训练潜在空间和仅编码几何信息的嵌入方法。我们进行了两个案例研究：一个涉及关注阻力系数的 3D 汽车模型，另一个涉及同时考虑阻力和升力系数的 3D 飞机模型。我们的结果表明，使用潜在向量作为 VDR 会显着恶化代理模型的预测。此外，在嵌入方法中增加VDR的维度可能不一定会改善预测，特别是当VDR包含更多与工程性能无关的信息时。因此，在选择用于代理建模的 VDR 时，必须谨慎使用从训练结构感知 DDGD 模型中获得的潜在向量，尽管训练完成后它们更容易访问。应注意与工程性能相关的基础物理。本文为结构感知 DDGD 的不同类型 VDR 用于代理建模的有效性提供了经验证据，从而有助于为 AI 生成的设计构建更好的代理模型。
Abstract Data-driven generative design (DDGD) methods utilize deep neural networks to create novel designs based on existing data. The structure-aware DDGD method can handle complex geometries and automate the assembly of separate components into systems, showing promise in facilitating creative designs. However, determining the appropriate vectorized design representation (VDR) to evaluate 3D shapes generated from the structure-aware DDGD model remains largely unexplored. To that end, we conducted a comparative analysis of surrogate models’ performance in predicting the engineering performance of 3D shapes using VDRs from two sources: the trained latent space of structure-aware DDGD models encoding structural and geometric information and an embedding method encoding only geometric information. We conducted two case studies: one involving 3D car models focusing on drag coefficients and the other involving 3D aircraft models considering both drag and lift coefficients. Our results demonstrate that using latent vectors as VDRs can significantly deteriorate surrogate models’ predictions. Moreover, increasing the dimensionality of the VDRs in the embedding method may not necessarily improve the prediction, especially when the VDRs contain more information irrelevant to the engineering performance. Therefore, when selecting VDRs for surrogate modeling, the latent vectors obtained from training structure-aware DDGD models must be used with caution, although they are more accessible once training is complete. The underlying physics associated with the engineering performance should be paid attention. This paper provides empirical evidence for the effectiveness of different types of VDRs of structure-aware DDGD for surrogate modeling, thus facilitating the construction of better surrogate models for AI-generated designs.