The additive manufacturing (AM) system today still suffers from low precision, low product stability, low automation level, and low data utilization. The urgent need for enhancements on these fronts leads to the various monitoring and control techniques for AM process. Currently, most existing AM monitoring and corresponding control methods are built on statistical models and used to refine the AM process under a parameter-tuning framework. They help regulate the steady-state or low-frequency behavior in AM process but do not address issues related to AM’s unsteady or dynamic behavior. These unsteady and dynamic behaviors are especially prevalent in high-speed, high-acceleration AM systems with moving parts (e.g., fused filament fabrication, directed energy deposition) and dynamic trajectory regions (e.g., corners). Therefore, in this paper, we proposed a monitoring and control solution that addresses printing inaccuracies with real-time thermal image monitoring. The method depends on a low-order analytical model describing the printed path boundary and an established non-contacting deposition direction. The thermal image observations are thus converted to the dynamic time-domain compensation signals to account for dynamic over-extrusion. The method is verified with a fused filament fabrication printer to alleviate its over-extrusion when printing corners.

This content is only available via PDF.
You do not currently have access to this content.