The Pennsylvania State University The Graduate School Department of Energy and Mineral Engineering

FIELD PERFORMANCE ANALYSIS AND OPTIMIZATION OF GAS CONDENSATE SYSTEMS USING ZERO-DIMENSIONAL RESERVOIR MODELS

A Thesis in Energy and Mineral Engineering by Pichit Vardcharragosad

2011 Pichit Vardcharragosad

Submitted in Partial Fulfillment of the Requirements for the Degree of

Master of Science

August 2011

The thesis of Pichit Vardcharragosad was reviewed and approved* by the following:

Luis F. Ayala Associate Professor of Petroleum and Natural Gas Engineering Thesis Advisor

R. Larry Grayson Professor of Energy and Mineral Engineering Graduate Program Officer of Energy and Mineral Engineering

Li Li Assistant Professor of Energy and Mineral Engineering

Yaw D. Yeboah Professor of Energy and Mineral Engineering Head of the Department of Energy and Mineral Engineering

*Signatures are on file in the Graduate School

iii

ABSTRACT
Field performance prediction is a crucial piece of information that all relevant parties have to use in their design and decision processes during the development and exploitation of a hydrocarbon reservoir. Field performance analysis is an engineering task which requires knowledge, time, and the right tools and models. Available tools such as commercial reservoir simulators might not always be the most efficient or the most optimum solution even if they make use of highly sophisticated or detailed models. This is because these sophisticated models often require more input data and longer running time than the less detailed ones. These problems become worse when availability of input data and working time are constrained. This thesis aims to develop a field performance model which will allow engineer to perform analysis and optimization tasks more effectively for the case of gas condensate reservoirs. A gas condensate is one of the many fluid types that can be found in conventional hydrocarbon reservoirs. The development of a two-phase condition below the dew point pressure can significantly increases the complexity in engineering performance calculation. The proposed tool utilizes both zero-dimensional reservoir model customized for gas condensate and pseudo component model. Results indicate that both models can provide fairly good prediction results while requiring much less input and running time. Microsoft Excel with built-in Visual Basic for Applications (VBA) is selected as the platform to develop this simulator due to the user-friendly interface, useful built-in features, and high flexibility to use and hard-code modification. The proposed model is able to successfully predict field performance while capturing all major fluid behavior characteristics of gas condensates as well as being capable of performing various optimization tasks effectively. Limitations of the implemented pseudo component model, such as negative solution gas-oil ratios at low reservoir pressure, are elaborated and discussed. The possible sources of error and associated preventive measures derived from the use of a gas

iv condensate tank model for the case volatile oil reservoirs are addressed. Further recommended studies on negative value of decline exponent variable and expanding current capability of the proposed model are also presented.

TABLE OF CONTENTS
List of Figures .......................................................................................................................... vii List of Tables ........................................................................................................................... ix Nomenclature ........................................................................................................................... x Acknowledgements .................................................................................................................. xv Chapter 1 Introduction ............................................................................................................. 1 Chapter 2 Background ............................................................................................................. 3 2.1 Gas Condensate Hydrocarbon Fluid .......................................................................... 3 2.2 Modified Black-Oil Model ......................................................................................... 5 2.3 Zero-Dimensional Reservoir Model .......................................................................... 7 2.4 Field Performance Prediction..................................................................................... 8 2.5 Visual Basic for Applications (VBA) ........................................................................ 12 Chapter 3 Problem Statement .................................................................................................. 13 Chapter 4 Model Description ................................................................................................... 15 4.1 Phase Behavior Model (PBM) ................................................................................... 16 4.1.1 Compressibility Factor .................................................................................... 17 4.1.2 Vapor-Liquid Equilibrium............................................................................... 21 4.1.3 Fluid Property Prediction ................................................................................ 27 4.1.4 Phase Stability Analysis .................................................................................. 37 4.2 Standard PVT Properties ............................................................................................ 42 4.2.1 Definitions, Mathematic Relationships, and Characteristics ........................... 43 4.2.2 Obtaining Standard PVT Properties from Laboratory PVT Reports............... 51 4.2.3 Obtaining Standard PVT Properties from a Phase Behavior Model ............... 60 4.3 Zero-Dimensional Reservoir Model .......................................................................... 70 4.3.1 Generalized Material Balance Equation .......................................................... 71 4.3.2 Material Balance Equation for a Gas Condensate Fluid ................................. 75 4.3.3 Phase Saturation Calculations ......................................................................... 80 4.3.4 Volumetric OGIP/OOIP Calculations ............................................................. 82 4.4 Flow Rates and Flowing Pressures Calculation ......................................................... 84 4.4.1 Inflow Performance Relationship (IPR) .......................................................... 85 4.4.2 Tubing Performance Relationships ................................................................. 91 4.4.3 Nodal Analysis ................................................................................................ 96 4.5 Field Performance Prediction..................................................................................... 99 4.5.1 Performance during Plateau Period ................................................................. 100 4.5.2 Performance during Decline Period ................................................................ 103 4.5.3 Annual Production Calculation ....................................................................... 106 4.6 Economic Analysis and Field Optimization............................................................... 109 4.6.1 Simplified Economic Model ........................................................................... 110 4.6.2 Field Optimization........................................................................................... 116

vi Chapter 5 Model Performance ................................................................................................. 117 5.1 Simulation of Standard PVT Black Oil Properties ................................................. 117 5.1.1 Simulated Standard PVT Properties ................................................................ 117 5.1.2 Limitations of Pseudo Component Model....................................................... 121 5.1.3 Impact on Standard PVT Properties ................................................................ 127 5.2 Zero-Dimensional Material Balance Calculations for Gas Condensates ................... 129 5.2.1 Simulation Results from Gas Condensate Tank Model .................................. 129 5.2.2 Misuse of Gas Condensate Tank Model in Volatile Oil Reservoir ................. 132 5.3 Field Performance Prediction..................................................................................... 137 5.3.1 Field Performance Prediction Results ............................................................. 137 5.3.2 Decline Trend Analysis ................................................................................... 143 5.4 Economic Analysis and Optimization........................................................................ 146 5.4.1 Field Economic Analysis ................................................................................ 146 5.4.2 Field Optimization........................................................................................... 149 5.5 Application for Other Production Situations.............................................................. 152 5.5.1 Application for Dry Gas / Wet Gas ................................................................. 153 5.5.2 Application for Gas Condensate with Producible (Mobile) Reservoir Oil ..... 155 Chapter 6 Summary and Conclusions ...................................................................................... 158 Appendix A Input Data Summary .......................................................................................... 163 Appendix B User Guide .......................................................................................................... 170

vii

LIST OF FIGURES
Figure 2-1: Phase Diagram of Typical Gas Condensate Reservoir.......................................... 3 Figure 2-2: Distributions of Pseudo Components among Phases in Modified Black-Oil Model ............................................................................................................................... 5 Figure 2-3: Graphical Representation of Zero-Dimensional Reservoir Model........................ 7 Figure 2-4: Typical Field Performance of Gas Condensate Gas and Oil Flow Rates vs. Time ................................................................................................................................. 9 Figure 2-5: Typical Field Performance of Gas Condensate Reservoir Pressure, Bottomhole Flowing Pressure and Wellhead Pressure vs. Time ..................................... 10 Figure 2-6: Typical Field Performance of Gas Condensate Cumulative Gas and Oil Production vs. Time ......................................................................................................... 11 Figure 4-1: Graphical Representation of Standard PVT Properties ......................................... 44 Figure 4-2: Typical Characteristic of Gas Formation Volume Factor ( ) and Volatilized Oil-Gas Ratio ( ) for Gas Condensate .......................................................................... 48 Figure 4-3: Typical Characteristic of Oil Formation Volume Factor ( ) and Solution Gas-Oil Ratio ( ) for Gas Condensate ........................................................................... 48 Figure 4-4: Graphical Representation of CVD Data used in Walsh-Towler Algorithm .......... 53 Figure 4-5: Graphical Representation of Nodal Analysis ........................................................ 96 Figure 4-6: Graphical Representation of Field Optimization .................................................. 116 Figure 5-1: Simulated Gas Formation Volume Factor and Volatilized Oil-Gas Ratio of Gas Condensate ................................................................................................................ 118 Figure 5-2: Simulated Oil Formation Volume Factor and Solution Gas-Oil Ratio of Gas Condensate ....................................................................................................................... 119 Figure 5-3: Simulated Specific Gravity of Reservoir Gas ....................................................... 120 Figure 5-4: Volumes of Surface Gas Pseudo Component in Reservoir Gas Reservoir Oil, and Cumulative Gas Production....................................................................................... 121 Figure 5-5: Volumes of Stock-Tank Oil Pseudo Component in Reservoir Gas, Reservoir Oil, and Cumulative Oil Production................................................................................. 122 Figure 5-6: Densities of Surface Gas and Stock-Tank Oil Pseudo Components at First Stage Separator, Second Stage Separator and Stock Tank Condition .............................. 124

viii Figure 5-7: Volumes of Stock-Tank Oil Pseudo Component in Reservoir Gas, Reservoir Oil, and Cumulative Oil Production in term of Gas-Equivalent ...................................... 125 Figure 5-8: Total Volumes of Stock-Tank Oil Pseudo Component and Surface Gas Pseudo Component in term of Gas-Equivalent ................................................................ 126 Figure 5-9: Simulated Production Results of Gas Condensate using Simplified Gas Condensate Tank Model .................................................................................................. 130 Figure 5-10: Phase Envelope and Reservoir Depletion Paths at Two Different Reservoir Temperatures.................................................................................................................... 133 Figure 5-11: Simulated Gas Formation Volume Factor and Volatilized Oil-Gas Ratio of Volatile Oil using Gas Condensate PVT Model .............................................................. 134 Figure 5-12: Simulated Oil Formation Volume Factor and Solution Gas-Oil Ratio of Volatile Oil using Gas Condensate PVT Model .............................................................. 134 Figure 5-13: Simulated Production Results of Volatile Oil using Simplified Gas Condensate Tank Model .................................................................................................. 135 Figure 5-14: Mole Fraction Behavior of Vapor Phase Molar Fraction ( ) for Gas Condensates and Volatile Oils ......................................................................................... 136 Figure 5-15: Cumulative Gas and Oil Production vs. Time..................................................... 138 Figure 5-16: Total Gas and Oil Flow Rates vs. Time .............................................................. 139 Figure 5-17: Reservoir Pressure, Bottomhole Flowing Pressure, and Wellhead Pressure vs. Time............................................................................................................................ 140 Figure 5-18: Gas Saturation and Specific Gravity of Reservoir Gas vs. Time ........................ 142 Figure 5-19: Total Gas Flow Rate ( ) vs. Cumulative Gas Production during Decline Period .................................................................................................. 143 Figure 5-20: Decline Rate ( ) vs. Cumulative Gas Production during Decline Period ( ) ........................................................................................................... 145 Figure 5-21: Annual Expenditure, Annual Total Revenue, and Cumulative Discounted Net Cash Flow vs. Production Time ................................................................................ 147 Figure 5-22: Net Present Value vs. Interest Rate ..................................................................... 148 Figure 5-23: Field Optimization Results.................................................................................. 149

ix

LIST OF TABLES
Table 4-1: Volume-Translation Coefficients for Pure Components (Whitson and Brule, 2000) ................................................................................................................................ 31 Table A-1: Pressures and Temperatures for Standard PVT Properties Calculation Subroutine ........................................................................................................................ 163 Table A-2: Physical Properties of Pure Components............................................................... 163 Table A-3: Binary Interaction Coefficients of Pure Components ............................................ 164 Table A-4: Volume Translation Coefficient of Pure Components .......................................... 164 Table A-5: Reservoir Input Data.............................................................................................. 165 Table A-6: Relative Permeability Input Data .......................................................................... 165 Table A-7: Standard PVT Properties ....................................................................................... 166 Table A-7: Standard PVT Properties (Cont.) ........................................................................... 167 Table A-8: Tubing Input Data.................................................................................................. 168 Table A-9: Economic Input Data ............................................................................................. 168 Table A-9: Economic Input Data (Cont.)................................................................................. 169 Table A-10: Field Performance Prediction Input ..................................................................... 169 Table A-11: Field Performance Optimization Input ................................................................ 169

NOMENCLATURE
Normal Symbol Definition Reservoir drainage area Hyperbolic decline exponent Co-volume parameter of i-th component Formation volume factor Gas formation volume factor Oil formation volume factor Two-phase gas formation volume factor Two-phase oil formation volume factor Overall molar faction of i-th component Formation (rock) compressibility Deitz shape factor Non-Darcy coefficient / Tubing Diameter Decline rate Expansivity of formation (rock) Expansivity of reservoir gas Expansivity of reservoir oil Expansivity of reservoir water Efficiency factor of tubing Fannings friction factor Fugacity of i-th component in vapor phase Fugacity of i-th component in liquid phase Moodys friction factor Molar fraction of vapor phase Molar fraction of liquid phase at reservoir condition Molar fraction of liquid phase Molar fraction of liquid phase at first-stage separator Molar fraction of liquid phase at first-stage separator produced from reservoir gas Molar fraction of liquid phase at first-stage separator produced from reservoir oil Molar fraction of liquid phase at second-stage separator Molar fraction of liquid phase at second-stage separator produced from reservoir gas Molar fraction of liquid phase at second-stage separator produced from reservoir oil Molar fraction of liquid phase at stock-tank condition Molar fraction of liquid phase at stock-tank condition produced from reservoir gas Molar fraction of liquid phase at stock-tank condition produced from reservoir oil Fugacity of i-th component in original fluid

xi Fugacity of i-th component in liquid-like phase Fugacity of i-th component in vapor-like phase Amount of surface gas pseudo component / Gas in place Amount of gas-equivalent pseudo component Amount of surface gas pseudo component in reservoir gas phase Amount of surface gas pseudo component in reservoir oil phase Amount of cumulative gas injection Amount of cumulative gas production Cumulative gas production at year Cumulative gas production at abandonment condition Cumulative gas production at end of plateau Cumulative gas recovery Incremental of cumulative gas production Annual gas production at year Incremental of gas recovery Reservoir thickness Elevation of upstream node Elevation of downstream node Difference in elevation of downstream and upstream node Absolute permeability of reservoir Effective permeability of reservoir gas Relative permeability of reservoir gas Relative permeability of reservoir oil Volatility ratio of i-th component Tubing length Temperature dependency coefficient of i-th component Molecular weight of vapor phase Molecular weight of gas at reservoir condition Molecular weight of i-th component Molecular weight of oil at reservoir condition Molecular weight of oil at stock-tank condition Molecular weight of oil at stock-tank condition produced from reservoir gas Molecular weight of oil at stock-tank condition produced from reservoir oil Molecular weight of liquid phase Molecular weight of remaining fluid inside PVT cell Number of component in multi-component hydrocarbon Mole fraction of excess gas removed from PVT cell Mole fraction of remaining gas inside PVT cell Mole fraction of remaining gas inside PVT cell plus excess gas Mole fraction of remaining oil inside PVT cell Mole fraction of remaining fluid inside PVT cell Amount of stock-tank oil pseudo component / Oil in place Amount of stock-tank oil pseudo component in reservoir gas phase Amount of stock-tank oil pseudo component in reservoir oil phase

xii Amount of cumulative oil production Cumulative oil production at year Cumulative oil production at abandonment condition Cumulative oil recovery Reynolds number Incremental of cumulative oil production Annual oil production at year Incremental of oil recovery Original gas in place Original oil in place Pressure Upstream pressure Downstream pressure Average pressure between upstream and downstream Critical pressure of i-th component Drawdown pressure inside the reservoir Pseudocritical pressure Reservoir pressure Reservoir pressure at abandonment condition Reservoir pressure at end of plateau Reduced pressure of i-th component Pressure at standard condition Bottomhole flowing pressure Bottomhole flowing pressure at end of plateau Wellhead pressure Minimum allowable wellhead pressure Pressure drop from initial reservoir pressure Total gas flow rate of the field Total gas flow rate of the field at abandonment condition Total gas flow rate of the field during plateau period Total oil flow rate of the field Total oil flow rate of the field at abandonment condition Gas flow rate per well Gas flow rate per well during plateau period Annual average gas flow rate of the field Annual average oil flow rate of the field Reservoir radius Wellbore radius Universal gas constant Gas-oil equivalent factor Fugacity ratio of i-th component Solution gas-oil ratio Solution gas-oil ratio at bubble point pressure Volatilized oil-gas ratio Volatilized oil-gas ratio at dew point pressure Target recovery factor at end of plateau

xiii Fugacity ratio of i-th component in liquid-like phase Fugacity ratio of i-th component in vapor-like phase Total skin factor Volume-translate coefficient of i-th component Mechanical skin factor Average reservoir gas saturation Minimum gas saturation Sum of the mole number of liquid-like phase Average reservoir oil saturation Sum of the mole number of vapor-like phase Average reservoir water saturation Connate water saturation Specific gravity of gas Production time Production time at abandonment condition Production time at end of plateau Temperature Temperature of upstream node Temperature of downstream node Pipe section average temperature Critical temperature of fluid Critical temperature of i-th component Pseudocritical temperature of Reduced temperature of i-th component Temperature at standard condition Fluid velocity Retrograde liquid volume fraction Amount of excess gas at reservoir condition Amount of remaining gas phase at reservoir condition Amount of remaining gas phase plus excess gas at reservoir condition Amount of remaining oil phase at reservoir condition Pore volume of reservoir Original volume of PVT cell Molar volume of phase a calculated from EOS Critical molar volume of i-th component Molar volume of vapor phase Molar volume of vapor phase calculated from EOS Molar volume of liquid phase Molar volume of liquid phase calculated from EOS Pseudocritical molar volume Amount of water pseudo component in reservoir water Amount of water influx Amount of cumulative water injection Amount of cumulative water production Molar fraction of surface gas pseudo component in reservoir oil Molar fraction of i-th component in liquid phase Molar fraction of stock-tank oil pseudo component in reservoir oil

xiv Molar fraction of surface gas pseudo component in reservoir gas Molar fraction of i-th component in vapor phase Molar fraction of stock-tank oil pseudo component in reservoir gas Molar fraction of i-th component in liquid-like phase Molar fraction of i-th component in vapor-like phase Molar number of i-th component in liquid-like phase Molar number of i-th component in vapor-like phase Compressibility factor of fluid Two-phase compressibility factor Compressibility factor of phase a Average compressibility factor Greek Symbol Definition Coefficient to adjust relative permeability of reservoir gas Turbulence parameter Specific gravity of gas Binary interaction coefficient between i-th and j-th component Tubing roughness Fluid viscosity Viscosity of vapor phase / Viscosity of reservoir gas Viscosity of i-th component at low pressure Viscosity of liquid phase Viscosity of liquid phase at low pressure Viscosity of reservoir oil Fluid density Density of vapor phase Density of gas phase at reservoir condition Density of liquid phase Density of oil phase at reservoir condition Density of oil phase at stock-tank condition Density of oil phase at stock-tank condition produced from reservoir gas Density of oil phase at stock-tank condition produced from reservoir oil Pseudo reduced density of liquid phase Density of remaining fluid inside PVT cell Molar density of reservoir gas Molar density of surface gas pseudo component Molar density of reservoir oil Molar density of stock-tank oil pseudo component Annual production time Average reservoir porosity Fugacity coefficient of i-th component in vapor phase Fugacity coefficient of i-th component Fugacity coefficient of i-th component in liquid phase Pitzers acentric factor of i-th component

xv

ACKNOWLEDGEMENTS
First and foremost I would like to thanks my advisor, Dr. Luis Ayala, for his continuous guidance, support and friendship throughout my graduate study. Without his encouragement and invaluable advice, this research would not have been completed. Additional thanks are extended to Dr. Larry Grayson, and Dr. Li Li for their interest and time in serving as my thesis committee. I would like to express my sincere appreciation to Dr. Turgay Ertekin and Dr. Russel Johns, and Dr. Zuleima Karpyn for the fundamental knowledge they have taught. I am also very grateful for educational environment that the faculty and staff of the Department of Energy and Mineral Engineering have created. I highly thank my sponsor, PTT Exploration and Production Company, for every support they have given. Many friends and colleagues have been very supportive. I would like to express my gratitude to Pipat Likanapaisal, Nithiwat Siripatrachai and Kanin Bodipat who always are good friends throughout my student life at Pennsylvania State University. I also thank all of my colleagues for making me have meaningful time and experience. Finally, but most deeply, I am forever in dept to my family, my father Phiraphong Vardcharragosad, my mother Pikun Tanarungreung, my sister Pungjai Keandoungchun, and sisters family, for their support, encouragement, and most importantly their tolerance.

Chapter 1 Introduction
Natural gas is a natural occurring gas which consisting of methane primarily. It plays a significant role in global economic as one of the main sources of energy. In 2009, world natural gas reserves equaled 6.29 Trillion Standard Cubic Feet (TCF) while world production reached 106 BCF for the year (EIA, 2011). Conventional reservoirs consist of five different fluid types: dry gas, wet gas, retrograde gas, volatile oil, and black oils (McCain, 1990). They are distinguished from each other based on the present of fluid phases inside the reservoir and at surface production facilities. Field development and investment decisions in petroleum and natural gas require an integration of expertise from various areas including geology, reservoir, drilling, completion, process, and economic. Location and size of reservoirs, production rates and time, total recoverable volumes, number of wells and platforms, drilling and completion techniques, processing facilities scheme, cost and revenue, etc. are examples of information required for adequate field development decisions. Field performance indicators consist of information regarding flow rates, pressures, and production time is very important for field development. If field performance indicators are satisfactorily predicted, the hydrocarbon field could be developed using the best possible exploitation strategy while optimizing its economic performance. If not, the field might end up with too many wells, processing facilities that are too large, or wrong equipment sizing which can jeopardize profits or even lead to significant losses of investors capital. In modern age, computer simulation is used to simulate various types of mathematical models which can couple geological, fluid property, reservoir, production network, processing

2 facilities, and economic information. Field performance could be predicted by integrating these models together. However, the required type of mathematical model needs to be carefully selected to be able to perform the calculation most effectively. For reservoir characterization, for example, the modeler might utilize either a fully dimensional numerical model - which can aptly capture all reservoir heterogeneities and geometry by discretizing it into many small grids -, or a zero-dimensional model - which assumes average reservoir and fluid properties across the domain. For fluid behavior characterization, the modeler might select either a fully compositional model based on the use of an equation of state and detailed fluid composition data, or a black-oil model - which uses the pseudo-component concept and relies on PVT laboratory results. Selection of those models generally depends on availability of input data, time constraint, and required accuracy of simulation results. In this study, a zero dimensional model coupled with a black-oil PVT fluid description is implemented for the study of field development optimization strategies in retrograde natural gas reservoirs.

