Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
Results
Machine Learning-Based Optimization of Well Locations and WAG Parameters under Geologic Uncertainty
Nwachukwu, Azor (The University of Texas at Austin) | Jeong, Hoonyoung (Bureau of Economic Geology) | Sun, Alexander (Bureau of Economic Geology) | Pyrcz, Michael (The University of Texas at Austin) | Lake, Larry W. (The University of Texas at Austin)
Abstract The effects of well locations and control parameters on reservoir responses is important during CO2 enhanced oil recovery (CO2-EOR) and water-alternating-gas (WAG) injection. Oil recovery and CO2 storage capacity typically vary with respect to corresponding changes in WAG ratio, slug size, and injector locations. The relationships between control parameters and response variables are usually studied using compositional simulators. However, the computational expense required to run such simulators could hinder their applicability to optimization procedures requiring many evaluations. Surrogate models (proxies) provide inexpensive alternatives for approximating reservoir responses. In this study, a machine learning-based proxy is developed to predict responses to changes in location and control parameters during WAG injection. Slight adjustments in injector well locations, WAG ratio and slug size could yield dramatic changes in the objective function responses. This complex relationship between control parameters and reservoir-wide responses makes data-driven methods an attractive option. We extend a recently developed machine learning approach in which the primary predictors are physical well locations, and water and gas injection rates, and the primary response is net present value (NPV). Because of the complexity of the response surface, we augmented the predictor variables with well-to-well pairwise connectivities, injector block permeabilities and porosities, and initial injector block saturations. Connectivities are represented by โdiffusive times of flightโ of the pressure front, which is computed using the Fast Marching Method. Training observations are obtained from a handful of compositional simulations. We then used the Extreme Gradient Boosting method (XGBoost) to build intelligent proxies for making predictions given any set of observations. We propose a hyperdimensional, simultaneous optimization of well locations and controls using a novel optimization scheme similar Mesh Adaptive Direct Search (MADS). Our optimization scheme was developed to fully take advantage of the speed and statistical output of the proxy. The proposed approach is demonstrated using a case study in which the underlying geology is uncertain. Results show significant correlation between proxy predictions and reservoir simulations, which validates application of the trained proxy. In addition, we demonstrate that simultaneous optimization of location and controls yields improvements over a sequential approach without any significant increase in computational cost.
Capillary Desaturation Curve Fundamentals
Yeganeh, Mohsen (ExxonMobil Research and Engineering Co.) | Hegner, Jessica (ExxonMobil Research and Engineering Co.) | Lewandowski, Eric (ExxonMobil Research and Engineering Co.) | Mohan, Aruna (ExxonMobil Research and Engineering Co.) | Lake, Larry W. (The University of Texas at Austin) | Cherney, Dan (ExxonMobil Research and Engineering Co.) | Jusufi, Arben (ExxonMobil Research and Engineering Co.) | Jaishankar, Aditya (ExxonMobil Research and Engineering Co.)
Abstract A capillary desaturation curve (CDC) depicts the relationship between residual oil saturation, Sor, (i.e. oil left behind in a well-swept permeable medium) and capillary number. A CDC is one of the most fundamental curves of oil recovery as it reveals flow conditions required for good oil displacement in porous media. Despite the importance of this critical curve, the fundamentals describing the physics of a CDC are still incomplete. We present a physical model to describe the capillary desaturation curve. The model balances the capillary pressure and applied viscous stresses caused by flow and takes advantage of contact angle hysteresis that occurs in porous media. It defines a critical oil ganglia length that depends inversely on capillary number and depends on porosity, permeability, and wettability. We have combined the critical oil ganglia expression and ganglia length distribution in porous media to arrive at an expression for the capillary desaturation curve. The model suggests that when a trapped oil ganglion is larger than the critical ganglia length, the applied pressure difference can mobilize the trapped oil ganglion. We describe the differences and similarities between our critical ganglia length expression and previously reported expressions. The model describing the relationship between residual oil saturation and capillary number was successfully verified with microfluidic experiments using various crude oils and displacing fluids. We have also demonstrated that the model applies to previously reported coreflood CDCs from sandstone and carbonate media. Extension of the model led to a single curve representation of variations in Sor with reduced pressure. This representation is independent of the chemistry of the displacing fluid.
Abstract This paper summarizes the current state of the ethane industry in the United States and explores the opportunity for using ethane for enhanced oil recovery. We show both simulation data and field examples to demonstrate that ethane is an excellent EOR injectant. After decades of research and field application, the use of CO2 as an EOR injectant has proven to be very successful. However, there are limited supplies of low cost CO2 available, and there are also significant drawbacks, especially corrosion, involving its use. The rich gasses and volatile oils developed by horizontal drilling and fracturing in the shale reservoirs have brought about an enormous increase in ethane production. Ethane prices have dropped substantially. In the U.S., ethane is no longer priced as a petrochemical feedstock, but is priced as fuel. Also, substantial quantities of ethane are currently being flared. Ethane-based EOR can supplement the very successful CO2-based EOR industry in the U.S. There simply isn't enough low-cost CO2 available to undertake all of the potential gas EOR projects in the U.S. The current abundance of low-cost ethane presents a significant opportunity to add new gas EOR projects. The ethane-based EOR opportunity can be summarized as follows; CO2-based EOR works well, and is well understood. Ethane is better than CO2 for EOR. Ethane is simpler than CO2 for EOR. Ethane is now inexpensive, and will likely stay inexpensive. Ethane-based EOR has become a viable option in the Lower 48. Large volumes of low-cost ethane are available. Recent additions to the growing ethane infrastructure now deliver ethane to locations where ethane-based EOR targets are plentiful.
- North America > United States > Texas (1.00)
- North America > United States > Oklahoma (1.00)
- North America > United States > Alaska > North Slope Borough (0.68)
- Materials > Chemicals > Commodity Chemicals > Petrochemicals (1.00)
- Government > Regional Government > North America Government > United States Government (1.00)
- Energy > Oil & Gas > Upstream (1.00)
- North America > United States > Texas > East Texas Salt Basin > East Texas Field > Woodbine Formation (0.99)
- North America > United States > Oklahoma > Northeast Oklahoma Platform Basin > Glenn Pool Field > Glenn Formation (0.99)
- North America > United States > Oklahoma > Northeast Oklahoma Platform Basin > Glenn Pool Field > Bartlesville Formation (0.99)
- (42 more...)
- Reservoir Description and Dynamics > Storage Reservoir Engineering > CO2 capture and sequestration (1.00)
- Reservoir Description and Dynamics > Improved and Enhanced Recovery > Gas-injection methods (1.00)
- Reservoir Description and Dynamics > Improved and Enhanced Recovery > Chemical flooding methods (1.00)
- (2 more...)