Oil production from tight oil reservoirs has become economically feasible because of the combination of horizontal drilling and multistage hydraulic fracturing. Optimal fracture design plays a critical role in successful economical production from a tight oil reservoir. However, many complex parameters such as fracture spacing and fracture half-length make fracturing treatments costly and uncertain. To improve fracture design, it is essential to determine reasonable ranges for these parameters and to evaluate their effects on well performance and economic feasibility. In traditional analytical and numerical simulation methods, many simplifications and assumptions are introduced for artificial fracture characterization and gas percolation mechanisms, and their implementation process remains complicated and computationally inefficient. Most previous studies on big data-driven fracturing parameter optimization have been based on only a single output, such as expected ultimate recovery, and few studies have integrated machine learning with evolutionary algorithms to optimize fracturing parameters based on time-series production prediction and economic objectives. This study proposed a novel approach, combining a data-driven model with evolutionary optimization algorithms to optimize fracturing parameters. We established a significant number of static and dynamic data sets representing the geological and developmental characteristics of tight oil reservoirs from numerical simulation. Four production-prediction models based on machine-learning methods—support vector machine, gradient-boosted decision tree, random forest, and multilayer perception—were constructed as mapping functions between static properties and dynamic production. Then, to optimize the fracturing parameters, the best machine-learning-based production predictive model was coupled with four evolutionary algorithms—genetic algorithm, differential evolution algorithm, simulated annealing algorithm, and particle swarm optimization—to investigate the highest net present value (NPV). The results show that among the four production-prediction models established, multilayer perception (MLP) has the best prediction performance. Among the evolutionary algorithms, particle swarm optimization (PSO) not only has the fastest convergence speed but also the highest net present value. The optimal fracturing parameters for the study area were identified. The hybrid MLP-PSO model represents a robust and convenient method to forecast the time-series production and to optimize fracturing parameters by reducing manual tuning.

Reservoir lithology identification is the basis for the exploration and development of complex lithological reservoirs. Efficient processing of well-logging data is the key to lithology identification. However, reservoir lithology identification through well-logging is still a challenge with conventional machine learning methods, such as Convolutional Neural Networks (CNN), and Long Short-term Memory (LSTM). To address this issue, a fully connected network (FCN) and LSTM were coupled for predicting reservoir lithology. The proposed algorithm (LSTM-FCN) is composed of two sections. One section uses FCN to extract the spatial properties, the other one captures feature selections by LSTM. Well-logging data from Hugoton Field is used to evaluate the performance. In this study, well-logging data, including Gamma-ray (GR), Resistivity (ILD_log10), Neutron-density porosity difference (DeltaPHI), Average neutron-density porosity(PHIND), and (Photoelectric effect) PE, are used for training and identifying lithology. For comparison, seven conventional methods are also proposed and trained, such as support vector machines (SVM), and random forest classifiers (RFC). The accuracy results indicate that the proposed architecture obtains better performance. After that, particle swarm optimization (PSO) is proposed to optimize hyper-parameters of LSTM-FCN. The investigation indicates the proposed PSO-LSTM-FCN model can enhance the performance of machine learning algorithms on identify the lithology of complex reservoirs.

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.

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.

In this paper, a composite sample (VES and SiO₂ nanoparticle) was used to overcome the deficiencies of polymer. The rheological character of the VES/nanoparticles hybrid and flow behavior in porous media were examined. It was found that SiO₂ nanoparticles exhibited viscosifying action and improved the oil tolerance. In addition, the VES solution without nanoparticles showed a lower capacity to recover oil, which might be attributed to the fact that wormlike micelles would be destroyed in crude oil. On the contrary, an enhanced oil recovery of 9.68% was achieved in the composited experiment for the VES sample with nanoparticles which is relatively stable with oil.

In this paper, the weight coefficient of influencing factors for chemical combination flooding is determined by using grey correlation theory. A calculation method for comprehensive evaluation score of chemical combination flooding is established. The influence factor grading and weight grading are combined in this method. By collecting and analysing chemical combination flooding field tests, a prediction method for chemical combination flooding is established by the exponential regression. It reflects the relationship between EOR of the chemical combination flooding and the comprehensive evaluation score. The new method is applied in 6 different reservoirs to evaluate the effect of chemical flooding. The new method determines the weight coefficient of influence factors for chemical combination flooding. It can quickly predict EOR factor of chemical flooding to provide a basis for chemical flooding planning in the field.