Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
Results
Summary Many horizontal oil wells will after a time start producing unwanted fluids. Autonomous inflow control valves may help to choke these unwanted fluids and consequently improve carbon efficiency. This paper publishes new experimental data describing how an autonomous inflow control valve manages medium-light oil (6 cp), water, and gas at reservoir conditions. A further objective is to evaluate how this valve might impact well performance under various conditions. To verify the single- and multiphase flow behavior of the valve, extensive flow loop experiments were performed. Initial testing was done in a model fluid laboratory, while a more extensive test was performed at reservoir conditions (i.e., with formation water, reservoir oil, and hydrocarbon gas at the given reservoir temperature and pressure). To explore and understand the impact of this valve for various reservoir scenarios, a simple conceptual reservoir model with realistic boundary conditions was used. At various differential pressures, the single-phase oil, water, and gas rates were measured. Performance at varying water and gas fractions was measured to get an improved understanding and knowledge of multiphase flow occurring in a well. The results show clearly that the valve will choke gas and water effectively, both at single-phase and multiphase flow conditions. The reservoir and model fluid evaluations show consistent results. The valve shows roughly a monotonic decreasing total rate with decreasing oil fraction, implying that the valve will always prioritize sections with the largest oil fraction. A mathematical model match of the valve performance is possible via the 10-parameter extended autonomous inflow control device (AICD) equation that enables practical evaluation of the valve in industry-standard reservoir simulators. Various scenarios are explored with a conceptual reservoir model, and the autonomous inflow control valve shows its capacity to reduce water production and enable a more gradual and controlled increase in gas/oil ratio for most scenarios. The autonomous inflow control valve shows its largest potential to reduce unwanted fluids and increase oil recovery when used in segmented reservoirs. In cases with uncertain aquifer and/or gas cap strength, or large variation in effective permeability, the valve will make an infill well more robust as it autonomously adapts to reality, chokes unwanted fluids, and consequently enables more carbon-efficient reservoir management.
- North America > United States (1.00)
- Asia > Middle East (0.94)
- Europe > Norway > North Sea > Northern North Sea (0.28)
- Europe > Norway > North Sea > Northern North Sea > South Viking Graben > Vana Basin > RL 088 BS > Block 25/4 > Alvheim Field > Lista Formation > Våle Formation (0.99)
- Europe > Norway > North Sea > Northern North Sea > South Viking Graben > Vana Basin > RL 088 BS > Block 25/4 > Alvheim Field > Lista Formation > A2 North Heimdal T60 Formation (0.99)
- Europe > Norway > North Sea > Northern North Sea > South Viking Graben > Vana Basin > RL 088 BS > Block 25/4 > Alvheim Field > Hermod Formation > Våle Formation (0.99)
- (37 more...)
An Insight into the Prediction of Scale Precipitation in Harsh Conditions Using Different Machine Learning Algorithms
Yousefzadeh, Reza (Amirkabir University of Technology (Tehran Polytechnic)) | Bemani, Amin (Amirkabir University of Technology (Tehran Polytechnic)) | Kazemi, Alireza (Sultan Qaboos University (Corresponding author)) | Ahmadi, Mohammad (Amirkabir University of Technology (Tehran Polytechnic))
Summary Scale precipitation in petroleum equipment is known as an important problem that causes damages in injection and production wells. Scale precipitation causes equipment corrosion and flow restriction and consequently a reduction in oil production. Due to this fact, the prediction of scale precipitation has vital importance among petroleum engineers. In the current work, different intelligent models, including the decision tree, random forest (RF), artificial neural network (ANN), K-nearest neighbors (KNN), convolutional neural network (CNN), support vector machine (SVM), ensemble learning, logistic regression, Naïve Bayes, and adaptive boosting (AdaBoost), are used to estimate scale formation as a function of pH and ionic compositions. Also, a sensitivity analysis is done to determine the most influential parameters on scale formation. The novelty of this work is to compare the performance of 10 different machine learning algorithms at modeling an extremely non-linear relationship between the inputs and the outputs in scale precipitation prediction. After determining the best models, they can be used to determine scale formation by manipulating the concentration of a variable in accordance with the result of the sensitivity analysis. Different classification metrics, including the accuracy, precision, F1-score, and recall, were used to compare the performance of the mentioned models. Results in the testing phase showed that the KNN and ensemble learning were the most accurate tools based on all performance metrics of solving the classification of scale/no-scale problem. As the output had an extremely non-linear behavior in terms of the inputs, an instance-based learning algorithm such as the KNN best suited the classification task in this study. This argumentation was backed by the classification results. Furthermore, the SVM, Naïve Bayes, and logistic regression performance metrics were not satisfactory in the prediction of scale formation. Note that the hyperparameters of the models were found by grid search and random search approaches. Finally, the sensitivity analysis showed that the variations in the concentration of Ca had the highest impact on scale precipitation.
