1. Experiments, CFD Simulation and Modeling of Sand Wear and Performance Degradation in ESPs 

Sand erosion is a great challenge to the reliability and life expectancy of electrical submersible pumps (ESP). Erosion also causes degradation of ESP performance. In this study, CFD simulations will be carried out with the selected sand particle sub closure relationships to investigate the sand particle distribution and trajectories in the ESP flow. Experiments will be conducted to observe the erosion rates under different flow conditions and sand sizes, properties and concentrations. The ESP performance will be measured. Stages with different designs and materials may be tested simultaneously as they are arranged in series in an ESP. The experimental results will be compared with the CFD simulations. Areas in an ESP vulnerable to sand erosion will be identified. Mechanistic model will be developed for the predictions of sand wear and performance degradation in the ESPs. Recommendations will be made for better handling of sand erosion problems. 

  1. High Oil Viscosity and Water/Oil Emulsion Effects on ESP Performance  

When oil and water are produced, ESP pumps are well known for creating very tight emulsions, which cannot be easily broken on the surface. Emulsion may also have a significant effect on the pump performance due to the dramatic increase of the effective viscosity near the inversion point. The emulsion creation is related to the selections of the pump type and stage number, and the estimation of electric power consumption. On the other hand, emulsion can give a low effective viscosity if high viscosity oil is dispersed in water. The objective of this research is to investigate the effects of W/O and O/W emulsions on the ESP stage and overall performances (flow, head and power). At the same time, the emulsions generated in an ESP stage are characterized. CFD simulations will be carried out and the results will be compared with experimental measurements. The closure laws will be evaluated. Different oil viscosities and interfacial tensions will be used in the experiments and simulations. 

  1. CFD Simulation, Experiments and Modeling of ESP Performance under Gassy Conditions 

TUALP has conducted extensive experimental studies on ESP performance with various gas involvements. The competition between turbulent homogenizing and centrifugal separation in an ESP stage under different flow conditions can be better understood with Computational Fluid Dynamics (CFD) simulation. This is especially important for hydrocarbon fluid production. The pressure and temperature distributions will also affect phase change. With the advances of computing power and simulation software, CFD becomes more powerful for simulations of single-phase and multiphase flows in turbine machinery and rotation equipment such as ESP. In this study, CFD simulation will be used to better understand the flow behavior and to develop mechanistic models for the prediction of ESP performance at different GVF, rpm and fluid properties. Experiments will be conducted to measure ESP performance under different GVF. The interfacial tension effect will be investigated with different surfactant concentrations. CFD simulation results will be compared with experimental data and observations under different flow conditions to identify the deficiencies of the closure laws used in the CFD simulations. CFD simulation can also assist with the design of experimental programs and the interpretation of experimental results. 

Contour surface of 30% gas void fraction corresponding to 10% GVF in ESP impeller 

  1. Transient Gas Lift Modeling 

Gas lift is widely used for high productivity wells and offshore productions. A gas-lift system needs to be designed and optimized considering the PVT and flow rate changes during the well’s entire production life. At decline stages of the oil fields, gas-lift instabilities such as “casing heading” and “density wave” may occur and cause fluctuations and production losses. Zhang et al. (2003, 2006) developed steady state unified models for gas/liquid two-phase and oil/water/gas three-phase flows in wells and pipelines of different inclination angles. In this study, the Zhang et al. models are further extended to transient hydrodynamics to simulate various onshore and offshore gas lift unsteady operations and instabilities. A graphical user interface (GUI) will also be developed using VB.net. The model will be validated through comparisons with the TUALP gas-lift experimental results, available field data and commercial transient simulator calculations. 

