| About this Abstract |
| Meeting |
2023 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2023)
|
| Symposium
|
2023 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2023)
|
| Presentation Title |
Hybrid Geometry/Property Autoencoders for Multi-Lattice Transitions |
| Author(s) |
Martha Baldwin, Nicholas A Meisel, Christopher McComb |
| On-Site Speaker (Planned) |
Martha Baldwin |
| Abstract Scope |
Additive manufacturing has revolutionized structural optimization. This may be achieved through multi-lattice structures, which have emerged as promising candidates for impact reduction and other applications. However, the performance of these structures relies on the details of mesostructural elements. Many current approaches use data-driven approaches to generate multi-lattice transition regions but focus largely on the geometry of the mesostructure. Therefore, it remains unclear whether the integration of functional performance factors in the generation of multi-lattice interpolations is beneficial beyond geometry alone. To address this issue, this work implements and evaluates a VAE to create a hybrid geometric/functional latent space for generating multi-lattice transition regions. This approach allows us to test the performance of the proposed method against geometry and property defined latent spaces. Our research aims to determine whether it is necessary to incorporate physical properties into the geometrically defined latent spaces to achieve robust multi-lattice structures. |
| Proceedings Inclusion? |
Definite: Post-meeting proceedings |