Code Repository for "Real-Time Prediction of Thermal History and Hardness in Laser Powder Bed Fusion Using Deep Learning"
Schüßler, Philipp
1; Schulze, Volker 1; Dietrich, Stefan
1
1 Institut für Angewandte Materialien – Werkstoffkunde (IAM-WK), Karlsruher Institut für Technologie (KIT)
1; Schulze, Volker 1; Dietrich, Stefan
11 Institut für Angewandte Materialien – Werkstoffkunde (IAM-WK), Karlsruher Institut für Technologie (KIT)
Abstract (englisch):
A PyTorch LSTM model that predicts the thermal history of individual measurement points during laser powder bed fusion (PBF-LB/M) additive manufacturing. The model uses teacher-forcing during training and supports both teacher-forcing and auto-regressive (inference) forward modes. An ensemble training workflow is included for uncertainty quantification.
Version v1.0.0
Full-text publication: https://doi.org/10.1016/j.commatsci.2026.115025
GitLab repository: https://gitlab.kit.edu/kit/iam-wk-public/iam-wk-fub-deep-learning-pbf-lb
Trained model dataset: https://doi.org/10.35097/37da9d66y4t27q55
Training-Validation-Testing Dataset: https://doi.org/10.35097/pmem1cb9gu1ck8xz
Full-text publication for the FEM simulation model: https://doi.org/10.1080/17452759.2023.2271455
Version v1.0.0
Full-text publication: https://doi.org/10.1016/j.commatsci.2026.115025
GitLab repository: https://gitlab.kit.edu/kit/iam-wk-public/iam-wk-fub-deep-learning-pbf-lb
Trained model dataset: https://doi.org/10.35097/37da9d66y4t27q55
Training-Validation-Testing Dataset: https://doi.org/10.35097/pmem1cb9gu1ck8xz
Full-text publication for the FEM simulation model: https://doi.org/10.1080/17452759.2023.2271455