Chapter 2 Background

2.1 Gas Condensate Hydrocarbon Fluid A gas condensate, retrograde gas condensate, or retrograde gas, is one of the five reservoir fluid types (McCain, 1990). The typical phase envelope of gas condensate reservoirs is shown in Figure 2-1. Gas condensates contain more intermediate and heavy hydrocarbon components more than dry gases or wet gases. As shown in Figure 2-1, their reservoir temperature is located in between the fluids critical temperature and their cricondentherm. The reservoir depletion path of a gas condensate fluid typically crosses the dew point line and a liquid phase appears at reservoir pressures lower than that of the dew point. The presence of liquid phase in the reservoir significantly increases the system complexity, even if this liquid phase does not flow and is very unlikely to be produced under normal production conditions.

Critical Point

Reservoir Pressure

Surface Depletion Path

Reservoir Depletion Path

Reservoir Temperature

Figure 2-1: Phase Diagram of Typical Gas Condensate Reservoir

4 The general characteristics of gas condensate reservoir fluid can be summarized as follows (Walsh and Lake, 2003): Initial Fluid Molecular Weight: 23 40 lb/lbmol Stock-Tank Oil Color: Clear to Orange Stock Tank Oil Gravity: 45 60 API C7-plus Mole Fraction: 0.01 0.12 Typical Reservoir Temperature: 150 300 F Typical Reservoir Pressure: 1500 9000 psia Volatilized Oil-Gas Ratio: 50 300 STB/MMSCF Primary Recovery of Original Gas In Place: 70% 85% Primary Recovery of Original Oil In Place: 30% - 60%

5 2.2 Modified Black-Oil Model A black-oil fluid model is a fluid characterization formulation which represents multicomponent hydrocarbon mixture in terms of two hydrocarbon pseudo components, namely the surface gas and stock-tank oil pseudo components. In a traditional black-oil model, the solubility of the surface gas pseudo component in the reservoir oil fluid phase is taken into account while the solubility of stock-tank oil pseudo component in reservoir gas phase is neglected. The modified black-oil model which also called two-phase two-pseudo component model does not neglect the stock tank oil solubility in the gaseous reservoir phase, thus including both solubility variables into the formulation. Figure 2-2 shows the distribution of surface gas and stock-tank oil pseudo components among reservoir gas and reservoir oil phases.

Reservoir Gas Phase

Surface Gas

Stock-Tank Oil

Reservoir Oil Phase

Surface Gas

Stock-Tank Oil

Figure 2-2: Distributions of Pseudo Components among Phases in Modified Black-Oil Model

The assumptions behind the modified black-oil PVT model can be summarized as follows (Walsh and Lake, 2003 and Whitson and Brule, 2000): There are two pseudo components which are surface gas and stock-tank oil. There are two fluid phases which are reservoir gas (vapor) and reservoir oil (liquid) phases.

6 Surface gas pseudo component is reservoir fluid which remains in gas phase at standard condition. Stock-tank oil pseudo component is reservoir fluid which remains in oil phase at standard condition. The reservoir gas phase, which is reservoir fluid remains in vapor phase at reservoir condition, consists of surface gas and stock-tank oil pseudo components. The reservoir oil phase, which is reservoir fluid remains in liquid phase at reservoir condition, consists of surface gas and stock-tank oil pseudo components. Properties of surface gas and stock-tank oil pseudo components remain the same throughout the reservoir depletion.

7 2.3 Zero-Dimensional Reservoir Model The Material Balance Equation (MBE) is a specialized type of mass balance equation that combines mass balance equations of all pseudo components present in the reservoir into single equation. The MBE is also called zero-dimensional reservoir model or tank model because it assumes that a reservoir behaves like a homogeneous tank with average rock and fluid properties across the domain. Pressure, temperature, and compositional gradients are thus neglected. MBEs can be derived from integrating diffusivity equations over space and time.

Figure 2-3: Graphical Representation of Zero-Dimensional Reservoir Model (Source: http://www.joe.co.jp/english/menu2-5.html)

The following assumptions are implemented in traditional in zero-dimensional reservoir models: Reservoir is isothermal Reservoir is under thermodynamic equilibrium condition There are no chemical and biological reaction in reservoir Capillary pressures of reservoir fluids are negligible Gravitational gradients in reservoir are negligible Pressure gradients in reservoir are negligible

8 2.4 Field Performance Prediction A field performance prediction consists in the calculations of pressures, flow rates, cumulative productions, and expected production times based on available reservoir, production network, and production constraint data. Field life is divided into three periods which are buildup, plateau, and decline periods (Ayala, 2009a), as depicted in Figure 2-4. During build-up period, gas flow rate per well ( ) is kept constant while number of wells continuously

increases until total maximum number of wells needed for field development is reached. During the plateau period, both gas flow rate per well ( total gas flow rate ( ) and number of wells are fixed; therefore, ) times number of wells) remains ) is kept constant at the minimum

) (equal to gas flow rate per well (

constant. During decline period, wellhead pressure ( allowable wellhead pressure (

). Under such conditions, reservoir pressure ( ) becomes too ) continuously declines until

low to maintain the target plateau rate, thus gas flow rate ( abandonment condition is reached.

9
Build-up
Plateau Decline

qgsc - Total Gas Flow Rate

Dew Point qgsc qosc

Production Time

Figure 2-4: Typical Field Performance of Gas Condensate Gas and Oil Flow Rates vs. Time

Figure 2-4 through Figure 2-6 show the typical field performance predictions for the development of a gas condensate reservoir. As indicated earlier, gas flow rate ( ) increases

with increasing production time during the build-up period because more wells are put on production. Then, it is kept constant until the end of the plateau period. During final decline period, gas flow rate ( ) continuously decreases with production time because reservoir

pressure ( ) becomes not enough to sustain the plateau rate. Above dew point conditions, oil flow rate ( ) produced at the surface becomes directly proportional to gas flow rate ( ).

However, once reservoir conditions reach the dew point, condensate production at the surface becomes a function of both gas flow rate ( ) and the volatilized oil-gas ratio ( ) at reservoir

conditions. Figure 2-4 shows that even if gas flow rate ( value during the plateau period, oil flow rate ( volatilized oil-gas ratio ( ) below the dew point.

) is maintained at a constant target

) can actually decreases because of decreasing

qosc - Total Oil Flow Rate

10

Build-up

Plateau

Decline

pr

Pressure

pwf

pwh

Minimum Allowable Wellhead Pressure

Production Time

Figure 2-5: Typical Field Performance of Gas Condensate Reservoir Pressure, Bottomhole Flowing Pressure and Wellhead Pressure vs. Time Figure 2-5 demonstrates that, during field development calculations, reservoir pressure ( ) decreases as production time increases because more oil and gas are being removed from the reservoir. Wellhead pressure ( gas flow rate ( pressure ( ) is also continuously decreased in time in order to maintain the

) per well during the build-up and plateau periods. After that, once wellhead

) reaches the minimum allowable wellhead pressure at surface conditions, the ) cannot be maintained any longer and the decline period starts. ) changes along with changes in reservoir pressure ( ) and

plateau gas flow rate (

Bottomhole flowing pressure ( wellhead pressure ( production tubing.

) in order to provide the required pressure drop within the reservoir and

11
Plateau
Decline Gp
Gp - Cumulative Gas Production Np - Cumulative Oil Production

Build-up

Np

Production Time

Figure 2-6: Typical Field Performance of Gas Condensate Cumulative Gas and Oil Production vs. Time

Figure 2-6 shows cumulative gas (

) and cumulative oil (

) production are directly

related to their corresponding flow rates. Above dew point conditions, both of them increase at the same pace. Below the dew point, however, cumulative oil production ( much slower rate compared to that of cumulative gas production ( ) builds up at a

) because of the increased ). Recovery factor of

reservoir condensation driven by a decreasing volatilized oil-gas ratio (

gas at abandonment condition would therefore become much higher than the recovery factor of oil or condensate because large amounts of condensate are left behind as immobile phase inside the reservoir.

12 2.5 Visual Basic for Applications (VBA) Visual Basic for Applications (VBA) is a programming language from Microsoft. The program is built into most MS-Office applications i.e. MS-Word, MS-Excel, MS-Access. Users can use VBA to create calculation subroutine and control user interface features such as menus, toolbars, worksheets, charts, etc (Walkenbach, 2007). VBA can only run within the host application, and not as a standalone application. VBA is functionally rich, and flexible. Because it is built into MS-Office applications, VBA subroutines will be able to execute so long as those applications are available on computer machines. MS-Excel with built in VBA is a very favorable platform for developing simulations. The main reasons are that most of engineers are familiar with MS-Excel application and MS-Excel itself is user-friendly software with many useful builtin features. Excels worksheets could be used as table to store input data. Simulation results could be easily stored in the tabular form and displayed on various types of built-in chart.

Chapter 3 Problem Statement

In the development of a petroleum and natural gas reservoir, projected field performance is the most important information required by all relevant people involved in the process of design, risk assessment, and decision-making process. Field performance analysis can require a significant amount of expertise and time, especially for more complex reservoir fluid system such as gas condensates. The use of the appropriate modeling approach is the key to analyze the field performance most efficiently. Full scale, fully dimensional, commercial simulators might not be able to yield the best or optimized solutions even if they are based on of highly sophisticated mathematical models. This is because more sophisticated and detailed models are subject to the availability of very detailed set of reservoir and fluid data, which is typically scarce, and time constraints and demands. In the analysis of gas condensates, for example, commercial simulators often rely on compositional modeling for fluid property calculations. Compositional models can accurately simulate reservoir fluid properties; however, it is sophisticated model and can take a relatively long time to run. For reservoir fluid flow characterization, commercial simulators generally rely on fully dimensional numerical models which could perfectly capture reservoir heterogeneities; yet, they can take a significant amount of time to construct, conceptualize, and execute. Again, the limited availability and uncertainty of required input data such as fluid composition, reservoir heterogeneities, capillary pressures, and relative permeabilities could significantly impact the reliability of the results obtained from these sophisticated models. This study aims at developing a model which can efficiently and inexpensively perform field performance analysis and optimization tasks for gas condensate reservoirs. The proposed model utilizes a zero-dimensional reservoir formulation coupled with a pseudo component or

14 black oil PVT formulation for fluid properties calculation. These models are relatively simple, but fast, reliable, and robust. Results show that the proposed model is able to predict field performance while faithfully capturing the most salient characteristics of gas condensate reservoirs. In addition, optimization on targeted variables can be accomplished without difficulty.

Chapter 4 Model Description

The proposed field performance predictor has been developed using Microsoft Excel with built-in Visual Basic for Applications (VBA) subroutines. Workflow begins with the simulation of standard black-oil PVT properties, which could be done either based on standard PVT laboratory results (such as the Constant Volume Expansion or CVD) or via a phase behavior model based on cubic equations of state. Next, field performance data is calculated by integrating a zerodimensional reservoir model, standard PVT fluid properties, well performance models for flow rates and pressure calculation, and production constraints. Based on this, an economic analysis can be performed based on simplified economic model. Finally, optimization on target variables can be carried out by evaluating field performance and net present value repeatedly for different and plausible production scenarios. The proposed simulation tool has been designed to simulate a single gas condensate reservoir based on the continuous drilling of identical wells placed at different locations of the reservoir area. Wellhead pressure is used to control gas flow rate. Reservoir pressure is used as the abandonment criteria. Optimization variables are target recovery factor at end of plateau and total number of wells. Those variables could be re-selected by simple modification in the VBA code. However, optimization variables can be made independent for a real field operation.

16 4.1 Phase Behavior Model (PBM) A phase behavior calculation (or a flash calculation) is used to predict the phase behavior of a reservoir fluid at an equilibrium condition. A standard phase behavior model consists of four main calculation modules; namely, compressibility factor calculations, vapor-liquid equilibrium calculations, fluid properties predictions, and phase stability analysis, which must be fully integrated to perform the flash calculation. The calculation starts with the determination of number of co-existent phases or phase stability analysis. If fluid is found in a single phase (stable) condition, fluid properties are calculated based on the available information on overall fluid composition. If fluid is found in a two-phase (unstable) condition, composition and molar fraction of each phase are determined using vapor-liquid equilibrium calculations. Then properties of each co-existent phase are calculated based on fluid composition of that phase (Ayala, 2009b). Input data consists of pressure, temperature, overall composition, physical properties, binary interaction coefficients, and volume translation coefficient of each pure component. PengRobinson Equation-of-State (PR EOS) is used to calculate Pressure-Volume-Temperature (PVT) relationship of the reservoir fluid (Peng and Robinson, 1976). Vapor-liquid equilibrium is assumed and an overall species material balance for a two-phase system is enforced. The output from a PBM subroutine consists of number of phases, molar fraction, composition, molecular weight, compressibility factor, density, adjusted density and viscosity of each fluid phase.

17 4.1.1 Compressibility Factor Compressibility factor or Z-factor is volumetric multiplier utilized to convert ideal gas volumes, as predicted by the ideal gas equation of state, to real gas volumes, as realized experimentally. Compressibility factor is a fundamental and very important variable because other fluid properties can be calculated based on compressibility factor data. Z-factor calculation subroutine is developed based on generalized formulation (Coats, 1985). Although PengRobinson EOS is utilized throughout this study, other EOSs could also be applied by implementing simple modifications outlined below When the fluid is in a single phase condition, overall composition will be inputted into generalized formula for the calculation of the single-phase compressibility factor. However, when the fluid is in two-phase condition, composition of each phase must be first calculated based on vapor-liquid equilibrium calculations in order to estimate the corresponding compressibility factors of each phase.

Generalized Formulation Compressibility factor depends on the chosen PVT relationship or equation of state (EOS). The generalized formula for cubic EOS proposed by Coats is utilized (Coats, 1985). This form can be applied for Redlich-Kwong (RK), Soave-Redlich-Kwong (SRK), and Peng-Robinson (PR) EOSs (Redlich, O. and Kwong, J.N.S. 1949, Soave, G. 1972, and Peng and Robinson, 1976).

18

Equation 4-1

where:

= number of components in the multi-component hydrocarbon = molar fraction of the i-th component = binary interaction coefficient between the i-th and j-th components = reduced pressure of the i-th component = = reduced temperature of the i-th component = = critical pressure of the i-th component {psia} = critical temperature of the i-th component {R} = pressure {psia} = temperature {R}

19

, which accounts for the temperature dependency built into the molecular attraction parameter, is calculated from Equation 4-2 for PR EOS and from Equation 4-3 for SRK EOS.

Equation 4-2

Equation 4-3

where: = Pitzers acentric factor of the i-th component

Pressure, temperature, molar fraction, and properties of pure components are input into the generalized EOS formula shown above which yields a cubic polynomial in Z. Analytical, semi-analytical, or numerical approach can be used to solve this cubic equation. In this work, the analytical approach is applied.

20 Z-Factor Selection Because of the nature of cubic equation, more than one root could be found for any given pressure, temperature, and fluid composition. As described by Danesh (p. 176), the following criteria are used for Z-factor selection (Danesh, 1998). If there is only one real root, Z-factor is equal to that root. If there is more than one real root, the following criteria must be applied. The intermediate root will always be rejected. If the minimum Z-factor is less than B, maximum Z-factor will be selected. If the minimum Z-factor is higher than B, The root that provides the lower Gibbs energy will be selected. Z-factor which is less than B must be rejected because when Z-factor is less than B, molar volume becomes smaller than the co-volume. For this reason, such Z-factor would have no physical meaning. For the last condition in the list above, Equation 4-4 is used to find the root with lower Gibbs energy. Following Danesh (1998), if the right hand side of this equation is positive, minimum Z-factor will be selected. Otherwise, the maximum Z-factor will be selected.

Equation 4-4

21 4.1.2 Vapor-Liquid Equilibrium Two main components are considered in order to predict properties of multi-component hydrocarbon in Vapor-Liquid Equilibrium (VLE) condition: material balance considerations and thermodynamic considerations. Iterative procedure is applied until the solution that satisfies both criteria can be determined.

Material Balance Considerations Rachford and Rice objective function, which is derived from enforcing an overall species mass balance in a two-phase multi-component system, is utilized to calculate molar fraction of each phase (Rachford and Rice, 1952):

Equation 4-5

where: = molar faction of i-th component = volatility ratio of i-th component = = molar fraction of i-th component in vapor phase = molar fraction of i-th component in liquid phase = molar fraction of vapor phase

22 After solving for from the objective function, molar fraction of liquid phase is

calculated from Equation 4-6, composition of vapor phase is calculated from Equation 4-7, and composition of liquid phase is calculated from Equation 4-8.

Equation 4-6

Equation 4-7

Equation 4-8

Thermodynamic Considerations According to the second law of thermodynamics, any system in equilibrium, such as a VLE condition, must have the maximum possible entropic state under the prevailing conditions. For such condition to be established, thermodynamics shows that net transfer of heat, momentum, and mass between both phases must be zero. Thus, temperature, pressure, and every species chemical potential in both phases must be equal to each other.

23 Chemical potential cannot be measured directly. However, equality of chemical potential can be represented by equality of fugacity between both phases. Fugacity is the pressure multiplier to correct non-ideality and to make ideal gas equation work for real gas during Gibbs energy calculations. In a VLE condition, fugacity of liquid phase must be equal to fugacity of vapor phase. Equation 4-9 is used to calculate fugacity for vapor phase while Equation 4-10 is used for liquid phase.

Equation 4-9

Equation 4-10

where: = fugacity of i-th component in vapor phase = fugacity of i-th component in liquid phase = fugacity coefficient of i-th component in vapor phase = fugacity coefficient of i-th component in liquid phase = molar fraction of i-th component in vapor phase = molar fraction of i-th component in liquid phase = pressure {psia}

24 For the generalized formula of cubic EOSs discussed above, fugacity coefficients can be calculated using Equation 4-11 (Coats, 1985) below. Definitions of parameters are the same as definitions used in Equation 4-1. It should be noted that coefficient of a liquid phase and phase. is equal to is equal to for calculating fugacity

for calculating fugacity coefficient of a vapor

Equation 4-11

Volatility ratio ( ) is equal to ratio between the gas composition composition to

and the liquid is equal

during an equilibrium condition. For a system with a VLE condition,

. By substituting Equation 4-9 and Equation 4-10 into definition of volatility ratio, volatility

ratio can be expressed in terms of fugacity coefficients as follows.

Equation 4-12

25 The Successive Substitution Method From material balance consideration, molar fraction of vapor phase and composition of each phase are functions of volatility ratios and overall composition. Volatility ratios themselves are also function of composition of each phase. Thus, an iterative procedure is needed in order to perform VLE prediction and honor the fugacity equality constraint. The following procedure is used to perform two-phase flash calculation (Whitson and Brule, 2000, p.52-55). First, initial guesses of volatility ratios are calculated using Equation 4-13 as proposed by Wilson (Wilson, 1968). Rachford and Rice objective function (Equation 4-5) is then solved using a standard Newton-Raphson iterative method. Then, the compositions of each phase are calculated using Equation 4-7 and Equation 4-8.

Equation 4-13

Next, the fugacity values of each component in both liquid and vapor phases are calculated using Equation 4-9 through Equation 4-11. Successive Substitution Method (SSM) is utilized to update volatility ratios (Equation 4-14) for a next iteration as shown below

Equation 4-14

26 where: = volatility ratio of i-th component at iteration level n = fugacity of i-th component in liquid phase at iteration level n = fugacity of i-th component in vapor phase at iteration level n

Once volatility ratios are updated, convergence criteria presented in Equation 4-15 must be checked. If the criteria are not satisfied, the procedure is repeated by solving Rachford and Rice objective function and recalculating phase compositions and resulting fugacities until convergence is attained.

Equation 4-15

The SSM algorithm is expected to have slow convergence rate near the critical point. To avoid this problem, accelerated SSM algorithm has been proposed. The algorithm proposed by Michelsen (Michelsen, 1982b) or the algorithm proposed by Merah et al (Merah et al, 1983) are examples of well-known ASSM algorithms.

27 4.1.3 Fluid Property Prediction

Molecular Weight Molecular weight of vapor and liquid phases are weighted average of molecular weight of all pure components, as shown below

Equation 4-16

Equation 4-17

where: = molecular weight of vapor phase {lb/lbmol} = molecular weight of liquid phase {lb/lbmol} = molecular weight of i-th component {lb/lbmol} = mole fraction of i-th component in vapor phase = mole fraction of i-th component in liquid phase = number of components in the multi-component hydrocarbon

28 Density Density of each phase is calculated from Equation 4-18 and Equation 4-19.

Equation 4-18

Equation 4-19

where: = density of vapor phase {lbm/ft3} = density of liquid phase {lbm/ft3} = molecular weight of vapor phase {lbm/lbmol} = molecular weight of liquid phase {lbm/lbmol} = molar volume of vapor phase {ft3/lbmol} = molar volume of liquid phase {ft3/lbmol }

29 Molar volume of each phase is calculated from real gas law (Equation 4-20), then, adjusted by using volume-translation technique.

Equation 4-20

where = calculated molar volume of phase a from EOS {ft3/lbmol} = compressibility factor of phase a = universal gas constant {10.732 psi-ft3/R-lbmol} = temperature {R} = pressure {psia}

As discussed by Whitson and Brule (p.51) and Danesh (p.141-143), calculated molar volume from real gas law can be adjusted by implementing volume-translation or volume-shift technique (Whitson and Brule, 2000 and Danesh, 1998). This technique improves volumetric calculation of liquid phase, which is the main problem of two-constant EOSs, without altering VLE prediction results. The volume translation technique, originally introduced by Martin and further developed by Penelous et al and Jhaveri and Youngren, can be summarized as follows (Martin, 1979, Penelus et al, 1982, and Jhaveri and Youngren, 1988):

30 Calculated molar volumes from the selected EOS are corrected by using Equation 4-21 and Equation 4-22. Component-dependent volume-shift parameters ( ) are calculated from Equation 4-23 and volume-translate coefficients are in Table 4-1.

Equation 4-21

Equation 4-22

where: = corrected molar volume of liquid phase = corrected molar volume of vapor phase = calculated molar volume of liquid phase from EOS = calculated molar volume of vapor phase from EOS = component-dependent volume-shift parameter = molar fraction of i-th component in liquid phase = molar fraction of i-th component in vapor phase = number of components in the multi-component hydrocarbon

31

Equation 4-23

where: = component-dependent volume-shift parameter = co-volume parameter of i-th component = volume-translate coefficient of i-th component

Table 4-1: Volume-Translation Coefficients for Pure Components (Whitson and Brule, 2000) Component N2 CO2 H2S C1 C2 C3 i-C4 n-C4 i-C5 n-C5 n-C6 n-C7 n-C8 n-C9 n-C10 PR EOS -0.1927 -0.0817 -0.1288 -0.1595 -0.1134 -0.0863 -0.0844 -0.0675 -0.0608 -0.0390 -0.0080 0.0033 0.0314 0.0408 0.0655 SRK EOS -0.0079 0.0833 0.0466 0.0234 0.0605 0.0825 0.0830 0.0975 0.1022 0.1209 0.1467 0.1554 0.1794 0.1868 0.2080

32 Viscosity Viscosity of vapor phase is calculated from the correlation proposed by Lee et al in 1966 (Equation 4-24 through Equation 4-27).

