Terrestrial tight oil has extremely strong diagenesis heterogeneity, so a large number of rock thin slices are needed to reveal the real microscopic pore-throat structure characteristics. In addition, difficult identification, high cost, long time, strong subjectivity and other problems exist in the identification of tight oil rock thin slices, and it is difficult to meet the needs of fine description and quantitative characterization of the reservoir. In this paper, a method for identifying the characteristics of rock thin slices in tight oil reservoirs based on the deep learning technique was proposed. The present work has the following steps: first, the image preprocessing technique was studied. The original image noise was removed by filtering, and the image pixel size was unified by a normalization technique to ensure the quality of samples; second, the self-labeling image data augmentation technique was constructed to solve the problem of sparse samples; third, the Mask R-CNN algorithm was introduced and improved to synchronize the segmentation and recognition of rock thin slice components in tight oil reservoirs; Finally, it was demonstrated through experiments that the SMR method has significant advantages in accuracy, execution speed and migration.

—Erosion and sedimentary landforms are linked through sediment transport pathways, which forms a source-to-sink system (S2S). The coupling relationship of different components in the clastic sediment system is emphasized by the S2S concept. A new method for characterizing the sedimentary process of continental rift basins is also provided. It has been proven that there is rich exploration potential in the third member of the Shahejie Formation (Es3) in the Shulu Slope of the Bohai Bay Basin in China but with relatively low production. With the complex structural background of the Es3 of the Shulu Slope, conventional research methods are ineffective in guiding the current development strategies. Therefore, this study adopts the S2S theory, and its elements in the study are characterized using core, logging, and seismic data. The results suggest that the S2S in this study area was supplied by the Ningjin Uplift in the western region, and a fan delta and lake sedimentary systems were formed by the sand transported through valleys and fault troughs. The S2S coupling model, “the Ningjin Uplift sand supply–fault trough, valley transport–fan delta, and shore–shallow lake sedimentation,” is established. This research field permits prediction of sand bodies in continental rift basins with similar structural backgrounds.

Hydrothermal activity is widespread in sedimentary basins, but its dual effects (chemistry and temperature) on shale reservoirs are rarely discussed. In this research, we systematically collected 33 well core samples from Q1 to Q9 units in Gulong Sag, Songliao Basin, and analyzed them using a variety of analytical techniques, including a field emission scanning electron microscopy (FE-SEM), an energy-dispersive spectrometer (EDS), X-ray diffraction (XRD), and stable C-O isotopes. Combined with the collected vitrinite reflectance (Ro), total organic carbon (TOC), and soluble hydrocarbon content data, which is the sum of free oil (pyrolysis S1) and sorb oil content (pyrolysis S2), the results show that (1) Q4 and Q8 units have large amounts of hydrothermal minerals, and its C-O isotope obviously shifts to negative, which implied those units are the main hydrothermal fluid influence area; (2) the hydrothermal activity occurred in the late depositional period of Q1–Q9 units such that its geochemistry has little effect on the proliferation of algae blooms, but its high temperature calculated by δ¹⁸O temperature formulas (around 208 °C) promoted the organic matter maturation process around Q4 and Q8; and (3) the overpressure caused by hydrothermal activity protected the shale reservoir and minimized the decrease in mineral reservoir brittleness index caused by hydrothermal fluid influence. We suggest that the shale reservoir affected by hydrothermal fluid will become a good geology “dessert”, and its upper and/or lower bounds can form an engineering “dessert” due to the precipitation of large amounts of brittle carbonate minerals.

Reservoir evaluation is essential in oil and gas exploration, influencing development decisions. Traditional classification methods are often limited by small sample sizes and low accuracy, restricting their effectiveness. To address this, we propose an intelligent classification method, GA-XGBoost, which integrates Genetic Algorithm (GA) optimization with Extreme Gradient Boosting (XGBoost) to enhance the classification accuracy in small-sample scenarios. The lithological, physical, and lithofacies characteristics of tight sandstone reservoirs are analyzed, and key evaluation parameters—including the mineral composition, porosity, permeability, oil saturation, and logging data (GR, SP, CAL, DEN, AC, LLS)—are selected. After data normalization, the GA-XGBoost model is developed and compared with SVM, XGBoost, and AdaBoost models. The experimental results demonstrate that GA-XGBoost achieves an 88.8% classification precision, outperforming traditional algorithms in both efficiency and accuracy. This study advances experiments on and the standardization of intelligent reservoir evaluations, providing a more reliable classification approach for tight sandstone reservoirs. Additionally, it contributes to the integration of geological exploration and computational intelligence, offering new insights into the application of machine learning in geosciences.