Advanced Numerical Modeling of Powder Bed Fusion: From Physics-Based Simulations to AI-Augmented Digital Twins

Area of Science:

• Additive Manufacturing (AM)
• Materials Science
• Computational Modeling

Background:

• Powder bed fusion (PBF) is a key additive manufacturing (AM) technology for high-resolution fabrication.
• The complexity of PBF necessitates advanced modeling for quality, reliability, and scalability.
• Current modeling approaches face challenges in capturing intricate process dynamics.

Purpose of the Study:

• To critically synthesize advances in modeling frameworks for PBF.
• To highlight the role of digital twins in PBF.
• To identify challenges and opportunities for PBF digital twin development.

Main Methods:

• Review of physics-based simulations, machine learning, and digital twin frameworks.
• Analysis of progress across micro, meso, and part scales.
• Emphasis on hybrid physics-data-driven approaches and real-time sensing integration.

Main Results:

• Significant progress in multi-scale modeling for PBF, from melt pool dynamics to residual stress.
• Growing importance of hybrid physics-data-driven methods for capturing process-structure-property (PSP) relationships.
• Digital twins, integrating sensing, multi-scale modeling, and AI, are foundational for advanced PBF.

Conclusions:

• Digital twins represent a transformative paradigm for PBF.
• Addressing challenges like computational cost and data scarcity is crucial for industry adoption.
• Future research should focus on scalable, interpretable, and industry-ready digital twin platforms for PBF.