Research on pore structure and classification evaluation of tight oil reservoirs based on fractal theory

The Chang 8 formation of the Yanchang Group, located in Yuancheng area of the Ordos Basin, is a typical tight oil reservoir in China. This reservoir is characterized by low porosity, low permeability, strong non-homogeneity, and significant difficulty in evaluating the reservoir parameters. To examine and investigate the microscopic pore structure characteristics of the Chang 8 formation, cast thin section and scanning electron microscopy techniques were utilized in this study, Moreover, tests on the core physical properties were conducted and the data from these tests were integrated into the analysis of basic characteristics of the reservoir rock mineralogy, pore permeability, and other fundamental characteristics. The shapes of piezomercury curves were systematically examined to study the characteristics and features of pore structures for the 17 samples. In accordance with the fractal dimension of the NMR T₂ spectrum, the reservoir was classified into four categories, and a conversion model delineating the correlation between the NMR T₂ spectrum and the capillary pressure curve was formulated through the application of the segmented power function method. This model was then implemented in the interpretation of NMR logging, facilitating the acquisition of a seamless pseudo-capillary pressure curve spanning the entire well section. Three essential parameters reflecting the microscopic pore structure, namely the expulsion pressure, median pressure, and sorting coefficient of the core samples, were extracted. The association between reservoir parameters and reservoir categorization was then determined through the application of a generalized regression neural network. The pseudo-capillary pressure curve reservoir parameters of the whole well section were processed to derive the classification profile of the reservoir, and the classification results demonstrated a strong alignment with those of the mercury injection experiments. This study highlights that the proposed method can provide crucial foundation for the investigations on pore structures in tight oil reservoirs and the evaluation of reservoir classification.