Total organic carbon (TOC) content is an important parameter for evaluating the abundance of organic matter in, and the hydrocarbon production capacity, of shale. Currently, no prediction method is applicable to all geological conditions, so exploring an efficient and accurate prediction method suitable for the study area is of great significance. In this study, for the shale of the Qingshankou Formation of the Gulong Sag in the Songliao Basin, TOC content prediction models using various machine learning algorithms are established and compared based on measured data, principal component analysis, and the particle swarm optimization algorithm. The results showed that GR, AC, DEN, CNL, LLS, and LLD are the most sensitive parameters using the Pearson correlation coefficient. The four principal components were also identified as input features through PCA processing. The XGBoost prediction model, established after selecting the parameters through PSO intelligence, had the highest accuracy with an $R^2$ and RMSE of 0.90 and 0.1545, respectively, which are superior to the values of the other models. This model is suitable for the prediction of TOC content and provides effective technical support for shale oil exploration and development in the study area.

With the rapid development of digital core technology, the acquisition of high-resolution rock thin section images has become crucial. Due to the limitation of optical principles, thin section imaging involves a contradiction between resolution and field of view. In order to solve this problem, this paper proposes a lightweight, fully aggregated network with multi-branch structure for super resolution of rock thin section images. The experimental results on the rock thin section dataset demonstrate that the improved method, called OmniSR-M, achieves significant enhancement compared to the original OmniSR method and also surpasses other state-of-the-art methods. OmniSR-M effectively recovers image details while maintaining its lightweight nature. Specifically, OmniSR-M reduces the number of parameters by 26.56% and the computation by 27.66% compared to OmniSR. Moreover, this paper quantitatively analyzes both the facies porosity rate and grain size features in the application scenario. The results show that the images generated by OmniSR-M successfully recover key information about the rock thin section.

With the deepening of oilfield development, logging data proliferate, and their complexity makes manual stratigraphic division both difficult and time-consuming. Aimed at the current network model widely used to solve the problem of stratigraphic delineation, which has problems such as not considering the multi-scale features of logging curves and insufficient accuracy, the YOLOv8x target detection algorithm in deep learning is utilized to detect the target strata, which has the ability to characterize the multi-scale features and can improve the efficiency and accuracy of the division. In order to better localize and identify targets, this paper proposes a new stratigraphic automatic division method, YOLOv8x-CAMDP, which introduces a CA (Coordinate Attention) mechanism module into the original YOLOv8x model to improve the model’s ability to identify stratigraphic interval boundaries. In addition, the CIOU loss function in the original YOLOv8x network model was replaced using the MDPIOU loss function to effectively improve the accuracy and efficiency of bounding box regression. Based on the logging data from the Xing 10 area pure oil zone, a thorough comparison of the YOLOv8x-CAMDP and YOLOv8x models’ training results is presented. The YOLOv8x-CAMDP model achieves a mean Average Precision (mAP) value of 98.7%, outperforming the YOLOv8x model by one percentage point. Moreover, the YOLOv8x-CAMDP model demonstrates greater precision in boundary division for each stratigraphic interval. The application of the YOLOv8x-CAMDP model to project implementation achieved significant results in stratigraphic division, reduced workload, and optimized manual division. These results not only confirm the practical value of the YOLOv8x-CAMDP model but also demonstrate the prospect and potential of its wide application.