Physics-Informed Machine Learning of Manufacturing Dynamics
Yuebin Guo
Rutgers University-New Brunswick，USA
Abstract
The complex dynamics of advanced manufacturing processes are manifested by non-linear effects, unknown physics, high dimensionality, and uncertainty, significantly impacting product quality in form accuracy and surface integrity. Understanding and predicting such complex manufacturing dynamics (e.g., metal pool dynamics in metal additive manufacturing) remains the central intractable problem for producing high-quality components or advanced materials. Compared with conventional physics-based modeling approaches such as finite element methods (FEM) and computational fluid dynamics (CFD), machine learning (ML) may leverage high-dimensional and online process data for real-time model updating, prediction, and process control. However, data-driven ML models are black-box, inherently computation-intensive, and storage-intensive. A deep knowledge gap exists between ML and physics-based models in predicting complex manufacturing dynamics. This talk presents a physics-informed machine learning (PIML) framework to integrate physical laws (e.g., governing partial differential equations) underpinning process dynamics with ML for efficient forward prediction, inverse learning, and model discovery. Case studies in additive and subtractive manufacturing are provided to demonstrate the wide applications of the PIML framework in diverse manufacturing operations.
Bio
Dr. Yuebin Guo is Henry Rutgers Distinguished Professor and Director of New Jersey Advanced Manufacturing Institute at Rutgers University-New Brunswick, USA. Prior to Rutgers, he served as the Assistant Director for Research Partnerships at the U.S. Advanced Manufacturing National Program Office. His research focuses on manufacturing processes, scientific machine learning, AI-driven digital twins, and surface integrity. He is a recipient of numerous awards, including the ASME William T. Ennor Manufacturing Technology Award, the SME Albert M. Sargent Progress Award, the Rutgers Board of Trustees Award for Excellence in Research, and the Alexander von Humboldt Research Award. He is an elected fellow of the American Society of Mechanical Engineers (ASME), the Society of Manufacturing Engineers (SME), and the International Academy for Production Engineering (CIRP).
