Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
Jia, Xinli
Abstract Based on historical production data, decline curve analysis (DCA) can be used to monitor production, identify potential problems, and predict well performance, life, and economics. Optimized production history matching is crucial to economic analysis on future operations and decision making. The widely used decline models such as Arps rate-time relations and their variations are based on fitting predefined equations and often times do not work for shale gas and oil wells since most of the production data from these wells exhibit fracture-dominated flow regimes and rarely reach late-time-flow regimes. This approach can mislead the trend of the decline curves and produce poor matches and unreliable production forecasts. A suitable data-driven model combining physical or operating parameters can be greatly beneficial and serve as the basis for decline analysis and prediction. This paper discusses a method for automatic history matching and decline analysis for shale production data based on machine learning, which can be effectively applied to production surveillance and process automation. This approach is based on time series (TS) analysis and neural networks (NNs), which was then extended to applications with operational parameters available, such as bottom-hole pressure (BHP). The proposed TS and NN models were applied to production data from a Barnett gas well. The historical production data was divided into two parts. The first part of the data is used to train the NN model, and the second part of the data is used to verify the accuracy of the prediction results from each input parameter. The results were analyzed and compared with classic Arps decline models.
- Geology > Petroleum Play Type > Unconventional Play (0.47)
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Mudrock > Shale (0.46)
- North America > United States > Wyoming > Green River Basin > Jonah Field (0.99)
- North America > United States > Texas > Fort Worth Basin > Barnett Shale Formation (0.99)
- North America > United States > Oklahoma > Anadarko Basin > Cana Woodford Shale Formation (0.99)
- (3 more...)
- Reservoir Description and Dynamics > Unconventional and Complex Reservoirs > Shale gas (1.00)
- Reservoir Description and Dynamics > Formation Evaluation & Management > Production forecasting (1.00)
- Production and Well Operations > Well & Reservoir Surveillance and Monitoring > Production data management (1.00)
- Data Science & Engineering Analytics > Information Management and Systems > Neural networks (1.00)
Abstract The development of unconventional hydrocarbons has become a significant resource, leading to growth of worldwide oil and natural gas supplies. Hydraulic fracturing has been successfully employed for unconventional oil and gas recovery for decades. In recent years, the rapid progress of technology has led to reduced gas prices and a shift in focus to liquid extraction. However, liquid flow, both in the wellbore and channels inside porous media or fractures, experiences more resistance compared to gas, resulting in significant pressure losses in the wellbore and fractures. Reservoir productivity also becomes more complex because of relative permeability effects. Forecasting production and estimating shale reserves is still not fully understood because of the limited knowledge of flow mechanics in ultralow-permeability rock. Many analytical, semi-analytical, and numerical models have been developed to better understand flow in ultralow-permeability rocks and hydraulic fractures. Because analytical models only apply to mostly dry gas reservoirs, numerical reservoir simulation is generally believed to be the most rigorous and accurate method for liquid-rich formations. However, the drawbacks of using reservoir simulation are substantial. Some examples include the significant data requirements, level of expertise required to set up the model, and the demanding turnaround times for meeting the design, optimization, and decision-making cycle deadlines. Also, because each engineer is responsible for a large number of wells, full-scale three-dimensional (3D) reservoir modeling is impossible for a majority of wells. Therefore, an approach is required that is less time-consuming than detailed reservoir simulation while still being sufficiently accurate to capture the physics of the process. It should be based on numerical modeling of multiphase flow in the interconnecting system of the wellbore and fractures, with the reservoir represented by a productivity index (PI) inflow model, as well as a physics-based pressure-volume-temperature (PVT) model for phase transition and phase equilibrium. The production decline and prediction should be analyzed based on reservoir depletion, relative permeabilities, and fracture conductivities. This paper describes a numerical fracture production model (FPM) based on the previously mentioned physics that can be used to simulate production resulting from reservoir depletion and analyze historical production data. The outcome of the model focuses on a few primary input parameters that are dedicated to predicting future production and quickly analyzing the parametric effects and economic value of fracture-stimulated condensate reservoirs. The model is validated using two commercially available software programs, as well as historical production data of an Eagle Ford play. The outputs are then used for history matching, sensitivity analysis, parameter optimization, and future production prediction.
