Make rational use of coal resources to improve the level of coal blending for coking

Thread Content

:) :) Coal rock blending technology is a technical theory and method based on coal petrology principles; it makes use of coal petrological testing methods and parameters to guide the management of coal used as raw material for coking. By adjusting other coal and coking process parameters, this technology enables the prediction and control of coke quality, thereby helping to steadily improve coke quality, make rational use of coal resources, and reduce production costs. The development of coal blending technology has led to the establishment of several recognized basic principles: first, coal is a heterogeneous material, a mixture of complex organic and inorganic substances. The properties of organic matter in coal vary, and thus their roles in coal blending also differ. Therefore, each type of coal is a natural blend of coals. Based on the changes that occur in coal during heating, those components of its organic matter that can melt and form active bonds during heating are considered to be cohesive active components ； Components that cannot be melted by heating and do not form active bonds are considered inert elements with no adhesive properties. Second, the quality of the active components in the coal is not uniform, which can be illustrated by a reflectance distribution graph. The quality of active ingredients can vary greatly; not only are there significant differences among different types of coal, but even within the same type of coal, the quality of its active ingredients can differ considerably. If the composition of the active components of different properties (referring to the vitrinite fraction) contained in a single coal is expressed in terms of reflectance, then the reflectance curve of the active components for each type of coal generally follows a normal distribution. This makes the reflectance distribution of coal vitrinite the only effective method for identifying blended coals. Thirdly, inert components, just like active components, are essential elements in coal blending; their content is another important indicator that determines the properties of the blended coal. Any reasonable coking coal blending scheme is a combination of active components of different qualities and quantities, along with an appropriate amount of inert components. Fourth, during the coking process, it is not a process in which coal particles melt together to form homogeneous coke; rather, it is through interfacial reactions and bonding between the coal particles that they come together to form coke lumps. (1) The application of coal petrology in the procurement management of coking coal. Currently, when choosing coal supply bases, coking plants take into account not only the compatibility among different types of coal but also their differences in cohesion, volatile matter, ash content, and sulfur content. In fact, the quality of caking and the level of volatiles are primarily determined by the degree of coal metamorphism and the lithofacial composition; sometimes, the degree of coal reduction can also cause abnormal caking properties. The ash content in coal originates mainly from the minerals present in it, and the degree of coal separability depends on the different symbiotic relationships between these minerals and the organic components. Furthermore, coal from the same coal supply base can exhibit very significant differences in properties due to its origin from different coal seams. By employing coal petrological methods and taking into account the geological characteristics of the coal fields from which coal is supplied, it is possible to identify coal supply bases that offer suitable coal types, stable coal quality, and reliable supply. (2) The application of coal petrology in the daily management of coking coal. Currently, many coal suppliers deliver coal by vehicle or from loading platforms, and their sources of coal are quite complex. Coal from different batches of the same manufacturer can sometimes exhibit significant differences in their petrographic characteristics, even though conventional tests indicate that they belong to the same type of coal. As a result, they play different roles in coal blending, and treating such coals as a single type has a very negative impact on the quality of coal used for coking. By conducting quantitative analysis of the microcomponents and measuring the vitrinite reflectance of the raw coal entering the plant, it is possible to determine whether each batch of coal is stable and normal. Among them, the vitrinite reflectance histogram can very intuitively reflect the coal blending phenomenon, and the approximate amount of abnormal coal mixed in can be determined from the graph ； Changes in the quantitative statistical results of microcomponents can also reflect changes in the coal source. (3) Application of coal petrology in coal yard management. Due to the significant increase in coke production capacity, there is a shortage of coking coal, and it is quite common for coking plants to use dozens of different types of coal. Due to the limitations in coal classification and the complexity of coal, using the simple method of stacking coking coal according to its classification grades results in a situation where the coal blending ratio remains unchanged while the quality of the coke changes significantly. To ensure stable iron production, it is necessary to operate at the minimum quality level for coke. The solution is to stack the coals according to the principle that the vitrinite reflectance of individual coal types is generally consistent, that the areas enclosed by their reflectance distribution graphs largely overlap, and that their caking properties and coking characteristics are similar. This approach helps to maintain stable coke quality, thereby enabling rational use of coal resources and reducing the costs associated with coal blending. (4) Application of coal petrology in optimizing coal blending. Studies show that, depending on the coal type structure of the individual coals used in blending, the characteristics of the reflectance distribution diagram for the optimal coal blend also vary. Therefore, the characteristics of the blended coal reflectance distribution map should also be used as an indicator for controlling coal blending. The specific approach is as follows: based on the reflectance distribution of each individual coal type as specified in the coal blending plan, a weighted average is used to determine the reflectance distribution of the blended coal. Efforts are made to adjust this distribution to match that of an ideal blended coal; it is important to avoid a reflectance distribution with significant dips, as this can ensure a smooth transition in the plasticity of the blended coal during coking and facilitate the formation of an appropriate optical microstructure in the coke, thereby optimizing its microscopic structure. The application of this method has enabled many enterprises to improve their coal blending quality, stabilize coke quality, and reduce the amount of high-quality coking coal used. (5) Application of coal petrology in predicting coke quality. Due to the numerous coal and rock parameters, the complexity of data processing, as well as the nonlinear relationship between coal quality and coke quality, it is difficult to meet the requirements of current coal blending using traditional empirical coal blending methods, empirical formulas, and linear models. The use of artificial neural networks and genetic algorithms in combination with computer technology, along with the collection of large amounts of data from coal blending and coking experiments, and the application of these new data processing methods to develop predictive models for coke quality, has been a major feature of coal and rock blending efforts in recent years. This coal blending method, which integrates coal-rock coal blending theory, modern mathematical processing techniques, and computer technology, has advanced coal blending technology from an empirical and qualitative stage to a new stage of science-based and numerically quantitative analysis. The use of coal-rock indicators has significantly improved the accuracy in predicting the coal blending level and coke quality parameters.