Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
The University of Oklahoma
Improving porosity and gamma-ray prediction for the Middle Jurassic Hugin sandstones in the southern Norwegian North Sea with the application of deep neural networks
Chopra, Satinder (SamiGeo) | Sharma, Ritesh Kumar (SamiGeo) | Marfurt, Kurt J. (The University of Oklahoma) | Zhang, Rongfeng (Geomodeling Technology Corp.) | Wen, Renjun (Geomodeling Technology Corp.)
Abstract The complete characterization of a reservoir requires accurate determination of properties such as the porosity, gamma ray, and density, among others. A common workflow is to predict the spatial distribution of properties measured by well logs to those that can be computed from the seismic data. In general, a high degree of scatter of data points is seen on crossplots between P-impedance and porosity, or P-impedance and gamma ray, suggesting great uncertainty in the determined relationship. Although for many rocks there is a well-established petrophysical model correlating the P-impedance to porosity, there is not a comparable model correlating the P-impedance to gamma ray. To address this issue, interpreters can use crossplots to graphically correlate two seismically derived variables to well measurements plotted in color. When there are more than two seismically derived variables, the interpreter can use multilinear regression or artificial neural network analysis that uses a percentage of the upscaled well data for training to establish an empirical relation with the input seismic data and then uses the remaining well data to validate the relationship. Once validated at the wells, this relationship can then be used to predict the desired reservoir property volumetrically. We have described the application of deep neural network (DNN) analysis for the determination of porosity and gamma ray over the Volve field in the southern Norwegian North Sea. After using several quality-control steps in the DNN workflow and observing encouraging results, we validate the final prediction of the porosity and gamma-ray properties using blind well correlation. The application of this workflow promises significant improvement to the reservoir property determination for fields that have good well control and exhibit lateral variations in the sought properties.
- Geology > Geological Subdiscipline (1.00)
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Sandstone (0.66)
- North America > United States > Texas > Permian Basin > Midland Basin > Pegasus Field > Pennsylvanian Formation (0.99)
- North America > United States > Texas > Permian Basin > Midland Basin > Pegasus Field > Ellenburger Formation (0.99)
- Europe > United Kingdom > North Sea > North Sea > Northern North Sea > Balder Formation (0.99)
- (36 more...)
Geomechanical Controls on Frac-Hits
Kumar, Dharmendra (The University of Oklahoma) | Ghassemi, Ahmad (The University of Oklahoma)
Abstract The communication among the horizontal wells or "frac-hits" issue have been reported in several field observations. These observations show that the "infill" well fractures could have a tendency to propagate towards the "parent" well depending on reservoir in-situ conditions and operational parameters. Drilling the horizontal wells in a "staggered" layout with both horizontal and vertical offset could be a mitigation strategy to prevent the "frac-hits" issue. In this study, we present a detailed geomechanical modeling and analysis of the proposed solution. For numerical modeling, we used our state-of-the-art fully coupled poroelastic model "GeoFrac-3D" which is based on the boundary element method for the rock matrix deformation/fracture propagation and the finite element method for the fracture fluid flow. The "GeoFrac-3D" simulator fully couples pore pressure to stresses and allows for dynamic modeling of production/injection and fracture propagation. The simulation results demonstrate that production from a "parent’ well causes a non-uniform reduction of the reservoir pore pressure around the production fractures, resulting in an anisotropic decrease of the reservoir total stresses, which could affect fracture propagation from the "infill" wells. We examine the optimal orientation and position of the "infill" well based on the numerical analysis to reduce the "frac-hits" issue in the horizontal well refracturing. The posibility of "frac-hits" can be reduced by optimizing the direction and locations of the "infill" wells, as well as re-pressurizing the "parent" well. The results suggest that arranging the horizontal wells in a "staggered" or "wine rack" arrangement decreases direct well interference and could increase the drainage volume.
- North America > United States > Oklahoma (1.00)
- North America > United States > Texas (0.94)
- North America > United States > California (0.94)
- North America > United States > Texas > West Gulf Coast Tertiary Basin > Eagle Ford Shale Formation (0.99)
- North America > United States > Texas > Sabinas - Rio Grande Basin > Eagle Ford Shale Formation (0.99)
- North America > United States > Texas > Maverick Basin > Eagle Ford Shale Formation (0.99)
- (7 more...)
