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 |
Hybrid Reinforcement Learning and Neural Network Framework for Optimizing Deposition Paths and Thermal Profiles in Solid Freeform Fabrication of Functionally Graded Materials Using Maxel (Material Voxel) Framework |
| Author(s) |
Bharat Dwivedi, Rajeev Dwivedi, Arihant Panwar, Sumant Joshi |
| On-Site Speaker (Planned) |
Bharat Dwivedi |
| Abstract Scope |
Solid Freeform Fabrication enables the production of functionally graded materials by allowing control over material composition within internal volume. However, the material deposition is often constrained by geometry of the path vs desired material distribution presenting many process related inefficiencies. To address this we are using field based approach to develop path geometry, additionally, remelting and remixing is employed to facilitate smoother transitions.
This research introduces an AI-agent-based framework utilizing a maxel (material voxel) representation to optimize deposition paths.Furthermore, we investigate the impact of substrate heat dissipation. The framework employs a hybrid agent combining Reinforcement Learning (RL) and Neural Networks (NN) to learn optimal deposition strategies. By simulating discrepancies between intended and actual material fields, we quantify deposition errors and analyze their propagation throughout the build process. Employing the shape factor as a parameter, our AI-agent framework predicts and adjusts thermal profiles to improve material properties. |
| Proceedings Inclusion? |
Planned: Post-meeting proceedings |