Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
Scenario
ABSTRACT Wind energy is considered to be of great importance for promoting energy transition and achieving net-zero carbon emission. Reliable modeling and monitoring of the near-subsurface geology are crucial for successful wind farm selection, construction, operation, and maintenance. For optimal characterization of shallow seafloor sediments, 2D ultrahigh-resolution (UHR) seismic survey and 1D cone-penetration testing (CPT) often are acquired, processed, interpreted, and integrated for building 3D ground models of essential geotechnical parameters such as friction. Such a task faces multiple challenges, such as limited CPT availability, strong noise contamination in UHR seismic data, and heavy manual efforts for completing the traditional workflows, particularly acoustic impedance inversion. This study accelerates the integration by a semisupervised learning workflow with three highlights. First, it enables geotechnical parameter estimation directly from UHR seismic data without impedance inversion. The second comes from the use of a pretrained feature engine to reduce the risk of overfitting while mapping massive UHR seismic data with sparse CPT measurements through deep learning. More importantly, it allows incorporating other geologic/geophysical information, such as a predefined structural model, to further constrain the machine learning and boost its generalization capability. Its values are validated through applications to the Dutch wind farm zone for estimating four geotechnical parameters, including cone-tip resistance, sleeve friction, pore-water pressure, and the derived friction ratio, in two example scenarios: (1) UHR seismic data only and (2) UHR seismic data and an 11-layer structural model. The results verify the feasibility of data-driven geotechnical parameter estimation. In addition to the two demonstrated scenarios, our workflow can be further customized for embedding more constraints, e.g., prestack seismic and elastic/static property models, given their availability in a wind farm of interest.
- Geophysics > Seismic Surveying > Seismic Interpretation (1.00)
- Geophysics > Seismic Surveying > Seismic Modeling > Velocity Modeling (0.46)
- Geophysics > Seismic Surveying > Seismic Processing > Seismic Migration (0.34)
ABSTRACT Seismic processing often involves suppressing multiples that are an inherent component of collected seismic data. Elaborate multiple prediction and subtraction schemes such as surface-related multiple removal have become standard in industry workflows. In cases of limited spatial sampling, low signal-to-noise ratio, or conservative subtraction of the predicted multiples, the processed data frequently suffer from residual multiples. To tackle these artifacts in the postmigration domain, practitioners often rely on Radon transform-based algorithms. However, such traditional approaches are both time-consuming and parameter dependent, making them relatively complex. In this work, we present a deep learning-based alternative that provides competitive results, while reducing the complexity of its usage, and, hence simplifying its applicability. Our proposed model demonstrates excellent performance when applied to complex field data, despite it being exclusively trained on synthetic data. Furthermore, extensive experiments show that our method can preserve the inherent characteristics of the data, avoiding undesired oversmoothed results, while removing the multiples from seismic offset or angle gathers. Finally, we conduct an in-depth analysis of the model, where we pinpoint the effects of the main hyperparameters on real data inference, and we probabilistically assess its performance from a Bayesian perspective. In this study, we put particular emphasis on helping the user reveal the inner workings of the neural network and attempt to unbox the model.
- Geophysics > Seismic Surveying > Seismic Processing (1.00)
- Geophysics > Seismic Surveying > Seismic Modeling > Velocity Modeling (0.46)
- North America > United States > Montana > Target Field (0.99)
- Europe > Norway > North Sea > Central North Sea > South Viking Graben > PL 046 > Block 15/9 > Volve Field > Shetland Group > Åsgard Formation (0.99)
- Europe > Norway > North Sea > Central North Sea > South Viking Graben > PL 046 > Block 15/9 > Volve Field > Shetland Group > Svarte Formation (0.99)
- (18 more...)
