About this Abstract |
| Meeting |
2025 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2025)
|
| Symposium
|
2025 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2025)
|
| Presentation Title |
Data-Efficient Design of Multistable, Robust Structures for Additive Manufacturing Using Bayesian Optimization |
| Author(s) |
Zihan Li, Minhee Kim, Andrea N Camacho-Betancourt, Iris V. Rivero |
| On-Site Speaker (Planned) |
Zihan Li |
| Abstract Scope |
Additive manufacturing (AM) enables the fabrication of complex, highly customized geometries. However, designing and fabricating structures with advanced functionalities, such as multistability and fail-safe mechanisms, remains challenging due to the significant time and cost required for high-fidelity simulations and iterative prototyping. In this study, we investigate the application of Bayesian Optimization (BO), an advanced machine learning framework, to accelerate the discovery of optimal AM-compatible designs with these properties.
BO strategically balances exploration of the design space with limited evaluations and exploitation near the best-performing designs, thereby minimizing the number of evaluations needed. While existing studies have demonstrated BO's potential in AM, most focus on static or simple structures. Here, we address the challenge of designing multistable structures that exhibit multiple stable configurations and reconfigure in response to external conditions. We present a BO-driven framework efficiently identifies high-performing designs while ensuring strength across all stable configurations. |
| Proceedings Inclusion? |
Planned: Post-meeting proceedings |