Application of Ensemble Learning Algorithm in Prediction and Control of Binder Quantity in Rotary Hearth Furnace

This paper proposes an efficient and excellent algorithm for intelligent control of binders based on the operational status of the rotary hearth furnace. The rotary hearth furnace technology is a new solid waste treatment method with several advantages over traditional approaches. However, due to complex equipment conditions, workers may fail to adjust binder values promptly, leading to reduced balling rate, increased waste material, and potential production halts due to waste material accumulation. To address these challenges, an innovative Ensemble Learning algorithm is introduced, integrating three classical machine learning algorithms tailored to specific features. The algorithm provides timely and precise binder regulation recommendations, optimizing the raw material mixing process and enhancing overall production efficiency in the rotary hearth furnace.