- Asia (1.00)
- Europe (0.93)
- North America > United States > Texas (0.46)
- Geology > Mineral > Sulfate (0.48)
- Geology > Geological Subdiscipline (0.46)
- Information Technology > Artificial Intelligence > Machine Learning > Performance Analysis > Accuracy (1.00)
- Information Technology > Artificial Intelligence > Machine Learning > Neural Networks (1.00)
- Information Technology > Artificial Intelligence > Machine Learning > Statistical Learning > Regression (0.88)
- (2 more...)
Experimental Investigation of the Effects of Fluid Viscosity on Electrical Submersible Pumps Performance
Monte Verde, W. (Center for Petroleum Studies, University of Campinas (Corresponding author)) | Kindermann, E. (School of Mechanical Engineering, University of Campinas) | Biazussi, J. L. (Center for Petroleum Studies, University of Campinas) | Estevam, V. (Center for Petroleum Studies, University of Campinas) | Foresti, B. P. (Petrobras) | Bannwart, A. C. (Center for Petroleum Studies, University of Campinas)
Summary Electrical submersible pumps (ESPs) are an important artificial lift method used in oil production. ESPs can provide high production flow rate, are flexible, and can be installed in highly deviated wells, subsea deepwater wells, or on the seabed. ESP performance is generally characterized by manufacturers using only water as fluid. However, oil properties are very different from water and significantly alter the pump's performance. Operating ESPs with viscous fluids leads to degraded pump performance. Therefore, knowing the ESP's performance when pumping viscous fluid is essential to properly design the production system. In this work, we present an experimental study of ESP performance operating with viscous flow. A total of six ESP models were tested, operating at four different rotational speeds and 11 viscosities, resulting in a comprehensive database of more than 5,800 operating conditions. This database contributes to the literature given the lack of available data. We also perform a phenomenological analysis on the influence of operational parameters, such as viscosity, rotational speed, specific speed, and rotational Reynolds number. The database and analyses performed are central for future models predicting the viscous performance of ESPs. The results from our investigation and tests showed that the increase in viscosity causes (1) a reduction in the head and (2) an increase in drive power, resulting in (3) a sharp decrease in efficiency. However, increasing rotational speed tends to mitigate this performance degradation. Efficiency and flow rate correction factors are virtually independent of the flow rate within the recommended operating region. This is not true for the head correction factor, which is not constant. The pump geometry seems to influence its performance as ESPs with higher specific speed are less impaired by viscous effects. The database obtained in the present work is available in the data repository of the University of Campinas, at the address presented by Monte Verde et al. (2022). Introduction Throughout the history of petroleum exploration, centrifugal pumps have proved to be effective for oil production.
