The Pelletizing Disc Control Based On The RBF NN

The pellet production is the important source of blast furnace pudding in the metallurgy raw material. It's position is also getting more and more important, the quality and output of pellet have direct influence to the blast furnace production. However, most domestic pelletization is the manual control process. The operator artificially controls the feed quantity, the watering quantity, the rotational speed ,and so on,various parameters,who relies on the former experience. The efficiency is low , time-consuming is long, and the pellet quality is affected in this kind of artificial operating. In recent years, the development of neural networks has had very big progress,the NN's industrial application has made the essential progress.As the structure of RBF NN is confirmed easily,fast convergence and the capability approaching the nonlinear function very well. In this paper, I adopt the RBF NN as the control model. The choice of quantity and position of hidden layer radial basis functions is very important and directly affects the goodness of fit of overall network approximation capabilities. In this paper, a new optimization algorithm based on genetic algorithm for RBF neural network is presented after traditional algorithms are thoroughly researched.This article makes the intelligent control model based on RBF artificial NN in view of disc pelletizing process realizing the autocontrol. This thesis gives up the complicated mathematic models which are based on the kinetics of pelletization during the investigating. Because these models are deduced from the certain conditions, and they emphasize particularly on the theoretic analysis. In the models every variable is hard to relate to the real producing parameter, and the model coefficient is hard to calculate. This article makes the pelletizing control model based on the RBF NN by way of summing up the mathematic model and manual control experience, the model learns the operator's experiential data, which can control the disc's rotational speed and the watering quantity parameters automatically according to the pelletizing situation and environment , realizing the autocontrol of pelletization.It saves the manpower, material resources,raises the working efficiency,makes the pellet production of more qualify.