|Theme||Visible||Selectable||Appearance||Zoom Range (now: 0)|
Zhang, Shaojie (Petroleum Engineering, University of North Dakota) | Li, Chuncheng (Petroleum Engineering, University of North Dakota) | Pu, Hui (Petroleum Engineering, University of North Dakota) | Ling, Kegang (Petroleum Engineering, University of North Dakota) | Sun, Runxuan (Petroleum Engineering, University of North Dakota) | Zhao, Julia Xiaojun (Department of Chemistry, University of North Dakota)
In this study, six core samples were obtained from the Middle Bakken Formation in North Dakota. Before the imbibition experiment, petrophysical analysis were conducted for the samples. XRD method was used to analyze the mineral composition. Nitrogen adsorption and SEM methods were combined to study the pore size distribution and microstructures. Then the authors performed brine imbibition and surfactant imbibition for six Bakken cores and two Berea sandstones. Before the experiment, the cores were fully filled with Bakken crude oil. The core plugs were then submerged into the brine and surfactant solutions with all-face-open (AFO) condition. Experiments of brine and surfactant imbibing into oil-filled cores were carried out with recording of recovered oil volume using imbibition cells. Different types of surfactants such as cationic, anionic, and nonionic, were tested in the study. Those experiments evaluate the oil displacement efficiencies of brine and different surfactants in Bakken rocks.
Fu, Hao (University of North Dakota, Grand Forks) | Long, Yifu (Missouri University of Science and Technology) | Wang, Sai (University of North Dakota, Grand Forks) | Wang, Yanbo (University of North Dakota, Grand Forks) | Yu, Peng (Beibu Gulf University, Qinzhou) | Ling, Kegang (University of North Dakota, Grand Forks)
Geological carbon sequestration through injecting large-scale carbon dioxide (CO2) into the deep saline aquifers represents a long-term storage of CO2. In the CO2 sequestration process, the injected CO2 is displacing water from the injection point and is expected to remain in the reservoir. Due to the nature of one phase displacing another phase in porous media, it is noted that different water saturation exists in the CO2 plume during the displacement. Water distribution in the plume will affect the size of the plume subsurface. Furthermore, the gravitational segregation between CO2 and water will cause overriding- tonguing during the injection and impact the shape of plume. To better understand the CO2 movement underground and development of CO2 plume, it is necessary to take the two-phase flow and gravity force effects into account when evaluating CO2 displacing water. The displacement of water by injecting CO2 is not a piston-like process in aquifer. Because water is the wetting phase and CO2 is the non-wetting phase when two phases flow in reservoir, water occupies the surface of matrix and small pores while CO2 resides in large pores and centers of pores. As a result, various water saturations distribute behind CO2 front during the displacement. The distribution is a function of fluid and rock properties, fluid-rock interaction, and injection operation. In this study, these factors are considered when developing new models to predict CO2 plume evolution during injection. Mass conservation, multiphase flow, and equation-of-states are applied in the derivation of the models, which guarantees a rigorous approach in the investigation. The modeling results indicate that CO2 does not displace water completely away from the plume. The shape of the CO2 front is controlled by the relative permeability of two phases and capillary pressure. Water saturation profile from CO2 injecting point to the displacement front shows that water saturation behind the CO2 front increases outwardly, and the change in saturation is non-linear. The injection rate impacts the sharpness of the CO2 front, thus leads to different gas plume sizes for same injection volume. The outward movement of the CO2 front decelerates as injection time goes on. The research illustrates that injection experiences two stages: transient and steady-state, in which the displacement behavior and the development of gas plume vary. Although the duration of transient stage is dictated by size of aquifer and is relatively short comparing with steady-state stage, its influence on the development of CO2 plume cannot be neglected when selecting gas compressor horsepower and determining injection rate.
