Pore space of tight sandstone samples exhibits fractal characteristics. Nuclear magnetic resonance is an effective method for pore size characterization. This paper focuses on fractal characteristics of pore size from nuclear magnetic resonance (NMR) of tight sandstone samples. The relationship between the fractal dimension from NMR with pore structure and water saturation is parameterized by analyzing experimental data. Based on it, a pore structure characterization and classification method for water-saturated tight sandstone and a water saturation prediction method in a gas-bearing sandstone reservoir have been proposed. To verify the models, the fractal dimension from NMR of 19 tight sandstone samples selected from the gas-bearing tight sandstone reservoir of Shahejie Formation in Nanpu Sag and that of 16 of them under different water saturation states are analyzed. The application result of new methods in the gas-bearing tight sandstone reservoir of Shahejie Formation in Nanpu Sag shows consistency with experimental data. This paper has facilitated the development of the NMR application by providing a non-electrical logging idea in reservoir quality evaluation and water saturation prediction. It provides a valuable scientific resource for reservoir engineering and petrophysics of unconventional reservoir types, such as tight sandstone, low porosity, and low permeability sandstone, shale, and carbonate rock reservoirs.

Oil saturation is an important parameter in shale-oil reservoir evaluation. However, due to complex wettability and pore construction, we find that conventional resistivity and nuclear magnetic resonance (NMR) methods do not perform well in calculating oil saturation in shale-oil reservoirs. Hence, we have developed a practical NMR-based method to calculate the oil saturation of the Lucaogou shale-oil Formation, Permian, in Jimusar Sag, Junggar Basin, China. First, we analyze the relationships among the wettability, oil saturation, and T2 distribution based on the theoretical formula and core analysis data. Results indicate that the ratio of the surface area wetted by water and oil is approximately equal to the ratio of water saturation and oil saturation. So we conclude that oil is mainly stored in relatively bigger pores and the surface relaxivity of the oil-wet surface is lower than that of the water-wet surface, resulting in long relaxation signals, that is, the long relaxation signals of NMR T2 spectra of shale-oil reservoirs are primarily attributed to oil signals. We have made a series of NMR measurements of as-received samples and confirm this point. Thus, we propose a T2 cutoff for water and oil to calculate the oil saturation, and we determine 6 ms as the T2 cutoff based on the oil saturation analysis of cores and NMR logs. Finally, we verify and make application of our method and acquire good results.

Tight gas, a category of unconventional natural gas, relies on advanced intelligent monitoring methods for their extraction. Conventional logging for reservoir evaluation relies on logging data and the manual setting of evaluation criteria to classify reservoirs. However, the complexity and heterogeneity of tight reservoirs pose challenges in accurately identifying target layers by using traditional well-logging techniques. Machine learning may hold the key to solving this problem, as it enables computers to learn without being explicitly programmed and manually adding rules. Therefore, it is possible to make reservoir evaluations using machine learning methods. In this paper, the reservoir quality index (RQI) and porous geometric parameters obtained from the optimized inversion of the spherical–tubular model are adopted to evaluate the reservoir. Then, three different machine learning approaches, the random forest (RF) algorithm, support vector machine (SVM) algorithm, and extreme gradient boosting (XGB) algorithm, are utilized for reservoir classification. The selected dataset covers more than 7000 samples from five wells. The data from four wells are arranged as the training dataset, and the data of the remaining one well is designed as the testing dataset to calculate the prediction accuracies of different machine learning algorithms. Among them, accuracies of RF, SVM, and XGB are all higher than 90%, and XGB owns the highest result by reaching 97%. Machine learning based approaches can greatly assist reservoir prediction by implementing the well-logging data. The research highlights the application of reservoir classification with a higher prediction accuracy by combining machine learning algorithms with NMR-logging-based pore structure characterization, which can provide a guideline for sweet spot identification within the tight formation. This not only optimizes resource extraction but also aligns with the global shift towards clean and renewable energy sources, promoting sustainability and reducing the carbon footprint associated with conventional energy production. In summary, the fusion of machine learning and NMR-logging-based reservoir evaluation plays a crucial role in advancing both energy efficiency and the transition to cleaner energy sources.

As the clay film developed in low‐permeability lithologic reservoirs absorbs oil, reservoirs become oil‐wet, which results in abnormally high resistivity oil‐water layers and water layers. Such layers bring great challenges to the identification of oil and water layers. In order to make clear the resistivity response characteristics and conductive mechanism in reservoirs with different wettability, cores were selected in Chang 8 formation, Yanchang group, Upper Triassic, the western Ordos basin of China. Experiments were conducted with these samples to simulate the process of oil displacing water, ageing and water displacing oil. In addition, the contact angles of thin slices after washing oil were also tested. The experimental results show that abnormally high resistivity cores are not completely water‐wet after washing oil. However, the formation factors of abnormally high resistivity cores and normal resistivity cores have little difference. In the process of oil displacing water, the relationship between normal resistivity core resistivity index and water saturation is linear in a log‐log plot, while that of abnormally high resistivity cores is convex. The resistivity of abnormally high resistivity cores remains unchanged during ageing process. Combining the analysis of nuclear magnetic resonance (NMR) T₂ spectra under different conditions, it can be inferred that wettability of abnormally high resistivity cores has become less water‐wet after oil displacing water, whereas the ageing process has little effect on the wettability. In the process of water displacing oil, the relation between abnormally high resistivity core resistivity index and water saturation is almost linear, which shows that the conductive structure of rock is not changed. Based on the analyses and discussions of these experimental results, a hydrocarbon accumulation mode and the corresponding conductive mechanism are proposed for low‐permeability reservoirs with complex wettability. It shows that abnormally high resistivity water layers are caused by the destruction of continuous conductive paths under oil‐wet condition. The researches and experiments are of great importance for understanding the conductive mechanism, identifying oil and water layers and evaluating water flooded formations in low‐permeability reservoirs with complex wettability.