About this Abstract |
Meeting |
MS&T22: Materials Science & Technology
|
Symposium
|
Additive Manufacturing: Equipment, Instrumentation and In-Situ Process Monitoring
|
Presentation Title |
Real-time, High-speed and High-resolution Multi- and Hyperspectral Imaging of Powder Bed Fusion |
Author(s) |
Steven Storck, Mark Foster, Nathan Drenkow, Brendan Croom, Milad Alemohammad, Christopher Stiles, Bobby Mueller, Michael Pekala, Mary Dafron, Ryan Carter, Dylan Madisetti |
On-Site Speaker (Planned) |
Steven Storck |
Abstract Scope |
Laser-powder-bed-fusion (L-PBF) has become a prime candidate for the aerospace and medical fields, which have challenging qualification requirements. Stringent qualification/certification standards are the main limiting factors for part acceptance, resulting in significant time and >60% of the cost to develop a final component. Real-time monitoring of L-PBF to detect defect formation and processing anomalies will significantly increase engineering certainty. This will significantly reduce qualification and certification costs and time, enabling AM in more applications and industries. Our team developed a custom high-speed hyperspectral sensor and high speed multicolor pyrometer that tracks the thermal signature of the laser melt pool at up to 500 KHz. Data is correlated with spatial porosity and analyzed with machine learning, correlating spectral characteristics to process anomalies. This talk will cover sensor development, detection of laser-induced defects and thermal anomalies using hyper- and multi-spectral imaging, machine learning analysis, and validation using 3-dimensional X-ray computed tomography analysis. |