Abstract Hydraulic fracturing has been successfully employed for unconventional oil and gas recovery for decades. The fracturing process is realized by injecting fluid, which contains the proppant materials used to keep the fracture open and productive, into a well at a high enough rate and pressure to crack open the formation. Hydraulic fracture-stimulated production plays an important role in unconventional hydrocarbon production. The distribution and transport of proppant significantly affect fracture conductivity and, in turn, the production rate and decline. It is widely recognized that the effective placement of proppant in a fracture has a dominant effect on a well's productivity, yet it is greatly underestimated owing to a lack of knowledge and practical means to deal with the transient proppant settling process inside the fracture. Existing hydraulic fracture models mostly simplify the proppant transport process or even totally neglect the effect. A common assumption is that the average proppant velocity is equal to the average carrier fluid velocity, and the settling velocity is calculated using Stokes' law, while some important forces exerting on proppant particles are not taken into account, which often leads to the overprediction of the effective fracture length by up to 300%. To effectively simulate the dynamic proppant settling inside the fracture requires the consideration of such factors, including the wall-effect lift force, drag, and leakage of the fluid into the formation. A numerical model has been developed to predict the transient transport and settling of proppants during hydraulic fracturing treatments and production to improve fracture conductivity. The model presented in this paper includes three stages. The initial stage simulates the homogenous phase behavior with a previously developed fracture injection and production model (FIPM) developed elsewhere. The FIPM can alternate between injection and production modes, with gas, oil, and water phases included, and gas-oil phase transition allowed. Therefore, leakoff and production into the formation are simulated based on the pressure and phase saturation fields, with gravitational, drag, and lift forces taken into account. During the final stage, a fracture-stimulated horizontal well in the Eagle Ford is used to validate the model, both for injection-induced water damage and production. During the initial proppant injection stage, the distribution of proppants can be considered to be homogenously dispersed in the injection fluid. Shortly after the injection started, because of geothermal effects, leakoff of injection fluid, and gravity, the proppant particles will settle, and the initial fracture conductivity profile is formed during this stage. This conductivity profile has a significant effect on the early-stage production rate. As production continues, the proppant particles will be shifted dynamically and will result in a change in the effective fracture length, width, and conductivity distribution over time. During the second stage, the leakoff and evaporation of the liquid phase is simulated as flow through porous media, taking the thermal gradient and evaporation latent heat into consideration. The settling and transport of transient solid proppant particles are modeled using the upstream scheme, accounting for the forces of gravity, drag, and wall-lift forces on the particles. Both injection and leakoff of fluids help determine the size and conductivity of fractures, as well as the water-envelope damage in the near-fracture region inside the formation. These can significantly impact the transient depletion of the reservoir, particularly during early production time. This effect is studied in terms of reservoir permeability, production rate, and phase distribution.
- North America > United States > Texas (0.94)
- Europe (0.66)
Abstract This paper presents a method for the fast evaluation of fracture-stimulated condensate reservoir economics. For the calculation of production decline in such reservoirs, an efficient numerical model with a three-phase transient analysis of pressure distribution was built and validated using the predictions from reservoir solvers and field data. This model solves for gas-, oil-, and water-flow parameters, accounting for the gas-oil phase transition, and has been realized in a numerical code and compared with predictions from commercial software and available field data, such as production-decline curves. The developed numerical model has been implemented in commercial software and used for the sensitivity analysis of reservoir productivity regarding changes of fracture size and spacing, as well as reservoir permeability in the fractured condensate reservoirs, with an account for multiphase reservoir flows and reservoir properties. A side-by-side comparison of predictions from two commercial reservoir simulators has shown that that this model accurately calculates transientpressure fields near the fractures and the productiondecline curve. The objective of the economic analysis and fracture optimization stage is reduced to finding the target function minimum in an N-dimensional parametric space using various constrained minimization techniques, including a Quasi-Monte Carlo analysis and the Active Set Method.
- North America > United States > West Virginia (0.29)
- North America > United States > Texas (0.29)
- North America > United States > West Virginia > Appalachian Basin > Marcellus Shale Formation (0.99)
- North America > United States > Virginia > Appalachian Basin > Marcellus Shale Formation (0.99)
- North America > United States > Pennsylvania > Appalachian Basin > Marcellus Shale Formation (0.99)
- (3 more...)