Simulation of Pressure- and Temperature-Dependent Fracturing Fluid Loss in Multi-Porosity Multi-Permeability Formations
Liu, Chao (Aramco Services Company: Aramco Research Center-Houston) | Phan, Dung (Aramco Services Company: Aramco Research Center-Houston) | Abousleiman, Younane (The University of Oklahoma)
Abstract In this paper, the multi-porosity multi-permeability porothermoelastic theory is used to derive the analytical solution to calculate the pressure- and temperature-dependent fracturing fluid loss. A triple-porosity triple-permeability source rock formation is selected as an example to illustrate the model. The effects of fracturing fluid temperature and natural fractures on the fluid loss rate are systematically illustrated. The model successfully accounts for the varying leak-off rates in the multi-permeability channels through the hydraulic fracture faces. Furthermore, thermal diffusion near the hydraulic fracture faces contributes to a variation of pore pressure whose gradient at hydraulic fracture faces directly controls the fracturing fluid leak-off rate. The model shows that thermal effects bring almost 27% variation in the leak-off rate. Comparison study indicates that the single porosity model without considering multi-permeability systems or thermal effects significantly underestimates the rate of fracturing fluid loss and predicts nearly 84% and 87% lower leak-off rate, compared to the dual-porosity dual-permeability and triple-porosity triple-permeability models, respectively. Two case studies using published laboratory measurements on naturally fractured Blue Ohio sandstone samples are conducted to show the performances of the model. It is shown that the model presented in this paper well captures the total leak-off volume during the pressure-dependent fluid loss measured from laboratory tests. Matching the analytical solution to the laboratory data also allows rocks’ double permeabilities to be estimated.
- Asia > Middle East (0.47)
- North America > United States > Ohio (0.25)
- Europe > Norway > Norwegian Sea (0.24)
- Geology > Geological Subdiscipline > Geomechanics (1.00)
- Geology > Rock Type > Sedimentary Rock > Clastic Rock (0.51)
- Well Drilling > Pressure Management > Well control (1.00)
- Well Drilling > Drilling Fluids and Materials > Drilling fluid selection and formulation (chemistry, properties) (1.00)
- Well Completion > Hydraulic Fracturing > Fracturing materials (fluids, proppant) (1.00)
- Reservoir Description and Dynamics > Reservoir Fluid Dynamics > Flow in porous media (1.00)
Seismic characterization of a Triassic-Jurassic deep geothermal sandstone reservoir, onshore Denmark, using unsupervised machine learning techniques
Chopra, Satinder (SamiGeo) | Sharma, Ritesh Kumar (SamiGeo) | Bredesen, Kenneth (Geological Survey of Denmark and Greenland (GEUS)) | Ha, Thang (The University of Oklahoma) | Marfurt, Kurt J. (The University of Oklahoma)
Abstract The Triassic-Jurassic deep sandstone reservoirs in onshore Denmark are known geothermal targets that can be exploited for sustainable and green energy for the next several decades. The economic development of such resources requires accurate characterization of the sandstone reservoir properties, namely, volume of clay, porosity, and permeability. The classic approach to achieving such objectives has been to integrate well-log and prestack seismic data with geologic information to obtain facies and reservoir property predictions in a Bayesian framework. Using this prestack inversion approach, we can obtain superior spatial and temporal variations within the target formation. We then examined whether unsupervised facies classification in the target units can provide additional information. We evaluated several machine learning techniques and found that generative topographic mapping further subdivided intervals mapped by the Bayesian framework into additional subunits.
- Europe > Denmark (1.00)
- North America > United States > Oklahoma (0.28)
- North America > United States > Texas > Permian Basin > Delaware Basin (0.99)
- North America > United States > Texas > Fort Worth Basin > Barnett Shale Formation (0.99)
- North America > United States > New Mexico > Permian Basin > Delaware Basin (0.99)
- (5 more...)
