| About this Abstract |
| Meeting |
2024 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2024)
|
| Symposium
|
2024 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2024)
|
| Presentation Title |
In-Situ Sensor Monitoring for Defect Prediction During Metal 3D-Printing of Aluminum Alloy Using Convolutional Neural Networks |
| Author(s) |
Sandesh Giri, Sen Liu, Sanam Gorgannejad, Peiyu Quan, Vivek Thampy, Nicholas Calta, Christopher J. Tassone |
| On-Site Speaker (Planned) |
Sandesh Giri |
| Abstract Scope |
Laser powder bed fusion (LPBF) 3D printing of alloy materials shows the potential for advanced manufacturing and environmental sustainability in the future. However, LPBF alloying process is prone to defects which may affect the structural integrity and mechanical properties of manufactured components. This research presents an approach for detecting and classifying porosity defects of Al6061 alloy through in-situ photodiode sensor monitoring and advanced in-situ synchrotron X-ray characterization. Continuous Wavelet Transforms (CWT) were applied to the photodiode sensor signals collected during LPBF process to create time-frequency scalogram representations. These CWT plots were then analyzed using Convolutional Neural Networks to identify and classify different types of pore regions, “No Pore”, “Gas Pore” and “Keyhole Pores”. This method demonstrated high reliability across various defect classifications, effectively detecting and providing valuable insights into the mechanisms of pore formation. This research shows the potential for cost-effective, scalable flaw detection and process monitoring in additive manufacturing. |
| Proceedings Inclusion? |
Definite: Post-meeting proceedings |