Machine learning technology is becoming increasingly prevalent in the petroleum industry, especially for reservoir characterization and drilling problems. The aim of this study is to present an alternative way to predict water saturation distribution in reservoirs with a machine learning method. In this study, we utilized Long Short-Term Memory (LSTM) to build a prediction model for forecast of water saturation distribution. The dataset deriving from monitoring and simulating of an actual reservoir was utilized for model training and testing. The data model after training was validated and utilized to forecast water saturation distribution, pressure distribution and oil production. We also compared standard Recurrent Neural Network (RNN) and Gated Recurrent Unit (GRU) which are popular machine learning methods with LSTM for better water saturation prediction. The results show that the LSTM method has a good performance on the water saturation prediction with overall AARD below 14.82%. Compared with other machine learning methods such as GRU and standard RNN, LSTM has better performance in calculation accuracy. This study presented an alternative way for quick and robust prediction of water saturation distribution in reservoir.

Machine learning is becoming prevalent increasingly for reservoir characteristics analysis in the petroleum industry. This investigation proposes an alternative way for evaluating interwell connectivity in oil fields utilizing machine learning. In this study, three-dimensional convolutional neural network (CNN) was utilized to establish a deep learning model, which can invert interwell connectivity combining with dynamic production data. Different from traditional methods that try to construct mathematical formulas to calculate the connectivity among wells basing on physical laws, deep learning model can capture autonomously the changing characteristics of dynamic production data by training continuously and provide a potential to characterize the interwell connectivity accurately without physical model. At the same time, the back propagation (BP) neural network has also been built to analyze the prediction performance, which are compared with CNN. The results demonstrate that CNN has better performance in predicting the connectivity with the overall AARD below 15.35%. Moreover, the connectivity predicted by CNN is closest to the real connectivity factor compared with some traditional methods. The evaluation method on interwell connectivity proposed by this paper provides effective guidance for the secondary development of both conventional and unconventional reservoirs.

Due to the coexistence of multiple types of reservoir bodies and widely distributed aquifer support in karst carbonate reservoirs, it remains a great challenge to understand the reservoir flow dynamics based on traditional capacitance–resistance (CRM) models and Darcy’s percolation theory. To solve this issue, an improved injector–producer-pair-based CRM model coupling the effect of active aquifer support was first developed and combined with the newly-developed Stochastic Simplex Approximate Gradient (StoSAG) optimization algorithm for accurate inter-well connectivity estimation in a waterflood operation. The improved CRM–StoSAG workflow was further applied for real-time production optimization to find the optimal water injection rate at each control step by maximizing the net present value of production. The case study conducted for a typical karst reservoir indicated that the proposed workflow can provide good insight into complex multi-phase flow behaviors in karst carbonate reservoirs. Low connectivity coefficient and time delay constant most likely refer to active aquifer support through a high-permeable flow channel. Moreover, the injector–producer pair may be interconnected by complex fissure zones when both the connectivity coefficient and time delay constant are relatively large.

Gas transport in shale gas reservoirs is largely affected by rock properties such as permeability. These properties are often sensitive to the in-situ stress state changes. Accurate modeling of shale gas transport in shale reservoir rocks considering the stress sensitive effects on rock petrophysical properties is important for successful shale gas extraction. Nonlinear elasticity in stress sensitive reservoir rocks depicts the nonlinear stress-strain relationship, yet it is not thoroughly studied in previous reservoir modeling works. In this study, an improved coupled flow and geomechanics model that considers nonlinear elasticity is proposed. The model is based on finite element methods, and the nonlinear elasticity in the model is validated with experimental data on shale samples selected from the Longmaxi Formation in Sichuan Basin China. Numerical results indicate that, in stress sensitive shale rocks, nonlinear elasticity affects shale permeability, shale porosity, and distributions of effective stress and pore pressure. Elastic modulus change is dependent on not only in-situ stress state but also stress history path. Without considering nonlinear elasticity, the modeling of shale rock permeability in Longmaxi Formation can overestimate permeability values by 1.6 to 53 times.

Production forecasting plays an important role in development plans during the entire period of petroleum exploration and development. Artificial intelligence has been extensively investigated in recent years because of its capacity to extensively analyze and interpret complex data. With the emergence of spatiotemporal models that can integrate graph convolutional networks (GCN) and recurrent neural networks (RNN), it is now possible to achieve multi-well production prediction by considering the impact of interactions between producers and historical production data simultaneously. Moreover, an accurate prediction not only depends on historical production data but also on the influence of neighboring injectors’ historical gas injection rate (GIR). Therefore, based on the assumption that introducing GIR can enhance prediction accuracy, this paper proposes a deep learning-based hybrid production forecasting model that is aimed at considering both the spatiotemporal characteristics of producers and the GIR of neighboring injectors. Specifically, we integrated spatiotemporal characteristics and GIR into an attribute-augmented spatiotemporal graph convolutional network (AST-GCN) and gated recurrent units (GRU) neural network to extract intricate temporal correlations from historical data. The method proposed in this paper has been successfully applied in a well pattern (including five producers and seven gas injectors) in a low-permeability carbonate reservoir in the Middle East. In single well production forecasting, the error of AST-GCN is 63.2%, 37.3%, and 16.1% lower in MedAE, MAE, and RMSE compared with GRU and 62.9%, 44.6%, and 28.9% lower compared with RNS. Similarly, the accuracy of AST-GCN is 15.9% and 35.8% higher than GRU and RNS in single well prediction. In well-pattern production forecasting, the error of AST-GCN is 41.2%, 64.2%, and 75.2% lower in RMSE, MAE, and MedAE compared with RNS, while the accuracy of AST-GCN is 29.3% higher. After different degrees of Gaussian noise are added to the actual data, the average change in AST-GCN is 3.3%, 0.4%, and 1.2% in MedAE, MAE, and RMSE, which indicates the robustness of the proposed model. The results show that the proposed model can consider the production data, gas injection data, and spatial correlation at the same time, which performs well in oil production forecasts.

Water–rock interactions can alter rock properties through chemical reactions during subsurface transport processes like geological CO₂ sequestration (GCS), matrix acidizing, and waterflooding in carbonate formations. Dynamic changes in rock properties cause a failure of waterflooding and GCS and could also dramatically affect the efficiency of the acidizing. Efficient numerical simulations are thus essential to the optimized design of those subsurface processes. In this paper, we develop a three-dimensional (3D) numerical model for simulating the coupled processes of fluid flow and chemical reactions in fractured carbonate formations. In the proposed model, we employ the Stokes–Brinkman equation for momentum balance, which is a single-domain formulation for modeling fluid flow in fractured porous media. We then couple the Stokes–Brinkman equation with reactive-transport equations. The model can be formulated to describe linear as well as radial flow. We employ a decoupling procedure that sequentially solves the Stokes–Brinkman equation and the reactive transport equations. Numerical experiments show that the proposed method can model the coupled processes of fluid flow, solute transport, chemical reactions, and alterations of rock properties in both linear and radial flow scenarios. The rock heterogeneity and the mineral volume fractions are two important factors that significantly affect the structure of conductive channels.