About this Abstract |
Meeting |
2022 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2022)
|
Symposium
|
2022 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2022)
|
Presentation Title |
Detecting Layerwise Build Defects Using Low-Cost Imaging and Machine Learning |
Author(s) |
Bradley Howell Jared, Devon Goodspeed, Dylan Lewis, Ethan Rummel, Shuchi Khurana |
On-Site Speaker (Planned) |
Devon Goodspeed |
Abstract Scope |
Process monitoring is a timely and necessary topic in metal laser-powder bed fusion as interest continues to cost effectively improve part confidence and simultaneously reduce inspection and qualification cycle times. The identification of build defects will be demonstrated using low-cost optical hardware retro-fitted onto a Farsoon FS271M powder bed system. The existing system captures static powder bed images during each layering step. It then utilizes customized machine learning algorithms to detect powder bed anomalies, re-coater streaking and hopping, and part swelling. After discussing the equipment and approach, its efficacy will be exhibited in 316L stainless steel using varying build artifacts intended to introduce build errors or to yield acceptable parts. The utility of the existing hardware and software in detailing excessive part deformations and internal porosity will also be explored. |
Proceedings Inclusion? |
Definite: Post-meeting proceedings |