AI-Driven Radiative Sensing for Optimizing Continuous Galvanizing Processes

Precise thermal control during galvannealing is essential for producing phase-specific, high-quality steel on continuous galvanizing lines (CGLs). However, conventional near-infrared pyrometry techniques are limited in their ability to capture the spectral features linked to phase-dependent radiative properties of the steel surface. Focusing on galvannealed interstitial-free steel, this study combines near- and mid-infrared spectral irradiance measurements with machine learning to build a physics-informed model that links spectral emissivity and pyrometry signals to strip temperature and Fe-Zn intermetallic phase formation. The results offer insights for developing real-time algorithms that enable accurate phase control and improved CGL process monitoring.

Authors:

Dr. Fatima Suleiman | University of Waterloo
 

Michiyo Kagaya | University Of Waterloo
 

Dr. Malo Lecorgne | University of Waterloo
 

Andrew Jackson Leach | University of Waterloo
 

Dr. Kyle Daun | University of Waterloo
 

Session Chairs:

Dan Baker | LIFT
 

Nikhil Kulkarni | Steel Dynamics Inc. - Jeffersonville Plant