About this Abstract |
Meeting |
MS&T24: Materials Science & Technology
|
Symposium
|
Additive Manufacturing Modeling, Simulation, and Machine Learning: Microstructure, Mechanics, and Process
|
Presentation Title |
Minimizing Layer-Level Thermal Variance in Electron Beam Powder Bed Fusion via Numerical Optimal Control |
Author(s) |
Mikhail Khrenov, William Frieden Templeton, Sneha Prabha Narra |
On-Site Speaker (Planned) |
Mikhail Khrenov |
Abstract Scope |
Electron beam powder bed fusion (EB-PBF) is capable of high power and rapid scanning, creating both opportunities for rapid processing and challenges in process planning. Rastering at km/s velocities is infeasible, leading to the adoption of random spot-melting. However, in spot-melting, the process never reaches thermal steady-state, resulting in variations between each point. While heuristics may mitigate some issues, they are not sufficient.
Instead, we formulate minimizing thermal variance in EB-PBF as an optimal control problem over a power field. We show that under conduction, minimizing variance is a convex quadratic program, guaranteeing global optimality. We solve this problem using tools we have developed for additive manufacturing optimal control (ADDOPT) and achieve an 87% reduction in variance compared to spot-melting. We then validate the solutions experimentally. This work serves as additional evidence for the power of applying optimal control to expand the capabilities of additive manufacturing. |