  1. Mechanistic Modeling of ESP Performance for Single-Phase and Gas-Liquid Flows 

Normally only the water performance curve is provided by centrifugal pump manufacturers. Pump performance needs to be predicted for different fluid properties including viscosity and density. Starting from the Euler equation for centrifugal pump, individual losses and overall performance of an electrical submersible pump (ESP) are mechanistically modeled for single-phase liquid flow. The model uses a best match flow rate at which the flow direction at the impeller outlet matches the designed flow direction. When the flow rate is lower or higher than the best match flow rate, the theoretical fluid velocity at the impeller outlet needs to be projected to the flow direction corresponding to the best match flow rate. If the projected velocity is higher than the continuity velocity, the difference may be lost due to recirculation in the impeller. For the friction losses in impeller and diffuser, the friction factors are adjusted considering the flow conditions and geometry of the pump. Losses due to change of flow direction through the pump and leakage are also included in the model. The mechanistic model is continuously improved with better closure relationships and validated with experimental results. 

Gas is commonly produced with oil (and water). Accurate prediction of ESP performance for gas-liquid flow is important for production and artificial lift system design and optimization. Assuming gas dispersion in liquid, the gas slippage velocity can be calculated based on the balance of centrifugal force and drag force on a gas bubble in a rotating ESP impeller. Then, the gas void fraction and mixture density in the impeller can be calculated and the ESP boosting pressure can be calculated. This mechanistic model considers effects of gas volumetric fraction, gas density (pressure effect), bubble size (related to turbulence, shear, interfacial tension, etc.), liquid viscosity, rotation speed, and flow rate as well as the pump geometry. 

  1. Transient Modeling of Plunger Lift 

Plunger lift is an effective method to unload liquid (condensate or water) from gas wells. It is also suitable for low to medium rate oil production from high GOR wells. However, current plunger lift operation is largely based on rules of thumb and experience. Plunger lift is a transient flow process. In this study, a mechanistic transient model and a computer program will be developed to simulate the plunger position and liquid/gas production in different phases of a plunger lift cycle, namely, upstroke, gas production, downstroke and pressure buildup. The model can be used for different well geometries and fluid properties and it covers the flow and pressure change from reservoir to separator. The back pressure Inflow Performance Relationship (IPR) will be used for the gas production from reservoir to the bottom hole. The flow and pressure drop through the pipeline between the lubricator and the separator will also be considered. Additionally, the fluctuation of liquid level in the casing-tubing annulus during the plunger lift cycle will be evaluated. Produced gas is distributed between the tubing and the casing-tubing annulus based on their area fractions during pressure buildup. Liquid leakage (or fall back) will also be considered. In this model, the performance of different plunger types can be incorporated. The computer program for the transient model will be written in Fortran. A VB.net GUI will be developed to test the model performance. The transient model will be validated with field data. 

  1. Artificial Lift Application in Unconventional Reservoir Wells 

Mishrif is a tight carbonate reservoir in Minagish field, West Kuwait and has natural fractures and highly viscous reservoir fluid. Oil viscosity is ranging from 40 cp to 60 cp with an oil density of about 20°API at reservoir conditions which has classified Mishrif as an unconventional reservoir. Worldwide, some of these reservoirs are facing a rapid pressure changes and production decline, therefore it required an Artificial Lift (AL) technology to overcome the pressure issue.  

Optimizing production and recovery by ensuring the proper AL technology for unconventional reservoir in conjunction with optimum surface pressure is required. Progressive Cavity Pump (PCP), Electrical Submersible Pump (ESP) and Sucker Rod Pump (SRP), etc. will be affected by the challenging conditions/environment where they are installed. Thus, the pump performance and run life are going to change but a quantitative and qualitative study is required for better understanding, proper planning, and reserve recovery.  

The objectives of this study include:    

  1. To improve the reliability, efficiency, and applicability of AL approaches for unconventional reservoir by a better understanding of the optimum AL technologies. 
  1. Development of selection and operation methodology of AL technology for unconventional reservoirs, support in providing a recommendation of pump setting depth and effect of DLS. 
  1. Study the effect of pressure changes and pump performance due to stimulation activities (acid matrix & fracture).  
  1. Problematic Case Characterization and Data Interpretation for Well Surveillance 

ESP well surveillance and smart well development require problematic case characterization and data interpretation methodologies. From surveillance data and simulation, equipment failures, flow assurance problems and abnormal operations can be classified. The trends and signatures of problematic cases can be characterized. Interpretation criteria and logics for anomaly identification in real-time data can be developed. 

