About this Abstract |
Meeting |
2025 TMS Annual Meeting & Exhibition
|
Symposium
|
Melt Processing, Casting and Recycling
|
Presentation Title |
Loading solid and liquid charges into melting furnaces is a critical step in the melting process, impacting melt efficiency in terms of cycle time and energy consumption. It also affects refractory integrity, burner efficiency, and liquid metal security. Determining the optimal timing for loading stages is contentious and often relies on standard practices within the cast-shop or other deterministic rules. Novamet's SmartMelt technology was developed to enhance the efficiency of aluminum melters by integrating a rapid and precise digital twin of the furnace. Utilizing real-time data acquisition, SmartMelt predicts the ideal loading times for both liquid and solid charges based on the specific scrap type. This optimization has demonstrated a potential efficiency improvement of up to 10%. Furthermore, it provides warnings for potential overheating incidents when thermocouples are not correctly inserted. Any delays or deviations from these recommendations are monitored and logged in a dedicated database, ensuring comprehensive process control and optimization. |
Author(s) |
Amin Rostamian, Marc Bertherat |
On-Site Speaker (Planned) |
Amin Rostamian |
Abstract Scope |
Loading solid and liquid charges into melting furnaces is a critical step in the melting process, impacting melt efficiency in terms of cycle time and energy consumption. It also affects refractory integrity, burner efficiency, and liquid metal security. Determining the optimal timing for loading stages is contentious and often relies on standard practices within the cast-shop or other deterministic rules. Novamet's SmartMelt technology was developed to enhance the efficiency of aluminum melters by integrating a rapid and precise digital twin of the furnace. Utilizing real-time data acquisition, SmartMelt predicts the ideal loading times for both liquid and solid charges based on the specific scrap type. This optimization has demonstrated a potential efficiency improvement of up to 10%. Furthermore, it provides warnings for potential overheating incidents. Any delays or deviations from these recommendations are monitored and logged in a dedicated database, ensuring comprehensive process control and optimization. |
Proceedings Inclusion? |
Planned: Light Metals |
Keywords |
Aluminum, Process Technology, Machine Learning |