The application of coke performance prediction of Kunming steel using coking coals blending methods of fuzzy sets metrics
ZHANG Xue-mei
CHENG Shi
WEI Ya-ling
LI Dong
ZHANG Kun-hua
Abstract：Treating vitrinite numeric reflectance distribution and ratio of reactive component and inertinite by coking coals blending methods of fuzzy sets metrics, all the elements in Pseudo-Strength Index (PSI) and Pseudo-Composition Balance Index (PCBI) of blending coal could be calculated, and then, Strength Index (SI) and Composition Balance Index (CBI) could be calculated.By using residual standard deviation, we can know the binary quadratic binomial expression is the best regression equation, taking SI、CBI as the independent variables, and using corresponding coke properties such as CRI, CSR, M40 and M10, the regression analysis was done.Use the MATLAB to make the corresponding visual surface and contour curves and then, predict the coke performance.The Predicted and actual values were compared, we can find that the goodness of fit was relatively high, so it was demonstrated that the fuzzy sets metrics is reliable for prediction of coke performance.
Keywords：coke propertiescoal blending methodcoke reactivitystrength indexcompositional balance index
Publication Date：2015-01-01
Pages：5( 1-5 )