What does this research mean for the field?

A physics-informed deep learning surrogate model can accurately predict real-time local thermal histories and hardness in laser powder bed fusion of steel, achieving a temperature RMSE of 2.6 ± 1.2 K. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To develop a fast, physics-informed deep learning surrogate model capable of real-time prediction of localized thermal histories and resulting hardness during laser powder bed fusion of AISI 4140 steel.

August 30, 2026Computational Materials ScienceOpen Access

Towards real-time prediction of thermal history and hardness in laser powder bed fusion using deep learning

View Full Paper

Ask AI

Bookmark

To develop a fast, physics-informed deep learning surrogate model capable of real-time prediction of localized thermal histories and resulting hardness during laser powder bed fusion of AISI 4140 steel.
Trained an autoregressive two-layer Long Short-Term Memory (LSTM) network ensemble using multiscale finite element simulation data capturing laser-material interactions and process parameters.
Calculated spatial hardness distributions by feeding the predicted local temperature–time histories into a non-isothermal Hollomon–Jaffe tempering model across varying geometries.
The two-layer LSTM ensemble achieved a temperature root-mean-square error (RMSE) of 2.6 ± 1.2 K on an independent test dataset.
The framework accurately reconstructed full thermal histories and successfully predicted local material hardness under dynamic process conditions and cross-sectional geometries.

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Computational study demonstrates real-time prediction of thermal history and hardness in steel 3D printing using an LSTM ensemble, suggesting pathways for in situ digital twins.

Cite This Study

Schüßler et al. (2026) studied this question.

synapsesocial.com/papers/6a93f0f56c1a8fb52e79d93b https://doi.org/10.1016/j.commatsci.2026.115025

View Full Paper

Ask AI

Bookmark

Implication

Computational study demonstrates real-time prediction of thermal history and hardness in steel 3D printing using an LSTM ensemble, suggesting pathways for in situ digital twins.

Towards real-time prediction of thermal history and hardness in laser powder bed fusion using deep learning

Key Points

Discussion

Implication

Cite This Study

Towards real-time prediction of thermal history and hardness in laser powder bed fusion using deep learning

Key Points

Discussion

Implication

Cite This Study

Also Consider

Also Consider