A practical list of problematic cases can be identified through analyses of available surveillance data, feedbacks from equipment manufacturers and producers, literature review, and research experiences of the Tulsa University Artificial Lift Projects (TUALP). For the problematic case characterization, two approaches can be used.  First, simultaneous measurements of different parameters at different points in a production system with known abnormal occurrences can be analyzed and the correspondent trends/signatures can be identified.  Secondly, the mechanistic models for multiphase flow in wells and artificial lift equipment can be used to simulate the parameter trends in the system under different flow conditions. Assuming an abnormal occurrence, the parameter trends at different measurement locations can be characterized. TUALP has developed some of the best mechanistic models for this purpose. The interpretation criteria and logics will be developed through analyses of data and model simulations under different flow conditions. 

TUALP RESEARCH PROJECTS


Experiments and CFD Simulation of ESP Sand Erosion

PI: Saul Gomez, Research Assistant, PhD Student

Objectives:
• Experimental observation of sand erosion and measurement of pump performance
• CFD simulations of sand particle distribution and trajectory in ESP
• Compare simulation and experimental results to identify vulnerable areas in an ESP

Abstract:
Sand erosion is a great challenge to the reliability and life expectancy of electrical submersible pumps (ESP). Erosion also causes degradation of ESP performance. In this study, CFD simulations will be carried out with the selected sand particle sub closure relationships to investigate the sand particle distribution and trajectories in the ESP flow. Experiments will be conducted to observe the erosion rates under different flow conditions and sand sizes, properties and concentrations. The ESP performance will be measured. Stages with different designs and materials may be tested simultaneously as they are arranged in series in an ESP. The experimental results will be compared with the CFD simulations. Areas in an ESP vulnerable to sand erosion will be identified. Recommendations will be made for better handling of sand erosion problems.


Oil/Water Flow and Emulsion Formation in Electrical Submersible Pumps

PI: Daniel Croce, Research Assistant, MSc Student

Objectives:
• Measure ESP stage performance at different water cuts
• Characterize emulsion from an ESP stage
• CFD simulation of oil and water flow in ESP and comparison with measured performance

Abstract:
When oil and water are produced, ESP pumps are well known for creating very tight emulsions, which cannot be easily broken on the surface. Emulsion may also have a significant effect on the pump performance due to the dramatic increase of the effective viscosity near the inversion point. The emulsion creation is related to the selections of the pump type and stage number, and the estimation of electric power consumption. On the other hand, emulsion can give a low effective viscosity if high viscosity oil is dispersed in water. The objective of this research is to investigate the effects of W/O and O/W emulsions on the ESP stage and overall performances (flow, head and power). At the same time, the emulsions generated in an ESP stage are characterized. CFD simulations will be carried out and the results will be compared with experimental measurements. The closure laws will be evaluated. Different oil viscosities and interfacial tensions will be used in the experiments and simulations.


CFD Simulation, Experiments and Modeling of ESP Performance under Gassy Conditions

Contour surface of 30% gas void fraction corresponding to 10% GVF

PI: Jianjun Zhu, Research Assistant, PhD Student

Objectives:
• CFD simulation of gas-liquid flow in ESP
• Measurements of ESP performance under different gas volumetric fractions (GVF)
• Mechanistic modeling of ESP performance under gassy conditions
• Compare simulation and modeling results with experimental observations and measurements

Abstract:
TUALP has conducted extensive experimental studies on ESP performance with various gas involvements. The competition between turbulent homogenizing and centrifugal separation in an ESP stage under different flow conditions can be better understood with Computational Fluid Dynamics (CFD) simulation. This is especially important for hydrocarbon fluid production. The pressure and temperature distributions will also affect phase change. With the advances of computing power and simulation software, CFD becomes more and more powerful for simulations of single-phase and multiphase flows in turbine machinery and rotation equipment such as ESP. In this study, CFD simulation will be used to better understand the flow behavior and to develop mechanistic models for the prediction of ESP performance at different GVF, rpm and fluid properties. Experiments will be conducted to measure ESP performance under different GVF. The interfacial tension effect will be investigated. CFD simulation results will be compared with experimental data and observations under different flow conditions to identify the deficiencies of the closure laws used in the CFD simulations. CFD simulation can also assist with the design of experimental programs and the interpretation of experimental results.
 


