An Innovative Data-Driven Model to Predict Hydrogen and Nitrogen During Vacuum Degassing

A data-driven predictive model was developed to calculate the fraction of dissolved hydrogen and nitrogen in the liquid steel during the vacuum degassing process. The model, trained on historical data automatically collected by the system, considers several relevant process variables as well as the output of a gas analysis system monitoring the VD offgas composition and flow. Based on this information, the system calculates when the dissolved gases target is achieved and suggests when to stop the degassing step, aiding the optimization of the VD process. The system was installed at a VD station and the achieved results are shown.

Authors:

Stefano Furlanetto | Danieli & C. Officine Meccaniche S.p.A.
 

Manuele Piazza | Danieli Automation S.p.A.
 

Sergio Comuzzo | Danieli & C. Officine Meccaniche S.p.A.
 

Luca Bortolin | Danieli & C. Officine Meccaniche S.p.A.
 

Giacomo Di Giorgio | Acciaierie Bertoli Safau S.p.A.
 

Luca Gemo | Acciaierie Bertoli Safau S.p.A.
 

Federico Bianco | Danieli & C. Officine Meccaniche S.p.A.
 

Session Chairs:

Tucker Keuter | Charter Steel - Saukville, Wisconsin
 

Shaojie Chen | EVRAZ Regina