Equation 4-24

Equation 4-25

Equation 4-26

Equation 4-27

where: = viscosity of vapor phase {cp} = density of vapor phase {lbm/ft3} = molecular weight of vapor phase {lbm/lbmol} = temperature {R}

33 The viscosity of a liquid phase is calculated from the correlation proposed by Lohrenz et al in 1964. The correlation is originally proposed by Jossi et al in 1962 for calculating viscosity of pure component. Lohrenz et al extend the use of original correlation to hydrocarbon mixtures. It should be noted that the formula in Lohrenz et als paper contains a typing error on coefficient 0.040758 for the cubic density term.

Equation 4-28

where: = viscosity of liquid phase {cp} = viscosity of liquid phase at low pressure {cp} = viscosity parameter of liquid phase (mixture) {cp-1} = pseudo reduced density of liquid phase

Viscosity of liquid phase at low pressure is calculated from Equation 4-29, Equation 4-30, and Equation 4-31. A conversion factor of 5.4402 is used to convert original units (K and atm) to oil field units (R and psia).

Equation 4-29

34

Equation 4-30

Equation 4-31 where: = viscosity of liquid phase at low pressure {cp} = molar fraction of i-th component in liquid phase = viscosity of i-th component at low pressure {cp} = viscosity parameter of i-th component {cp-1} = reduce temperature of i-th component ( = temperature {R} = critical temperature of i-th component {R} = critical pressure of i-th component {psia} = molecular weight of i-th component {lbm/lbmol} = number of components )

Viscosity parameter of liquid phase is calculated from Equation 4-32 to Equation 4-35.

Equation 4-32

35

Equation 4-33

Equation 4-34

Equation 4-35

where: = viscosity parameter of liquid phase (mixture) {cp-1} = pseudocritical temperature of liquid phase {R} = critical temperature of i-th component {R} = pseudocritical pressure of liquid phase {psia} = critical pressure of i-th component {psia} = molecular weight of liquid phase {lbm/lbmol} = molecular weight of i-th component {lbm/lbmol} = molar fraction of i-th component in liquid phase = number of components

36 Pseudo reduced density of the liquid phase is calculated from Equation 4-36 and Equation 4-37 shown below.

Equation 4-36

Equation 4-37

where: = pseudo reduced density of liquid phase = density of liquid phase {lbm/ft3} = molecular weight of liquid phase {lbm/lbmol} = pseudocritical molar volume of liquid phase {ft3/lbmol} = critical molar volume of i-th component {ft3/lbmol} = molar fraction of i-th component in liquid phase = number of components

37 4.1.4 Phase Stability Analysis The ability to predict whether the system is in single phase (stable) or multiple phases (unstable) is crucial in a VLE or flash calculation. Whitson and Brule (p.55-61) discuss the graphical representation as well as numerical algorithm of phase stability analysis based on the studies by Baker et al and Michelsen (Whitson and Brule, 2000; Baker et al, 1982; Michelsen, 1982a). These studies explain how the Gibbs tangent-plane criteria can effectively be used to analyze the phase stability problem. The phase stability analysis subroutine utilized by this study has been developed based on these calculation procedures, which can be summarized in the 11 steps outlined below.

Step 1: Calculate the mixture fugacity from overall composition using Equation 4-9 / Equation 4-10 and Equation 4-11. The Z-factor yielding the lowest Gibbs energy should be utilized for the calculation of mixture fugacity.

Step 2: Use Wilsons equation to estimate initial

values (Equation 4-13).

Step 3: Calculate second-phase mole number, estimated K values.

, using the mixture composition and the

Equation 4-38

Equation 4-39

38 where: = mole number of i-th component in vapor-like phase = mole number of i-th component in liquid-like phase = mole fraction of i-th component = volatility ratio of i-th component

Step 4: Sum the mole numbers of vapor-like phase ( ) and liquid-like phase ( ).

Equation 4-40

Equation 4-41

Step 5: Normalize the mole numbers to get the mole fraction of i-th component in vaporlike phase, and liquid-like phases,

Equation 4-42

Equation 4-43

39 Step 6: Calculate the fugacity of vapor-like and liquid-like phases based on the calculated mole fraction from Step 5. Equation 4-9 , Equation 4-10 and Equation 4-11 are utilized.

Step 7: Calculate the fugacity ratio corrections for successive substitution update of the values.

Equation 4-44

Equation 4-45

where: = fugacity ratio calculation of i-th component in vapor-like phase = fugacity ratio calculation of i-th component in liquid-like phase = fugacity of i-th component in original fluid = fugacity of i-th component in vapor-like phase = fugacity of i-th component in liquid-like phase = Sum the mole numbers of vapor-like phase = Sum the mole numbers of liquid-like phase

40 Step 8: Check whether convergence criteria is achieved

Equation 4-46

Step 9: If convergence is not obtained, update

values

Equation 4-47

Step10: Apply criterion to check whether a trivial solution has been obtained

Equation 4-48

Step 11: If a trivial solution is not indicated, go to Step 3 for the next iteration.

41 The following criteria are used to interpret the results from this numerical algorithm: If the tests on both vapor-like and liquid-like phases satisfy trivial solution criterion, the system of interest is stable (single phase) If sum of the mole numbers on both vapor-like and liquid-like phases is less than or equal to 1.0, the system of interest is stable (single phase). If one of the pseudo phases satisfies trivial solution criterion and sum of the mole numbers of the other pseudo phase is less than or equal to 1.0, the system is stable (single phase). Otherwise, the system is unstable; both vapor and liquid phases coexist.

42 4.2 Standard PVT Properties The standard PVT properties used to describe a two-phase, two-pseudo component fluid model (black oil model) relies on the definition and calculation of four basic properties, namely: gas formation volume factor ( ratio ( ), and solution gas-oil ratio ( ), oil formation volume factor ( ), volatilized oil-gas

). These PVT properties are required inputs for a zero-

dimensional reservoir model. In this study, these required PVT properties can be obtained from either a laboratory fluid analysis, typically a Constant Volume Depletion (CVD) test, or from a phase behavior model (PBM) calculation. If the PVT/CVD laboratory report is available, the resulting PVT properties are calculated using Walsh-Towler algorithm (Walsh and Lake, 2003). A template has been prepared using MS-Excel worksheet for this purpose. In the absence of a PVT lab report, a PBM calculation is implemented which combines Walsh-Tolwer method with the work of Thararoop in 2007 (Thararoop, 2007). This PBM subroutine does not only extend the flexibility of the main simulator significantly, but also provide very useful information about fluid properties which could help in thoroughly analyzing the depletion characteristics of the given gas condensate fluid. The specific gravity of reservoir gas is required for flow rate and flowing pressure calculations, as it will be discussed below. The specific gravity of a reservoir gas can be obtained from either the laboratory fluid analysis or from molecular weight calculations derived from PBM. If the lab analysis is available, compositions of the produced wellstreams reported in the experimental depletion study based on the Constant Volume Depletion (CVD) test are used to calculate molecular weight of reservoir gas. If the lab report is unavailable, the molecular weight of the reservoir gas is obtained directly from flash/PBM calculation results. Specific gravity of reservoir gas is equal to molecular weight of reservoir gas divided by molecular weight of air.

43 4.2.1 Definitions, Mathematic Relationships, and Characteristics A clear understanding of the definitions of standard PVT black oil properties that are used to characterize two-phase, two-pseudo component fluid models is crucial for their meaningful calculation and prediction. These definitions, mathematic relationships, and their most significant features have been summarized below (Walsh and Lake, 2003; Whitson and Brule, 2000).

Definitions Figure 4-1shows the graphical representation of the definitions of the standard PVT properties used in the formulation of two-phase, two-pseudo component fluid model (or modified black-oil model). In this figure, the gas phase at reservoir condition ( ) results from the mixing of certain amounts of surface gas ( ) and stock-tank oil ( ) pseudo components. The oil )

phase at reservoir condition ( ) results from the mixing of certain amounts of surface gas ( and stock-tank oil (

) pseudo components. The produced gas phase at surface condition ( )

(not shown in the figure) would consists of the combination of surface gas pseudo component produced from gas phase at reservoir condition ( from oil phase at reservoir condition ( ) and surface gas pseudo component liberated

). By the same token, the produced oil phase at surface

condition ( ) (not shown in the figure) consists of stock-tank oil pseudo component produced from oil phase at reservoir condition ( gas phase at reservoir condition ( ). ) and stock-tank oil pseudo component condensed from

44
Reservoir Condition PR, TR Vg

Surface Condition
Psc, Tsc

Vg Bg = -----------

Nfg Rv = -----------

Vo

Gfg

Gfg

Gfg

Nfg Reservoir Gas

Reservoir Oil Surface Gas Stock-Tank Oil Nfo Gfo

Gfo

Vo Bo = ----------Nfo

Rs = ----------Nfo

Figure 4-1: Graphical Representation of Standard PVT Properties

Based on the pseudo component definitions described above, the definitions of the associated black oil properties can be straightforwardly presented. For example, the f ormation volume factor for the gas ( ) would be basically defined as ratio between volume of gas phase at

reservoir condition ( ) and volume of surface gas pseudo component produced from that reservoir gas, evaluated at surface conditions ( ). Formation volume factor of oil ( ) is

defined as ratio between volume of oil phase at reservoir condition ( ) and volume of stock-tank oil pseudo component produced from that reservoir oil, evaluated at surface condition ( Volatilized oil-gas ratio ( volume of surface gas ( ) is defined as ratio between volume of stock-tank oil ( ).

) and

) pseudo components produced from the same reservoir gas ( ), ) is defined as ratio between volume of

evaluated at surface condition. Solution gas-oil ratio ( surface gas ( ) and volume of stock-tank oil (

) pseudo components produced from the same

reservoir oil ( ), evaluated at surface condition. Mathematically, Equation 4-49 through

45 Equation 4-52 summarize, in oil field units, the standard PVT properties based on these definitions and the nomenclature presented in Figure 4-1.

Equation 4-49

Equation 4-50

Equation 4-51

Equation 4-52

It follows from the preceding discussion that reservoir fluid compositions can be calculated for the envisioned pseudo binary mixture. For example, the molar fraction of surface gas pseudo component in the gas phase at reservoir conditions, defined as , should be directly

related to the value of Rv. Molar fraction of stock-tank oil pseudo component in gas phase at reservoir condition would be defined as . Clearly, + = 1. For the oil reservoir phase, the

molar fraction of surface gas pseudo component in the oil phase at reservoir conditions would be , and should be directly related to the value of Rs The molar fraction of stock-tank oil pseudo

46 component in oil phase at reservoir condition is thus defined as formulas are summarized in Equation 4-53 through Equation 4-56. . Clearly, + = 1. Their

Equation 4-53

Equation 4-54

Equation 4-55

Equation 4-56

Mathematic Relationships If only one mole of reservoir fluid is considered, volumes at reservoir condition, , can be represented by molar density at reservoir condition, Similarly, volumes at surface condition, , , , and and and

, respectively.

, can be represented by molar ,

fraction of pseudo component in reservoir fluid and molar density at surface condition, , , and

, respectively. If we substitute these definitions into equations for

47 standard PVT properties and substitute densities of gases with real gas equation, the following expressions can be derived.

Equation 4-57

Equation 4-58

Equation 4-59

Equation 4-60

Depletion Characteristics Figure 4-2 and Figure 4-3 show the typical depletion behavior of the standard PVT properties for the case of a gas condensate reservoir fluid. Similar behavior can be found in the work by Walsh and Lake (Walsh and Lake, 2003, p.493) for the case of field-data derived properties.

48

Rv
Bg - Gas Formation Volume Factor

Bg

Dew Point Pressure Reservoir Pressure

Figure 4-2: Typical Characteristic of Gas Formation Volume Factor ( and Volatilized Oil-Gas Ratio ( ) for Gas Condensate

Bo - Oil Formation Volume Factor

Bo

Rs

Dew Point Pressure

Reservoir Pressure

Figure 4-3: Typical Characteristic of Oil Formation Volume Factor ( and Solution Gas-Oil Ratio ( ) for Gas Condensate

Rs - Solution Gas-Oil Ratio

Rv - Volatilized Oil-Gas Ratio

49 As shown in Figure 4-2, gas formation volume factors ( ) are expected to increase with

decreasing reservoir pressure ( ) because the denominator, , in Equation 4-57 approaches zero. Volatilized oil-gas ratio ( ) will remain constant because all parameters in Equation 4-59 remain , , and result from the constant composition of gas phase in

the same. Constant values of

the reservoir. Once dew point conditions are reached, Figure 4-2 also shows that the volatilized oil-gas ratio ( decreasing ) is expected to decrease with decreasing reservoir pressure, mainly because of and increasing values in Equation 4-59. Driven by the condensate drop out that

develops in the reservoir below dew point conditions, the reservoir gas will start to contain less heavy hydrocarbon molecules that can be produced as condensate at surface condition. As a result, the fraction of stock-tank oil ( ) in the reservoir gas decreases while fraction of surface gas ( ) increases ( + = 1). As pressure depletion progresses, and if it gets low enough, the ) trend would be reversed.

volatilized oil-gas ratio (

Figure 4-3 illustrates that at reservoir pressure above the dew point there is no liquid phase at reservoir condition and therefore no calculations of and can be directly performed ) is

from their definitions. Once dew point conditions are crossed, oil formation volume factor ( expected to decrease with decreasing reservoir pressure mainly because of increasing and

values in Equation 4-58. As pressure decreases, more surface gas pseudo component will be liberated from the oil phase. As a result, the molar fraction of stock-tank oil pseudo component in oil phase ( ) becomes higher and the density of oil phase at reservoir condition ( increases. Similarly, the solution gas-oil ration ( ) also

) will be expected to decrease with decreasing

reservoir pressure because of the increased molar fraction of stock-tank oil pseudo component in oil phase ( ) and decreasing molar fraction of surface gas pseudo component in oil phase ( Equation 4-60. Even though oil formation volume factors ( ) in )

) and solution gas-oil ratios (

cannot be calculated directly because of the lack of an actual liquid phase at reservoir from their

50 definitions, Walsh and Lake suggest employing the following relationships for oil formation volume factor ( ) and solution gas-oil ratio ( ) as place-holder values above the dew point:

Equation 4-61

Equation 4-62

51 4.2.2 Obtaining Standard PVT Properties from Laboratory PVT Reports In a laboratory PVT test, a representative sample of the reservoir fluid is subjected to a series of depletion steps that try to closely mimic or reproduce the expected pressure depletion path followed by the fluid during reservoir production. Temperature of the test is maintained constant and equal to prevailing reservoir temperature. Resulting volumes of each phase (liquid and vapor) are recorded along with the pressure at which the record is made. Fluid composition and physical properties of the produced fluids are also analyzed. The typical standardized PVT tests conducted for gas condensate fluids are the Constant Composition Expansion (CCE) and Constant Volume Depletion (CVD) tests. Details of these PVT tests can be found in many petroleum engineering textbooks (McCain, 1990; Denesh, 1998; Whitson and Brule, 2000, Walsh and Lake, 2003); thus, they will be discussed very briefly in this manuscript. In a CCE test, the reservoir fluid sample is placed inside a PVT cell and is pressurized to a pressure equal to initial reservoir pressure, while maintaining a constant temperature inside the PVT cell equal to reservoir temperature. Pressure inside the cell is then decreased to a next lower pressure level by isothermal expansion. The new volume of each phase is recorded. This process continues until abandonment pressure conditions are reached. In the CCE testing process, no fluid is taken out the cell and therefore the overall composition of reservoir fluid inside the PVT cell remains constant while the volumes and densities of each the co-existing phases below dew point conditions do change with cell pressure. In a CVD test, a reservoir fluid sample will be placed inside the PVT cell and pressurized to the dew point pressure, while the temperature of the PVT cell is kept constant at reservoir temperature. Then, pressure of the cell will be lowered to the next pressure level by isothermal expansion. After that, a portion of gas phase inside the cell is produced (i.e., removed out of the cell) so that the cells volume is restored back to the original cell volume at dew point conditions.

52 The volume that the liquid phase occupies inside the PVT cell is recorded and the excess (produced) gas analyzed. Depletion study which provides the resulting cumulative production data at every pressure level is recorded and is used during the calculation of the standard PVT properties from laboratory PVT fluid test report. In this study, a calculation template is prepared in MS-Excel worksheet. The WalshTowler algorithm is implemented to convert the results from the CVD experiments into the standard table of PVT properties for a gas condensate fluid. Walsh-Towler algorithm is summarized below.

Walsh-Towler Algorithm Walsh-Towler algorithm is one of the methods used to calculate standard PVT properties for gas condensate based on CVD testing results (Walsh and Towler, 1995; Walsh and Lake, 2003). This algorithm is relatively simple because it based on enforcing material balance constraints around the PVT cell at every pressure level during the PVT lab test. The algorithm was originally proposed by Walsh and Towler in 1995 and was later modified by Walsh and Lake in 2003. By directly using data from a CVD report, this algorithm is implicitly assuming that actual field separator conditions of the surface production system is the same as those surface condition used during the CVD PVT test. It also assumes that only the gas phase at reservoir condition can be recovered and that any condensate drops out inside the reservoir will remain immobile during reservoir life. One of the constraints of using this method is the availability of cumulative production data at surface conditions because such data is not always performed or reported for every CVD experiment. If such cumulative production data at surface conditions is not available in the CVD report, it is customarily recommended to implement surface flash calculations using Standings

53 K-values to reproduce them (Walsh and Lake, 2003). The algorithm also requires a high accuracy and reliability of the CVD report in order to obtain a healthy and physically meaningful set of derived standard PVT properties. It can be demonstrated that small error in the data reported by a CVD test can result in PVT property values which are physically impossible (e.g., negative values). And even when the data reported by the CVD report is highly reliable, the Walsh and Towler algorithm can still lead to unphysical values for standard PVT properties. This limitation results from combining the two-phase two-pseudo component (black oil) model with material balance calculation around the PVT cell. This limitation will be discussed in detail in Chapter 5. Walsh-Towler algorithm consists of six sequential steps which must be fully completed at every given pressure level before moving to the next pressure. One pre-calculation is also needed before starting the algorithm. The variables and their nomenclature employed in the sequence of calculations are graphically illustrated in Figure 4-4.

Reservoir Condition PR PDew PR < PDew VEG,j

Surface Condition
Gpj Npj Vg,j Bg = ----------Nfg,j

VT

Vg,j Vo,j Gfg,j

Rv = -----------

Gfg,j

Gfg,j

Reservoir Gas
Reservoir Oil Surface Gas Stock-Tank Oil

Nfg,j

Gfo,j Nfo,j

Vo,j Bo = ----------Nfo,j

Gfo,j Rs = ----------Nfo,j

Figure 4-4: Graphical Representation of CVD Data used in Walsh-Towler Algorithm

54 Pre-calculation: In this step, the total cumulative volumes of surface gas ( ) and stocktank oil ( ) pseudo components produced from the reservoir fluid, and the resulting volume of PVT cell ( ) are calculated for the dew point condition. The volume of surface gas pseudo

component ( ) is calculated from the summation of cumulative gas recovery from 1 st stage separator, 2nd stage separator, and stock tank for all available pressures - from dew point conditions to the last reported (abandonment) pressure. The volume of stock-tank oil pseudo components ( ) is equal to cumulative oil recovery from stock tank for all available and reported pressures (dew point to abandonment). These data are obtained from the calculated cumulative recovery reported in the depletion table. PVT cells volume is calculated from the definition of gas formation volume factor (Equation 4-63). The gas formation volume factor ( ) is calculated from Equation 4-57.

Compressibility factor of gas phase ( ) can be obtained from the CVD report. Mole fraction of surface gas pseudo component in the reservoir gas ( ) is equal to equivalent at the dew point ( divided by the volume of gas

) which is usually taken as 1000 MSCF.

Equation 4-63

Volatilized oil-gas ratio at dew point ( formation volume factor (

) is calculated from Equation 4-64, while oil ) are calculated from Equation 4-61

) and solution gas-oil ratio (

and Equation 4-62, respectively.

55

Equation 4-64

Step 1: Find

and

: Starting at the dew point, the volume of surface gas pseudo ) at each pressure is calculated from the summation

component released from the excess gas (

of cumulative gas recovery from 1st stage separator, 2nd stage separator, and stock tank. Volume of and stock-tank oil pseudo component released from the same excess gas ( ) at each pressure

is equal to cumulative oil recovery from stock tank. These data are obtained from the calculated cumulative recovery reported in the depletion table. Incremental of and from pressure level

j-1 to pressure level j are calculated from Equation 4-65 and Equation 4-66. Please note that pressure level j begins from zero at the dew point (j=0). equal to zero. , , , and are also

Equation 4-65

Equation 4-66

Step 2: Find

and

: Total volume of surface gas ( ) and stock-tank oil ( ) pseudo

components released from both reservoir gas and reservoir oil at pressure level j are calculated from Equation 4-67 and Equation 4-68. It should be noted that pressure level j begins from zero at the dew point (j=0), and and are equal to and , respectively.

56

Equation 4-67

Equation 4-68

Step 3: Find ( (

and

: Volume of oil phase at reservoir condition at pressure level j

) is calculated from Equation 4-69. Retrograde liquid volume fraction at pressure level j ), can be obtained from CVD report. Volume of gas phase after excess gas removal at ) is calculated from Equation 4-70. Note that pressure

reservoir condition at pressure level j (

level j begins at zero at dew point conditions (j=0)

Equation 4-69

Equation 4-70

Step 4: Find

, and

: Molar fraction of reservoir fluid which remains in the

PVT cell at pressure level j ( phase compressibility factor (

) is calculated from Equation 4-71. For this calculation, two) data can be obtained from the CVD report. Molar fraction of ) is calculated from Equation ) is calculated

excess gas which is removed from PVT cell at pressure level j (

4-72. Molar fraction of gas phase which remain in PVT cell at pressure level j (

from Equation 4-73. Compressibility factor of gas ( ) is also obtained from the CVD report.

57 Please note that pressure level j begins from zero (j=0) at the dew point. point are equal to 1.0 while at dew point is equal to zero. and at dew

Equation 4-71

Equation 4-72

Equation 4-73

Step 5: Find

and

: Volume of surface gas pseudo component produced from ) is calculated from Equation 4-74. Volume of stock-tank ) is calculated from

reservoir gas at pressure level j (

pseudo component produced from reservoir gas at pressure level j (

Equation 4-75. It is important to note that pressure level j begins from zero at the dew point (j=0). and at dew point pressure are equal to and , respectively.

Equation 4-74

58

Equation 4-75

Step 6: Find

and

: Volume of surface gas pseudo component produced from ) is calculated from Equation 4-76. Volume of stock-tank oil ) is calculated from

reservoir oil at pressure level j (

pseudo component produced from reservoir oil at pressure level j ( Equation 4-77.