Wang, Sai (University of North Dakota) | Han, Juan (University of North Dakota) | Wang, Yanbo (University of North Dakota) | Ling, Kegang (University of North Dakota) | Jia, Bao (University of North Dakota) | Wang, Hongsheng (Virginia Tech) | Long, Yifu (Missouri University of Science and Technology)
The low recovery of oil from the tight liquid-rich formations is still a main challenge for the tight reservoir. Thus, in order to break the chains and remove the obstacle such as the low recovery factor in the Bakken tight formation, even though the horizontal drilling and hydraulic fracturing technologies were already well applied in this field, the supercritical CO2 flooding was proposed as an immense potential recovery method for the production improvement. In this research, we conducted a series of CO2 flooding experiments under various injection pressure (2500psi, 2800psi, 3000psi, 3500psi), to investigate the recovery potential of the core sample from Bakken tight formation. Also, the NMR analysis was processed of the core samples flooded with CO2 agent under the above injection pressure variables. The result comparison demonstrates that, with the supercritical CO2 injection pressure increase, the recovery factor gets incremental trend from 8.8% up to 33% recovery. Also, the macro pore and natural fracture system were proved to contribute more on the recovery potential. After reaching the miscible phase between the CO2 and oil in the sample, the hydrocarbon existed in the micro pores start the contribution to the recovery potential. Thus, The CO2 was identified as a potential recovery agent and the supercritical CO2 EOR method was proposed as the potential recovery technology due to the high recovery factor obtained in the immiscible and miscible processes.
The supersonic separation is a new approach to dehydrate the natural gas in recent years. In the conventional structure, the straight tube is typically combined with a cyclone to create a strong vortex flow. The shock wave usually occurs near the swirling device in the supersonic separator, which can make the flow unstable and decrease the separation efficiency. Due to removing the negative effects of the shockwave, a newtype helical guide blade is designed as the swirling device, installed in the separate straight tube in the supersonic cyclone separator. The flow characteristics in the supersonic separator was investigated and the geometry structure was optimized by performing the computational fluid dynamics modeling methods. The optimization results showed that the model with a converging tube of 190 mm length, a diverging tube of half-cone angle of 5 and a single blade installed in the middle position, is the best supersonic separator model in the dehydration process, which can create the most stable flow field and achieve the optimum separation. In addition, when the outlet back pressure in the diffuser tube is 1 Mpa 1.5 Mpa, the separation performance will be better.
In recent years, the exploration and production of oil and gas from Bakken formation in Williston Basin have proceeded quickly due to the application of multi-stage fracturing technology in horizontal wells. Knowledge of the rock elastic moduli is important for the horizontal drilling and hydraulic fracturing. Although static moduli obtained by tri-axial compression test are accurate, the procedures are cost expensive and time consuming. Therefore, developing correlation to predict static moduli from dynamic moduli, which is calculated from sonic wave velocities, is meaningful in cutting cost and it makes the unconventional oil and gas exploration and production more efficient.
Literature review indicates such a correlation is not available for Bakken formation. This may be attributed to the extremely low success rate in Bakken core sample preparation and not enough published data to develop correlation to relate dynamic moduli to static moduli. This study measures and compares the moduli obtained from sonic wave velocity tests with deformation tests (tri-axial compression tests) for the samples taken from Bakken formation of Williston Basin, North Dakota, USA. The results show that the dynamic moduli of Bakken samples are considerably different from the static moduli measured by tri-axial compression tests. Correlations are developed based on the static and dynamic moduli of 117 Bakken core samples. The cores used in this study were taken from the core areas of Bakken formation in Williston Basin. Therefore, they are representatives of the Bakken reservoir rock. These correlations can be used to evaluate the uncertainty of Bakken formation elastic moduli estimated from the seismic and/or well log data and adjust to static moduli at a lower cost comparing with conducting static tests. The correlations are crucial to understand the rock geomechanical properties and forecast reservoir performance when no core sample is available for direct measurement of static moduli.
Gaither Draw Unit is a heterogeneous and tight formation with an average permeability less than 0.1 mD. After more than 1.7 MMSTB water injection, there was no clear indication or benefit of the injected water from any producer. However, knowing the distribution of the injected water is critical for future well planning and quantifying the efficiency of injection. The objective of this study is to show how the Capacitance-Resistance Model (CRM) was used on this field and validated using other independent methods.
