Research Of Status Prediction And Modeling Of Gas Flow Distribution Relationship For Blast Furnace

Abstract:Blast furnace (BF) ironmaking is a key link in the process of iron and steel smelting. The stable and smooth BF status is a guarantee for the optimal BF operation and BF-life. Meanwhile, to adjust and confirmed the reasonable distribution of the gas flow is always the main goal of the BF operation. Since the actual BF production process is of high temperature, dust, airtight and complexity characteristics, the inner states of the BF are hard to measure directly, which brings the great of difficulties to judge the status and the gas flow distribution of BF effectively.Firstly, aiming at the difficulty of predicting BF status, a prediction method for the BF status is proposed based on D-S evidence theory, combining fuzzy expert inference and posterior probability least squares support vector machine. Then, considering the mechanism and expert experience of BF smelting, the subjective evidences are extracted based on the fuzzy expert reasoning, while a posterior probability least squares support vector machine model is developed to extract objective evidences. And the subjective and objective evidences are fused based on the D-S evidence theory, so as to predict the BF status precisely.Secondly, a relation model for CO utilization and gas flow distribution is based on least squares support vector regression, which is aimed at solving the problem of the lack of effectively method for modeling the BF production index and gas flow distribution. The factors associated with BF gas flow distribution modeling are determined by the analysis of mechanism and the grey correlation analysis method. Then the characteristics of the various factors, which are the input and output parameters of the relation model, are extracted based on image processing, charging model and the statistical analysis. Based on the above, the relation model of CO utilization and gas flow distribution is established based on the least squares support vector regression, whose parameters are determined by cross validation method. Finally, the models of BF status prediction and gas flow distribution relationship are simulated based on the actual operation data of1#BF process of a steel plant, which are built in this thesis. And the simulation results indicated that the proposed methods are effective and reliable, so they can be used to provide theoretical support for blast furnace operation and production index optimization, promote the BF energy saving, improve the economic benefits of iron and steel enterprises.