Keze Lin

For over four decades, carbon dioxide (CO₂) has been instrumental in enhancing oil extraction through advanced recovery techniques. One such method, water alternating gas (WAG) injection, while effective, grapples with limitations like gas channeling and gravity segregation. To tackle the aforementioned issues, this paper proposes an upgrade coupling method named alkaline-surfactant-polymer alternating gas (ASPAG). ASP flooding and CO₂ are injected alternately into the reservoir to enhance the recovery of the WAG process. The uniqueness of this method lies in the fact that polymers could help profile modification, CO₂ would miscible mix with oil, and alkaline surfactant would reduce oil–water interfacial tension (IFT). To analyze the feasibility of ASPAG, a couples model considering both gas flooding and ASP flooding processes is established by using the CMG-STARS (Version 2021) to study the performance of ASPAG and compare the recovery among ASPAG, WAG, and ASP flooding. Our research delved into the ASPAG’s adaptability across reservoirs varying in average permeability, interlayer heterogeneity, formation rhythmicity, and fluid properties. Key findings include that ASPAG surpasses the conventional WAG in sweep and displacement efficiency, elevating oil recovery by 12–17%, and in comparison to ASP, ASPAG bolsters displacement efficiency, leading to a 9–11% increase in oil recovery. The primary flooding mechanism of ASPAG stems from the ASP slug’s ability to diminish the interfacial tension, enhancing the oil and water mobility ratio, which is particularly efficient in medium-high permeability layers. Through sensitivity analysis, ASPAG is best suited for mid-high-permeability reservoirs characterized by low crude oil viscosity and a composite reverse sedimentary rhythm. This study offers invaluable insights into the underlying mechanisms and critical parameters that influence the alkaline-surfactant-polymer alternating gas method’s success for enhanced oil recovery. Furthermore, it unveils an innovative strategy to boost oil recovery in medium-to-high-permeability reservoirs.

Polymer flooding has proved an effective technique to improve development efficiency in heterogeneous reservoirs. Previously, operators usually employed continuous injection of large polymer slugs. However, problems existed including premature injection profile reversal during development, ineffective circulation of polymer solution in low-permeability zones, and excessive polymer consumption in the late development stage. The paper proposed a combination method, named as polymer-alternating-water (PAW), to solve the issues of traditional polymer flooding. The characteristic of this method is injecting water slugs between polymer slugs during the polymer flooding process. At present, research on the PAW technique is limited, with the operational parameters, oil recovery mechanisms, and applicability under various reservoir conditions remaining unclear. In this work, a numerical polymer flooding model is developed using the commercial CMG-STARS reservoir simulation module to investigate the oil displacement performance of PAW. Numerical simulations are performed to determine the optimal parameters for maximizing oil recovery factor. The results indicated compared with continuous polymer flooding, PAW shows multiple peaks in daily oil production rate, with a significantly decreased decline rate, resulting in a 1.47% increase in recovery factor. Under different reservoir conditions including mean permeability, heterogeneity, and crude oil viscosity, PAW can achieve further improvements on the basis of continuous polymer flooding. Polymer adsorption and injection concentration significantly impact the recovery factor, requiring further optimization for field applications. In this study, better polymer flooding performance was achieved when the number of alternating cycles of PAW was 2, and the injected alternating water slug volume was 50%. This study contributes to a deeper understanding of the key mechanisms and parameters in PAW enhanced oil recovery process, thereby providing guidance for the further optimization and field application of this technique.

The dynagraph card plays a crucial role in evaluating oilfield pumping systems’ performance. Nevertheless, classifying dynagraph cards can be quite difficult because certain operating conditions may exhibit similar patterns. Conventional classification approaches mainly involve labor-intensive manual analysis of these cards, leading to subjectivity, prolonged processing times, and vulnerability to human prejudices. In response to this challenge, our study introduces a novel approach that leverages transfer learning and the Swin Transformer model for classifying dynagraph cards across various operating conditions in rod pumping systems. Initially, the Swin Transformer model undergoes pre-training using the ImageNet-22k dataset. Subsequently, we fine-tune the model’s weights using actual dynagraph card datasets, facilitating direct classification analysis with dynagraph cards as input variables. The adoption of transfer learning significantly reduces the training time while enhancing the accuracy of condition diagnosis. To assess the effectiveness of our proposed method, we conducted a comparative evaluation against conventional models like ResNet50, DenseNet121, LeNet, and ViT. The findings demonstrate that our approach outperforms other methods, achieving an accuracy of 96%, thereby improving classification accuracy by 3–4%. Therefore, our approach, based on transfer learning and the Swin Transformer model, provides a better solution for practical problems involving similar dynagraph cards. It meets the requirements of oil field operations, enhancing economic benefits and work efficiency.

Hydraulic fracturing is a crucial method in shale oil development, and predicting production after hydraulic fracturing is one of the challenges in shale oil development. Conventional methods for predicting production include analytical methods and numerical simulation methods, but these methods involve many parameters, have high uncertainty, and are time-consuming and costly. With the development of shale oil development, there are more and more sample data on the geological parameters, engineering parameters, and development parameters of shale oil hydraulic fracturing, making it possible to use machine learning methods to predict production after hydraulic fracturing. This article first analyzes the impact of different parameters on initial production and recoverable reserves based on field data from Chang-7 shale oil in the Ordos Basin of China. Then, using the Particle Swarm Optimization (PSO) algorithm and the Gradient Boosting Decision Tree (GBDT) algorithm, machine learning models for initial production and recoverable reserves are established. The Local Interpretable Model-Agnostic Explanations (LIME) and Shapley Additive exPlanations (SHAP) explanation methods are used to explain the models. The study found that initial production is highly correlated with parameters such as the number of fracturing stages and fracturing fluid volume, while recoverable reserves are significantly related to parameters such as well spacing, area, and reserver-controlled. The PSO-GBDT model established in this study has an accuracy of over 85% and can be used for production prediction and subsequent parameter optimization research. By comparing the LIME and SHAP local explanation methods, it is shown that different explanation methods can obtain reasonable and credible local explanation results. This article establishes a high-precision shale oil well production prediction model and two model interpretation methods, which could provide technical support for shale oil well production prediction and production analysis.