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 |
Additive Manufacturing Process State Inference with Convolutional LSTM |
Author(s) |
Vivek Patel, Richard Anarfi, Benjamin Kwapong, Kenneth Fletcher, Todd Sparks, Aaron Flood |
On-Site Speaker (Planned) |
Vivek Patel |
Abstract Scope |
Simulation of the melt pool physics during directed energy deposition (DED) or laser powder bed fusion (LBPF) processes can be a computationally onerous task. If such a model is needed for real time control, the problem can be intractable, leading to the usage of reduced-fidelity models. Machine learning (ML) offers methods to use high-fidelity models as a training data source to infer solution results quickly. In this paper we propose a ML method based on convolutional long short-term memory (LSTM) network trained on rendered images from a metal AM process simulation and CAM data. The convolutional LSTM architectures bring together time series processing and computer vision by introducing a convolutional recurrent cell in a LSTM layer. The proposed method is used to predict images of an AM process given tool path and process information. Such a prediction can be an important first step in a ML-based process control schema. |
Proceedings Inclusion? |
Definite: Post-meeting proceedings |