SFNN model for control and prediction hot metal silicon content in BF

A new intelligent control and prediction model, based on SFNN, was developed to approach the change of hot metal silicon content in BF ironmaking system, where, noise influencing the system was thoroughly considered. Nonuniform function fuzzification, fuzzy rules, fuzzy inference and special defuzzification are included in the SFNN model, presented in this paper. The result of control and prediction to hot metal silicon content is very successful. Data, used in this paper, was collected online from No.1 BF at Laiwu Iron and Steel Group Co.