The CRM model describes the connectivity and the degree of fluid storage quantitatively between injectors and producers from production and injection rates. Rooted in material balance, signals from injectors to producers can be captured in the CRM. Using constrained nonlinear multivariable optimization techniques, the connectivity is estimated in the selected portion of the field through signal analysis on injection and production rates. In this tight formation, the whole field is divided into seven regions with one injection well and surrounding producers to conduct CRM analysis. We further use integrated but independent approaches to validate the results from CRM. The validation includes full field modeling and history match and fluid level measurement using echometering technology.
This paper focuses on a real field water flooding project in Gaither Draw Units(GDU). CRM is used to detect reservoir heterogeneity through quantifying communication between injectors and producers, and attains a production match. The fitting results of connectivity through CRM indicate permeability regional heterogeneity, which is consistent with full field modelling. The history matched full field model presents the saturation distribution showing that the majority of injected water mainly saturates the surrounding regions of injectors, and the low transmissibility slows down the pressure dissipation. Overall, the comprehensive interpretation obtained through these three independent methods is consistent, and is very useful in planning infill well drilling and future development plan for the Gaither Draw Units.
This paper shows that it is critical to integrate different sources of data in reservoir management through a field case study. The experience and observations from this asset can be applied to other tight formations being developed with water flooding projects.
Flow pattern of a multi-phase flow refers to the spatial distribution of the phase along transport conduit when liquid and gas flow simultaneously. The determination of flow patterns is a fundamental problem in two-phase flow analysis, and an accurate model for gas-liquid flow pattern prediction is critical for any multiphase flow characterization as the model is used in many applications in petroleum engineering. The objective of this study is to present a new model based on machine learning techniques and more than 8000 laboratory multi-phase flow tests.
The flow pattern is affected by fluid properties, in-situ flow rates of liquid and gas, and flow conduit geometry and mechanical properties. Laboratory data since 1950s have been collected and more than 8000 data points had been obtained. However, the actual flow conditions are significantly different with any laboratory settings. Therefore, several dimensionless variables are derived to characterize these data points first. Then machine learning techniques were applied on these dimensionless variables to develop the flow pattern prediction models. Applying hydraulic fundamentals and dimensional analysis, we developed dimensionless numbers to reduce number of freedom dimensions. These dimensionless variables are easy to use for upscaling and have physical meanings. We converted the collected data from actual laboratory measurement to the variations of these dimensionless variables. Machine learning techniques on the dimensionless variable significantly improved their predictive accuracy. Currently the best matching on these laboratory data was about 80% using the most recently developed semi-analytical models. Using machine learning techniques, we improved the matching quality to more than 90% on the experimental data.
This paper applies machine learning techniques on flow pattern prediction, which has tremendous practical usages and scientific merits. The developed model is better than current existing semi-analytical or classical correlations in matching the laboratory database.
Han, Guoqing (China University of Petroleum, Beijing) | Ma, Gaoqiang (China University of Petroleum, Beijing) | Zheng, Jian (Oilfield E&P Department, Sinopec) | Gao, Yue (Sinopec Research Institute of Safety Engineering) | Zhang, He (University of North Dakota) | Ling, Kegang
Nowadays, unloading gas wells with coiled tubing is a common application to the field. However, it still lacks of adequate understanding of dynamic behavior of the unloading process. This paper investigates the process of liquid unloading by gas lift with coiled tubing under transient condition. This unloading process can be divided into three stages: liquid rising in tubing, liquid slug production, and liquid production by entrainment. In each stage, the mass and linear momentum conversation equations are applied as governing equations. The components of each stage includes coiled tubing, coiled tubing-tubing annulus, liquid slug, gas bubble, and liquid film. Empirical correlations have been used including surface gas injection choke, check valve, friction factor, the relationship between the gas bubble and the liquid slug velocity, inflow performance relationship, and blackoil fluid properties. From the above, the dynamic model coupling real-time change of inflow performance relationship. Using the LU factorization and Euler's method to solve the proposed dynamic model in time domain. Among all these variables, the most important ones include gas injection rates, pressures at various locations, length of the liquid slug and gas bubble, and various velocities. By the simulation efforts, the mechanism of liquid unloading process is revealed.