Fault surface objects from fault probability volumes using active contours
Mora, Jose Pedro (The University of Oklahoma) | Bedle, Heather (The University of Oklahoma) | Marfurt, Kurt J. (The University of Oklahoma)
Seismic surveys provide an invaluable source for understanding depositional histories and structural configurations, important in a range of applications from hydrocarbon accumulations, to geothermal, and carbon sequestration studies. Although manual fault interpretation by a skilled interpreter using knowledge of lithology and structural style usually provides the most accurate fault surfaces, time constraints rarely allow an interpreter to pick every line in a large 3D seismic survey. For good quality data, horizon mapping is easily accelerated using auto-trackers that follow continuous reflectors, stopping at interpreter-posted discontinuities. Fault digitization is a more complex task, where the interpreter manually picks a grid of fault sticks which are then linked to create a three-dimensional fault mesh. Interpreters often employ coherence images to help them pick a set of fault samples, fault by fault, and line by line, or more commonly, on every n line. The continuity of coherence images can be enhanced using a variety of nonlinear filters or replaced altogether using convolutional neural networks. In this study, we use active contours to convert fault probability volumes into a set of fault objects. This study employs a semi-automatic approach that scans high probability fault locations and moves from low to high probability values to fit the fault image’s shape. This approach reduces the time to map faults in a seismic section since active contours act as an auto tracker of fault probability or coherence volumes.
- Geophysics > Seismic Surveying > Seismic Processing (1.00)
- Geophysics > Seismic Surveying > Seismic Interpretation (1.00)
- Geophysics > Seismic Surveying > Surface Seismic Acquisition (0.92)
Manuscript Title: Mechanical and Microstructural Studies of Volcanic Ash Beds in Unconventional Reservoirs
Acosta, Juan C (The University of Oklahoma) | Curtis, Mark E (The University of Oklahoma) | Sondergeld, Carl H (The University of Oklahoma) | Rai, Chandra S (The University of Oklahoma)
Abstract Volcanic ash beds are thin layers commonly observed in the Eagle Ford, Niobrara and, Vaca Muerta formations. Because of their differences in composition, sedimentary structures, and diagenetic alteration, they exhibit a significant contrast in mechanical properties with respect to surrounding formation layers. This can impact hydraulic fracturing, affecting fracture propagation and fracture geometry. Quantifying the mechanical properties of ash beds becomes significant; however, it is a challenge with traditional testing methods. Common logging fails to identify the ash beds, and core plug testing is not possible because of their friability. In this study, nanoindentation was used to measure the mechanical properties (Young's modulus, creep, and anisotropy) in Eagle Ford ash beds, and to determine the contrast with the formation matrix properties. Two separate ash beds of high clay and plagioclase composition were epoxied in an aluminum tray and left for 48 hours curing time. Horizontal and vertical samples of ash beds were acquired and mounted on a metal stub, followed by polishing and broad beam ion milling. Adjacent samples were also prepared for high-resolution Scanning Electron Microscope (SEM) microstructural analysis. The Young's modulus in ash beds ranged from 12 to 24 GPa, with the horizontal direction Young's modulus being slightly greater than that of the vertical samples. The Young's modulus contrast with adjacent layers was calculated to be 1:2 with clay-rich zones and 1:4 with calcite rich zones. The creep deformation rate was three times higher for ash beds compared to other zones. Using Backus averaging, it was determined that the presence of ash beds can increase the anisotropy in the formation by 15-25%. SEM results showed a variation in microstructure between the ash beds with evidence of diagenetic conversion of rhyolitic material into clays. Key differences between the two ash beds were due to the presence of plagioclase and the occurrence of porosity within kaolinite. Overall porosity varied between the two ash beds and adjacent carbonate layers showing a significant increase in porosity. Understanding the moduli contrast between adjacent layers can improve the hydraulic fracturing design when ash beds are encountered. In addition, the presence of these beds can lead to proppant embedment and loss in fracture connectivity. These results can be used for improving geomechanical models.
