The Heat State Prediction Of Blast Furnace Based On The Stove Heat-index And RBF

Steel industry is the pillar of the national economy and blast furnace is the important part of it. The object of iron making is high quality, low loss, high yield and longevity. In order to achieve this goal, we have to control the blast furnace safely, stably, evenly and regularly.The Silicon content in the molten iron is regarded as a major prediction object based on manufacturing data of blast furnace. The entire model includes the static forecast and dynamic forecast.The static forecast model calculates real-time parameter,mass balance and thermal equilibrium of the blast furnace. We take the linear relations between the stove heat coefficient and the content of silicon in molten iron into account, then choosing three types of the stove heat-index. All this can help predicting the heat state of blast furnace and the temperature of the molten iron.In the department of dynamic forecasts, we analyze the correlation coefficient, the time relationship and the influence factor, then determining input parameters of RBF. We add self-connection neural unit to RBF NN which made RBF NN have the capacity of memorizing the past-time data. It also can enhance the astringency of RBF network in time series and speed up the training time.The features and operation of artificial neural network are introduced in detail. It analyzes the architecture and training algorithm of RBF.Finally, through simulation using Matlab, we respectively established simulation model in the stove heat-index of forecast part and RBF forecast part .After test, we find that the stove heat-index model have a good effect when the furnace conditions is stable.The RBF network has very good auto-adapted effect. It is also better in precision of prediction and speed of training. Taking the advantages of the two methodes at the same time, the content of silicon predicting in molten iron and the heat state of blast furnace are gained, which provide the fine condition to improve the quality of steel and stabilize manufacturing craftwork.