ABSTRACT
In this paper, the sample data are collected from a blast furnace of Baosteel. Firstly, we divide the train samples basing on the theory of fuzzy C-means clustering (FCM). Then we make using of the multi-LSSVM to train the samples divided. Finally, we combine with the model after trained and the predicted samples to forecast the change of furnace temperature. The simulation result validates the feasibility and effectiveness of the model.
2 FUZZY C-MEANS CLUSTERING
