Intelligent prediction of silicon content in hot metal of blast furnace based on neural network time series model

In order to accurately predict the silicon content of molten iron in blast furnace ironmaking process,aiming at the characteristics of nonlinear,time-varying,high-dimensional and large time delay in blast furnace ironmaking process,a sample set for predicting the silicon content of molten iron based on time series was constructed,and the autocorrelation of the silicon content of molten iron in time series was analyzed. The time series weighted moving average method was used to pretreat the sample data,and the neural network time series model was introduced to excavate the quantitative relationship between the silicon content of molten iron at the history time and the silicon content of molten iron at the current time. After the adaptive adjustment of weights and thresholds for several times, the intelligent prediction of the silicon content of molten iron in blast furnace was realized. The simulation test shows that the absolute error of the model is less than 0. 2%, the confidence is about 95%, and the prediction accuracy is high,which can be applied to practice.