Search

Understanding the two-phase displacement behaviors of oil and water in porous media under different reservoir development modes for enhanced oil recovery is essential. In this paper, the influence of development measures, such as increasing the injection rate, changing the inlet/outlet position, increasing the water viscosity, and reducing the surface tension coefficient, on oil–water dynamic behaviors was studied using a numerical simulation based on the study of the formation of a high-water-cut channel by water flooding at different injection rates. The results show that blockage and restart occur during displacement in the pore–throat channel and during staggered displacement in different pore channels. With an increase in the injection rate, the recovery increases first and then decreases. All the different development measures can increase the swept area and recovery factor. The recovery factor increases significantly by reducing the surface tension coefficient at medium/high injection rates (≥0.01 m/s) and by increasing the viscosity of the water at low injection rates (

In recent years, the development of deep learning-based remaining useful life (RUL) prediction methods of bearings has flourished because of their high accuracy, easy implementation, and lack of reliance on a priori knowledge. However, there are two challenging issues concerning the prediction accuracy of existing methods. The run-to-failure sequential data and its RUL labels are almost inaccessible in real-world scenarios. Meanwhile, the existing models usually capture the general degradation trend of bearings while ignoring the local information, which restricts the model performance. To tackle the aforementioned problems, we propose a novel health indicator derived from the original vibration signals by combining principal components analysis with Euclidean distance metric, which was motivated by the desire to resolve the dependency on RUL labels. Then, we design a novel self-attention augmented convolution GRU network (SACGNet) to predict the RUL. Combining a self-attention mechanism with a convolution framework can both adaptively assign greater weights to more important information and focus on local information. Furthermore, Gated Recurrent Units are used to parse the long-term dependencies in weighted features such that SACGNet can utilize the important weighted features and focus on local features to improve the prognostic accuracy. The experimental results on the PHM 2012 Challenge dataset and the XJTU-SY bearing dataset have demonstrated that our proposed method is superior to the state of the art.