Equation 4-76

Equation 4-77

After completing all six steps outline above for the given pressure level, Equation 4-49 through Equation 4-52 are now directly used to calculate the standard PVT properties. All applicable unit conversion factors must be checked and adjusted properly. The calculation process is systematically repeated for all pressure levels until all reported data in the CVD report have been considered and abandonment conditions have been reached. Standard PVT properties at pressures higher than the dew point are calculated based on the properties at dew point pressure. Gas formation volume factor ( ) is the product of gas

formation volume factor at dew point pressure and relative volume obtained directly from CCE testing results. The relative volume is the ratio between total volume of hydrocarbon at reservoir conditions and the volume at saturated conditions. For under-saturated gas condensate system,

59 relative volume is equal to the ratio between Volatized oil-gas ratio ( formation volume factor ( at specified pressure and at dew point pressure.

) is equal to volatilized oil gas ratio at dew point pressure. Oil ) and solution gas-oil ratio ( ) are calculated from Equation 4-61

and Equation 4-62, respectively. Finally, it is very important to mention that, in Walsh-Towler algorithm, volumes of pseudo components produced from the reservoir oil (step 6) do not actually come from direct surface measurement. In a CVD test, the oil inside the cell is never produced (is assumed immobile) so surface data for produced oil is not available.. Instead, these values are indirectly calculated based on the enforcement of mass balance constraints around the PVT cell. Therefore, actual oil formation volume factor ( ) and solution gas-oil ratio ( ) calculated from actual

surface flashes of the reservoir fluid might be significantly different from the ones estimated using these indirectly calculated surface volumes. If the calculated and resulting from the

application of this algorithm do not agree with the physically acceptable trends or values, the results should be disregarded and the laboratory results have to be adjusted.

60 4.2.3 Obtaining Standard PVT Properties from a Phase Behavior Model Another method for simulating standard PVT properties for gas condensate is to utilize Phase Behavior Model (PBM). This method is based on combination of the algorithm used in Walsh-Towler method and the work of Thararoop in 2007. The general idea of this method is to substitute CVD testing results with the outputs from flash calculation. Mass balance around PVT cell, which is used to obtain the properties of reservoir oil in Walsh-Towler algorithm, is replaced with an actual flash calculation performed for both the reservoir gas and oil phases. Chapter 5 will discuss about the impact from these changes in more detail. Input data required for this method include initial reservoir condition, surface separator conditions, initial reservoir fluid composition, physical properties, binary interaction coefficients, and volume translation coefficients of pure components. The simulation algorithm consists of nine calculation steps and a pre-calculation. Parameters used in those equations were represented graphically in Figure 4-4.

Pre-calculation: First, dew point pressure is determined using a phase stability calculation. Then, mole of initial reservoir fluid inside PVT cell ( ), volume of PVT cell ( ),

volume of surface gas ( ) and stock-tank oil ( ) pseudo components are evaluated at dew point condition. The dew point pressure is determined by performing Phase Stability Analysis. Stability of initial reservoir fluid is continuously evaluated at different pressure levels, while temperature is controlled at reservoir temperature. Pressure level starts at initial reservoir pressure; then, it is continuously decreased by 1.0 psi interval until the initial reservoir fluid becomes unstable. The last pressure level that initial reservoir fluid is in stable condition is the dew point pressure. A

61 direct calculation of saturation pressure at the prevailing reservoir temperature could be also alternatively employed (Whitson and Brule, 2000). The initial amount of mole of the reservoir fluid sample inside PVT cell ( ) is

calculated from Equation 4-78. Standard condition is set to be 14.7 psia and 520 R. Volume of initial reservoir fluid in term of gas equivalent ( as the basis for the calculation. ) is assumed to be 1.0 MMSCF which is used

Equation 4-78

The associated volume of PVT cell ( weight ( ) and density (

) is calculated from Equation 4-79. Molecular

) are obtained by performing flash calculation on initial reservoir

fluid composition at the dew point condition.

Equation 4-79

The molar fractions of surface gas (

) and stock-tank oil ( ) pseudo components in

reservoir fluid are calculated from Equation 4-80 and Equation 4-81. Molar fraction of liquid phase at first-stage separator ( ) is obtained by performing flash calculation on initial

reservoir fluid composition at first-stage separator condition. Molar fraction of liquid phase at second-stage separator ( ) is obtained by performing flash calculation on liquid composition

62 from first-stage separator at second-stage separator condition. Molar fraction of liquid phase at stock-tank condition ( ) is obtained by performing flash calculation on liquid composition

from second-stage separator at stock-tank condition.

Equation 4-80

Equation 4-81

Total volume of surface gas ( ) and stock-tank oil ( ) pseudo components initially present in the reservoir fluid are calculated from Equation 4-82 and Equation 4-83. Value of 379.56 is molar volume of gases at standard condition which is constant. Molecular weight ( ) and density ( ) of oil at stock-tank condition are obtained from flash calculation

results at stock-tank condition. Please note that these values (G and N) are not being obtained by cumulative adding cumulative production values at every pressure level, as done in the original Walsh and Tower algorithm. Chapter 5 will present a discussion on this regard and justification.

Equation 4-82

Equation 4-83

63 The gas formation volume factor ( gas ratio at dew point ( ) is calculated from Equation 4-84. Volatilized oil-

) is calculated from Equation 4-64 by implementing the proper unit ) and solution gas-oil ratio ( ) are calculated

conversion factor. Oil formation volume factor (

from Equation 4-61 and Equation 4-62, respectively.

Equation 4-84

Step 1: Find

and

: Moles of gas phase present at reservoir conditions before ) is calculated from Equation 4-85. ) is calculated from ) is

the removal of excess gas at every pressure level j (

Moles of oil phase remaining at reservoir conditions at pressure level j (

Equation 4-86. Molar fraction of gas phase at reservoir condition at pressure level j (

obtained from performing flash calculation on overall composition from pressure level j-1, at pressure level j. Note that pressure level j begins from zero (j=0) at the dew point. equal to , is equal to zero, and is equal to . is

Equation 4-85

Equation 4-86

64 Step 2: Find and : The volume that the gas phase occupies at reservoir ) is calculated ) is

condition before the removal of the excess gas at every pressure level j (

from Equation 4-87. The volume of reservoir oil phase present at pressure level j (

calculated from Equation 4-88. Molecular weight and density of gas and oil phases at reservoir condition at pressure level j ( , , , ) are obtained by performing flash

calculation on overall composition from pressure level j-1, at pressure level j. Note that pressure level j begins from zero at the dew point (j=0). is equal to and is equal to zero.

Equation 4-87

Equation 4-88

Step 3: Find pressure level j (

and

: The volume of reservoir gas phase after excess gas removal at

) is calculated from Equation 4-89. Volume of excess gas at reservoir ) is then calculated from Equation 4-90.

condition at pressure level j (

Equation 4-89

65

Equation 4-90

Step 4: Find

and

: Remaining moles of gas phase at reservoir condition after ) is calculated from Equation 4-91. Moles of ) is then calculated from Equation 4-92.

excess gas removal at every pressure level j (

excess gas which are removed at pressure level j (

Density and molecular weight are the same as those in Equation 4-87.

Equation 4-91

Equation 4-92

Step 5: Find stock-tank oil (

and

: The molar fractions or compositions of surface gas (

) and

) pseudo components in the reservoir gas at every pressure level j are

calculated from Equation 4-93 and Equation 4-94. The fraction of liquid phase at first-stage separator recovered from reservoir gas at pressure level j ( ) is obtained by performing flash

calculation on composition of reservoir gas at pressure level j, at first-stage separator condition. The fraction of liquid phase at second-stage separator recovered from reservoir gas at pressure level j ( ) is obtained by performing flash calculation on liquid composition from first-stage

66 separator at second-stage separator condition. The fraction of liquid phase at stock-tank condition recovered from reservoir gas at pressure level j ( ) is obtained by performing flash

calculation on liquid composition from second-stage separator at stock-tank condition.

Equation 4-93

Equation 4-94

Step 6: Find

and

: Volume of surface gas (

) and stock-tank oil (

pseudo components in reservoir gas at pressure level j are calculated from Equation 4-95 and Equation 4-96. The value of 379.56 is molar volume of gases at standard condition which is a constant for ideal gases. Molecular weight ( ) and density ( ) of oil at stock-tank

condition recovered from reservoir gas at pressure level j are obtained from flash calculation results at stock-tank condition in Step 5.

Equation 4-95

Equation 4-96

67

Step 7: Find (

and

: The molar fractions of surface gas (

) and stock-tank oil

) pseudo components in the reservoir oil at every pressure level j are calculated from

Equation 4-97 and Equation 4-98. The fraction of liquid phase at first-stage separator recovered from reservoir oil at pressure level j ( ) is obtained by performing flash calculation on

composition of reservoir oil at pressure level j, at first-stage separator condition. The fraction of liquid phase at second-stage separator recovered from reservoir oil at pressure level j ( ) is

obtained by performing flash calculation on liquid composition from first-stage separator at second-stage separator condition. The fraction of liquid phase at stock-tank condition recovered from reservoir oil at pressure level j ( ) is obtained by performing flash calculation on liquid

composition from second-stage separator at stock-tank condition.

Equation 4-97

Equation 4-98

Step 8: Find

and

: The volume of surface gas (

) and stock-tank oil (

pseudo components in reservoir oil at pressure level j are calculated from Equation 4-99 and Equation 4-100. The value of 379.56 is molar volume of gas at standard condition which is constant. Molecular weight ( ) and density ( ) of oil at stock-tank condition recovered

from reservoir oil at pressure level j are obtained from flash calculation results at stock-tank condition in Step 7.

68

Equation 4-99

Equation 4-100

Step 9: Find level j (

and

: Remaining moles of reservoir fluid inside PVT cell at pressure

) is calculated from Equation 4-101. Overall composition of i-th component insider ) after gas removal is updated by implementing Equation 4-102. is equal to . Liquid

PVT cell at pressure level j (

Note that pressure level j begins from zero (j=0) at the dew point. composition ( ) and vapor composition (

) of i-th component at pressure level j are obtained

by performing flash calculation on overall composition from pressure level j-1, at pressure level j.

Equation 4-101

Equation 4-102

After completing all nine steps outlined above at every given pressure level, Equation 4-49 through Equation 4-52 will be used to directly calculate standard PVT properties. All unit

69 conversion factors must be checked and properly adjusted. This calculation process must be continuously repeated for the every pressure level until abandonment pressure is reached. Standard PVT properties at pressures higher than the dew point are calculated based on available properties at dew point pressure. Gas formation volume factor ( ) is calculated from

gas formation volume factor at dew point pressure using Equation 4-103. The ratio between ( ) at dew point pressure and ( ) at specified pressures above the dew point is equivalent to

ratio between volume of reservoir gas ( ) at specified pressures above the dew point and volume of reservoir gas ( ) at dew point pressure. Volatized oil-gas ratio ( gas ratio at dew point pressure ( ( ). Oil formation volume factor ( ) is equal to volatilized oil) and solution gas-oil ratio

) are calculated from Equation 4-61 and Equation 4-62, respectively.

Equation 4-103

70 4.3 Zero-Dimensional Reservoir Model The Material Balance Equation (MBE) (also known as zero-dimensional reservoir model or tank model) is a mass balance statement that combines mass balance equations of all pseudo components present in the reservoir fluid. The assumptions behind a tank model have been already addressed in Section 2.3. Walsh and Lake (2003) have presented a generalized form of material balance equation that could be used for the analysis of depletion performance for all five types from reservoir fluids, based on the work originally published by Walsh (1995). They also developed the MBE specialized for gas condensate fluids by simplifying the generalized MBE for the conditions particular to these kind of fluids. Section 4.3.1 discusses and presents the GMBE proposed by Walsh as implemented in this study. In zero-dimensional reservoir model, cumulative productions of pseudo components and saturations of reservoir fluids are calculated as functions of reservoir pressure, standard PVT properties, and initial reservoir condition. This model treats a reservoir as a homogeneous tank; thus only average reservoir pressure and average PVT properties are required as the model inputs. In this study, a VBA subroutine has been developed to simulate cumulative oil and gas productions as well as their saturations as a function of reservoir pressure, by implementing the MBE specialized for gas condensate fluids. Most of the time, the MBE is used to simulate the results explicitly as a function of time and depletion. However, if some target outputs are specified, such as cumulative recovery at end of plateau, an iterative procedure would need to be implemented in order to honor the additional constraint.

71 4.3.1 Generalized Material Balance Equation Generalized Material Balance Equation (GMBE) is the most generalized form of Material Balance Equation which can be applied to all types of reservoir fluids. Walsh and Lake derived the GMBE by combining mass balance equation of pseudo components, surface gas, stock-tank oil and stock-tank water, with the saturation constraint and standard PVT properties described in section 4.2 (Walsh and Lake, 2003). The following assumptions are assumed in addition to the general assumptions for zero-dimensional reservoir model. Reservoir consists of surface gas, stock-tank oil, and stock-tank water pseudo components Reservoir consists of gas, oil, and water phases. Surface gas pseudo component is in reservoir gas and oil phases. Stock-tank oil pseudo component is in reservoir gas and oil phases. Stock-tank water pseudo component is in reservoir water phase. Surfaces gas, stock-tank oil, and stock-tank water can be produced Surface gas and stock-tank water can be injected into the reservoir Water phase can enter into reservoir by water influx from aquifer GMBE can be manipulated into many different forms. One of the most useful forms of the GMBE is shown in Equation 4-104. The terms on the left-hand side represents net reservoir expansion terms while terms on the right-had side represents net reservoir withdrawal. Net reservoir expansion consists of net reservoir gas expansion, net reservoir oil expansion, net reservoir water expansion, net formation expansion, and water influx. Net reservoir withdrawal consists of net gas and oil withdrawal, and net water withdrawal.

72

Equation 4-104

where: = volume of surface gas pseudo component in reservoir gas at initial condition {SCF} = volume of stock-tank oil pseudo component in reservoir oil at initial condition {STB} = volume of water component in reservoir water at initial condition {STB} = pore volume at initial condition {RB} = volume of water influx {RB} = cumulative gas production {SCF} = cumulative gas injection {SCF} = cumulative oil production {STB} = cumulative water production {STB} = cumulative water injection {SCF} = expansivity of reservoir gas {RB/SCF} = expansivity of reservoir oil {RB/STB} = expansivity of reservoir water {RB/STB} = expansivity of formation (rock) {Dimensionless} = gas formation volume factor {RB/SCF} = oil formation volume factor {RB/STB}

73 = water formation volume factor {RB/STB} = volatilized oil-gas ratio {STB/SCF} = solution gas-oil ratio {SCF/STB}

Expansivity of reservoir fluid is defined as the total expansion of a unit mass of reservoir fluid between two reservoir pressures at the same reservoir temperature. Expansivities of reservoir gas, reservoir oil, and reservoir water are calculated from Equation 4-105, Equation 4-106, and Equation 4-107, respectively. Expansivity of formation (rock) is defined in a different form from fluid expansivity and is calculated in terms of formation (rock) compressibility as indicated by Equation 4-108.

Equation 4-105

Equation 4-106

Equation 4-107

Equation 4-108

74 where: = two-phase gas formation volume factor {RB/SCF} = two-phase oil formation volume factor {RB/SCF} = formation (rock) compressibility {psi-1} = pressure drop from initial reservoir pressure {psi}

The two-phase formation volume factor implemented above is defined as the ratio between total volume of reservoir fluid (gas and oil phases) and total volume of the pseudo component. Two-phase formation volume factor of gas ( ) and oil ( ) are calculated from

Equation 4-109 and Equation 4-110, respectively. If reservoir is a single phase gas reservoir, twophase gas formation volume factor ( ) will be equal to gas formation volume factor ( ) while

two-phase oil formation will remain undefined. Similarly, if reservoir is single phase oil reservoir, two-phase oil formation volume factor ( ) will be equal to oil formation volume factor ( )

while the two-phase gas formation volume factor will remain undefined.

Equation 4-109

Equation 4-110

75 4.3.2 Material Balance Equation for a Gas Condensate Fluid GMBE can be simplified significantly when condensate drop out, developed below dew point saturation conditions in the reservoir, is considered immobile. The immobile condensate assumption is a fairly reasonable one for gas condensates; however, it cannot be applied for other types of reservoir fluid (Walsh and Lake, 2003). The Simplified Gas Condensate Tank model, SGCT, is derived from Generalized Material Balance Equation with the following additional assumptions: Reservoir is under-saturated at initial reservoir pressure Expansivities of water and formation are negligible There is no water influx, water production, and water injection There is no gas injection Condensate drop out in the reservoir is immobile

Gas Condensate Performance Below Dew Point At initial undersaturated conditions, the volume of surface gas pseudo component in reservoir gas at initial condition is equal to the Original Gas In Place (OGIP or G) while and the volume of stock-tank oil pseudo component in reservoir oil at initial condition is equal to zero. Equation 4-111 is the SGCT model after applying all these additional assumptions:

Equation 4-111

76 This SGCT model can be further manipulated in order to obtain a more useful form by dividing it through by and substituting by . After that, finite difference

approximation is applied, resulting in expressions for the calculation of incremental oil and gas production. As a result, Equation 4-112 through Equation 4-118 are a set of equations that can be used to calculate reservoir performance from SGCT model.

Equation 4-112

Equation 4-113

Equation 4-114

Equation 4-115

Equation 4-116

77

Equation 4-117

Equation 4-118

where = incremental gas recovery from pressure level = incremental oil recovery from pressure level = cumulative gas recovery at pressure level = cumulative oil recovery at pressure level = gas formation volume factor at pressure level {RB/SCF} = oil formation volume factor at pressure level {RB/STB} = volatilized oil-gas ratio at pressure level {STB/SCF} = solution gas-oil ratio at pressure level {SCF/STB} = two-phase gas formation volume factor at pressure level {RB/SCF} = two-phase oil formation volume factor at pressure level {RB/SCF} to to

78 It is important to note that this set of equations (Equation 4-112 through Equation 4-118) only applies for pressure below the dew point. Pressure level j begins at the first pressure below the dew point. and at pressure level j-1 are cumulative results from the calculation

above the dew point. The standard PVT properties are either obtained from any of the procedures described in section 4.2 or are given as input data. Two-phase formation volume factors are calculated from Equation 4-109 and Equation 4-110. Calculation of SGCT model must be fully completed at one pressure level before moving onto the next pressure level.

Gas Condensate Performance Above Dew Point At pressure higher than dew point pressure, SGCT model should be further modified by substituting and into Equation 4-111. As a result, the MBE for an

under-saturated gas condensate collapses to the typical MBE for a wet gas, shown in Equation 4-119. A set of equations which mimic the calculation procedure of SGCT model below dew point can be developed by using a finite difference approach. Resulting performance prediction equations are shown in Equation 4-120 to Equation 4-123.

Equation 4-119

Equation 4-120

79

Equation 4-121

Equation 4-122

Equation 4-123

It should be noted that this set of equation (Equation 4-120 to Equation 4-123) only applies for pressures above dew point. The pressure level j begins from zero at initial reservoir and end at the last pressure above the dew point. , , , and are equal to

zero at the initial reservoir pressure. The calculation has to be completed at one pressure level before moving to the next pressure level.

80 4.3.3 Phase Saturation Calculations One of the methods to derive a phase saturation equation is to combine the mass balance equation for the stock-tank oil pseudo-component with the saturation equation constraint in order to eliminate gas saturation parameter ( ) and then combine the resulting

equation and volumetric OGIP calculation equation so that the pore volume variable is eliminated. The resulting saturation equation for the oil phase is shown in Equation 4-124.

Equation 4-124

where: = average reservoir oil saturation = average reservoir gas saturation = average reservoir water saturation = cumulative oil production {STB} = original oil in place {STB} = average reservoir porosity = gas formation volume factor {RB/SCF} = oil formation volume factor {RB/STB} = volatilized oil-gas ratio {STB/SCF} = subscript for initial condition

81 For the typical gas condensate reservoir, initial oil saturation ( ) is equal to zero

because its initial pressure is typically found above dew point conditions. If formation/rock expansion is neglected, porosity of the porous medium would remain constant at the value of initial porosity ( ). If net water withdrawal, influx, and expansion are neglected, water ).

saturation would remain constant at the value of connate water saturation ( Thus, for such conditions, Equation 4-124 can be significantly simplified to Equation 4-125.

Equation 4-125

For saturation calculation of gas condensate reservoir, if reservoir pressure is equal to or higher than dew point pressure, average reservoir oil saturation is equal to average reservoir oil saturation is calculated from Equation 4-125. . Otherwise,

82 4.3.4 Volumetric OGIP/OOIP Calculations Original Gas In Place ( ) and Original Oil In Place ( ) can be calculated using

volumetric method. Equation 4-128 and Equation 4-129 are used to calculate OGIP and OOIP, respectively, based on the premise that both surface gas and stock-tank oil pseudo components can be found in the reservoir gas and reservoir oil phases. In these equations, multiplication of drainage area, reservoir thickness, and porosity represent the pore volume of the reservoir. The conversion factor of 7758 is used to convert acre-ft unit to RB unit. The first and the second terms inside the bracket of Equation 4-126 represent volumes of surface gas pseudo component in reservoir oil and reservoir gas phases, per reservoir pore volume, respectively. The first and the second terms inside the bracket of Equation 4-127 represents volume of stock-tank oil pseudo component in reservoir oil and reservoir gas phases, per reservoir pore volume, respectively.

Equation 4-126

Equation 4-127

where: = reservoir drainage area {acre} = reservoir thickness {ft} = reservoir porosity

83 = initial oil saturation = initial gas saturation = oil formation volume factor at initial condition {RB/STB} = gas formation volume factor at initial condition {RB/SCF} = solution gas-oil ratio at initial condition {SCF/STB} = volatilized oil-gas ratio at initial condition {STB/SCF}

For gas condensate reservoir which do not initially have an oil phase (

), Equation

4-126 and Equation 4-127 can be simplified into Equation 4-128 and Equation 4-129, respectively. Reservoir properties which are drainage area ( ), thickness ( ), porosity ( ), and initial water saturation ( volume factor ( ) must be known. Fluid properties which are initial gas formation ) are obtained from standard PVT

) and initial volatilized oil-gas ratio (

property estimations.

Equation 4-128

Equation 4-129

84 4.4 Flow Rates and Flowing Pressures Calculation In this study, flow rates and flowing pressures calculations are based on the implementation of Inflow Performance Relationships (IPR) and Tubing Performance Relationships (TPR). The IPR equation relates flow rates from reservoir into the wellbore with the difference between reservoir pressure and bottomhole flowing pressure. The TPR equation relates wellbore flow rates between a bottomhole to a surface location in terms of the difference between bottomhole flowing pressure and wellhead pressure. Because of the immobile condensate assumption, oil flow rates can be calculated as a function of gas flow rates and volatilized oil-gas ratio. Moreover, because a pipeline equation based on the homogeneous flow assumption (single pseudo phase) can be used as the TPR equation, by implementing an appropriate tubing efficiency factor, gas flow rate is the only parameter that needs to be determined. In field performance prediction, either the desired gas flow rate or the target wellhead pressure will be specified. If gas flow rate is specified, the IPR equation will be used to explicitly calculate bottomhole flowing pressure, and the TPR equation will be used to explicitly calculate wellhead pressure. If wellhead pressure is specified, bottomhole flowing pressure and gas flow rate are simultaneously solved by using nodal analysis method to determine the solution of the IPR and TPR system of equations. In this study, two subroutines have been developed for gas flow rate and bottomhole flowing pressure calculations based on IPR equations. Similarly, two subroutines have been developed for gas flow rate and wellhead pressure calculations based on TRP equations. For the numerical solution, nodal analysis method is also implemented by using bisection iterative procedure.

