Prediction modeling study for blast furnace hot metal temperature based on T-S fuzzy neural network model

According to the shortage of the positively relation of blast furnace temperature and hot metal silicon rather than strict linear relationship, the subjectivity of the modelling by mechanism, and mechanism model difficultly to establish the connotative mathematic relation between every variables, the mass of the sample data was processed through preprocessing and feature extraction based on the theory of data mining, and then blast furnace hot metal temperature was deemed as the research object, blast furnace hot metal temperature prediction model was established based on T-S fuzzy neural network. And the model and T-S fuzzy regression model and BP neural network model were compared; Through using some blast furnace data for model test, simulation results show that T-S fuzzy neural network model performance is superior to other models.