Combining Discrete Element Method and Artificial Neural Network to Predict the Particle Segregation Behaviors at Bell-Less Top Blast Furnace

Abstract

The inevitable particle segregation behaviors of granular materials are undesirable in the industrial processes. It will directly affect the heat transfer efficiency between gas and solid phases during the ironmaking process of blast furnace. Therefore, the understanding and prediction on the particle segregation behaviors are vital to optimize the process and improve the production efficiency. This work first employs the Discrete Element Method (DEM) to investigate the effects of the parameters, such as the particle density, the mean diameter, the mass of particles, the particles mass ratio, and the angle of chute, on the mass segregation and size segregation behaviors at the bell-less top blast furnace with the serial type hopper. Then, the Artificial Neural Network (ANN) model is proposed to predict the aforementioned behaviors in the radial direction based on the numerical data. The results show that the predicted segregation behaviors by the established ANN model are in a good agreement with the simulated results.