Dynamic model construction method for coal quality control in complex-structural deposits

Introduction. Sustainable development of the coal mine is inextricably linked to quality control of the coal produced. Three main factors influence coal quality: geological, technological and industrial. The process of these factors influencing is varied: mean values of field quality indicators are generally known from exploration data, and during coal mining quality values may change. Keeping the required production balance in terms of the quality of coal produced is an immediate task of the coal-mining enterprise in terms of the strategy of complex development of solid mineral deposits. It should be noted that in scientific works concerning the quality of mined mineral, the question of the reliability and consistency of incoming geological exploration and mining-technological information and its interpretation was not considered sufficiently thoroughly. Materials and methods of research. In order to solve the existing problem of changing the quality of coal extracted, a multi-level model structure is being created, including coal quality map models, models for presenting knowledge of the coal mining process, а first approximation model for monitoring the quality of coal mined and a probability-graphical model for clarifying existing new knowledge on the uncertainty and unreliability of data presented in the form of the Bayesian network. Research results. On the basis of the created graph representing the change in the quality of coal extracted, the production problem was solved in evaluating the definition and understanding of which indicators have influenced the change in the quality of coal extracted in the form of the Bayesian network and the Algebraic Bayesian network. They are an acyclic graph with the top «Influence of major mining and technological indicators on the quality of coal extracted», which in turn determines the next level of peaks «Numerical estimation (obtained from the previous stage of model construction), allowing to determine the allowable level of extracted coal» and «Quality of mining operations» a set of these representations is fundamental for the complex indicator of quality of extracted coal. In addition to these indicators, vertices are introduced about the existence of certain deviations. As a result, the post-test probability of estimating the influence of the change in the allowable level of coal and the quality of mining operations on the complex indicator of extracted coal is P (Hi | E) = 0.025. Discussion. We can conclude that the posterior probability P (Hi | E) = 0.025 is less than a priori P (Н1) = 0.057 because it includes two basic indicators of express testing and dynamic geometry. This makes it possible to speak about the influence of newly obtained information on the complex indicator of the extracted coal, which appears as the front of exploration and exploitation works to produce real-time mining geometric models of the deposit. These, in turn, serve as the basis for operational planning of coal production in terms of quantity and quality. Conclusion. The study found that the quality of mining operations has a significant impact on the integrated quality of coal produced, which in turn forms the main indicator of dynamic geometry. The study is complicated by the fact that there is a need to establish the regularities of the spatial placement of components and their variability at the spent and developed sites of the deposit. Resume. The article presents the results of research in the form of a method of construction of dynamic model of quality management of mined coal on complex-structural deposits with consideration of existing data and knowledge uncertainties. The application of the Bayesian network has been found to be useful for a more detailed study and refinement of existing new knowledge on the process of quality management of extracted coal under conditions of uncertainty and unreliability. With the help of the Bayesian network, it has been determined that the complex quality of coal produced is influenced by the quality of mining operations and numerical estimates of the acceptable quality level of coal extracted. The study revealed that the quality of mining operations has a significant impact on the complex indicator of the quality of extracted coal, which in turn forms the main indicator of dynamic geometry. The study is complicated by the fact that there is a need to establish the regularities of the spatial placement of components and their variability at the spent and developed sites of the deposit. The results of the research may be useful in creating the next level of dynamic model, namely the development of the Bayesian network for a real-time operational quality management process for coal production, which is one of the components of sustainable development of the coal mining enterprise. Proposals for practical application and direction for future research. Further research and application of the results of the work should be continued towards the creation of the next level of dynamic model for the development of the Bayesian network for the process of operational quality management of hard-to-obtain coal structural coal deposits in real mode. Funding: The work was carried out within the framework of the State Order No. 075-00412-22 PR. Theme 1 (2022-2024). Methodological foundations of the strategy for the integrated development of reserves of solid mineral deposits in the dynamics of the development of mining systems (FUWE-2022-0005), reg. №1021062010531-8-1.5.1.