Method for predicting coke quality and optimizing coal blending ratio for tamping coking

Abstract

The invention relates to a method for predicting coke quality and optimizing a coal blending ratio for tamping coking. The technical scheme of the invention comprises the steps: 1, building a single coal quality index database and a coke quality index database; 2, building a cooperation coal quality index prediction model; 3, building a coke quality index prediction model through employing the technology of support vector machines; 4, building a fitness mathematic model; 5, searching an optimal coal blending ratio through a dynamic vaccine extraction immunization genetic algorithm and the fitness mathematic model; 6, carrying out coal blending according to the optimal coal blending ratio, and carrying out charging and coking after tamping. The method employs the technology of support vector machines for modeling, is high in model prediction speed, is higher in prediction precision, can quickly obtain high coking quality and low-cost coal blending ratio through the dynamic vaccine extraction immunization genetic algorithm, achieves quick optimization of the coal blending ratio, improves the prediction precision of coking quality, improves the coking quality, and reduces the coal blending cost.

Description

technical field [0001] The invention belongs to the technical field of optimizing coal blending ratio. Specifically relates to a method for predicting coke quality and optimizing coal blending ratio for tamping coking. Background technique [0002] my country is rich in coking coal resources, but high-quality coking coal resources are tight and extremely unevenly distributed. As a method that can increase the amount of high-volatile, weakly caking or even non-caking coal in coal blending to expand the raw coal for coking, the stamping coking process has become a mature coking process and is widely used at home and abroad. use. [0003] The large-scale blast furnace and the development of oxygen-enriched coal injection technology put forward higher requirements for the quality of coke. Moreover, with the increase in the number of small coal mines, the types of single coal used in coking plants have increased and coal blending schemes have changed frequently. How to quickly...

AI Technical Summary

Problems solved by technology

The disadvantage of using this method is that the prediction of coke quality is slow, the accuracy is not high and the coke quality cannot be improved

This method has high calculation accuracy and fast convergence speed, and can minimize the cost of coal blending while ensuring that the coke quality meets the needs of users. Did not achieve the purpose of improving coke quality

[0008] Although the above-mentioned coke quality prediction method and coking-coal blending ratio optimization method have many advantages, they still have the following deficiencies. First, they do not consider the control of a large amount of low-metamorphic coal and high-metamorphic coal blending during tamping coking coal blending; two. The first is that it does not consider the guiding role of the vitrinite random reflectance distribution interval in coal blending and coking; the third is that it considers the accurate prediction of coke quality without considering the cost of coal blending; the fourth is that it considers the maximum reduction of coal blending costs without Consider the problem of improving coke quality

Images