An improved WM fuzzy modeling method for blast furnace gas system

The blast furnace gas is an important secondary energy for the iron and steel production. Establishing an effective model to describe the state of BFG system is of great significant to maintain the system balance and stability. Considering the strong coupling characteristics of the blast furnace gas system and the high level noises in the industrial data, a simplex unscented Kalman filter-based Wang-Mendel modeling method is proposed in this paper to improve the accuracy and generalization ability of the fuzzy model. In the proposed method, the maximum posterior estimation of the center value in each fuzzy rule space is calculated in the probability density distribution perspectives by incorporating the Kalman filtering method into rule extraction process, which eliminates the influence of noises effectively and improves the accuracy of the fuzzy model. The experimental results by using the Lorenz time series with noises and the industrial data demonstrated that the proposed method could extract the fuzzy rules from the data accurately and had good performance for blast furnace gas system modeling.