Model predictive control of the refining stage in an electric arc furnace using soft sensors

Blažič, Aljaž (avtor), Škrjanc, Igor (avtor), Logar, Vito (avtor)

In this paper, a novel Model Predictive Control (MPC) based on soft sensors is developed to optimize the refining stage in Electric Arc Furnaces (EAFs). Given the significant energy consumption and environmental impact associated with EAF operations, there is a critical need for more precise and efficient control strategies. The proposed MPC approach utilizes real-time soft sensors to accurately determine critical parameters, i.e., bath temperature and dissolved oxygen content. By replacing operators’ subjective judgments with continuous, data-driven feedback, this methodology facilitates precise control adjustments throughout the refining process. The implementation of this MPC has shown that the duration of the refining stage has been reduced by approximately 80 seconds, which directly translates into shorter tap-to-tap times, lower energy consumption and improved operational efficiency. In addition, consistently maintaining optimal bath conditions ensures improved steel quality, greater reliability, and higher overall productivity of the EAF process.

• Jezik: Angleški jezik
• Ključne besede: model predictive control, bath temperature, electric arc furnace, dissolved oxygen
• Vrsta gradiva: Članek v reviji
• Tipologija: 1.08 - Objavljeni znanstveni prispevek na konferenci
• Organizacija: FE - Fakulteta za elektrotehniko
• Status publikacije: Objavljeno
• Različica publikacije: Objavljena publikacija
• Leto izida: 2025
• Št. strani: 8 str.
• PID: 20.500.12556/RUL-172523
• UDK: 621.365.2:681.5
• ISSN pri članku: 1742-6596
• DOI: 10.1088/1742-6596/3050/1/012010
• COBISS.SI-ID: 246338563
• Datum objave v RUL: 08.09.2025

Gradivo je del zbornika

• Naslov: 6th European Academic Symposium on EAF Steelmaking 2025 (EASES 2025)
• COBISS.SI-ID: 246334211

Gradivo je del revije

• Naslov: Journal of physics : Conference series
• Založnik: Institute of Physics Publishing
• ISSN: 1742-6596
• COBISS.SI-ID: 2035044

Licence

• Licenca: CC BY 4.0, Creative Commons Priznanje avtorstva 4.0 Mednarodna
• Povezava: http://creativecommons.org/licenses/by/4.0/deed.sl
• Opis: To je standardna licenca Creative Commons, ki daje uporabnikom največ možnosti za nadaljnjo uporabo dela, pri čemer morajo navesti avtorja.

Sekundarni jezik

• Jezik: Slovenski jezik
• Ključne besede: prediktivno vodenje, temperatura kopeli, elektroobločna peč, raztopljen kisik

Projekti

• Financer: ARIS - Javna agencija za znanstvenoraziskovalno in inovacijsko dejavnost Republike Slovenije
• Številka projekta: P2-0219
• Naslov: Modeliranje, simulacija in vodenje procesov