Interpretable Deep Learning Techniques and Meta Information for Automated Decision Making in Ironmaking Plants
(Room 317)
16 May 22
2:30 PM
-
3:00 PM
Modeling complex processes like ironmaking is a demanding task. Approaches based on machine learning and especially deep learning (DL) offer a considerable option. Using the latest artificial intelligence (AI) achievements, a multi-step time series DL framework was developed to forecast key performance indicators, e.g., the silicon content and temperature of hot metal from a blast furnace. However, DL methods are often seen as black boxes, which hinder their contribution to expand process knowledge as well as their integration into control systems. Enhancing the DL model with advanced, explainable AI techniques and meta information makes them more transparent, enabling optimized decision-making in ironmaking.