3D Shape Synthesis for Conceptual Design and Optimization Using Variational Autoencoders

Abstract

We propose a data-driven 3D shape design method that can learn a generative model from a corpus of existing designs, and use this model to produce a wide range of new designs. The approach learns an encoding of the samples in the training corpus using an unsupervised variational autoencoder-decoder architecture, without the need for an explicit parametric representation of the original designs. To facilitate the generation of smooth final surfaces, we develop a 3D shape representation based on a distance transformation of the original 3D data, rather than using the commonly utilized binary voxel representation. Once established, the generator maps the latent space representations to the high-dimensional distance transformation fields, which are then automatically surfaced to produce 3D representations amenable to physics simulations or other objective function evaluation modules. We demonstrate our approach for the computational design of gliders that are optimized to attain prescribed performance scores. Our results show that when combined with genetic optimization, the proposed approach can generate a rich set of candidate concept designs that achieve prescribed functional goals, even when the original dataset has only a few or no solutions that achieve these goals.

TopologyGAN: Topology Optimization Using Generative Adversarial Networks Based on Physical Fields Over the Initial Domain

“…In recent years, new data-driven methods for topology optimization have been proposed to accelerate the process. Deep learning methods have shown promise in efficiently producing near-optimal results with respect to shape similarity as well as compliance with negligible run-time cost [34][35][36][37][38][39][40][41]. Theory-guided machine learning methods use domain-specific theories to establish the mapping between the design variables and the external boundary conditions [42][43][44].…”

Data-driven topology design using a deep generative model

“…Data-driven topology design may seem to be similar to the above studies, particularly the studies of Oh et al (2019), Guo et al (2018), andZhang et al (2019a). However, its novelty can be clearly explained using estimation of distribution algorithm (EDA) (Larrañaga and Lozano 2001).…”

“…In addition, a style transfer network (Gatys et al 2016) was used to reduce the noise included in the material distributions generated by the VAE. Zhang et al (2019a) proposed a structural design method for the three-dimensional shape of a glider. In their study, a VAE is trained using airplane models registered in a threedimensional structure database (Wu et al 2015), and the latent space of the trained VAE is exploited using a GA in a manner similar to that of Guo et al (2018).…”

“…Furthermore, data-driven topology design can be clearly distinguished from the studies of Oh et al (2019), Guo et al (2018), andZhang et al (2019a), from the viewpoint of an EDA. That is, the former can be regarded as a type of EDA, whereas the latter cannot.…”

Design Strategy Network: A deep hierarchical framework to represent generative design strategies in complex action spaces

“…Recently, there has been significant research in this area inspired by the point cloud representation of 3D models [60][61][62]. Studies have further enhanced these ideas' capabilities by introducing a deep-set formulation of neural networks that can accommodate such data's complexities [60,[63][64][65]. Deep sets present the concept of invariant neural networks that transform each element of a set using multiple layers individually.…”