Sinter Quality Forecast System Based On The Integrated Model And Industrial Applications

As the main raw material for blast furnace, the quality of sinter has a direct influence on the state of blast furnace and quality of final products, sintering process is essential in the procedure of the iron and steel works. Since the sintering process is a dynamic system with long circuit, multivariable and complex mechanism, and the sinter quality inspection has a large time delay, it is hard to give a assessment to the sintering process according to the quality inspection results. It is meaningful to predict the sinter quality precisely by developing a sinter quality prediction system.On the basis of the mechanism and characteristics of the sintering process, to solve the problem that the sinter quality can not be precisely predicted by only using proportion information without the sintering state and operation parameters, A kind of integrated prediction model based on grey system theory and radial basic function (RBF) neural network is put forward in the paper. The main study achievements include:Firstly, proportion information, statistical values of sintering state parameters, operational parameters and sinter quality detection data are analysed and key influence factors of sinter quality are worked out, afterwards, the RBF neural networks model is created based on the filtered data of key influence factors and sinter quality detection data. Then, the equal dimensional new information GM(1,1) prediction model is established on the basis of the detection data of sinter quality. Finally, with the information entropy theory, the integrated model which combine the GM(1,1) model and RBF neural networks model are created to predict the accurate values of ferrum content (TFe), alkalinity (R) and drum index (Ro).On the basis of history data, the strategy proposed for modeling is validate effective in this paper. Meanwhile, in order to verify its value of application prospect, Sinter quality prediction system is developed based on a sintering process in one domestic iron and steel enterprise. The online running results shows that:the system achieve the exact prediction of sinter quality, and give a accurate guide information to the optimization control of the sintering process.