85 4.4.1 Inflow Performance Relationship (IPR) The pseudo steady state (PSS) flow rate from the reservoir into the wellbore of radius flow in cylindrical-shape reservoir, closed boundary can be calculated using Equation 4-130 (Lee et al, 2003).

Equation 4-130

For other reservoir shapes, PSS flow rates can be calculated by applying shape factor concept (Deitz, 1965). Flow rate calculation with Deitz shape factor ( 4-131. The Deitz shape factor ( ) is shown in Equation

) is equal to 31.62 for a circular drainage area with a well

located at the center of the reservoir, and equal to 30.88 for a square drainage area with a well located at the center of the reservoir.

Equation 4-131

For the two-phase two-pseudo component model, the surface gas pseudo-component is recovered from both reservoir gas and reservoir oil (as dissolved gas); while stock-tank oil pseudo component is recovered from reservoir oil and reservoir gas (as volatilized oil). By introducing the concept of phase mobilities and phase relative permeabilities, Equation 4-131 can then be modified to calculate flow rates for this two-phase two-pseudo component model. Gas flow rate

86 ( ) and oil flow rate ( ) can be calculated from Equation 4-132 and Equation 4-133,

respectively.

Equation 4-132

Equation 4-133

where: = flow rate of surface gas {SCF/D} = flow rate of stock-tank oil {STB/D} = absolute permeability of reservoir {md} = relative permeability of reservoir gas = relative permeability of reservoir oil = average reservoir pressure {psia} = bottomhole flowing pressure {psia} = reservoir thickness {ft} = reservoir drainage area per well {acre} = wellbore radius {ft} = Detiz Shape Factor = total skin factor

87 = viscosity of reservoir gas {cp} = viscosity of reservoir oil {cp} = oil formation volume factor {RB/STB} = gas formation volume factor {RB/SCF} = solution gas-oil ratio {SCF/STB} = volatilized oil-gas ratio {STB/SCF}

The first terms in the last bracket of Equation 4-132 and Equation 4-133 represent fluid production that comes from the reservoir oil, while the second terms represent fluid production that comes from the reservoir gas. However, for a gas condensate reservoir, reservoir oil is typically assumed to be immobile ( ). Thus, gas flow rate ( ) and oil flow rate ( )

equations are simplified into Equation 4-134 and Equation 4-135, respectively

Equation 4-134

Equation 4-135

88 In Equation 4-134 and Equation 4-135, only gas and oil flow rates ( bottomhole flowing pressure ( and ) or

) could be specified. Absolute reservoir permeability ( ),

reservoir thickness ( ), drainage area per well ( ), wellbore radius ( ), and Deitz shape factor ( ) are required reservoir data. Gas formation volume factor ( ), volatilized oil-gas ratio ( ),

and specific gravity of gas (

) are functions of average reservoir pressure ( ). Average

reservoir pressure ( ) is obtained from the SGCT (zero-dimensional material balance) subroutine. Gas viscosity ( ) is calculated from correlation proposed by Lee et al in 1966

presented as Equation 4-24 through Equation 4-27. Even though the relative permeability to oil ( ) can be safely assumed to remain zero

or close to zero during depletion of a gas condensate reservoir, the relative permeability to gas is not expected to remain equal to one in the presence of condensate. Typically, the mobility of the gas phase and thus its relative permeability are expected to decrease with increased condensate drop out. The relative permeability of the gas phase will be further hindered if average water saturation in the reservoir increases because of the presence of an active water drive. Relative permeability of reservoir gas ( ) is a function of the average reservoir gas saturation (

). In this study, there are two input options for relative permeability data. The first option is to input gas saturations and their corresponding values manually in a tabular form. Such

data could be obtained from core study results performed in a laboratory. The second option is to use a correlation for three-phase relative permeability. Any relative permeability model can be utilized; however, Naar, Henderson, and Wygals model (Ertekin et al, 2001) has been used as a default model in this work. The correlation is shown in Equation 4-136.

Equation 4-136

89 where: = average reservoir gas saturation = connate water saturation = average reservoir gas saturation = coefficient to adjust relative permeability of reservoir gas, typically one when ) is expected to take value of one at the end point ( = 1-

Gas saturation ( ) can change from a minimum gas saturation ( saturation ( gas saturation ( coefficient, ( ). Relative permeability of reservoir gas ( ) or lower. Connate water saturation (

) to a maximum gas

) is equal to zero at the minimum ) is a required reservoir data. The

, is used to adjust relative permeability anchor point at the initial gas saturation ). If at the initial gas saturation, or if there is no further core or lab data adjustment coefficient value is set to 1.0.

available indicating otherwise,

Skin factor is dimensionless pressure drop around the wellbore, which accounts for the differences between reservoir models analytical assumptions and actual conditions in reservoir flow. Total skin factor ( ) consists of mechanical skin ( ) and non-Darcy skin ( ).

Mechanical skin can be estimated from pressure transient analysis or from other analogous approaches such as type curve matching. Non-Darcy coefficient ( ) can be obtained from the analysis of multi-rate well test, or analogous approaches, or from Equation 4-138 (Lee et al, 2003. Eq. 3.19) if its required input data is known

90 Equation 4-137

Equation 4-138

where: = mechanical skin factor = non-Darcy coefficient {Day/SCF} = flow rate of surface gas {SCF/D} = turbulence parameter = effective permeability of reservoir gas ( = molecular weight of reservoir gas {lbm/lbmol} = pressure at standard condition {14.7 psia} = reservoir thickness {ft} = wellbore radius {ft} = temperature at standard condition {520 R} = viscosity of reservoir gas at bottomhole flowing pressure {cp} ) {md}

Total skin factor ( ) is a function of gas flow rate, because of the presence of the nonDarcy skin component, while gas flow rate also is function of skin factor. Thus, Equation 4-134 cannot be solved explicitly for gas flow rate. When the non-Darcy component is expected to be significant, Equation 4-134 can be recast in terms of a quadratic expression in gas flow rate ( )

91 which can be solved analytically. Alternatively, Equation 4-134 can be solved directly by implementing an iterative numerical approach. The latter is the approach employed in this study.

4.4.2 Tubing Performance Relationships For a gas condensate reservoir, condensate drop out is not only expected to occur at reservoir conditions but also along the surface depletion path as produced fluids make their way to the surface. For the case of the two-phase two-pseudo component model, the amount of reservoir condensation can be estimated for the isothermal reservoir conditions using the concept of volatilized oil-gas ratio. The amount of condensate at the surface is estimated using the surface pseudo-component concept. However, the table of black oil standard PVT properties provides no information about how much condensate can be expected as a function of both pressure and temperature changes inside the well tubing during the wellstream fluid travel from reservoir to surface conditions. The difference in the values of volatilized oil-gas ratio between for any two points of pressure inside the tubing would provide a measure of gas condensation but assuming that those two points are found at separator temperature conditions. To overcome this problem, this study invokes the homogeneous flow assumption at tubing conditions and applies the wellknown expressions for the flow of gases in a pipeline, adjusted according to an appropriate value of tubing efficiency. The homogeneous single phase flow equation used to calculate gas flow rate ( ) for a given downstream pressure ( ), and vice versa is shown in Equation 4-139. Pipeline ) is defined to account for the extra pressure drop that should be

efficiency factor (

expected due to presence of liquid phase.

92

Equation 4-139

Equation 4-140

where: = gas flow rate {SCF/D} = upstream pressure {psia} = downstream pressure {psia} = pressure at standard condition {14.7 psia} = pipe section average temperature {R} = temperature at standard condition {520 R} = average compressibility factor = Fannings fraction factor = specific gravity of gas = tubing diameter {inch} = tubing length {mile} = difference in elevation of downstream and upstream {ft} = efficiency factor of tubing

93 Upstream pressure ( ), tubing diameter ( ), tubing length ( ), elevation at upstream ( ) and downstream ( calculated as ) of the tubing are required input data. Difference in elevation is . Tubing temperature changes are assumed to follow the geothermal ) is assumed to be equal to an average between

gradient. An average temperature (

temperatures at upstream ( ) and downstream ( ) nodes for any given tubing section. Average pressure ( ) is calculated from Equation 4-141. Average compressibility factor ( ) and average temperature ( ). ) is

calculated at average pressure (

Equation 4-141

In Equation 4-140 above, the Fannings friction factor ( ) is equal to a quarter of Moodys friction factor ( ). The Moodys fraction factor ( ) can be calculated from Equation

4-142 (Colebrook, 1939). This equation is solved using iterative procedure because Moodys friction factor appears implicitly on both sides of the equation. The Reynolds number is calculated from Equation 4-143. In situ fluid density ( ) and velocity ( ) of gas are evaluated at average pressure and temperature using real gas equation of state. Gas viscosity is calculated from the correlation proposed by Lee et al in 1966.

Equation 4-142

94

Equation 4-143

where: = Moodys friction factor = tubing diameter {inch} = tubing roughness {ft} = Reynolds number = fluid density {lbm/ft3} = fluid velocity {ft/sec} = fluid viscosity {cp}

The concept of tubing efficiency factor (

) is applied to for the detrimental

presence of a liquid phase on pipe performance. For gas condensate fluid, amount of liquid phase in pipeline is partially related the change in volatilized oil-gas ratio ( ) and tubing flowing

pressure, which both of them are also functions of reservoir pressure. Therefore, the efficiency factor would also implicitly depend on reservoir pressure. In this study, a single value of tubing efficiency factor, assumed to be representative of the entire tubing flow, is employed. The efficiency factor should be calculated in such a way that the calculated flow rate and pressure would match actual production data for the period of interest. During calculations, the tubing is divided into several pipe sections. The calculation is performed section by section, starting at the bottommost section and marching towards the topmost section of the tubing. The upstream pressure of the bottommost section is the bottomhole flowing pressure, while the upstream pressure of the upper section is the calculated downstream

95 pressure from the lower section. Wellhead pressure represents the downstream pressure of the topmost section of the tubing. TPR calculations can be performed in two ways. One possibility is calculating wellhead pressure ( ) as a function of gas flow rate ( ). For such scenario, downstream pressure ( )

in gas flow equation is sequentially solved starting at the bottommost section until reaching the topmost section of the tubing. For each section, an iterative process is needed to determine downstream pressure ( ) because downstream pressure ( ) affects the value of average compressibility factor ( ) and Reynolds number ( ). The downstream pressure ( ) of the

topmost section of the tubing is the desired target. The second possibility for TPR calculations is finding the corresponding gas flow rate ( ) for a given wellhead pressure ( ). In this case, initial guess of gas flow rate ( ) is ) are

calculated using well-known Weymouth correlation. Then, wellhead pressures ( calculated for the initial guess of gas flow rate ( incremental wellhead pressure ( The gas flow rate (

). Based on the ensuing sensitivity analysis, ) can be estimated. )

) and incremental gas flow rate (

) is now updated based on the derivative of gas flow rate (

and the difference between calculated and specified wellhead pressures (

). This procedure is

repeated until the difference between calculated and specified wellhead pressure is less than a prescribed tolerance value.

96 4.4.3 Nodal Analysis In petroleum engineering, nodal analysis is used to determine pressure and flow rate at some node or location of interest in the production system. In this study, sand-face location is the selected node or location of interest. Incoming gas flow rate to the node (inflow) can be calculated from inflow performance relationships (IPR), while outgoing gas rate from the node (outflow) can be calculated using tubing performance relationships (TPR). Nodal analysis is then performed to determine pressure ( ) and flow rate ( ) that satisfies both IPR and TPR

relations at the same time which correspond to the point where inflow and outflow curve cross each other, as depicted in Figure 4-5.

Inflow Performance Relationship

Pwf - Bottomhole Flowing Pressure

Pwf and qgsc that satisfy both IPR & TPR

Tubing Performance Relationship

qgsc - Gas Flow Rate

IPR

TPR

Figure 4-5: Graphical Representation of Nodal Analysis

Figure 4-5 shows the graphical representation of nodal analysis method. IPR curve is constructed using Equation 4-134, and TPR curve is constructed using Equation 4-139. The

97 intersection between IPR and TPR curves represents the solution of the nodal analysis problem. As it can be seen from the figure, as the bottomhole flowing pressure ( ) is lower than the

intersection pressure, more gas would be able to flow from reservoir into the wellbore; however, the tubing would not be able to deliver all of gas to the surface for the given pressure drop. In contrast, if the bottomhole flowing pressure ( ) is higher than the intersection pressure, less

gas is able to flow from reservoir into the wellbore even though the tubing would be able to deliver more than that. Both scenarios are not physically possible for a steady state condition where the flow from the reservoir and in the tubing must be equal to each other. There is only one possible flow condition which is found at the intersection point. At the intersection pressure, all of gas which flows from the reservoir into the wellbore can be delivered to the surface production system. The solution of the IPR and TRP point of intersection requires an iterative procedure. In this study, the bi-section method is applied. Reservoir pressure ( ) and wellhead pressure ( are the independent variables while bottomhole flowing pressure ( ) and gas flow rate ( ) )

are the dependent variables. Initial guess of bottomhole flowing pressure (

) is taken as an

average between the higher pressure boundary which is equal to reservoir pressure ( ) and lower pressure boundary which is equal to wellhead pressure ( rates ( ) for the first iteration. Then, gas flow

) from IPR and TPR correlations are calculated for the current guess of bottomhole ). If gas flow rate calculated from the IPR curve is less than gas flow rate

flowing pressure (

from TPR, then the higher pressure boundary is replaced by current bottomhole flowing pressure ( ( ). If not, the lower pressure boundary is replaced by current bottomhole flowing pressure ) instead. New iteration starts by averaging the bottomhole flowing pressure ( ) from the

updated higher and lower pressure boundaries. The iterative process is repeated until the

98 difference between calculated gas flow rate ( tolerance. ) from IPR and TPR is less than a prescribed

99 4.5 Field Performance Prediction Field performance predictions are based on the integration of reservoir tank models with flow rates and flowing pressure (IPR/TRP) models, subjected to a given set of production constraints, with the goal of predicting pressures (reservoir, wellhead, bottomhole), flow rate, and cumulative production evolution with respect to reservoir production time. The production time variable, which is eliminated during the development of reservoir tank or zero-dimensional models, is now calculated by combining tank models and flow rate models (IPR/TRP) in the same calculation. In this work, substantial drilling capacity is assumed; thus build-up period is neglected. Field life is divided into two periods: the plateau period and decline period. All wells are put on production at the beginning of plateau period. During the plateau period, gas flow rate ( kept constant by adjusting wellhead pressure ( ( ) is

). During the decline period, wellhead pressure

) is kept constant at the minimum allowable wellhead pressure. The field is abandoned when

the reservoir pressure reaches the specified abandonment pressure. The following sections described the algorithm for predicting field performance in a step-by-step procedure.

100 4.5.1 Performance during Plateau Period During plateau period, the delivered gas flow rate ( ) from the field is kept constant.

In order to achieve this, in spite of decreasing pressure and gas flow capacity in the reservoir, wellhead pressures ( ) are adjusted throughout the plateau period in order to honor the

constant production specification. Naturally, the reservoir must have the capacity (in terms of reserves or OGIP) of delivering such target plateau rate throughout the plateau period. Therefore, the gas flow rate during plateau period ( ) is constrained by the feasibility of productive

reservoir performance at the end of plateau period. The following procedure is used to predict the field performance during plateau period.

Estimation of Cumulative Production at End of Plateau (

Cumulative production at end of plateau is calculated from target recovery factor at end of plateau ( ) (which is considered a given or specified variable provided to the model)

times original gas in place ( ) (which must be known before field performance predictions can be undertaken).

Estimation of Reservoir Pressure at End of Plateau (

Reservoir pressure at end of plateau ( production at end of plateau (

) is calculated from cumulative gas

) and the SGCT model. In the SGCT model, reservoir ) is the dependent )

pressure ( ) is an independent variable, and cumulative gas recovery ( variable. Cumulative gas production (

) is calculated from cumulative gas recovery (

times original gas in place ( ). Therefore, finding reservoir pressure ( ) at a given cumulative

101 gas production ( ) requires an iterative process or pressure that is build around matching the ). Bi-section iterative method is utilized

prescribed gas production at the end of plateau ( in this study.

Estimation of Plateau Gas Flow Rate (

Plateau gas flow rate (

) is calculated using the nodal analysis method. Because in ) fixes the gas rate at the end of

this calculation the minimum allowable wellhead pressure (

plateau, nodal analysis is applied at the prevailing conditions at the end of the plateau period. For the IPR calculation, for example, the reservoir pressure ( ) is equal to reservoir pressure at end of plateau ( ). For the TPR calculation, wellhead pressure ( ) is fixed to be equal to

minimum allowable wellhead pressure ( gas flow rate per well ( Total plateau gas flow rate ( number of wells.

). Intersection of IPR and TPR curves are the plateau ).

) and bottomhole flowing pressure at end of plateau ( ) is equal to the plateau rate per well (

) times total

Estimation of Reservoir Pressure ( ), Bottomhole Flowing Pressure ( Pressure ( ) during plateau period

), and Wellhead

During the plateau period, reservoir pressure ( ) decreases continuously from initial reservoir pressure ( ) to reservoir pressure at end of plateau ( ) based on the ) are calculated as a

predictions from the SGCT model. Bottomhole flowing pressures ( function of plateau gas flow rate per well (

) at different reservoir pressures ( ) by ) are calculated as a

implementing the IPR equation (Equation 4-134). Wellhead pressures (

102 function of plateau gas flow rate per well ( ( ) by implementing the TPR calculation. ) at different bottomhole flowing pressures

Estimation of Gas Flow Rate (

) and Oil Flow Rate (

Total gas flow rate production ( prescribed plateau gas flow rate ( calculated from total gas flow rate ( ratios (

) during plateau period is constant and equal to the ) for the entire field. Total oil flow rates ( ) are

) times volatilized oil-gas ratios (

). Volatilized oil-gas

) at different reservoir pressures ( ) are linearly interpolated from the table of standard

PVT properties.

Estimation of Cumulative Gas Production ( Cumulative gas production (

) and Cumulative Oil Production (

) ) times

) is calculated as cumulative gas recovery (

original gas in place ( ). Cumulative oil production ( recoveries ( (

) is calculated as cumulative oil ) and cumulative oil

) times original oil in place ( ). Cumulative gas (

) recoveries at different reservoir pressures ( ) are calculated from the SGCT model.

Estimation of Production Time ( ) Production times ( ) at different reservoir pressures during the plateau period are calculated from cumulative gas productions ( ( ( ). Plateau time ( ) divided by total plateau gas flow rate

) is equal to cumulative gas production at end of plateau ).

) over total plateau gas flow rate (

103 4.5.2 Performance during Decline Period During decline period, wellhead pressure ( minimum allowable wellhead pressure ( ) is kept at a constant value equal to the )

). As a consequence of this, total gas flow rate (

would continuously decrease as a function of reservoir pressure ( ). The following procedure is used to predict field performance during decline period.

Estimation of Reservoir Pressure ( Pressure ( ) during decline

), Bottomhole Flowing Pressure (

), and Wellhead

During the decline period, reservoir pressure ( ) continues to decreases from reservoir pressure at the end of plateau ( ) to abandonment reservoir pressure ( ) based on

predictions from the SGCT model. Bottomhole flowing pressures (

) are calculated using

nodal analysis. In the IPR calculation, reservoir pressures ( ) change with production. Wellhead pressure is fixed ( ) in the TPR calculation. Intersections of IPR and TPR curves at different ) and gas flow rates

reservoir pressures ( ) yield the resulting bottomhole flowing pressures ( per well ( ) for the decline period.

Estimation of Gas Flow Rate (

) and Oil Flow Rate (

Gas flow rates per well ( with bottomhole flowing pressures ( flow rate ( rate (

) at different reservoir pressures ( ) are calculated along ) using the nodal analysis discussed earlier. Total gas ) times total number of wells. Oil flow ) times the volatilized oil-gas ratio

) is equal to gas flow rate per well (

) is calculated from total gas flow rate (

104 ( ).Volatilized oil-gas ratios ( ) at different reservoir pressures ( ) are linearly interpolated

from the table of standard PVT properties.

Estimation of Cumulative Gas Production ( Cumulative gas productions (

) and Cumulative Oil Production (

) and cumulative oil production (

) calculation for

decline period follows the same protocol discussed for the plateau period.

Estimation of Decline Rate ( i) If the reservoir decline behavior is assumed to be an exponential decline, the decline rate ( ) is expected to remain constant throughout decline period. Overall decline rate ( ) is ) and

calculated from difference between total gas flow rates at end of plateau ( abandonment ( (

) over difference between cumulative gas production at end of plateau ).

) and abandonment (

Equation 4-144

Estimation of Production Time ( ) Production times ( ) at different reservoir pressures ( ) are calculated from Equation 4-145. The calculation is based on the exponential decline assumption and constant decline rate

105 ( ) discussed earlier. Abandonment time ( ) is calculated from Equation 4-145 at for the ).

condition at which flow rate equals abandonment total gas flow rate (

Equation 4-145

106 4.5.3 Annual Production Calculation Annual hydrocarbon production is the principal input for economic evaluation. Annual production volumes during plateau period can be straightforwardly calculated because rates are constant for the case of gas production. For the decline period, however, calculating annual production is more involved because flow rates are time dependent. Iterative procedures can be used to find exact values of annual production during decline period; however, because of the exponential decline assumption, produced volumes can be determined directly, as outlined below for annual gas and oil production calculations. Annual production time ( ) is defined as Equation 4-146. It converts production time ( ) from day unit to year unit. Annual production time ( ) begin from zero at initial condition.

Equation 4-146

Annual Gas Production Annual gas production at year gas production at year ( ( ) is calculated from difference between cumulative ( ).

) and cumulative gas production at year

Equation 4-147

107 For plateau period ( ), cumulative gas production at year ( ) is calculated

from Equation 4-148. For decline period ( year ( ) is calculated from Equation 4-149.

), cumulative gas production at

Equation 4-148

Equation 4-149

Annual Oil Production Annual oil production at year cumulative oil production at year Cumulative oil production ( gas production ( ). ( ( ) is calculated from the difference between ( ).