ABSTRACT Wind energy is considered to be of great importance for promoting energy transition and achieving net-zero carbon emission. Reliable modeling and monitoring of the near-subsurface geology are crucial for successful wind farm selection, construction, operation, and maintenance. For optimal characterization of shallow seafloor sediments, 2D ultrahigh-resolution (UHR) seismic survey and 1D cone-penetration testing (CPT) often are acquired, processed, interpreted, and integrated for building 3D ground models of essential geotechnical parameters such as friction. Such a task faces multiple challenges, such as limited CPT availability, strong noise contamination in UHR seismic data, and heavy manual efforts for completing the traditional workflows, particularly acoustic impedance inversion. This study accelerates the integration by a semisupervised learning workflow with three highlights. First, it enables geotechnical parameter estimation directly from UHR seismic data without impedance inversion. The second comes from the use of a pretrained feature engine to reduce the risk of overfitting while mapping massive UHR seismic data with sparse CPT measurements through deep learning. More importantly, it allows incorporating other geologic/geophysical information, such as a predefined structural model, to further constrain the machine learning and boost its generalization capability. Its values are validated through applications to the Dutch wind farm zone for estimating four geotechnical parameters, including cone-tip resistance, sleeve friction, pore-water pressure, and the derived friction ratio, in two example scenarios: (1) UHR seismic data only and (2) UHR seismic data and an 11-layer structural model. The results verify the feasibility of data-driven geotechnical parameter estimation. In addition to the two demonstrated scenarios, our workflow can be further customized for embedding more constraints, e.g., prestack seismic and elastic/static property models, given their availability in a wind farm of interest.
- Geophysics > Seismic Surveying > Seismic Interpretation (1.00)
- Geophysics > Seismic Surveying > Seismic Modeling > Velocity Modeling (0.46)
- Geophysics > Seismic Surveying > Seismic Processing > Seismic Migration (0.34)
ABSTRACT Seismic processing often involves suppressing multiples that are an inherent component of collected seismic data. Elaborate multiple prediction and subtraction schemes such as surface-related multiple removal have become standard in industry workflows. In cases of limited spatial sampling, low signal-to-noise ratio, or conservative subtraction of the predicted multiples, the processed data frequently suffer from residual multiples. To tackle these artifacts in the postmigration domain, practitioners often rely on Radon transform-based algorithms. However, such traditional approaches are both time-consuming and parameter dependent, making them relatively complex. In this work, we present a deep learning-based alternative that provides competitive results, while reducing the complexity of its usage, and, hence simplifying its applicability. Our proposed model demonstrates excellent performance when applied to complex field data, despite it being exclusively trained on synthetic data. Furthermore, extensive experiments show that our method can preserve the inherent characteristics of the data, avoiding undesired oversmoothed results, while removing the multiples from seismic offset or angle gathers. Finally, we conduct an in-depth analysis of the model, where we pinpoint the effects of the main hyperparameters on real data inference, and we probabilistically assess its performance from a Bayesian perspective. In this study, we put particular emphasis on helping the user reveal the inner workings of the neural network and attempt to unbox the model.
- Geophysics > Seismic Surveying > Seismic Processing (1.00)
- Geophysics > Seismic Surveying > Seismic Modeling > Velocity Modeling (0.46)
- North America > United States > Montana > Target Field (0.99)
- Europe > Norway > North Sea > Central North Sea > South Viking Graben > PL 046 > Block 15/9 > Volve Field > Shetland Group > Åsgard Formation (0.99)
- Europe > Norway > North Sea > Central North Sea > South Viking Graben > PL 046 > Block 15/9 > Volve Field > Shetland Group > Svarte Formation (0.99)
- (18 more...)
Impact of weather routing on a post-panamax bulk carrier equipped with different wind propulsion systems was studied. The studied wind propulsion systems were rotor sails, wing sails, and suction wings and the studied route was a typical bulk carrier route between China and Brazil. A 4 degrees of freedom (4-DoF) performance prediction program was used to generate a performance model of the ship with different wind propulsion devices and the performance models were used in the voyage optimization. The voyage optimization simulations were carried out using historical weather data during the years 2015 to 2019 with one departure per week on both ways. It was found that there is a significant reduction in fuel consumption when weather routing is used on wind propulsion ships, but the magnitudes of the weather routing benefit vary from system to system depending on its versatility. Higher benefits from weather routing were found first for rotor sails, then for suction wings, and finally for wing sails.
