Published Paper Details:

The steelmaking industry strives for efficient and cost-effective production while achieving desired steel qualities. This paper investigates the application of machine learning (ML) for optimizing ferroalloy usage in the steelmaking process. We explore how ML models can be leveraged to predict the required amount of ferroalloys for specific steel grades based on various process parameters and historical data. This approach offers the potential to minimize ferroalloy consumption while maintaining steel quality consistency. The paper discusses the benefits of ML-based optimization, including reduced production costs, minimized environmental impact, and improved process control. We explore potential ML algorithms suitable for this task and outline the development process for an ML-based ferroalloy optimization system. Finally, the discussion addresses the challenges and future directions for implementing ML-based optimization in steelmaking.