Research on data driven modeling method of grinding process based on RBF neural network

摘要

Beneficiation is a complex industrial process, current beneficiation methods are carried out by the difference in the nature of the minerals and gangues inside the ore, which needs the separation of gangue and minerals by grinding process. The production and investment consumption of grinding accounts for a large proportion of the total consumption of the dressing plant, and the grinding process is a key process for providing raw materials for mineral sorting. Therefore, the design and operation of the grinding process directly affects the economic indicators of the dressing plant. In this paper, the research is conducted on the background of a certain dressing plant, and the mechanism of the grinding process is analyzed in order to analyze the state of the grinding process and the parameter variables. Aiming at the situation that the ore is a mixed ore of various ores, the influence of different mineral contents on the results is fully considered. The mathematical model of the grinding process yield and the particle size distribution characteristics of the grinding products is established by RBF neural network. Simulation results demonstrate the effectiveness of the model.