Sequence division and coal facies research methods of the deep No. 8 coal seam in the Daning-Jixian Block, eastern margin of Ordos basin.

With the large-scale commercial development of deep coalbed methane in the Ordos Basin, the refined research on deep coal seams (with a burial depth > 2000 m) has begun to attract much attention. In this study, the No. 8 coal seam of the Taiyuan Formation in the Daning-Jixian Block on the eastern margin of the Ordos Basin was taken as the research object. By combining sequence stratigraphy theory and coal facies research methods, and using the Support Vector Regression (SVR) model optimized by the Sparrow Search Algorithm (SSA) as well as the clustering algorithm, SSA-SVR models based on logging parameters for calculating the vitrinite-inertinite ratio (V/I ratio) and the Texture Preservation Index (TPI) value, along with a coal facies division scheme suitable for the Daning-Jixian Block, was established. The results demonstrate that the unbiased Mean Absolute Percentage Error (MAPE) of the V/I-SSA-SVR model is 18.73%, the unbiased Root Mean Square Error (RMSE) is 1.36, and the unbiased Coefficient of Determination (R²) is 88.69%. For the TPI - SSA - SVR model, the unbiased MAPE is 17.90%, the unbiased RMSE is 1.12, and the unbiased R² is 86.39%.Through clustering analysis of nearly 5000 sets of V/I ratios and TPI values calculated by SSA-SVR models in the study area, The TPI values and V/I ratios were each divided into four categories. By combining these two parameters, a classification scheme for coal facies research in the Daning-Jixian Block was finally developed. Based on the modeling results, this study proposed a sequence stratigraphy division method for the deep No. 8 coal seam in the Daning-Jixian Block. By using the V/I-SSA-SVR model, the vertical V/I ratios of the No. 8 coal seam in individual wells was calculated. Combined with lithological changes, the water depth transition interfaces were identified. Based on this method, the Daning-Jixian Block can be divided into 5 fifth-order sequences. Furthermore, by combining the TPI-SSA-SVR model and the established coal facies classification scheme, this study finally reconstructed the coal-forming environments of each fifth-order sequence during the formation of the No. 8 coal seam in the Daning-Jixian Block. For the first time, this study conducted refined research on China's first commercial breakthrough block for deep coalbed methane, filling the gap in the research on deep coal sequences and coal facies. It has established coal facies parameter boundary values and a division scheme exclusive to the deep No. 8 coal seam in the Daning-Jixian Block, innovatively proposed the "V/I-SSA-SVR sequence division method", and provided guidance and reference for subsequent exploration and development of deep coalbed methane.