) and cumulative oil production at year

) is calculated using linear interpolation with respect to cumulative

Equation 4-150

Instantaneous and Annual Average Flow Rates Instantaneous flow rates of gas ( ) and oil ( ) are calculated using linear

interpolation with respect to cumulative gas production ( ( ) and oil (

). Annual average flow rates of gas ) and the annual oil

) are calculated from the annual gas production (

108 production ( ) divided by 365.25, respectively. In general, annual average rates can be used as

close approximation of instantaneous rates. They can also be converted to annual production volumes easily. Because of this, annual average rates are more meaningful to report than instantaneous flow rates for the purpose of economic analysis.

109 4.6 Economic Analysis and Field Optimization Economic analysis is the analytical method that quantifies economic performance or monetary value of a field investment project and provides a meaningful metric for the optimization of field operations. Economic model which used in this study is based on typical cash flow before tax regime (Mian, 2002). An economic evaluation subroutine has been developed to perform the economic analysis in this study. It consists of three main parts: compilation of production data, calculation of net present value (NPV), and calculation of rate of return (ROR). The economic subroutine has been constructed independently of the subroutine for field performance prediction. In this way, economic analysis can be performed either with or without calculating new field performance data. This is especially useful when performing parametric studies on economic parameters rather than field operational constraints. An optimization subroutine was also developed to find the recommended target recovery factor at end of plateau to be contracted and total recommended number of wells to be drilled for optimal field development. The optimization module requires the use of a variety of values of target recovery factor and total number of wells that need to be economically screened. The subroutine would repeatedly call the field performance prediction subroutine for a number of different values of target plateau recovery factor and total number of wells. Economic analysis is then performed for each development scenario and calculated NPVs are stored into an optimization table or matrix. Production profiles corresponding to each of the investigated target recovery factor and total numbers of well combinations are stored in the same worksheet. A separate subroutine is available for the formulation of economic analysis decoupled from field performance calculations. This subroutine would recall the stored production profiles which were generated earlier during the economic evaluation discussed above. This subroutine is

110 useful because the generation of production profiles is the step in the analysis that requiring the most significant computational time, and its availability allows to re-evaluate the sensitivity of field optimization to different values of economic parameter(s). The subroutine is especially useful when there are no changes in field operational constraints.

4.6.1 Simplified Economic Model Net annual productions are calculated from annual production volumes times the net hydrocarbon interest (Equation 4-151). Annual production volumes can be obtained from the annual production calculation outlined in Section 4.5.3. The net hydrocarbon fraction or interest is the fraction of the hydrocarbon production which is earned by the operator or investor. The remaining portion of hydrocarbon belongs to the owner of land or the mineral rights.

Equation 4-151

Annual revenues are calculated by multiplying the annual production volume with the estimated price of the product (Equation 4-152). The estimated price of gas has to be provided in the units of $ per MSCF while the price of oil or condensate has to be provided in the unit of $ per STB. Generally, gas price is quoted in the unit of $ per MMBTU, which can be converted to $ per MSCF by multiplying $/MMBTU times the gas heating value {BTU/SCF} and dividing through by 1000. Total annual revenue becomes the sum of annual revenue from gas and oil (Equation 4-153).

111

Equation 4-152

Equation 4-153

Capital expenditure (CAPEX) is a group of one-time costs which occur in order to make production possible. Total CAPEX consists of fixed CAPEX such as platform costs, flowline costs, and production facilities costs, and variable CAPEX includes the variable costs of operations such as drilling and completion (D&C). D&C cost is equal to D&C cost per well times total number of wells (Equation 4-155). For this simplified economic model, CAPEX is assumed to occur at the beginning of the project.

Equation 4-154

Equation 4-155

Operating expenditure (OPEX) is a group of costs which occur periodically in order to maintain the day-to-day operation. OPEX may include maintenance cost, utilities cost, overhead cost, production cost, etc. In this work, OPEX has to be given in the unit of $ per month. Annual OPEX is calculated from monthly OPEX times 12 months (Equation 4-156).

112

Equation 4-156

Severance tax is a government or state tax which is imposed on the production of nonrenewable resources such as oil and natural gas. Ad Valorem tax is a tax which imposed at the time of transaction. In this economic model, severance taxes for oil and gas have to be given in terms of a percentage of the total oil and gas revenue, respectively. Annual severance taxes are calculated from annual revenues times the severance tax rate (Equation 4-157). Ad Valorem tax has to be given in percentage of total revenue. Annual Ad Valorem tax is calculated from annual total revenue times the Ad Valorem tax rate (Equation 4-158). Annual total tax is summation of severance taxes and Ad Valorem tax (Equation 4-159).

Equation 4-157

Equation 4-158

Equation 4-159

113 Annual expenditure is the sum of total CAPEX, annual OPEX, and annual total tax (Equation 4-160). Annual net cash flow is the difference between annual total revenue and annual expenditure (Equation 4-161). Cumulative net cash flow at year cash flow from the beginning ( ) to year . is the summation of annual net

Equation 4-160

Equation 4-161

Equation 4-162

Commodity prices and operating cost can be escalated according to the economic inflation. In this simplified economic model, gas price, oil price, and monthly OPEX are escalated independently. The escalation is applied from the start to the end of the project. Todays product prices, monthly OPEX, and their escalation rates have to be provided to the model. Future prices and monthly OPEX are calculated from Equation 4-163 and Equation 4-164. Un-escalated economic analysis, which is required for some official reports, can be evaluated by specifying all escalation rates to be equal to zero.

114

Equation 4-163

Equation 4-164

Annual net cash flow in the future is discounted to todays equivalent value using concept of time value of money. Annual discounted net cash flow at year is calculated using Equation

4-165. Annual net cash flow at year can be obtained from Equation 4-161. The effective interest rate, which is interest rate applied on an annual basis, has to be provided. Cumulative discounted net cash flow at year ( ) to year . is the summation of annual discounted net cash flow from the beginning

Equation 4-165

Equation 4-166

115 Net present value ( The rate of return ( ) is equal to cumulative discounted net cash flow at abandonment.

) is the interest rate which results in zero NPV. The NPV profile is

calculated from evaluating NPV for a variety of interest rates. In this model, the interest rates are varied from 5% to 40%. However, the range can be adjusted depending on the analyst and desired economic results.

116 4.6.2 Field Optimization In this study, target recovery factor at end of plateau and total number of wells are targets for optimization because they can be controlled by the operator of the field. Net present values ( ) at differences target recovery factor and total number of wells are evaluated and

compared. The combination of target recovery factor and total number of wells which providing the global maximum NPV is the desired result.

Optimum NPV

Net Present Value

Total Number of Wells Target Recovery Factor at End of Plateau

Figure 4-6: Graphical Representation of Field Optimization

Production profiles of oil and gas at differences target recovery factor and total number of wells are calculated based on field performance prediction concept (Section 4.5). The predicted results are then imported into the simplified economic model (Section 4.6.1). The production scenario which provides the highest NPV is selected.

Chapter 5 Model Performance

5.1 Simulation of Standard PVT Black Oil Properties A model has been developed to simulate standard PVT properties from a phase behavior protocol. Input data presented in Appendix A is set to the subroutine and standard black oil PVT properties - gas formation volume factor ( oil-gas ratio ( ), and solution gas-oil ratio ( ), oil formation volume factor ( ), volatilized

) - are calculated based on fluid compositional

data. Section 5.1.1 shows calculation results from the procedure described in Section 4.2.3. Section 5.1.2 discusses the limitations inherent to representing a multi-component hydrocarbon mixture as a binary pseudo-component fluid model. Section 5.1.3 elaborates about the impact of such limitations on the values and behavior of standard PVT properties.

5.1.1 Simulated Standard PVT Properties The standard PVT properties calculated from data set in Appendix A are shown in Figure 5-1 and Figure 5-2. The values of these properties depend on pressure (reservoir and surface), temperature (reservoir and surface), and original composition of in-situ reservoir fluid. The trends depicted in Figure 5-1 and Figure 5-2 are consistent with the typical phase behavior of gas condensate reservoir fluids discussed in section 4.2.1 and the standard PVT properties for the Anschutz Ranch East rich-gas condensate presented by Walsh and Lake (2003).

118
10
Bg - Gas Formation Volume Factor (RB/MSCF)

250 Rv
Rv - Volatilized Oil-Gas Ratio (STB/MMSCF)

200

150

100

Bg Dew Point Pressure 0 1000 2000 3000 4000

50

0 5000

Reservoir Pressure (psia) Bg (RB/MSCF) Rv (STB/MMSCF)

Figure 5-1: Simulated Gas Formation Volume Factor and Volatilized Oil-Gas Ratio of Gas Condensate

For this reservoir fluid, gas formation volume factor (

) equals 1.040 RB/MSCF at the

initial reservoir pressure of 4000 psia. As shown in Figure 5-1, its value increases with decreasing reservoir pressure with an increased slope at lower reservoir pressures. At final abandonment pressure of 500 psia, gas formation volume factor ( Volatilized oil-gas ratio ( ) reached the value of 7.816 RB/MSCF.

) remains initially constant at the value of 205.9 STB/MMSCF for

reservoir pressures higher than the fluids dew point. Its value starts decreasing as soon as the pressure decreases below dew point conditions. The decreasing slope of volatilized oil-gas ratio ( ) is largest at conditions around dew point conditions and lower as reservoir pressure ) starts increasing

decreases. As the pressure continues to decrease, volatilized oil-gas ratio ( again because of re-vaporization of oil. Volatilized oil-gas ratio (

) reaches a minimum value of

63.5 STB/MMSCF around 1100 psia before increasing up to 80.8 STB/MMSCF at the final pressure of 500 psia.

119
7
Bo - Oil Formation Volume Factor (RB/STB)

7000
Rs - Solution Gas-Oil Ratio (SCF/STB)

6 5 4 3 2 Bo Rs Dew Point Pressure 0 1000 2000 3000 4000

6000 5000 4000 3000 2000

1
0

1000
0 5000

Reservoir Pressure (psia) Bo (RB/STB) Rs (SCF/STB)

Figure 5-2: Simulated Oil Formation Volume Factor and Solution Gas-Oil Ratio of Gas Condensate

At reservoir pressure higher than dew point pressure, oil formation volume factor ( and solution gas-oil ratio (

) are not actually defined because there is no free liquid phase in the ) and solution gas-oil ratio ( ) values above

reservoir. However, oil formation volume factor (

the dew point can be calculated from Equation 4-61 and Equation 4-62. Those relationships are defined force the Material Balance Equation (MBE) for gas condensate to collapse to the MBE for wet gas at pressures above the dew point. As shown in Figure 5-2, oil formation volume factor ( ) starts at 2.858 RB/STB around dew point conditions and monotonically decreases to a value ) begins at

of 1.206 RB/STB at the final pressure of 500 psia. Similarly, solution gas-oil ratio (

2067 SCF/STB around dew point conditions and then monotonically decreases to a value of 61.5 SCF/STB at the final pressure.

120
1.4
SG - Specific Gravity of Reservoir Gas

1.2 1.0 0.8 0.6 0.4

SG

0.2
0.0 0 500 1000 1500 2000 2500 Reservoir Pressure (psia) Specific Gravity of Reservoir Gas Dew Point Pressure 3000 3500 4000 4500

Figure 5-3: Simulated Specific Gravity of Reservoir Gas

In addition, specific gravity (

) of reservoir gas is also estimated. The results of this ) of reservoir gas is

estimation are shown in Figure 5-3. Above the dew point, specify gravity (

constant and equal to 1.147. After that, its value decreases with decreasing reservoir pressure to the minimum value of 0.876 at 1150 psia and then slightly increases back to 0.919 at the final pressure of 500 psia due to condensate re-vaporization.

121 5.1.2 Limitations of Pseudo Component Model The calculated volumes of surface gas and stock-tank oil pseudo components in reservoir gas, reservoir oil, and cumulative gas production below dew point pressure are displayed in Figure 5-4 and Figure 5-5, as they were estimated during the simulation of a CVD experiment using a phase behavior model. They are calculated from 1000 MSCF of equivalent gas at dew point pressure.

1000
G - Surface Gas Pseudo Component (MSCF)

900 800 700 600 500 400 300 200

861 735

Gfg+Gfo+Gp

847

Gfg

Gp

100
0 0

121 5 500 1000 1500

Gfo

Dew Point Pressure 2500 3000 3500

2000

Reservoir Pressure (psia)

Gfg (MSCF)

Gfo (MSCF)

Gp (MSCF)

Gfg +Gfo + Gp (MSCF)

Figure 5-4: Volumes of Surface Gas Pseudo Component in Reservoir Gas Reservoir Oil, and Cumulative Gas Production

As shown in Figure 5-4, the amount of surface gas pseudo component remaining in the reservoir gas phase ( ) decreases from a value of 847 MSCF at the dew point to a value of 121

MSCF at the final pressure because of continuous gas production and retrograde condensation of reservoir gas. Amounts of surface gas pseudo component in the reservoir oil phase ( ) starts at

122 zero at dew point pressure because there is no reservoir oil. Then, it increases to the maximum volume of 92 MSCF at 2700 psia before decreasing to 5 MSCF at 500 psia. This reversing trend is dominated by the changing amounts of reservoir oil volume or oil saturation during depletion. Cumulative gas production ( ), which is the volume of surface gas pseudo component

recovered from the production of excess gas out of the PVT cell, increases from zero at dew point pressure to 735 MSCF at the final pressure.

200
N - Stock-Tank Oil Pseudo Component (STB)

180 160 140 120 100 80 60 40 88 67 Nfg Nfo 165 Nfg+Nfo+Np

174

20
0 0 10 500 1000 1500

Np 2000 2500

Dew Point Pressure 3000 3500

Reservoir Pressure (psia) Nfg (STB) Nfo (STB) Np (STB) Nfg +Nfo + Np (STB)

Figure 5-5: Volumes of Stock-Tank Oil Pseudo Component in Reservoir Gas, Reservoir Oil, and Cumulative Oil Production

Figure 5-5 shows that the amount of stock-tank oil pseudo-component in the reservoir gas phase ( ) decreases from a value of 174 STB at dew point pressure to a value of 10 STB at the

final pressure point due to gas production and retrograde condensation of reservoir gas. The amount of stock-tank oil pseudo-component in reservoir oil ( ) increases from zero at dew

point pressure because of the lack of a reservoir oil phase and reaches a maximum volume of 96

123 STB at 1450 psia before decreasing to 88 STB at the final pressure. This reversing trend mainly stems from changes in reservoir oil saturation in the reservoir during depletion. Cumulative oil production ( ), which is the volume of stock-tank oil pseudo-component recovered from the

excess gas produced from the PVT cell, increases from zero at dew point pressure to 67 STB at the final pressure. Let us now consider how the basic statements of species conservation are being honored during the simulated depletion. The sum of the amounts of surface gas pseudo-component in reservoir gas, in reservoir oil, and cumulative gas production ( ) in Figure 5-4

should be always equal to the total amount of surface gas pseudo component originally present in reservoir gas at dew point conditions if the surface gas pseudo-component was to be fully conserved throughout the simulated experiment. However, Figure 5-4 clearly shows that the total amount of surface gas pseudo component ( ) is not constant throughout the

simulated process; instead, it increases from 847 MSCF at the dew point to 861 MSCF at the final pressure. This suggests that total mass of surface gas pseudo component actually increases with decreasing reservoir pressure which is physically impossible as is in violation of mass conservation. Similarly, the fact that the total amount of stock-tank oil pseudo component ( ) decreases from 174 STB at dew point conditions to 165 STB at the final

pressure, suggests that total mass of stock-tank oil pseudo-component actually decreases with decreasing reservoir pressure. These unphysical trends are caused by one of the key assumptions used in pseudo component model, which establishes that the properties of the surface gas and stock-tank oil pseudo-components are supposed equal and unchanging during depletion. In reality, the properties of pseudo components, which are actually two multi-component mixtures in their own right, do change throughout reservoir depletion. Compositions of reservoir gas and reservoir oil change continuously because of retrograde condensation. In addition, there are three points of surface separation; first stage separator, second stage separator, and stock tank.

124 Figure 5-6 displays the estimated densities of surface gas and stock-tank oil pseudo components, as calculated by the Phase Behavior Model. They clearly suggest that fluid properties of these pseudo-components are not necessarily constant, although they do not change dramatically during depletion.

3.0 Nfo at STO 2.4 Gfg at Sep1 1.8 Gfo at Sep1 1.2 Gfg at Sep2 0.6 Gfg at STO 0 500 Gfo at STO 1000 1500 2000 2500 3000 Gfo at Sep2 Nfg at STO

50

40

30

20

10

0.0

0 3500

Reservoir Pressure (psia)

Gfg at Sep 1 (lbm/ft3) Gfo at Sep 2 (lbm/ft3) Gfg at Sep 2 (lbm/ft3) Gfo at STO (lbm/ft3) Gfg at STO (lbm/ft3) Nfg at STO (lbm/ft3) Gfo at Sep 1 (lbm/ft3) Nfo at STO (lbm/ft3)

Figure 5-6: Densities of Surface Gas and Stock-Tank Oil Pseudo Components at First Stage Separator, Second Stage Separator and Stock Tank Condition

Figure 5-7 shows the amount of stock-tank oil pseudo-component in the reservoir gas phase ( ), reservoir oil phase ( ) and cumulative production ( ) in term of gas-equivalent

volume. The conversion from oil volume to equivalent-gas volume is based on molar equivalency. Gas-oil equivalency factor ( ) is calculated from Equation 5-1. Densities and

molecular weights of stock-tank oil from reservoir gas, reservoir oil, and cumulative oil production are directly obtained from Phase Behavior Model.

Density of Stock-Tank Oil (lbm/ft3)

Density of Surface Gas (lbm/ft3)

125

Equation 5-1

180 160
GE - Surface Gas Equivalent (MSCF)

Nfg+Nfo+Np 139

153

140 120 100 80 60

Nfg 67 62 Ngo Dew Point Pressure 3000 3500

40
20

10
0 0 500 1000 1500 2000 Reservoir Pressure (psia)

Np 2500

Nfg

Nfo

Np

Nfg+Nfo+Np

Figure 5-7: Volumes of Stock-Tank Oil Pseudo Component in Reservoir Gas, Reservoir Oil, and Cumulative Oil Production in term of Gas-Equivalent

Figure 5-8 shows the volume of surface gas ( (

), volume of stock-tank oil

) expressed in term of gas-equivalent volume, and total volume from both

surface gas and stock-tank oil volumes. The total volumes are constant and equal to 1000 MSCF which is the original volume at dew point conditions. These results clearly prove that pseudo component model is able to honor overall material balance, but not species material balance.

126
1000 900
GE - Surface Gas Equivalent (MSCF)

TOTAL 861 Gfg+Gfo+Gp

800 700 600 500 400 300 200

847

Nfg+Nfo+Gp 139 153 1000 1500 2000 2500

100
0 0

Dew Point Pressure 3500

500

3000

Reservoir Pressure (psia)

Nfg+Nfo+Np

Gfg+Gfo+Gp

TOTAL

Figure 5-8: Total Volumes of Stock-Tank Oil Pseudo Component and Surface Gas Pseudo Component in term of Gas-Equivalent

127 5.1.3 Impact on Standard PVT Properties The limitations of representing a multi-component hydrocarbon mixture using a binary pseudo-component model, as discussed in the preceding section, would definitely have an effect on the calculation of the standard PVT black-oil properties. As discussed below, solution gasoil ratios ( ( ) could become negative at low reservoir pressures; or oil formation volume factors

) could become over-estimated. The Walsh-Towler algorithm applies concept of mass balance around the PVT cell by

calculating amounts of pseudo components in the reservoir oil at any pressure level j (

and

) from the differences between total pseudo-component amounts from the previous pressure level ( and ) and the summation of pseudoand

component amounts in the reservoir gas phase and cumulative production ( ).

The combination of this mass balance concept and simulated results from a phase behavior model can lead to negative values of surface gas pseudo component in oil phase ( and over-estimated values of stock-tank pseudo component in the oil phase ( )

) when reservoir

depletion is extensive (i.e., at low pressures). For example, at final abandonment pressure of 500 psia and using the data of Figure 5-4, the value of surface gas pseudo component in oil phase ( ), which is equal to 5 MSCF based on a rigorous flash calculation, would be equal to -9

MSCF (847 735 121 = -9) when this mass balance concept is applied. In Figure 5-5, the value of stock-tank oil pseudo component ( ), which is equal to 88 STB from a flash calculation,

would be equal to 97 STB (174 67 10 = 97) using the same mass balance concept. Thus, the resulting solution gas-oil ratio ( ) calculated from Walsh-Towler algorithm at this condition

would become equal to -96 SCF/STB, while 61.5 SCF/STB is the simulated value from flash

128 calculations as shown in Figure 5-2. Similarly, oil formation volume factor ( ) calculated from

Walsh-Towler algorithm would be equal to 1.082 RB/STB, while 1.206 RB/STB is the simulated value from flash calculations as shown in Figure 5-2.

129 5.2 Zero-Dimensional Material Balance Calculations for Gas Condensates A reservoir model has been developed in order to predict cumulative gas production, cumulative oil production, production gas-oil ratio, and average reservoir gas saturation based on the procedure described in Material Balance Equation for Gas Condensates and Saturation Calculation sections discussed above. Input data in Appendix A is used by the model, which generates its outputs as function of reservoir pressure. Section 5.2.1 discusses the simulation results and section 5.2.2 discusses the significant pitfalls of misusing the proposed gas condensate fluid tank model for performance prediction for other near-critical fluid: the volatile oil reservoir.

