| About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Additive Manufacturing Modeling, Simulation and Machine Learning
|
| Presentation Title |
Topology of cellular structures with the targeted
non-linear mechanical response
|
| Author(s) |
Sushan Nakarmi, Jeffery Allen Leiding, Kwan-Soo Lee, Nitin Pandurang Daphalapurkar |
| On-Site Speaker (Planned) |
Sushan Nakarmi |
| Abstract Scope |
With the advent of advanced additive manufacturing capabilities, it is feasible to create sophisticated cellular structures using diverse materials such as polymers, metals, and ceramics. By altering the topology of the unit cell, the mechanical properties like modulus and strength can be modulated. Nevertheless, identifying an optimized topology within an enormous design space that would precisely deliver the targeted non-linear material response is challenging. We have developed a data-driven machine-learning (ML) approach capable of inverse designing the topology of a cellular structure based on the intended material response in linear and non-linear regimes. Our method involves (a) generating new topologies using cellular automata, enabling a database of cellular structures for assessing the structure-property relationship, and (b) inverse designing the unit cell topology from desired non-linear stress-strain data using a generative ML framework. This work has potential applications in identifying novel structures for optimized energy absorption, damping attenuation, and impact mitigation. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Additive Manufacturing, Computational Materials Science & Engineering, Machine Learning |