About this Abstract |
Meeting |
2020 TMS Annual Meeting & Exhibition
|
Symposium
|
Additive Manufacturing: Advanced Characterization with Synchrotron, Neutron, and In Situ Laboratory-scale Techniques
|
Presentation Title |
A Machine-Agnostic Approach to Layer-wise Process Monitoring and Control of Powder Bed Additive Manufacturing Technologies |
Author(s) |
Luke Scime, Derek Siddel, Vincent Paquit |
On-Site Speaker (Planned) |
Luke Scime |
Abstract Scope |
It is well-accepted that robust in-situ process monitoring and control schema are required for the broad adoption of Additive Manufacturing technologies, particularly in industries producing safety-critical components. With new metal Additive machines entering the market every year, it is impractical to develop process monitoring solutions specific to each machine. Therefore, we propose a fully machine and camera agnostic algorithm for the layer-wise classification and localization of anomalies in powder bed systems. This is achieved through a novel Convolutional Neural Network (Machine Learning) architecture that automatically adapts to the input data while maintaining the ability to share learned knowledge across Additive platforms. This capability is demonstrated for multiple binder jet, laser fusion, and electron beam fusion machines. The dynamic architecture also enables fusion of different sensing modalities, including multispectral imaging and combining layer-wise images with melt pool-scale data. Finally, the utility of this algorithm in a process control scenario is demonstrated. |
Proceedings Inclusion? |
Planned: Supplemental Proceedings volume |