- North America > United States > Texas (0.47)
- South America > Argentina > Neuquén Province > Neuquén (0.24)
- Geology > Mineral > Silicate > Phyllosilicate (1.00)
- Geology > Geological Subdiscipline > Volcanology (1.00)
- Geology > Geological Subdiscipline > Geomechanics (1.00)
- North America > United States > Texas > West Gulf Coast Tertiary Basin > Eagle Ford Shale Formation (0.99)
- North America > United States > Texas > Sabinas - Rio Grande Basin > Eagle Ford Shale Formation (0.99)
- North America > United States > Texas > Maverick Basin > Eagle Ford Shale Formation (0.99)
- (2 more...)
A review of seismic attribute taxonomies, discussion of their historical use, and presentation of a seismic attribute communication framework using data analysis concepts
Dewett, Dustin T. (The University of Oklahoma) | Pigott, John D. (The University of Oklahoma) | Marfurt, Kurt J. (The University of Oklahoma)
Abstract Beginning in the 1970s, seismic attributes have grown from a few simple measurements of wavelet amplitude, frequency, and phase to an expanded attribute toolbox that measures not only wavelet properties but also their context within the 3D seismic volume. When using multiple seismic attributes, the interpreter must understand not only each individual attribute but also the relationships between them. Researchers communicate these relationships via seismic attribute taxonomies, which group attributes by their signal property, mathematical formulation, or their interpretive value. The first attempts to organize seismic attributes began in the 1990s, and with new attributes and their increasing breadth of applications, continues to this day. Most scientific papers that use seismic attributes focus on a specific application, new algorithms, or a novel interpretation workflow, rather than how a specific attribute fits within the greater whole, leading to confusion for the less experienced interpreter. We have analyzed more than 2100 citing works, identified the 231 papers that discuss the taxonomies specifically, and found out how the authors use those citations. The result is a list of more than a dozen seismic attribute classification systems, which we reduce to a smaller subset by including only those that apply to general use. An optimal seismic attribute taxonomy should not only be useful to the interpretation community today, but it should also adapt to the ever-changing needs of the profession, including changes appropriate for their use in modern machine-learning algorithms. The adaptability of prior work to modern workflows remains a shortcoming. However, as we develop our work in two parts — the first covering the evolution of seismic attribute taxonomies and their use through time and the second proposing a new seismic attribute communication framework for the larger community — we link attributes together via data analysis principles and provide an extensible model as the profession and research expand.
- Asia (0.67)
- North America > United States > Oklahoma (0.28)
- Geology > Geological Subdiscipline > Stratigraphy (1.00)
- Geology > Rock Type (0.93)
- Oceania > Australia > Western Australia > North West Shelf > Carnarvon Basin > Carnarvon Basin > Dampier Basin > Rankin Platform > Greater Gorgon Development Area > Block WA-268-P > Greater Gorgon Field > Gorgon Field (0.99)
- Oceania > Australia > Western Australia > North West Shelf > Carnarvon Basin > Carnarvon Basin > Carnarvon Basin > Rankin Platform > Greater Gorgon Development Area > Block WA-268-P > Greater Gorgon Field > Gorgon Field (0.99)
- Oceania > Australia > Western Australia > North West Shelf > Carnarvon Basin > Alpha Arch > Dampier Basin > Rankin Platform > Greater Gorgon Development Area > Block WA-268-P > Greater Gorgon Field > Gorgon Field (0.99)
- (3 more...)