Experimental Study of Oil/Water/Gas Three-Phase ESP Performance

PI: Hattan Banjar, Research Assistant, PhD Student

Objectives:
• Measurements of ESP performance under oil/water/gas three-phase flow conditions with different GVF and water cuts
• Mechanistic modeling of three-phase ESP performance
• CFD simulation of dispersed three-phase flow in ESP
• Compare simulation and modeling results with experimental observations and measurements

Abstract:
Oil, water and gas are often produced simultaneously onshore or offshore. When an ESP pumps oil and water with entrained gas, the pump performance may display significant difference from liquid/gas two-phase performance. In oil/water/gas three-phase flow through an ESP, either oil or water is continuous and the other liquid phase and gas are dispersed. The dispersed droplets and bubbles may interact with each other. If water is dispersed, water droplets will move toward the outside due to the centrifugal effect and its higher density. If oil is dispersed in water, the oil droplets will likely distribute differently. An experimental study can also provide three-phase ESP performance metrics and help understand flow behavior. Mechanistic modeling and preliminary CFD simulation can be carried out using closure laws selected from literature.
 


Artificial Lift Conditioning for Deviated and Horizontal Wells

PI’s: Dr. Xuhui Liu, Visiting Scholar from Yantze University, China
        Dr. Fachun Liang, Visiting Scholar from China University of Petroleum, China

Objectives:
• Downhole slugging modeling for multiphase flow from wells with different orientations
• Investigate downhole separation strategies with special separator, rat hole or special well heel geometries

Abstract:
Slugging is a major challenge to the performance of different pumps such as ESP and rod pump. The alternative arrivals of liquid slug and gas pocket of slug flow make separation difficult and cause significant gas entrainment which deteriorates pump performance. This problem is exacerbated in highly deviated and horizontal wells, where slug and gas pocket lengths tend to be much longer. In this study, multiphase flow in deviated and horizontal wells with distributed influxes will be modeled. The slugging behavior will be analyzed in order to understand the liquid loading and artificial lift challenges. Different strategies for slugging mitigation and gas-liquid separation will be investigated, such as special separator design and rat hole arrangement. Appropriate drilling and completion can also be proposed for better artificial lift integration in deviated and horizontal well production systems.


Transient Gas Lift Modeling

PI: Fahad Al-Mudairis, Research Assistant, MSc Student

Objectives:
• Develop a transient model to simulate various onshore and offshore gas lift applications
• Integrate current mechanistic models for local multiphase flow behavior predictions
• Analyze gas lift instabilities and compare with commercial simulators

Abstract:
Gas lift is widely used for high productivity wells and offshore productions. A gas-lift system needs to be designed and optimized considering the PVT and flow rate changes during the well’s entire production life. At decline stages of the oil fields, gas-lift instabilities such as “casing heading” and “density wave” may occur and cause fluctuations and production losses. In this study, generic PVT and transient mass conservation models will be developed to simulate various onshore and offshore gas lift applications. The local multiphase flow behavior (momentum conservation) is predicted through integration of the most recent mechanistic models such as the Zhang et al. unified model for oil/water/gas three-phase flows in wells and pipelines of different inclination angles. A graphical user interface (GUI) will also be developed using VB.net. The model will be validated through comparisons with the TUALP gas-lift experimental results, available field data and commercial transient simulator calculations.


