Classification Evaluation Method for Complex Reservoirsin Xinglong Structural Belt

The reservoir of Linhe formation in the Xinglong tectonic belt of the Hetao basin is controlled by low geothermal temperature and late deep burial, with obvious weak diagenetic characteristics, strong heterogeneity of the reservoir, and complex pore structure. With the continuous deepening of exploration, oil testing data shows that some oil testing layers have similar or identical physical properties, but there are significant differences in production. The applicability of reservoir classification evaluation methods based on traditional pore and permeability parameters is limited and the method cannot meet the precise evaluation needs of reservoir classification. This article combines nuclear magnetic resonance (NMR) and conventional logging data to comprehensively utilize the geometric mean of NMR T2 spectra, different pore ratios, conventional porosity, permeability, mud content and other parameters that characterize reservoir pore structure. Parameters such as reservoir pore structure evaluation index, reservoir quality factor index, and macroscopic effective pore permeability index are constructed. The application of pore structure evaluation index and macroscopic effective pore permeability index has formed a reservoir classification evaluation method with nuclear magnetic pore structure evaluation as the core. Using reservoir quality factor index and macroscopic effective pore permeability index, a reservoir classification method based on machine learning algorithm and multi parameter fusion as the core has been established, effectively solving the problem of low accuracy in reservoir classification evaluation due to the lack of nuclear magnetic logging. The established complex reservoir grading evaluation method has been applied on a large scale in the Linhe formation of the Xinglong tectonic belt, effectively guiding the oil layer testing and selection in the area, and achieving good application results.