- South America > Brazil (0.68)
- North America > United States > Texas (0.28)
An Insight into the Prediction of Scale Precipitation in Harsh Conditions Using Different Machine Learning Algorithms
Yousefzadeh, Reza (Amirkabir University of Technology (Tehran Polytechnic)) | Bemani, Amin (Amirkabir University of Technology (Tehran Polytechnic)) | Kazemi, Alireza (Sultan Qaboos University (Corresponding author)) | Ahmadi, Mohammad (Amirkabir University of Technology (Tehran Polytechnic))
Summary Scale precipitation in petroleum equipment is known as an important problem that causes damages in injection and production wells. Scale precipitation causes equipment corrosion and flow restriction and consequently a reduction in oil production. Due to this fact, the prediction of scale precipitation has vital importance among petroleum engineers. In the current work, different intelligent models, including the decision tree, random forest (RF), artificial neural network (ANN), K-nearest neighbors (KNN), convolutional neural network (CNN), support vector machine (SVM), ensemble learning, logistic regression, Naïve Bayes, and adaptive boosting (AdaBoost), are used to estimate scale formation as a function of pH and ionic compositions. Also, a sensitivity analysis is done to determine the most influential parameters on scale formation. The novelty of this work is to compare the performance of 10 different machine learning algorithms at modeling an extremely non-linear relationship between the inputs and the outputs in scale precipitation prediction. After determining the best models, they can be used to determine scale formation by manipulating the concentration of a variable in accordance with the result of the sensitivity analysis. Different classification metrics, including the accuracy, precision, F1-score, and recall, were used to compare the performance of the mentioned models. Results in the testing phase showed that the KNN and ensemble learning were the most accurate tools based on all performance metrics of solving the classification of scale/no-scale problem. As the output had an extremely non-linear behavior in terms of the inputs, an instance-based learning algorithm such as the KNN best suited the classification task in this study. This argumentation was backed by the classification results. Furthermore, the SVM, Naïve Bayes, and logistic regression performance metrics were not satisfactory in the prediction of scale formation. Note that the hyperparameters of the models were found by grid search and random search approaches. Finally, the sensitivity analysis showed that the variations in the concentration of Ca had the highest impact on scale precipitation.
- Asia (1.00)
- Europe (0.93)
- North America > United States > Texas (0.46)
- Geology > Mineral > Sulfate (0.48)
- Geology > Geological Subdiscipline (0.46)
- Information Technology > Artificial Intelligence > Machine Learning > Performance Analysis > Accuracy (1.00)
- Information Technology > Artificial Intelligence > Machine Learning > Neural Networks (1.00)
- Information Technology > Artificial Intelligence > Machine Learning > Statistical Learning > Regression (0.88)
- (2 more...)
Experimental Investigation of the Effects of Fluid Viscosity on Electrical Submersible Pumps Performance
Monte Verde, W. (Center for Petroleum Studies, University of Campinas (Corresponding author)) | Kindermann, E. (School of Mechanical Engineering, University of Campinas) | Biazussi, J. L. (Center for Petroleum Studies, University of Campinas) | Estevam, V. (Center for Petroleum Studies, University of Campinas) | Foresti, B. P. (Petrobras) | Bannwart, A. C. (Center for Petroleum Studies, University of Campinas)
Summary Electrical submersible pumps (ESPs) are an important artificial lift method used in oil production. ESPs can provide high production flow rate, are flexible, and can be installed in highly deviated wells, subsea deepwater wells, or on the seabed. ESP performance is generally characterized by manufacturers using only water as fluid. However, oil properties are very different from water and significantly alter the pump’s performance. Operating ESPs with viscous fluids leads to degraded pump performance. Therefore, knowing the ESP’s performance when pumping viscous fluid is essential to properly design the production system. In this work, we present an experimental study of ESP performance operating with viscous flow. A total of six ESP models were tested, operating at four different rotational speeds and 11 viscosities, resulting in a comprehensive database of more than 5,800 operating conditions. This database contributes to the literature given the lack of available data. We also perform a phenomenological analysis on the influence of operational parameters, such as viscosity, rotational speed, specific speed, and rotational Reynolds number. The database and analyses performed are central for future models predicting the viscous performance of ESPs. The results from our investigation and tests showed that the increase in viscosity causes (1) a reduction in the head and (2) an increase in drive power, resulting in (3) a sharp decrease in efficiency. However, increasing rotational speed tends to mitigate this performance degradation. Efficiency and flow rate correction factors are virtually independent of the flow rate within the recommended operating region. This is not true for the head correction factor, which is not constant. The pump geometry seems to influence its performance as ESPs with higher specific speed are less impaired by viscous effects. The database obtained in the present work is available in the data repository of the University of Campinas, at the address presented by Monte Verde et al. (2022).
- South America > Brazil (0.68)
- North America > United States > Texas (0.28)