Mechanistic Modeling of ESP Performance for Single-Phase and Gas-Liquid Flows

PI: Holden Zhang, TUALP Director

Objective:
• Develop mechanistic model to predict individual losses and overall performance of ESP based on overall pump geometry for different rotation speed, fluid viscosity and density
• Develop mechanistic model to predict ESP performance under different gas-liquid flow conditions, including gas volumetric fraction, gas density (pressure), liquid viscosity and interfacial tension

Abstract:
Normally only the water performance curve is provided by centrifugal pump manufacturers. Pump performance needs to be predicted for different rotation speed and fluids with different viscosities and densities. Starting from the Euler equation for centrifugal pump, individual losses and overall performance of an electrical submersible pump (ESP) are mechanistically modeled for single-phase liquid flow. The model uses a best match flow rate at which the flow direction at the impeller outlet matches the designed flow direction. When the flow rate is lower or higher than the best match flow rate, the theoretical fluid velocity at the impeller outlet needs to be projected to the flow direction corresponding to the best match flow rate. If the projected velocity is higher than the continuity velocity, the difference may be lost due to recirculation in the impeller. For the friction losses in impeller and diffuser, the friction factors are adjusted considering the flow conditions and geometry of the pump. Losses due to change of flow direction through the pump and leakage are also included in the model. The mechanistic model is continuously improved with better closure relationships and validated with experimental results.

Gas is commonly produced with oil (and water). Accurate prediction of ESP performance for gas-liquid flow is important for production and artificial lift system design and optimization. Assuming gas dispersion in liquid, the gas slippage velocity can be calculated based on the balance of centrifugal force and drag force on a gas bubble in a rotating ESP impeller. Then, the gas void fraction and mixture density in the impeller can be calculated and the ESP boosting pressure can be calculated. This mechanistic model considers effects of gas volumetric fraction, gas density (or density difference between gas and liquid, reflecting pressure effect), bubble size (as a result of turbulence, shear, interfacial tension, etc.), liquid viscosity, rotation speed, and flow rate as well as the pump geometry.


Transient Modeling of Plunger Lift for Liquid Unloading from Gas Wells and Oil Production from High GOR Wells

PI’s: Kun Bo, Visiting Scholar, Jilin University, China
        Holden Zhang, TUALP Director

Objectives:
• Review previous models developed at TUALP and other institutions
• Develop transient model to simulate plunger lift process and predict plunger lift performance
• Develop operation procedure for plunger lift optimization

Abstract:
Plunger lift is an effective method to unload liquid (condensate or water) from gas wells. It is also suitable for low to medium rate oil production from high GOR wells. However, current plunger lift operation is largely based on rules of thumb and experience. Plunger lift is a transient flow process. In this study, a mechanistic transient model and a computer program will be developed to simulate the plunger position and liquid and gas production in different phases of a plunger lift cycle, namely, upstroke, gas production, downstroke and pressure buildup. The model can be used for different well geometries and fluid properties and it covers the flow and pressure change from reservoir to separator. The back pressure Inflow Performance Relationship (IPR) will be used for the gas production from reservoir to the bottom hole. The flow and pressure drop through the pipeline between the lubricator and the separator will also be considered. Additionally, the fluctuation of liquid level in the casing-tubing annulus during the plunger lift cycle will be evaluated. Produced gas is distributed between the tubing and the casing-tubing annulus based on their area fractions during pressure buildup. Liquid leakage (or fall back) will also be considered. In this model, the performance of different plunger types can be incorporated. The computer program for the transient model will be written in Fortran. A VB.net GUI will be developed to test the model performance. The transient model will be validated with field data.


Modeling of Artificial Lift Integration in Production System

PI: Holden Zhang, TUALP Director

Objectives:
• Oil/water/gas flow modeling in wells and pipelines
• Artificial lift performance prediction under different flow conditions
• Development of artificial lift analyzing tool

Abstract:
Oil, water and gas may be produced simultaneously with different combinations. The multiphase flow behavior needs to be predicted for production system design. The multiphase flow conditions are also required for the selection of suitable artificial lift method. The objective of this study is to develop an artificial lift analyzing tool for the prediction of flow behavior during production and transportation of gas, oil and water through wells and pipelines at different inclination angles (-90° to 90° from horizontal) based on Zhang et al. (2003, 2006) unified model. Different artificial lift methods can be evaluated with the local multiphase flow conditions (pressure, temperature, flow pattern, etc.). Their performances can be analyzed through integration with the production system.