5.2.1 Simulation Results from Gas Condensate Tank Model Figure 5-9 plots the results of gas condensate tank model for the reservoir scenario described in Appendix A. Initial reservoir pressure is 4000 psia, dew point pressure is 3031 psia, and abandonment pressure is 750 psia. In Figure 5-9, cumulative gas recovery ( cumulative oil recovery ( ) and

) increase with decreasing reservoir pressure. The recovery slopes

are identical for conditions above dew point pressure; however, the increasing trend or slope of cumulative oil recovery becomes significantly flatter than that of gas for pressures below the dew point. At abandonment pressure conditions, in this example, cumulative gas and oil recoveries are equal to 80.6% and 45.7%, respectively. Gas saturation ( ) remains constant at 79% ( ) above the dew point. As reservoir pressure decreases below the dew point, gas saturation ( ) decreases to the minimum value of 66.6% at 2350 psia, and then slightly increases back to 69.5% at the abandonment pressure. Production gas-oil ratio ( volatilized oil-gas ratio ( ), which is the inverse of

) for a gas condensate system whose reservoir condensate remains

immobile, is constant at 4.857 MSCF/STB above the dew point. As reservoir pressure decreases

130 below dew point conditions, production gas-oil ratio ( ) increases to the maximum value of

15.756 MSCF/STB at 1050 psia, and then slightly decreases back to 14.863 MSCF/STB at the abandonment pressure.

OOIP Recovery (%) / OGIP Recovery (%) Production GOR (MSCF/STB) / Gas Saturation (%)

100 90 80 70 60 50 40 30 20 GOR Dew Point Pressure Np/N Sg Gp/G

10
0 0 1000 2000 3000 4000 5000 Reservoir Pressure (psia)

Gp/G (%)

Np/N (%)

GOR (MSCF/STB)

Sg (%)

Figure 5-9: Simulated Production Results of Gas Condensate using Simplified Gas Condensate Tank Model

Qualitatively, the production trends calculated from the gas condensate zero-dimensional model fully agree with the typical and expected trends observed in the field and through fullydimensional numerical reservoir simulation (Walsh and Lake, 2003). A standard numerical simulator also suggests, and field experience corroborates, that ultimate gas recovery - i.e., cumulative gas recovery at abandonment conditions - is much higher than ultimate oil recovery. The reason is that, after dew point conditions are reached, stock-tank oil is continuously being left behind as immobile condensate trapped in the reservoir. Gas saturation ( ) remains constant

131 above the dew point because there is no oil phase in the reservoir and water saturation ( ) is

assumed to be constant when no water encroachment is acting on the reservoir system. Gas saturation ( ) decreases as reservoir pressure decreases around dew point conditions because of the new presence of reservoir oil phase in the reservoir. Condensate dropout rate is maximum at conditions near the dew point. When gas condensate is rich enough and pressure is low enough, gas saturation can slightly increase with reservoir pressure. Gas phase production, oil phase expansion, and retrograde condensation near to dew point conditions tend to decrease gas saturation ( ). Expansion of the reservoir gas and re-vaporization of condensate at low pressures tend to increase gas saturation ( ). Production gas-oil ratio ( ) is constant above the dew

point because composition of produced reservoir gas remains unchanged for such conditions. Below the dew point, production gas-oil ratio ( ) increases with decreasing reservoir pressure

because of retrograde condensation and the changing nature of the produced reservoir gas. Production gas-oil ratio ( vaporization. ) can slightly decreases at low reservoir pressure due to liquid re-

132 5.2.2 Misuse of Gas Condensate Tank Model in Volatile Oil Reservoir Simulation results from the Simplified Gas Condensate Tank (SGCT) model must be crosschecked against the typical phase and depletion behavior of gas condensate reservoirs outlined in the preceding section. If the reservoir fluid is not a gas condensate, prediction results will be inconsistent with those typical PVT property and depletion behavior. For example, if the SGCT model is inadvertently used for the analysis of a different type of near-critical fluid, such as a volatile oil reservoir, calculated gas saturation ( ) profiles will start from zero at saturation conditions (which would actually represent a bubble point) and monotonically increase with decreasing reservoir pressure below saturation conditions. Such gas saturation profile is significantly different from the typical -profile for gas condensates which must starts at one if ) and decreases with decreasing

no water is present (or at initial gas saturation equal to

reservoir pressure at conditions below the dew point. In this case, all simulation results must be disregarded because the assumptions used for the gas condensate tank model (SGCT) are not applicable for volatile oil reservoir. To illustrate the differences between the phase and flow behavior between gas condensates and volatile oils, reservoir temperature can be manipulated. A volatile oil reservoir behavior can be obtained using the same fluid characterization and composition presented in Appendix A, but with a reservoir temperature reduced from 300 F to 190 F. Figure 5-10 plots the phase envelope of reservoir fluid and the two different reservoir depletion paths for the two reservoir temperatures under consideration. The depletion path at the reservoir temperature of 300 F represents the path of the gas condensate ( ), while the depletion path at the reservoir ). Both fluids are near critical

temperature of 190 F represents the path of the volatile oil (

fluids but they are found at the opposite sides of the critical point. Simulation results for the gas condensate were discussed in Figure 5-1 through Figure 5-9 in the preceding section.

133
4000 3500 3000 2500 2000 1500 1000 500 0 -300 -200 -100 0 100 200 300 400 500 Reservoir Temperature (F) Critical Point

Reservoir Pressure (psia)

Reservoir Depletion Path at 190 F "Volatile Oil"

Reservoir Depletion Path at 300 F "Gas Condensate"

Figure 5-10: Phase Envelope and Reservoir Depletion Paths at Two Different Reservoir Temperatures

The resulting standard PVT properties of the volatile oil, which are calculated using the PVT model for a gas condensate, are shown in Figure 5-11 and Figure 5-12. The characteristics of gas formation volume factor ( ( ), and solution gas-oil ratio ( ), oil formation volume factor ( ), volatilized oil-gas ratio

) are similar to the typical characteristics of gas condensates.

However, these results cannot be used and are physically meaningless for an actual volatile oil because their calculation has been based on assuming that the reservoir oil, which is the main hydrocarbon phase produced in a volatile oil reservoir, remains immobile in the gas condensate PVT cell.

134
10.0
Bg - Gas Formation Volume Factor (RB/MSCF)

250
Rv - Volatilized Oil-Gas Ratio (STB/MMSCF) Rs - Solution Gas-Oil Ratio (SCF/STB)

Rv 8.0 200

6.0

150

4.0

100

2.0

Bg Saturated Pressure 0 1000 2000 3000 4000

50

0.0 Reservoir Pressure (psia) Bg (RB/MSCF)

5000

Rv (STB/MMSCF)

Figure 5-11: Simulated Gas Formation Volume Factor and Volatilized Oil-Gas Ratio of Volatile Oil using Gas Condensate PVT Model

5.0
Bo - Oil Formation Volume Factor (RB/STB)

7000 Bo Rs 6000 5000 4000 3000 2000

4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0

0.5
0.0 0 1000 2000 3000 Reservoir Pressure (psia) Bo (RB/STB) Rs (SCF/STB) Saturated Pressure 4000

1000
0

5000

Figure 5-12: Simulated Oil Formation Volume Factor and Solution Gas-Oil Ratio of Volatile Oil using Gas Condensate PVT Model

135 Production data predictions for the volatile oil, as simulated by the SGCT model using the PVT properties in Figure 5-11 and Figure 5-12, are shown in Figure 5-13. Characteristics of cumulative gas recovery ( ( ), cumulative oil recovery ( ), and production gas-oil ratio

) are similar to those simulated for gas condensate but the gas saturation ( ) trend is

significantly different, as discussed above. The typical gas saturation ( ) of condensate should start at initial gas saturation ( ) at conditions above saturation (dew point) conditions

because there should be no liquid hydrocarbon in that state. However, gas saturation ( ) plotted in Figure 5-13 approaches zero around the saturation pressure, which indicates that oil saturation ( ) does not approach zero but rather approaches a maximum value ( all simulation results must be disregarded. ). Therefore,

OOIP Recovery (%) / OGIP Recovery (%) Production GOR (MSCF/STB) / Gas Saturation (%)

100 90 80 70 60 50 40 30 20 Sg GOR Gp/G

Np/N
Saturation Pressure

10
0 0 1000 2000 3000

4000

5000

Reservoir Pressure (psia)

Gp/G (%)

Np/N (%)

GOR (MSCF/STB)

Sg (%)

Figure 5-13: Simulated Production Results of Volatile Oil using Simplified Gas Condensate Tank Model

136 In short, models developed for gas condensate fluids should not be recklessly used without proper precautions and crosschecks. A crosschecking process has to be carried out either before or after having generated the simulation results. Before running the simulation, for example, the calculated fractions of vapor and liquid phases inside the PVT cell and along the reservoir depletion path should be analyzed. Molar fraction of vapor phase ( ) for a gas

condensate must approach 1.0 as the saturation pressure is approached because there should be no liquid at a dew point line, and molar fraction of vapor phase ( ) for a volatile oil must approach

zero as the saturation pressure is approached because there should be no vapor at a bubble point line. This situation is illustrated in Figure 5-14. After running the simulation, the resulting gas saturation ( ), for example, should be crosschecked as well.

1.0 0.9
Fng - Malar Fraction of Vapor Phase

Gas Condensate

0.8 0.7 0.6 0.5 0.4 0.3 0.2 Volatile Oil

0.1
0.0 0 500 1000 1500 2000 2500 3000 3500 4000 Reservoir Pressure (psia)

Fng of Gas Condensate

Fng of Volatile Oil

Figure 5-14: Mole Fraction Behavior of Vapor Phase Molar Fraction ( for Gas Condensates and Volatile Oils

137 5.3 Field Performance Prediction Field performance prediction calculations were described in Section 4.5 and have been developed in a stand-alone VBA application. The model is able to predict pressures, flow rates, cumulative productions, and expected production time by integrating a zero-dimensional reservoir model, a flow rate and pressure (IPR/TBR) model, and production constraints together. The input data set used by the model is presented in Appendix A and the generated results are discussed in Section 5.3.1. In addition, the appearance of a slightly negative hyperbolic decline coefficient during the decline period, which is very rare to encounter in conventional decline curve analysis, is observed from this simulation results. This is discussed in Section 5.3.2. Note that, in this study, field performance predictions begin at a plateau period because built-up has been neglected by assuming that the operator has a substantial drilling capacity available to develop the field.

5.3.1 Field Performance Prediction Results Predictions for cumulative gas production ( ) and cumulative oil ( ) production vs. ) linearly

production time ( ) are displayed in Figure 5-15. Cumulative gas production ( increases during plateau period because plateau gas flow rate (

) is being maintained.

Plateau period ends when the 55% target recovery factor is reached. Cumulative gas production at end of plateau ( condition ( ) is equal to 216 BSCF. Cumulative gas production at abandonment ) is equal to 317 BSCF, which is equivalent 81% gas recovery factor. ) also linearly increases at the beginning because of constant ) during plateau period and constant volatilized oil-gas ration ( )

Cumulative oil production ( plateau gas flow rate (

above the dew point. When reservoir pressure goes below the dew point, cumulative oil

138 production ( ) increases at lower rate than cumulative gas ( ) production due to a decreasing ) is

volatilized oil-gas ratio (

). Cumulative oil production at abandonment condition (

equal to 37 MMSTB, which is equivalent to a 46% condensate recovery factor. These results indicate, as expected, that total recovery factor of oil/condensate is significantly less than total recovery factor of gas in typical depletion operations for gas condensate fluids. This is expected because a large portion of original oil in place is being left as an immobile condensate phase inside the reservoir.

400
Gp - Cumulative Gas Production (BSCF)

350 300 250 200 150 100 50 0 0

Gp

70 60 50

Np

40 30 20 10

Dew Point 1 2

End of Plateau 3 4 5 6 7

Production Time (Year) Gp (BSCF) Np (MMSTB)

Figure 5-15: Cumulative Gas and Oil Production vs. Time

Figure 5-16 plots total gas flow rate ( time ( ). Total gas flow rate (

) and total oil flow rate (

) vs. production ) of 213

) is maintained at the plateau gas flow rate (

MMSCF/D during the plateau period. During the decline period, reservoir pressure is not enough to maintain the plateau gas flow rate ( ), thus total gas flow rate ( ) declines with

Np - Cumulative Oil Production (MMSTB)

G = 393 BSCF N = 81 MMSTB Rvi = 206 STB/MMSCF

80

139 decreasing reservoir pressure. At abandonment condition, total gas flow rate ( MMSCF/D. Total oil flow rate ( flow rate ( ) is equal to 19

) is constant at 44 MSTB/D above the dew point. Total oil ) below the dew point

) declines because of decreasing volatilize oil-gas ratio (

and because of declining total gas flow rate ( when the volatized oil-gas ratio (

) during decline period. At low pressures, even ) continues to

) slightly increases, total oil flow rate (

decline. This is because increasing volatilized oil-gas ratio ( decreasing total gas flow rate ( rate ( ) equals 1.3 MSTB/D.

) is not enough to compensate for

) at that condition. At abandonment conditions, total oil flow

250
qgsc - Total Gas Flow Rate (MMSCF/D)

50 qgsc
qosc - Total Oil Flow Rate (MSTB/D)

200

40

150

30

100

qosc

20

50 Dew Point 0 1 2 End of Plateau 3 4 5 6 7

10

Production Time (Year) q_gsc (MMSCF/D) q_osc (MSTB/D)

Figure 5-16: Total Gas and Oil Flow Rates vs. Time

Figure 5-17 plots reservoir pressure ( ), bottomhole flowing pressure ( wellhead pressure (

), and

) vs. production time ( ). During plateau period, production time ( ) is ) variable because total gas flow rate ( )

directly proportion of cumulative gas recovery (

140 and original gas in place ( ) are constant. Thus, the relationship between reservoir pressure ( ) and production time ( ) during plateau period is similar to the relationship between reservoir pressure ( ) and cumulative gas recovery ( ) calculated from the gas condensate tank

model. Reservoir pressure ( ) decreases from 4000 psia at initial condition to 3031 psia at the dew point. After that, reservoir pressure ( ) decreases at a slower rate as a result of the implementation of a two-phase mode of operation in the gas condensate tank model. This behavior could be observed in Figure 5-9 as well. In addition, reservoir pressure ( ) during the decline period decreases at an even slower pace thanks to a declining gas flow rate ( Reservoir pressure ( ) reaches abandonment pressure of 750 psia after 6.23 year of production. ).

4500 4000

3500
Pressure (psia)
3000 2500 2000 1500 1000 500 0 0 Dew Point 1 2 pwf pr

pwh
End of Plateau 3 4 5 6 7

Production Time (Year) Pr (psia) Pwf (psia) Pwh (psia)

Figure 5-17: Reservoir Pressure, Bottomhole Flowing Pressure, and Wellhead Pressure vs. Time

141 Figure 5-17 reveals that bottomhole flowing pressure ( ) decreases from 3440 psia at )

initial conditions to 668 psia at abandonment conditions. Bottomhole flowing pressure ( depends on reservoir pressure ( ) and drawdown pressure (

), which is the pressure drop ) decreases from ) of 550 psia at end

required to produce hydrocarbons out of the reservoir. Wellhead pressure ( 2438 psia at initial condition to minimum allowable wellhead pressure ( of plateau. After that, wellhead pressure ( Wellhead pressure (

) is maintained at constant level of 550 psia. ) and pressure drop inside

) varies with bottomhole flowing pressure (

the tubing, which is a pressure drop that must be maintained in the tubing in order to produce hydrocarbon out of the wellbore. Drawdown pressure ( bottomhole flowing pressure ( ), which is the difference between reservoir pressure ( ) and ), is relatively constant above the dew point because most of )

the variables in IPR equation (Equation 4-134) are relatively constant. Drawdown pressure (

found within the range 545 to 560 psia during this period. Below the dew point, drawdown pressure ( ) increases with decreasing reservoir pressure ( ) mainly because of decreasing ). At reservoir pressures close to the dew point, relative

relative permeability of gas ( permeability of gas (

) decreases significantly due to decreasing gas saturation ( ), while the ) times gas formation volume factor ( ) decreases slightly. As a

product of gas viscosity (

result, drawdown pressure ( (

) has to increase in order to maintain plateau gas flow rate

) so that it can compensate for the decreased mobility of the gas phase. As reservoir ) is relatively constant because ) )

pressure continues to decrease, relative permeability of gas (

gas saturation ( ) is relatively stable (see Figure 5-18), while the product of gas viscosity ( times gas formation volume factor ( ) slightly increases. As a result, drawdown pressure (

has to continue to increase in order to maintain the plateau gas flow rate (

). During

142 decline period, drawdown pressure ( ( ). ) is lower mainly because of declining total gas flow rate

1.00

1.5

0.80
Gas Saturation (Fraction)

Gas Saturation

1.3

0.60 Specific Gravity of Reservoir Gas

1.1

0.40

0.9

0.20 End of Plateau 3 4 5 6 7

0.7

0.00 0

Dew Point 1 2

0.5

Production Time (Year) Gas Saturation SG of Reservoir Gas

Figure 5-18: Gas Saturation and Specific Gravity of Reservoir Gas vs. Time

Pressure drop inside the tubing is the difference between bottomhole flowing pressure ( ) and wellhead pressure ( ). Pressure drop inside the tubing is relatively constant above

the dew point because most of the variables in TPR equation (Equation 4-139) are relatively constant. This pressure drop ranges between the values of 870 to 1000 psi during this above-dewpoint period. During decline period, pressure drop inside the tubing continues to decrease owing to decreasing total gas flow rate ( ). At abandonment condition, pressure drop inside the

tubing reaches the minimum value of 118 psi.

Specific Gravity of Reservoir Gas

143 5.3.2 Decline Trend Analysis Total gas flow rate ( ) and cumulative gas production during decline period (

) is calculated by combining gas condensate tank model and nodal analysis as described in Section 4.5. These results, for the scenario of interest, are plotted in Figure 5-19. The exponential decline trend which calculated from exponential decline equation is also plotted into the same figure. Total gas flow rate ( Equation 5-2 and the decline rate ( 4-144. ) from exponential decline trend is calculated from

) of exponential decline trend is calculate from Equation

Equation 5-2

250

qgsc - Total Gas Flow Rate (MMSCF/D)

200

150

Exponential Decline Trend Di = 1.92 * 10-3 {Day-1}

100 Field Performance Prediction Data 50

0 0 20 40 60 80 100 120 (Gp - Gpplateau) - Cumulative Gas Production during Decline Period (BSCF)

Field Performance Preidction Data

Exponentail Decline Trend (b = 0.0)

Figure 5-19: Total Gas Flow Rate ( ) vs. Cumulative Gas Production during Decline Period

144 From Figure 5-19, calculated results from field performance prediction and exponential decline equation agree with each other, which is not surprising. The assumptions used in this simulator, including pseudo steady state flow condition, no water production, no water injection, no water influx, and constant wellhead pressure during decline period, are favorable assumptions for an exponential decline. Therefore, the exponential decline is a fairly good assumption for calculating production time ( ) based on total gas flow rate ( ( ). However, detailed analysis of the decline trend shows decline rate ( ) actually varies although very slightly. If exponential decline is assumed between each decline interval, decline rate ( ) of each interval can be calculated from Equation 5-3. The calculation results which are plotted in Figure 5-20 show that decline rate ( ) increases with increasing cumulative gas production ( ) for most of the time. In other words, they suggest that decline rates are slightly ) and cumulative gas production

increasing in time, which in turn implies having a hyperbolic decline exponent ( ) of negative value. In conventional decline curve analysis, decline rates ( ) are always expected to dampen in time and thus hyperbolic decline coefficients ( ) are always expected to be positive ( ). A negative value for the decline exponent ( ) is extremely rare scenario. It became apparent during this study that this decline behavior was coupled with the appearance/ disappearance of the condensate phase and related property changes during decline, as suggested by Figure 5-20. In this figure, decline rates slightly increase with reservoir production but they reach a maximum after which they start to decrease. This event seems to closely follow the behavior of GOR presented in Figure 5-9. It is suggested that condensate effects and the inherent evolution of fluid properties in time impose the increase in the decline rate at earlier times during the decline period but this trend is reversed around the moment GOR reaches its maximum. It is important to note that these decline rate changes are not very large or significant; therefore the exponential decline assumption still remains largely valid for engineering evaluation purposes. It

145 should also stressed that conventional decline curve analysis is based on the assumption of production at a constant bottomhole pressure ( ) which is not strictly valid for the scenario

under consideration as displayed in Figure 5-17 during decline.

Equation 5-3

2.2E-03 2.1E-03
D - Decline Rate (1/Day)

2.0E-03 Decline Rate of 1.92 * 10-3 {Day-1} 1.9E-03 1.8E-03 1.7E-03

1.6E-03 0
(Gp -

20
Gpplateau) -

40

60

80

100

120

Cumulative Gas Production during Decline Period (BSCF) Decline Rate (1/Day)

Figure 5-20: Decline Rate ( ) vs. Cumulative Gas Production during Decline Period ( )

146 5.4 Economic Analysis and Optimization An economic evaluation model has been developed to calculate net present value (NPV) and rate of return (ROR) based on the simplified economic model described in Section 4.6.1. Results from field performance prediction from the previous section and economic parameters from Appendix A are inputted into this economic model, and the generated results are shown and discussed in Section 5.4.1. For field optimization studies, results of the sensitivity analysis of NPVs for different target recovery at end of plateau and total number of wells are displayed and elaborated upon in Section 5.4.2.

5.4.1 Field Economic Analysis Figure 5-21 presents the predictions for annual expenditure, annual revenue, and cumulative discounted net cash flow vs. production time for the reservoir exploitation scenario under consideration. Annual expenditure is equal to 1480 Million $ at the beginning. It is very large because all CAPEX, including drilling and completion cost, platform cost, pipeline cost, and production facilities cost, is spent at that time. Annual expenditure drops drastically to 116 Million $ in the 1st year of production because it consists of OPEX and taxes only and they are relatively small when compared to initial CAPEX. Annual expenditure continuously decreases because ad valorem and severance taxes decrease resulting from decreasing in annual revenue. Annual expenditure at the last year of production is equal to 8.21 Million $. Annual revenue is equal to 1373 Million $ in the 1st year of production. Then, it continuously declines due to decreasing gas and oil flow rates. Annual revenue at the last year of production is equal to15.70 Million $.

147
2000 1500 Annual Revenue

Monetary Value (Million $)

1000

500
0 (500) (1000) (1500) (2000) 0 1 2 3 4 5 6 7 Production Period (Year) Annual Expenditure Cumulative Discounted Net Cash Flow

Annual Expenditure

Annual Revenue

Cumulative Discounted Net Cash Flow

Figure 5-21: Annual Expenditure, Annual Total Revenue, and Cumulative Discounted Net Cash Flow vs. Production Time

Figure 5-21 shows that the cumulative discounted net cash flow is equal to (-1480) Million $ at the beginning of the project. Cumulative discounted net cash flow increases every year because annual revenue is higher than annual expenditures throughout the production period. However, the rate of net cash increase decreases at late time because net cash flow is lower and time discount factor is higher. The cumulative discounted net cash flow at the last year of production, which is equivalent to the projects NPV, is equal to 1189 Million $. Figure 5-21 also shows that cumulative discounted net cash flow turns from negative value to positive value between the first year and the second year of production. Thus, the payback period, which is the time period required for cumulative discounted net cash flow to be equal to zero, is between one to two years.

148
2000 ROR = 53% NPV @ 12% Interest Rate = 1189 Million $

NPV - Net Present Value (Million $)

1500

1000

500

(500) 0.00 0.10 0.20 0.30 0.40 0.50 0.60 Interest Rate (Fraction)

Figure 5-22: Net Present Value vs. Interest Rate

Figure 5-22 shows projects NPVs at different interest or discount rates ranging from 5% to 60%. The NPV decreases with increasing interest or discount rate because future net cash flow is more heavily discounted and penalized. The discount interest rate has less impact on projects expenditures than on its revenue because most of the investment is spent at the beginning of the project while most of the revenue is actually generated later in time. The NPV is equal to 1590 Million $ at interest rate of 5% and monotonically decreases to 245 Million $ at interest rate of 40%. Rate of return (ROR), which is the interest rate that results in zero NPV, is equal to 53%.