An in-depth analysis of logarithmic data transformation and per-class normalization in machine learning: Application to unsupervised classification of a turbidite system in the Canterbury Basin, New Zealand, and supervised classification of salt in the Eugene Island minibasin, Gulf of Mexico
Ha, Thang N. (The University of Oklahoma) | Lubo-Robles, David (The University of Oklahoma) | Marfurt, Kurt J. (The University of Oklahoma) | Wallet, Bradley C. (Aramco Service Company)
Abstract In a machine-learning workflow, data normalization is a crucial step that compensates for the large variation in data ranges and averages associated with different types of input measured with different units. However, most machine-learning implementations do not provide data normalization beyond the z-score algorithm, which subtracts the mean from the distribution and then scales the result by dividing by the standard deviation. Although the z-score converts data with Gaussian behavior to have the same shape and size, many of our seismic attribute volumes exhibit log-normal, or even more complicated, distributions. Because many machine-learning applications are based on Gaussian statistics, we have evaluated the impact of more sophisticated data normalization techniques on the resulting classification. To do so, we provide an in-depth analysis of data normalization in machine-learning classifications by formulating and applying a logarithmic data transformation scheme to the unsupervised classifications (including principal component analysis, independent component analysis, self-organizing maps, and generative topographic mapping) of a turbidite channel system in the Canterbury Basin, New Zealand, as well as implementing a per-class normalization scheme to the supervised probabilistic neural network (PNN) classification of salt in the Eugene Island minibasin, Gulf of Mexico. Compared to the simple z-score normalization, a single logarithmic transformation applied to each input attribute significantly increases the spread of the resulting clusters (and the corresponding color contrast), thereby enhancing subtle details in projection and unsupervised classification. However, this same uniform transformation produces less-confident results in supervised classification using PNNs. We find that more accurate supervised classifications can be found by applying class-dependent normalization for each input attribute.
- Oceania > New Zealand > South Island > South Pacific Ocean (0.61)
- North America > United States > Gulf of Mexico > Central GOM (0.61)
- North America > United States > Oklahoma (0.46)
- Geology > Rock Type > Sedimentary Rock > Clastic Rock (1.00)
- Geology > Geological Subdiscipline (1.00)
- Geology > Sedimentary Geology > Depositional Environment > Marine Environment > Deep Water Marine Environment (0.61)
- Geophysics > Seismic Surveying > Seismic Processing (1.00)
- Geophysics > Seismic Surveying > Seismic Interpretation (1.00)
- Geophysics > Seismic Surveying > Seismic Modeling > Velocity Modeling (0.46)
- Oceania > New Zealand > South Island > South Pacific Ocean > Canterbury Basin (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)
- (26 more...)
Determination of Pore Fluid Salinity in Tight Rocks Without Fluid Extraction
Odiachi, Judah (The University of Oklahoma) | Tinni, Ali (The University of Oklahoma) | Sondergeld, Carl H. (The University of Oklahoma) | Rai, Chandra S. (The University of Oklahoma)
Abstract Pore fluid salinity plays an important role during hydrocarbon reserves estimations from electrical resistivity logs and rock fluid interaction studies. Classical methods used to measure pore fluid salinity involve laboratory analysis of brine produced during hydrocarbon production or brine extracted from core samples. These classical methods are not applicable in many tight formations due to the small volume of water producible from these formations. This study presents a methodology to determine pore fluid salinity in tight rocks overcoming the limitations presented by classical methods. Recently, methods to determine pore fluid salinity based on the dielectric properties of saturated rocks have been developed, but these methods also present several limitations. The method presented in this paper relies on the relationship between dielectric loss frequency dispersion and pore fluid salinity. To validate this novel method, we measured the dielectric loss frequency dispersion on 18 samples including Berea sandstone, Wolfcamp, Eagle Ford and Meramec shale formations saturated with known salinity brines. These samples cover a porosity range of 6.6 to 18 %, clay content of 3 to 35% and TOC of 0 to 6.1%. Also, we measured the dielectric loss frequency dispersion on 2 wax preserved Eagle Ford shale samples at different locations of the puck from the same depth, showing consistency in salinity predictions. To expand the application of this method for in-situ salinity determination from dielectric logs, we developed calibration curves for salinity prediction at varying temperatures. Measurements on the resaturated samples show that the dielectric dispersion loss method is reliable within 10 % of the true value. Measurements at several locations on wax preserved Eagle Ford shale samples support the consistency of the method. The dielectric loss dispersion method presents the advantage to be applicable in laboratory and downhole conditions without requiring the acquisition of formation brine samples. Introduction The determination of pore fluid salinity is a necessary step during the evaluation of hydrocarbon reservoirs. Brine salinity exerts a primary control on the response of electrical logs which are often used to compute water saturation and hydrocarbon pore volume (Archie 1942, Clinch 2011, Newsham 2018). Brine salinity also affects other logging tools such as the nuclear magnetic resonance (NMR) tool. The NMR tool measures the total amount of hydrogen protons in the pore space, thus an increase in salinity will lead to a reduction in the total amount of hydrogen per unit pore volume, hence a reduction in the NMR porosity (Kleinberg and Vinegar 1996, Sondergeld 2016). In addition to its controls on the response of several logging tools, brine salinity plays an important role in many fluid flow processes as well as rock fluid interactions especially important in tight rocks such as shales. Zhou (2015) and Padin et al., (2018) have shown that brine salinity control water mobility in shales by the phenomenon of osmosis. Ewy and Stankovic (2010) have shown that brine salinity also controls the mechanical stability of clay rich formations.