- North America > United States (0.93)
- Asia (0.68)
- Transportation > Marine (1.00)
- Leisure & Entertainment > Sports > Sailing (1.00)
- Transportation > Freight & Logistics Services > Shipping > Dry Bulk Carrier (0.75)
A Machine Learning Approach for Gas Kick Identification
Obi, C. E. (Texas A&M University (Corresponding author)) | Falola, Y. (Texas A&M University) | Manikonda, K. (Texas A&M University) | Hasan, A. R. (Texas A&M University) | Hassan, I. G. (Texas A&M University Qatar) | Rahman, M. A. (Texas A&M University Qatar)
Summary Warning signs of a possible kick during drilling operations can either be primary (flow rate increase and pit gain) or secondary (drilling break and pump pressure decrease). Drillers rely on pressure data at the surface to determine in-situ downhole conditions while drilling. The surface pressure reading is always available and accessible. However, understanding or interpretation of this data is often ambiguous. This study analyzes significant kick symptoms in the wellbore annulus both under static (shut in) and dynamic (drilling/circulating) conditions. We used both supervised and unsupervised learning techniques for flow regime identification and kick prognosis. These include an artificial neural network (ANN), support vector machine (SVM), K-nearest neighbor (KNN), decision trees, K-means clustering, and agglomerative clustering. We trained these machine learning models to detect kick symptoms from the gas evolution data collected between the point of kick initiation and the wellhead. All the machine learning techniques used in this work made excellent predictions with accuracy greater than or equal to 90%. For the supervised learning, the decision tree gave the overall best results, with an accuracy of 96% for air influx cases and 98% for carbon dioxide influx cases in both static and dynamic scenarios. For unsupervised learning, K-means clustering was the best, with Silhouette scores ranging from about 0.4 to 0.8. The mass rate per hydraulic diameter and the mixture viscosity yielded the best types of clusters. This is because they account for the fluid properties, flow rate, and flow geometry. Although computationally demanding, the machine learning models can use the surface/downhole pressure data to relay annular flow patterns while drilling. There have been several recent advances in drilling automation. However, this is still limited to gas kick identification and handling. This work provides an alternative and easily accessible primary kick detection tool for drillers based on data at the surface. It also relates this surface data to certain annular flow regime patterns to better tell the downhole story while drilling.
- Europe (1.00)
- Asia > Middle East (0.93)
- North America > United States > Texas (0.46)
- Information Technology > Artificial Intelligence > Machine Learning > Statistical Learning > Support Vector Machines (1.00)
- Information Technology > Artificial Intelligence > Machine Learning > Statistical Learning > Clustering (1.00)
- Information Technology > Artificial Intelligence > Machine Learning > Neural Networks > Deep Learning (1.00)
We never seem to have enough data to analyze the complexity of the subsurface. The geologist would love to have hundreds of cores, but the manager will let him have just one. The geophysicist's dream is to acquire a new 3D with long offsets, complete azimuth coverage, and 5-m bins, but she is more likely to be stuck with a couple of 2D lines. The idea in this tutorial is to show how to create new data starting from what is available so we can make deductions, analyze alternative scenarios, and decide whether a certain variation in lithologic properties would have an impact on seismic. Specifically, I will show how to create new well-log data.
- Information Technology > Knowledge Management (0.40)
- Information Technology > Communications > Collaboration (0.40)
The increasing use of sparse acquisitions in seismic data acquisition offers advantages in cost and time savings. However, it results in irregularly sampled seismic data, adversely impacting the quality of the final images. In this paper, we propose the ResFFT-CAE network, a convolutional neural network with residual blocks based on the Fourier transform. Incorporating residual blocks allows the network to extract both high- and low-frequency features from the seismic data. The high-frequency features capture detailed information, while the low-frequency features integrate the overall data structure, facilitating superior recovery of irregularly sampled seismic data in the trace and shot domains. We evaluated the performance of the ResFFT-CAE network on both synthetic and field data. On synthetic data, we compared the ResFFT-CAE network with the compressive sensing (CS) method utilizing the curvelet transform. For field data, we conducted comparisons with other neural networks, including the convolutional autoencoder (CAE) and U-Net. The results demonstrated that the ResFFT-CAE network consistently outperformed other approaches in all scenarios. It produced images of superior quality, characterized by lower residuals and reduced distortions. Furthermore, when evaluating model generalization, tests using models trained on synthetic data also exhibited promising results. In conclusion, the ResFFT-CAE network shows great promise as a highly efficient tool for the regularizing irregularly sampled seismic data. Its excellent performance suggests potential applications in the preconditioning of seismic data analysis and processing flows.