Gas lift is commonly constrainted by gas availability. This is pioneer study on liquid unloading with coiled tubing is critical in design phase to optimize the usage of injected gas as well as choose appropriate pump and cost-saving for electricity expense.
Sucker rod pumps provide mechanical energy to lift oil from bottom hole to surface when oil wells do not have enough energy to produce the oil through natural flow. It is efficient, simple, easy to be operated, and can be applied to slim holes, multiple completions, and high-temperature and viscous oils. The disadvantages include excessive friction in crooked holes, solid-sensitive problems, low efficiency in gassy wells, limited depth, and bulky volume. The load on the rod is one of the key factors that dictate the maintenance frequency of pumping unit, energy consumed to lift the fluid, and the optimization of pumping system operating parameters.
The cyclic load applied on the rod causes the fatigue and finally the failure of the rod if not designed properly. The rod load is a function of friction force, plunger acceleration/deceleration, weights of plunger, fluid being lifted, and sucker rods string, and the pressures above and below on plunger. Literature review indicates that a model to accurately calculate the load of a pumping cycle is highly desired. In this study, we couple the wellbore with reservoir performance to better analyze the dynamics of pump system, which yields more accurate results.
In this study, force balance during the pumping cycle is analyzed. Friction force due to the movement of the plunger and the rod, buoyant force, and gravity force are included in the modeling. The effects of acceleration and deceleration of the plunger on rod are considered. The sensitivity of pumping speed is investigated. This study proposed a more general model comparing with former researches because more factors that affect the load applying on rod are included. Including the friction force due to the viscous fluid is critical to rod load analysis in pumping heavy oil.
The proposed model is significant to the cyclic fatigue and failure analysis of rod in sucker rod pumping system. It can be used to predict the possible failure point for rod string by analyzing load along the whole string. It is also a useful tool to design the tapered rod string to minimize the maximum rod load while achieving optimum rod string life. Therefore, an optimization of sucker rod pumping system is implemented by balancing the tradeoff between the maximum rod string life (or rod size) and the minimum rod load (or lowest energy consumption).
Zhong, Xun (Department of Petroleum Engineering, University of North Dakota) | Wang, Yuhe (Department of Petroleum Engineering, Texas A&M University at Qatar) | Pu, Hui (Department of Petroleum Engineering, University of North Dakota) | Li, Wei (College of Petroleum Engineering, Northeast Petroleum University) | Yin, Shize (Missouri University of Science and Technology) | Ling, Kegang (Department of Petroleum Engineering, University of North Dakota)
Chemical enhanced oil recovery(EOR) such as polymer flooding and alkali/surfactant/polymer(ASP) flooding have been applied throughout the world for more than several decades. However, few large-scale successes of these technologies have been reported, except in China. The annual crude oil production rate by chemical EOR in Daqing Oilfield has been kept over 10.0 million tonnes (~73.5 million barrels) per year for 16 consecutive years. Considerable experience has gained and lessons have learned on large-scale chemical flooding, including major factors that influence the recovery factor and methods to increase the oil recovery; measures to obtain the highest economic efficiency; and how to minimize the costs.
To date, incremental oil recovered by polymer flooding is over 10.0% OOIP. Due to some disadvantages of polymer flooding, like the existence of inaccessible pore volume and lack of the EOR mechanism of ultralow interfacial tension, some substitute technologies are needed, and ASP flooding is a promising one, which was reported and proved to have an ability of recovering more than 20.0% extra oil over water flooding. However, negative scaling problems caused by alkali are severe and problematic, as a result, weak-alkali ASP (WASP) and even surfactant-polymer (SP) flooding were developed or are under development. To alleviate the bad effets of scale and further improve the performance of ASP flooding, some new technologies such as refracturing and application of new anti-scaling chemicals like compound scale removal agent were applied. SP flooding is a theoretically feasible EOR technology, but there are few field test data available, and more effective surfactants are required.
In this paper, the development of commonly used chemical flooding technologies, normal problems that may merge during commercial implementation, together with the updated solutions are included. All the valuable experience obtained from commercial-scale application of chemical flooding in Daqing Oilfield is not only of great significance for the expansion of Daqing Oilfield itself, but also worth learning by other countries.