Physics-Informed Surrogates for Temperature Prediction of Multi-Tracks in Laser Powder Bed Fusion

2025-02-03

AI Summary

To address the challenge of efficiently and accurately predicting the transient 3D temperature field in multi-track laser powder bed fusion (LPBF), this work proposes a hybrid surrogate model integrating physics-informed constraints with deep operator learning. We introduce a novel sequential physics-informed neural network (PINN) training strategy to overcome the prohibitive training complexity of DeepONet in multi-track scenarios. Furthermore, we develop an end-to-end, parametric thermal history solver, rigorously validated against finite-difference simulations, enabling real-time temperature field modeling for both single- and multi-track scan paths. The model achieves accuracy comparable to high-fidelity numerical simulations while offering orders-of-magnitude speedup. It enables rapid, interpretable assessment of how scan trajectories and laser parameters influence thermal history, thereby establishing a generalizable, physically grounded digital twin foundation for LPBF process optimization.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Large Multimodal Models (LMMs)
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Modeling plays a critical role in additive manufacturing (AM), enabling a deeper understanding of underlying processes. Parametric solutions for such models are of great importance, enabling the optimization of production processes and considerable cost reductions. However, the complexity of the problem and diversity of spatio-temporal scales involved in the process pose significant challenges for traditional numerical methods. Surrogate models offer a powerful alternative by accelerating simulations and facilitating real-time monitoring and control. The present study presents an operator learning approach that relies on the deep operator network (DeepONet) and physics-informed neural networks (PINN) to predict the three-dimensional temperature distribution during melting and consolidation in laser powder bed fusion (LPBF). Parametric solutions for both single-track and multi-track scenarios with respect to tool path are obtained. To address the challenges in obtaining parametric solutions for multi-track scenarios using DeepONet architecture, a sequential PINN approach is proposed to efficiently manage the increased training complexity inherent in those scenarios. The accuracy and consistency of the model are verified against finite-difference computations. The developed surrogate allows us to efficiently analyze the effect of scanning paths and laser parameters on the thermal history.

Problem

• Predict temperature in laser powder bed fusion
• Model multi-track scenarios efficiently
• Optimize production using physics-informed surrogates

Innovation

• Physics-informed neural networks
• Deep operator network
• Sequential PINN approach