- North America > United States (0.28)
- South America > Brazil (0.28)
Model is any device or constructs that represent an approximation of a field situation or real world situation Cite error: Invalid ref tag; invalid names, e.g. Mathematical or numerical model is solving an equation or set of equations that describes the behavior of the real-world system (or at least some components of it). Most of the principal governing equations for groundwater systems are differential equation, which are distinguished by the presence of at least one differential in the equation. The solution to a differential equation (at least in cases where the dependent variable is contained in a differential) is an algebraic equation. Equation 1 requires transforming into an algebraic equation.
- Information Technology > Knowledge Management (0.40)
- Information Technology > Communications > Collaboration (0.40)
Compositional and Numerical Modeling of the HnP Technique by Using Both Dry and Rich Gas to Increase Production and Reserves in Shale Oil Wells in Vaca Muerta
Sancet, G. Fondevila (Capex SA) | Ponce, J. (Capex SA) | Gilardone, C. (FDC de Argentina SRL) | Canel, C. (FDC de Argentina SRL) | Albuquerque, L. (FDC de Argentina SRL) | Cardozo, J. (FDC de Argentina SRL)
Abstract To optimize oil recovery from Vaca Muerta (°API 40, GOR 617 scf/stb), lab tests were conducted to assess miscible gas injection. This unconventional formation with average initial pressures of 8500 psia and fluid bubble pressures between 1800 and 3200 psia, shows a significative oil decline. The formation's low transmissibility suggests that Huff and Puff gas injection is the best recovery option. This method not only reduces the decline rate but also improves wellbore flow. This work considers from PVT tests of the fluid to production forecasts by numerical simulation. The findings will be key to a pilot design. PVT tests were conducted to represent fluid behavior under reservoir conditions. Swelling tests were also performed to analyze the mixtures between the original fluid and injected gas, and two types of gas were studied: dry gas (from the Turboexpander plant outlet) and rich gas (primary separator). During the swelling tests, a known gas was injected into the original fluid at bubble pressure and reservoir temperature, increasing the pressure until total miscibility was achieved. This process was repeated with new gas fractions, determining properties and saturation pressures for each mixture. The study and the pilot test were conducted in the Agua del Cajón Area, in central-eastern Neuquén. The previously detailed laboratory tests served as a basis for characterizing the fluids and adjusting the state equation models that simulate their thermodynamic behavior. Once defined, several runs were carried out in the numerical simulator to optimize oil recovery efficacy through gas injection in the reservoir, considering various gas injection compositions (dry or rich gas), for the pilot well implementation design. The behavior of the wells with the new fluids to be produced was also evaluated, as well as the separation conditions for each injection alternative, seeking to optimize operating conditions to maximize oil recovery. Extremely interesting results were obtained showing an increase in RF and observing the change in the original fluid's behavior from light black oil to a "near critical" fluid after gas injection. This promotes the application of this technique to increase associated production and optimize the development of this type of reserves. A pilot will be designed to be implemented in a current well in the area to evaluate performance and model adjustment.
- North America > United States (0.93)
- South America > Argentina > Neuquén Province > Neuquén (0.48)
- Geology > Geological Subdiscipline (0.93)
- Geology > Petroleum Play Type > Unconventional Play (0.67)
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Mudrock > Shale (0.50)
- South America > Argentina > Patagonia > Neuquén > Neuquen Basin > Vaca Muerta Shale Formation (0.99)
- North America > United States > Texas > Permian Basin > Yeso Formation (0.94)
- North America > United States > Texas > Permian Basin > Yates Formation (0.94)
- (28 more...)
- Reservoir Description and Dynamics > Unconventional and Complex Reservoirs (1.00)
- Reservoir Description and Dynamics > Improved and Enhanced Recovery > Gas-injection methods (1.00)
- Production and Well Operations > Well & Reservoir Surveillance and Monitoring (1.00)
- Production and Well Operations > Artificial Lift Systems > Gas lift (1.00)