Secondary metallurgy is the most value adding step in the steelmaking process. As more sophisticated steel grades are developed, better process control is required to adjust and monitor properties of molten steel in secondary metallurgy. Improved process control also serves to increase productivity and profitability of steel mills. In steelmaking, process control is achieved through feedback from sensors. Unfortunately, existing state-of-the-art sensors for molten steel are consumable, unreliable, and provide limited measurements.

In the steelmaking industry, process control models need to be based on a sound physical understanding of the process but should also account for many uncertainties due to the nature and complexity of the environment in which the process is carried out. As a result, it is crucial to extract useful process control information from the raw data stream acquired by the industrial sensors.

In the steelmaking industry, process control models need to be based on a sound physical understanding of the process but should also account for many uncertainties due to the nature and complexity of the environment in which the process is carried out. As a result, it is crucial to extract useful process control information from the raw data stream acquired by the industrial sensors.