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 |
Defects Classification via Hierarchical Graph Convolutional Network in L-PBF Additive Manufacturing |
Author(s) |
Anyi Li, Jia Liu, Shuai Shao, Nima Shamsaei |
On-Site Speaker (Planned) |
Jia Liu |
Abstract Scope |
Three typical types of defects, i.e., lack of fusions, keyholes, and gas-entrapped pores, characterized by various features (e.g., area, volume, perimeter, etc.), are generated under different fabrication conditions of laser beam powder bed fusion (L-PBF) processes. However, there is a lack of recognized approaches to automatically and accurately classify the defects in L-PBF components. This work presents a novel hierarchical graph convolutional network (H-GCN) to classify different types of defects by a cascading GCN structure with a low-level features (defect features) layer and a high-level features (fabrication conditions) layer. Such an H-GCN not only leverages the information from different hierarchies to classify the defects but also explores the impact of fabrication conditions on defect features. The H-GCN is evaluated through simulation and a real case study with L-PBF defect datasets and compared with the neural network (NN) and GCN. |
Proceedings Inclusion? |
Definite: Post-meeting proceedings |