Blast Furnace Condition Recognizing in the Ironmaking Process Based on Prior Knowledge and Platt Scaling Probability

This paper proposes a method for recognizing blast furnace conditions based on prior knowledge and probability calculation. Firstly, we introduce a coupled multivariate correlation analysis method grounded in prior classification knowledge to extract key variables. Subsequently, a recognition method is proposed based on Nu-Support Vector Classification and Platt Scaling Probability. The categories of blast furnace conditions are recognized by calculating the probabilities of var-ious conditions using process datasets. Experiments demonstrate that the proposed method accurately recognizes blast furnace conditions, meets on-site detection accuracy requirements, and provides reliable auxiliary decision-making information for the operational optimization of the blast furnace.