An Analysis Method for Influence of BF Operating Parameters on GUR Based on Convergent Cross Mapping Algorithm

Abstract

A blast furnace (BF), which includes a variety of complex physical changes and chemical reactions, has many operating parameters that affect its smelting state. However, the relationship between operation parameters and state parameters is complex and the influence of time delay between operation parameters and state parameters is difficult to be quantified. There-fore, this paper proposes a convergent cross mapping algorithm (CCM) incorporating delay analysis to solve this problem. First, using the CCM algorithm to calculate the correlation coefficients between operation parameters and state parameters, and judge the causal relationship between each operating parameter and state parameter by the convergence of the CCM coefficient. Then, determine the specific delay time and time delay interval of BF operating variables to state variables through fusion time delay analysis. Considering that the selection of embedding dimension has an important influence on the result of causality analysis. This paper uses the cross-validation method to determine the optimal embedding dimension. Finally, by comparing the mechanism analysis and the experimental results, the effectiveness of this method in determining causality and delay time of BF variables is verified.