About this Abstract |
Meeting |
MS&T21: Materials Science & Technology
|
Symposium
|
Additive Manufacturing: Large-Scale Metal Additive Manufacturing
|
Presentation Title |
Thermo-mechanical FEM Modeling and Machine Learning of Distortion on Overhang Structure in Laser Powder Bed Fusion Additive Manufacturing |
Author(s) |
Xuesong Gao, Tyler High, Jesse Zhu, Wei Zhang, Hyeyun Song |
On-Site Speaker (Planned) |
Xuesong Gao |
Abstract Scope |
Net or near-net shape parts produced by additive manufacturing typically possess complex structures, e.g., overhang which is disposed to distortion problems. To address this, two kinds of thermo-mechanical models were developed. The first one used a lumped layer method where multiple layers were treated as a single “building block”. The distortion formation was calculated on the full scale part and the model exhibited high accuracy and efficiency. Another model was based on a moving heat source method where each pass and layer was simulated directly. The model geometry was scaled down by 10 times and most processing parameters were accounted for, such as power and scanning speed, etc. To validate these models, overhang structures were printed with in-situ temperature and distortion measurements. A machine learning program was developed to correlate the measured temperature distribution to the measured distortion. The roles of thermal stress on distortion formation were revealed. |