- North America > United States > Texas > West Gulf Coast Tertiary Basin > Eagle Ford Shale Formation (0.99)
- North America > United States > Texas > Sabinas - Rio Grande Basin > Eagle Ford Shale Formation (0.99)
- North America > United States > Texas > Permian Basin > Wolfcamp Formation (0.99)
- (4 more...)
Impact of EOR Huff-n-Puff on Rock Microstructure
Mamoudou, Sidi (The University of Oklahoma) | Tinni, Ali (The University of Oklahoma) | Curtis, Mark (The University of Oklahoma) | Sondergeld, Carl H. (The University of Oklahoma) | Rai, Chandra S. (The University of Oklahoma)
Abstract Field and laboratory studies have shown that EOR Huff-n-Puff (HnP) in tight rocks can mobilize additional hydrocarbon. However, these previous studies have not investigated potential changes to the microstructure of shales due to the EOR process. The present study focuses on the study of microstructural changes during EOR HnP on shale samples from the Duvernay and Montney. To evaluate how EOR operations could potentially alter the microstructure of unconventional formations, HnP experiments were conducted with a mixture of methane and ethane (C1:C2/72:28mol%) were performed on crushed samples (7-8 mm) from Duvernay and Montney (oil window). These experiments were carried on samples in their preserved state (without recombination with dead oil) at 1000 psi above minimum miscibility pressure (MMP) and 150 °F using 1-hour soak and 1-hour production time. Changes to the microstructure were evaluated by the integration of measurements from Nuclear Magnetic Resonance (NMR), Hydrocarbon Analyzer with Kinetics (HAWK) pyrolysis, Mercury Injection Capillary Pressure (MICP), isothermal adsorption based on Brunauer Emmett Teller theory (BET) and SEM (Scanning Electron Microscopy). Significant microstructural alterations were observed after the HnP experiments. We observed an increase in the proportion of pore throat size between 3-10 nm radius and an increase in surface area by a factor of 2 in the Duvernay sample. SEM images also confirm MICP and BET observations where the organic matter pore size increased by a factor of 2 in the same sample after EOR. Oil composition analysis from HAWK shows that hydrocarbons up to C24 were efficiently mobilized during the process for both samples. However, in the Duvernay sample heavier hydrocarbons up to C27 were produced. This experimental study shows that gas injection leads to measurable changes in the microstructure and organic component of the rock. It also gives an insight into reservoir selection for potential field HnP candidate. Introduction The successes reported by EOG in the Eagle Ford have renewed interest in EOR HnP; additional 30-70% recovery could be unlocked after primary production (Hoffman, 2018). The EOR HnP process is designed to release hydrocarbons through miscible injection, injected gas moves into the fractures and diffuses through the rock at the matrix/fracture interface (Fragaso et al., 2015; Sheng 2017); driven by pressure the oil swells and its viscosity reduces, as the pressure gradient decreases the oil is moved from pores to the fractures (Hawthorne et al., 2014).
- North America > Canada > Alberta (0.48)
- North America > United States > Oklahoma > Cleveland County > Norman (0.15)
- Research Report > New Finding (1.00)
- Research Report > Experimental Study (0.89)
- Geology > Geological Subdiscipline > Geochemistry (0.49)
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Mudrock > Shale (0.47)
- North America > Canada > Saskatchewan > Williston Basin > Bakken Shale Formation (0.99)
- North America > Canada > Manitoba > Williston Basin > Bakken Shale Formation (0.99)
- North America > United States > South Dakota > Williston Basin > Bakken Shale Formation (0.98)
- (5 more...)