A well-testing method for parameter evaluation of multiple fractured horizontal wells with non-uniform fractures in shale oil reservoirs

The precondition of well testing interpretation is to determine the appropriate well testing model. In numerous attempts in the past, automatic classification and identification of well testing plots have been limited to fully connected neural networks (FCNN). Compared with FCNN, the convolutional neural network (CNN) has a better performance in the domain of image recognition. Utilizing the newly proposed CNN, we develop a new automatic identification approach to evaluate the type of well testing curves. The field data in tight reservoirs such as the Ordos Basin exhibit various well test models. With those models, the corresponding well test curves are chosen as training samples. One-hot encoding, Xavier normal initialization, regularization technique, and Adam algorithm are combined to optimize the established model. The evaluation results show that the CNN has a better result when the ReLU function is used. For the learning rate and dropout rate, the optimized values respectively are 0.005 and 0.4. Meanwhile, when the number of training samples was greater than 2000, the performance of the established CNN tended to be stable. Compared with the FCNN of similar structure, the CNN is more suitable for classification of well testing plots. What is more, the practical application shows that the CNN can successfully classify 21 of the 25 cases.

An improved method for predicting CO2 minimum miscibility pressure based on artificial neural network

Asphaltene deposition is a common phenomenon during CO₂ flooding in ultralow permeability reservoirs. The deposited asphaltene occupies the pore volume and decreases permeability, resulting in serious formation damage and pore well productivity. It is urgent to investigate the asphaltene deposition mechanisms, adverse effects, and preventive measures. However, few asphaltene deposition investigations have been systematically conducted by now. In this research, the asphaltene precipitation mechanisms and adverse effects were comprehensively investigated by using experimental and numerical methods. To study the effects of pressure, asphaltene content, and temperature on asphaltene precipitation qualitatively and quantitatively, the microscope visible detection experiment and the PVT cell static experiment were firstly conducted. The adverse effects on porosity and permeability resulted from asphaltene deposition were also studied by the core flooding experiment. Secondly, simulation models of asphaltene precipitation and deposition were developed and validated by experimental data. Finally, a case study from Changqing oil field was presented to analyze the asphaltene deposition characteristic and preventive measures. The experimental results showed that the asphaltene precipitation increases with the increased pressure before reaching the minimum miscible pressure (MMP) and gets the peak value around the MMP, while decreases slowly. The asphaltene precipitation increases with the increased temperature and asphaltene content. The variation trend of adverse effects on porosity and permeability resulted from asphaltene deposition is similar to that of asphaltene precipitation under the influence of pressure, asphaltene content, and temperature. The case study shows that the water-altering-gas (WAG) with high injection rate suffers more serious asphaltene deposition compared with the WAG with low injection rate, for the asphaltene precipitation increases as the increased pressure before reaching the MMP. The CO₂ continuous injection with high injection rate is the worst choice, for low sweep efficiency and the most severe formation damage. Thus, the WAG with optimal injection rate was proposed to maintain well productivity and to reduce formation damage resulted from asphaltene deposition during developing ultralow permeability reservoirs.