Method of forecasting quality of molten iron of blast furnace and system thereof

Abstract

The invention discloses a method of forecasting the quality of the molten iron. In the method, a study on the nonlinear correlations between the influence variables of the quality of the molten iron and the quality of the molten iron is firstly conducted using the maximum information coefficient method, and screenings on the influence variables of the quality of the molten iron are conducted based on the above correlation analysis results, and a time-lag relationship of the influence factors and the quality of the molten iron is determined, corresponding history data are extracted from a database and are used as the training set of a forecast model, a dynamic neural network method is used to conduct training for the forecast model to achieve the rolling optimizations on the model, and accordingly the forecast of the quality of the molten iron is achieved. Based on the above forecasting method, the method further uses the LABVIEW and MATLAB hybrid programming to build a set of forecast system of the quality of the molten iron, the forecast on the quality of the molten iron and the monitoring of the quality of molten iron and its relevant influencing variables is achieved, thus a certain guiding function for the field personnel in the operation process of the blast furnace ironmaking is provided.

Description

technical field [0001] The invention relates to the field of automatic control of blast furnace ironmaking industrial process, in particular to a blast furnace molten iron quality forecasting method and system thereof. Background technique [0002] Blast furnace ironmaking is an important link in the steel production process. It is the upstream process of steel production and is the CO 2 The main link of emission is also the process with the largest energy consumption, and there is a huge room for optimization. Controlling the blast furnace to maintain a reasonable furnace temperature and maintaining the long-term stable operation of the blast furnace is the key to achieving high-efficiency, high-quality, and low-consumption blast furnace production. However, the blast furnace is an environment with high temperature, high pressure, strong corrosion, strong interference, coexistence of multiple physical fields, and simultaneous chemical reactions and transfer effects. The bl...

AI Technical Summary

Problems solved by technology

[0007] Most of the methods of the above patents and their related documents are artificially selected based on experience or some linear correlation analysis methods are used to select relevant variables as input for modeling. If the introduced variables have little effect on the silicon content of molten iron that needs to be predicted, it may Bring a certain degree of interference, thus affecting the prediction accuracy

Moreover, the parameters of the model in the above-mentioned patents and related documents are basically fixed after the training is completed. As the running time goes by, the samples change, and the static model cannot adapt to the sample changes, which may lead to a certain degree of prediction accuracy. Decline

Images