149 5.4.2 Field Optimization Figure 5-23 is a composite figure that shows field optimization results in both tabular and graphical forms. Target recovery factors at end of plateau are varied between 30% and 75% while total number of wells required for development is varied between 3 and 30 wells. It is readily realized from this figure that the optimized NPV can be placed at 1326 Million $ for the combination of 30% target recovery factor and the use of 15 wells for field development.

1,400

Net Present Value (Million $)

1,200 1,000 800 600 400 30 200 0 0.30 0.40 0.50 0.60 3 0.70 12 21 Total Number of Wells

Target Recovery Factor at End of Plateau

NPV (Million $) 3 6 9 12 15 18 21 24 27 30

0.30 639 1,077 1,244 1,310 1,326 1,313 1,284 1,241 1,195 1,139

0.35 601 1,052 1,226 1,296 1,315 1,304 1,277 1,236 1,190 1,136

Target Recovery Factor at End of Plateau 0.40 0.45 0.50 0.55 0.60 550 482 394 281 136 1,016 966 895 801 667 1,199 1,161 1,106 1,028 917 1,275 1,244 1,199 1,134 1,040 1,298 1,273 1,233 1,179 1,098 1,290 1,269 1,238 1,189 1,120 1,264 1,247 1,217 1,177 1,117 1,225 1,209 1,186 1,151 1,096 1,182 1,167 1,147 1,113 1,065 1,129 1,115 1,097 1,070 1,024

0.65 -50 480 753 899 975 1,010 1,018 1,010 985 954

0.70 -283 210 499 670 771 827 852 856 847 827

0.75 -579 -194 76 262 388 469 521 550 561 559

Total Number of Well

Figure 5-23: Field Optimization Results

150 If a target plateau recovery factor is fixed due to contractual obligations or market saturation demands, optimum number of wells could be determined. When too few wells are drilled to develop a hydrocarbon deposit, resulting field flow rates will be smaller and the required production period needed to reach abandonment will be prolonged. Thus, annual revenue from a distant future will be dramatically discounted, which will result in lower NPVs. In contrast, if total number of wells used to develop the reservoir is too high, the additional revenue that would be obtained from production acceleration will not be able to compensate or offset the significantly increased drilling and completion costs. As a result, NPV will not be maximized under either scenario. When the total number of wells is fixed, an optimum target recovery factor could not be found. Under the current model, NPV will always increases with decreasing target recovery factor. This is because CAPEX does not change with total flow rates in this economic model. If target recovery factor is lowered, initial flow rates will be higher and annual revenue will be accelerated. However, production facilities costs, required to handle such an increased volume of fluids, remain constant in this study. As a result, NPV will always be better for lower target plateau recovery factor because the production is accelerated and the revenue can be received earlier in the life of the field. In actual field applications, there is always a maximum fluid volume that can be reasonably handled at the surface and accepted by the market. This justifies the need for a plateau period. In the limiting case when target plateau recovery becomes zero, which effectively eliminates the plateau period, initial flow rates for the decline period will be extremely large. This is good news for the simplified economic model but bad news in real applications because those fluid volumes might not be able to be marketed effectively and the required surface facilities would become extremely expensive. In addition, facilities designed to handle such large volumes just during the first year alone would become awfully overdesigned for the rest of the reservoir life - a situation that is far from optimal. In order to be able to actually determine a more

151 realistic and optimum target recovery factor, the CAPEX model has to be adjusted. Surface production facilities cost and/or flowline costs have to be made functions of maximum expected flow rates by introducing either a continuous- or step- cost function in the CAPEX model.

152 5.5 Application for Other Production Situations This section discusses recommendations on how to extend the capabilities of field performance simulator for gas condensate fluids in order to tackle other possible production scenarios. Two other scenarios, such as dry gas / wet gas production and gas condensates with producible reservoir oil, are elaborated upon. The fundamental differences and the required modifications to current gas condensate scenario build in the present model, as well as the expected results from such modified models, are the main areas of discussion in this section.

153 5.5.1 Application for Dry Gas / Wet Gas Even though the proposed model has been specifically tailored to the analysis of gas condensate reservoir fluids, it could be actually be applied to any other natural gas reservoir, such as dry gases and wet gases, with few modifications. In dry gas and wet gas reservoirs, the hydrocarbon fluid is always found in a 100% vapor phase state throughout their isothermal reservoir depletion path. One of the main differences between dry gas and wet gas is that, along the surface depletion path, dry gas will also stay in a 100% vapor phase condition, while wet gas will experience two-phase condition or condensate dropout as it flows through the surface production system. The procedure used to calculate standard PVT properties of gas condensate at the dew point could be applied for the dry gas and wet gas. Characteristic of gas formation volume factor ( ) is expected to be the same for dry gases, wet gases, and gas condensates. Volatilized oil-gas ) for a dry gas will be zero, while volatilized oil-gas ratio ( ) and solution gas-oil ratio ( ) of the wet gas will be ) will not be defined or

ratio (

constant. Oil formation volume factor (

calculated in dry gas and wet gas because there is no presence of reservoir oil phase along the reservoir depletion path. In addition, specific gravity of reservoir gas will be constant for both dry gas and wet gas. For the reservoir zero-dimensional model, the gas condensate tank model above the dew point could be applied directly to dry gases and wet gases. However, because there is only surface-gas in the reservoir gas phase, cumulative oil recovery ( ) and original oil in place

( ) for a dry gas will be zero. Based on the same assumptions as gas condensate tank model, gas saturation ( ) in dry gas and wet gas will be constant at initial gas saturation ( In IPR calculations, relative permeability of gas ( saturation is constant. Gas flow rate ( ).

) will be constant because gas ) and the

) will depend on drawdown pressure (

154 multiplication of gas viscosity ( ) and gas formation volume factor ( ). Oil flow rate ( ) for

dry gas will be zero because volatilized oil-gas ratio ( will be equal to gas flow rate (

) is zero. Oil flow rate (

) for wet gas

) times volatilized oil-gas ratio (

) which is constant. In TPR

calculations, pressure drop inside the tubing of dry gas and wet gas will be relatively stable comparing to gas condensate because specific gravity of reservoir gas are constant. For the case of the dry-gas, there is no liquid drop out in the tubing. Field performance prediction, economic analysis, and field optimization for dry gases and wet gases are carried out using the same procedure as for a gas condensate above the dew point. Cumulative gas recovery ( ) at abandonment condition of dry gas is expected to be higher

than wet gases and gas condensates because lighter hydrocarbons exhibit larger expansivity coefficients or isothermal compressibility values and there is no obstruction to fluid flow due to the presence of a liquid hydrocarbon phase. Cumulative oil recovery ( ) at abandonment

condition for wet gases is expected to be much higher than gas condensate because the reservoir system does not leave any immobile condensate inside the reservoir. In term of NPV, dry and wet gases are expected to generate less NPV than gas condensates because they produce much less oil which can be sold for a much higher commodity price. In terms of optimization of target recovery factor at end of plateau and total number of wells, dry and wet gases are expected to exhibit similar characteristics as those of gas condensates.

155 5.5.2 Application for Gas Condensate with Producible (Mobile) Reservoir Oil One of the significant assumptions of the gas condensate tank model is that reservoir oil phase is immobile. However, this assumption is not always valid, especially around the wellbore where oil saturation might build up to a high enough value so that the relative permeability to oil might not be equal to zero anymore. This situation would increase complexity of gas condensate system. Standard PVT properties simulated from procedures described in either section 4.2.2 or 4.2.3 will carry certain degree of error toward the final results because those algorithms assumed that the reservoir oil remained immobile. However, there is no practical approach for modifying those procedures in order to fully represent the producible reservoir oil scenario. In order to simulate the producible reservoir oil situation within a standard PVT properties calculation algorithm, the ratio between excess gas and excess oil which should be removed from the PVT cell during the constant volume expansion has to be given. This ratio depends on the expected relative mobility ratio between gas and oil phases which are the functions of their relative permeabilities and fluid properties. Relative permeability depends on saturation fraction and saturation fraction could be obtained from saturation equation; thus, relative permeability is also the function of fluid properties. Moreover, the shape of the relative permeability curves is also rock-dependent and not solely fluid property dependent. In short, there is no simple method to fully and reliably represent producible reservoir oil situation. In the reservoir zero-dimensional model, the gas condensate tank model has to be modified by substituting Equation 4-115 with Equation 5-4 (below), substituting Equation 4-118 with Equation 5-5 (below), and calculating instantaneous production GOR at pressure level j ( ) using Equation 5-6 (below). Relative permeability in Equation 5-6 might be evaluated from

156 saturation value calculated at the preceding pressure level j-1. The definition of each parameter in these equations can be found in the gas condensate tank model section.

Equation 5-4

Equation 5-5

Equation 5-6

For IPR calculations, gas flow rate (

) and oil flow rate (

) should be calculated by

implementing Equation 4-132 and Equation 4-133 instead. For TPR calculations, the single phase gas flow equation is still applied; but pipe efficiency factor ( because of additional liquid phase flow into wellbore. Procedures used to perform field performance prediction, economic analysis, and field optimization remain unchanged. Cumulative gas recovery ( ) at abandonment condition for ) is expected to be lower

producible reservoir oil scenario is expected to be close to the value obtained from immobile reservoir oil scenario. In contrast, cumulative oil recovery ( ) at abandonment condition for

producible reservoir oil scenario is expected to be higher than the value obtained from immobile reservoir scenario because less oil is left immobile inside the reservoir. In term of NPV

157 estimations, larger NPVs are expected from producible reservoir oil scenario because reservoir oil could yield more stock-tank oil, which is more expensive product, than reservoir gas. For field optimization, the similar characteristic is expected whether reservoir oil is producible or not.

Chapter 6 Summary and Conclusions

A model able to perform field performance analysis and optimization of exploitation strategies for a gas condensate reservoir has been successfully developed. The model has been constructed using Microsoft Excel with built-in Visual Basic for Applications (VBA) program. The model includes a fluid property calculation subroutine which estimates standard PVT properties based on a binary pseudo-component model using fluid compositional information as input. The subroutine demonstrates to produce reliable standard PVT properties typical of gas condensate fluid phase behavior. Limitations of the pseudo-component model, such as the generation of negative solution gas-oil ratio ( ) at low reservoir pressure, are clearly shown and

explained using the simulated results produced by the fluid property calculation subroutine. A zero-dimensional reservoir model based on the generalized material balance equation for gas condensates has also been developed. The results from this gas condensate tank model are able to mimic the typical reservoir performance data found in gas condensate fields. Possible sources of error and misinterpretation from using a gas condensate tank model in the analysis of other near critical fluids, such as volatile oils, are discussed and recommended crosschecking procedure that should be implemented before and after running the simulation model is addressed. A field performance prediction that couples zero-dimensional reservoir models with nodal analysis concepts has been successfully developed for the screening of field development strategies. The field performance prediction tool has also been coupled with an economic model which enables the prediction of optimum field development strategies. The reservoir model demonstrates to provide results which are consistent with reservoir depletion behavior for gas

159 condensates. A discussion of the observed decline trend analysis has been included to shed some light on the possibility of obtaining slightly negative hyperbolic decline coefficients in gas condensate fields. The economic evaluation subroutine was implemented based on simplified economic model. The subroutine is used to optimize target development variables, such as target recovery factor at end of plateau and total number of wells required to optimally develop the field. This proposed model is shown to be able to perform economic analysis and field optimization effectively. Economic and optimization results are analyzed in detail. Limitations of the economic model assumptions are addressed and discussed. The proposed model is suitable for real field applications when either input data or working time is constrained. For example, this model is appropriate to be used to simulate field performance data for feasibility study of gas condensate reservoirs because, during that phase of field development, reservoir data is usually limited and all available data are highly uncertain; thus constructing highly sophisticated model is impractical. In addition, this model can effectively simulate field performance data for numerous production scenarios, which is a very important factor to cope with the high uncertainty found in that period. Another proper application is to use this proposed model to perform project evaluations for new asset acquisitions because, during the acquisition process, the evaluation of each project has to be completed within a short period of time; thus utilizing the less complicated model, which takes less time to construct and execute, is more feasible, even if there are plenty of reservoir and field development data. Additional recommendations for avenues for future work are also provided for the improvement of the reliability and the capabilities of the proposed field performance model. The first recommendation is to validate the accuracy of the models outputs with the outputs calculating from full scale full dimensional numerical simulator using both hypothetical data and actual field data. With the hypothetical data validation, the error resulting from a singlehomogeneous tank assumption (neglecting all gradients) can be analyzed. With the actual field

160 data validation, the impact of heterogeneity on naturally occurring reservoirs can be determined. The second recommendation is to perform sensitivity analysis to investigate how the variation in economic outputs can be attributed to the variation in reservoir and field development input variables. With this understanding, the limited resources (time and man-hour) can be properly spent to evaluate the expected values and uncertainties of the variables with highly economic impact. In most cases, product prices, development costs, and reserves are expected to be the variables with the highest impact on the projects economic. The other recommendations include the handling of production from dry gas and wet gas reservoirs and the modeling of gas condensates with producible (mobile) reservoir oil. Further avenues include allowing the model to handle multiple reservoirs and multiple types of producing wells, which would make the model applicable to be used in complex reservoir structures and marginal fields development.

Bibliography
Ayala, L.F. 2009a. Lecture Notes, PNG530 Natural Gas Engineering, Department of Energy and Mineral Engineering, Pennsylvania State University. Ayala, L.F. 2009b. Phase Behavior Model Instruction, PNG520 Phase Behavior of Petroleum Fluids, Department of Energy and Mineral Engineering, Pennsylvania State University. Baker, L.E., Pierce, A.C., and Luks, K.D. 1982. Gibbs Energy Analysis of Phase Equilibria. SPEJ. 22 (5): 731-742 Coats, K.H. 1985. Simulation of Gas Condensate Reservoir Performance. Paper SPE 10512, Journal of Petroleum Technology. October: 1870-1886. Colebrook, C.F. 1939. Turbulent Flow in Pipes, with Particular Reference to the Transition Region Between the Smooth and Rough Pipe Laws. Journal of the Institution of Civil Engineers. February: 133-156 Danesh, A. 1998. PVT and Phase Behavior of Petroleum Reservoir Fluids: Elsevier Deitz, D.N. 1965. Determination of Average Reservoir Pressure from Build-Up Surveys. Journal of Petroleum Technology. 17 (8): 955-959 Ertekin, T., Abou-Kassem, J.H., and King, G.R. 2001. Basic Applied Reservoir Simulation, SPE Textbook Series Vol.7: Society of Petroleum Engineer Jhaveri, B.S. and Youngren, G.K. 1988. Three-Parameter Modification of the Peng-Robinson Equation of State to Improve Volumetric Predictions. SPERE. 3 (3): 1033-1040 Jossi, J.A., Stiel, L.I., and Thodos, G. 1962. The Viscosity of Pure Substances in the Dense Gaseous and Liquid Phases. AIChE Journal. 8 (1): 59-63 Lee, A.L., Gonzalez M.H., and Eakin, B.E. 1966. The Viscosity of Natural Gases. Journal of Petroleum Technology. 18 (8): 997-1000 Lee, J., Rollins, J.B., and Spivey J.P. 2003. Pressure Transient Testing, SPE Textbook Series Vol.9: Society of Petroleum Engineer Lohrenz, J., Bray, B.G., and Clark, C.R. 1964. Calculating Viscosities of Reservoir Fluids from Their Compositions. Journal of Petroleum Technology. 16 (10): 1171-1176 Martin, J.J. 1979. Cubic Equations of State-Which?. Industrial and Engineering Chemical Fundamentals. 18 (2): 81-97 McCain, W.D., Jr. 1990. The Properties of Petroleum Fluid, 2nd Ed: PennWell Mehra, R. K., Heidemann, R. A. and Aziz, K. 1983. An Accelerated Successive Substitution Algorithm. The Canadian Journal of Chemical Engineering. 61 (4): 590-593

162 Mian, M.A. 2002. Project Economics and Decision Analysis, Volume I: Deterministic Models: PennWell Michelsen, M.L. 1982a. The Isothermal Flash Problem. Part I. Stability. Fluid Phase Equilibria. 9 (1): 1-19 Michelsen, M.L. 1982b. The Isothermal Flash Problem. Part II. Phase-Split Calculation. Fluid Phase Equilibria. 9 (1): 21-40 Peneloux, A., Rauzy E., and Freze R. 1982. A Consistent Correction for Redlich-Kwong-Soave Volumes. Fluid Phase Equilibria. 8 (1): 7-23 Peng, D.Y. and Robinson, D.B. 1976 A New Two-Constant Equation of State. Industrial and Engineering Chemical Fundamentals. 15 (1): 59-64 Rachford, H.H. and Rice, J.D. 1952. Procedure for Use of Electronic Digital Computers in Calculating Flash Vaporization Hydrocarbon Equilibrium. Papar SPE 952327-G, Petroleum Transaction, AIME 195: 327-328 Redlich, O. and Kwong, J.N.S. 1949. On the Thermodynamics of Solutions. V. An Equation of State. Fugacities of Gaseous Solutions. Chemical Reviews. 44 (1): 233-244 Soave, G. 1972. Equilibrium Constants from a Modified Redlich-Kwong Equation of State. Chemical Engineering Science 27 (6): 1197-1203 Thararoop, P. 2007. Computer Project No.1: Analysis of Gas Condensate Reservoir Exploitation, PNG530 Natural Gas Engineering, Department of Energy and Mineral Engineering, Pennsylvania State University. Walkenbach, J. 2007. Excel 2007 Power Programming with VBA. Wiley Publishing, Inc Walsh, M.P. 1995. A Generalized Approach to Reservoir Material Balance Calculations. Journal of Canadian Petroleum Technology. 34 (1) Walsh, M.P. and Lake, L.W. 2003. A Generalized Approach to Primary Hydrocarbon Recovery: Elsevier Walsh, M.P. and Towler, B.F. 1995. Method Computes PVT Properties for Gas Condensates. Oil and Gas Journal. July: 89-83 Wilson, G. 1968. A Modified Redlich-Kwong EOS, Application to General Physical Data Calculations. Paper No 15C presented at the AIChE 65th National Meeting Whitson, C.H. and Brule, M.R. 2000. Phase Behavior, SPE Monograph Series Volume 20. Society of Petroleum Engineer

Appendix A Input Data Summary

Table A-1: Pressures and Temperatures for Standard PVT Properties Calculation Subroutine
Pressure (psia) Reservoir Condition 1st Stage Separator 2nd Stage Separator Stock Tank Condition 4000 500 150 100 Temperature (F) 300 90 65 60

Table A-2: Physical Properties of Pure Components

Component N2 C1 C2 C3 i-C4 n-C4 i-C5 n-C5 n-C6 n-C10 CO2 Mole Fraction 0.0223 0.6568 0.1170 0.0587 0.0127 0.0168 0.0071 0.0071 0.0138 0.0832 0.0045 Critical Pressure (psia) 493.10 666.40 706.50 616.00 527.90 550.60 490.40 488.60 436.90 305.20 1071.00 Critical Temperature (R) 227.49 343.33 549.92 666.06 734.46 765.62 829.10 845.80 913.60 1112.00 547.91 Acentric Factor 0.037 0.010 0.098 0.152 0.185 0.200 0.228 0.251 0.299 0.490 0.267 Molecular Weight (lbm/lbmol) 28.013 16.043 30.070 44.097 58.123 58.123 72.150 72.150 86.177 142.285 44.010 Critical Volume (ft3/lbm) 0.0510 0.0988 0.0783 0.0727 0.0714 0.0703 0.0679 0.0675 0.0688 0.0679 0.0344

164 Table A-3: Binary Interaction Coefficients of Pure Components

ij's N2 C1 C2 C3 i-C4 n-C4 i-C5 n-C5 n-C6 n-C10 CO2 N2 0.0000 0.0180 0.0390 0.0460 0.0470 0.0470 0.0480 0.0480 0.0000 0.0000 0.0000 C1 0.0180 0.0000 0.0050 0.0100 0.0145 0.0145 0.0182 0.0182 0.0000 0.0000 0.0000 C2 0.0390 0.0050 0.0000 0.0017 0.0032 0.0032 0.0048 0.0048 0.0000 0.0000 0.0000 C3 0.0460 0.0100 0.0017 0.0000 0.0012 0.0012 0.0024 0.0024 0.0000 0.0000 0.0000 i-C4 0.0470 0.0145 0.0032 0.0012 0.0000 0.0000 0.0008 0.0008 0.0000 0.0000 0.0000 n-C4 0.0470 0.0145 0.0032 0.0012 0.0000 0.0000 0.0008 0.0008 0.0000 0.0000 0.0000 i-C5 0.0480 0.0182 0.0048 0.0024 0.0008 0.0008 0.0000 0.0000 0.0000 0.0000 0.0000 n-C5 0.0480 0.0182 0.0048 0.0024 0.0008 0.0008 0.0000 0.0000 0.0000 0.0000 0.0000 n-C6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 n-C10 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 CO2 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000

Table A-4: Volume Translation Coefficient of Pure Components

Component N2 C1 C2 C3 i-C4 n-C4 i-C5 n-C5 n-C6 n-C10 CO2 Si for PR EOS -0.19270 -0.15950 -0.11340 -0.08630 -0.08440 -0.06750 -0.06080 -0.03900 -0.00800 0.06550 -0.08170 Si for SRK EOS -0.00790 0.02340 0.06050 0.08250 0.08300 0.09750 0.10220 0.12090 0.14670 0.20800 0.08330

165 Table A-5: Reservoir Input Data

Reservoir Number 1 Initial Reservoir Pressure (psia) 4000 Abandonment Reservoir Pressure (psia) 750 Reservoir Drainage Area (Acres) 5000 Dew Point Reservoir Pressure (psia) 3031 Reservoir Temperature (Deg F) 300 Temperature Gradient (Deg F / ft) 0.0205 Connate Water Saturation (Frac) 0.210

Temperature Surface (Deg F) 60 Original Gas Equivalent In Place (BSCF) 393.15

Reservoir Depth (ft) 6560

Reservoir Thickness (ft) 55

Reservoir Pososity (Frac) 0.200

Absolute Permeability (md) 80

Wellbore Radius (ft) 0.375

Shape Factor

Mechanical Skin 10

Non-Darcy Flow Coeff (D/MSCF) 0.0011

30.88

Table A-6: Relative Permeability Input Data

Reservoir Number 1 Connate Water Saturation 0.20 Connate Gas Saturation 0.20 Maximum Gas Saturation 0.80 Krg at Maximum Gas Saturation 0.80

166 Table A-7: Standard PVT Properties

Pressure (psia) 4000 3950 3900 3850 3800 3750 3700 3650 3600 3550 3500 3450 3400 3350 3300 3250 3200 3150 3100 3050 3031 3000 2950 2900 2850 2800 2750 2700 2650 2600 2550 2500 2450 2400 2350 2300 Bo (RB/STB) 5.052 5.089 5.127 5.166 5.206 5.248 5.291 5.336 5.382 5.429 5.479 5.530 5.583 5.638 5.695 5.754 5.815 5.879 5.946 6.015 6.042 2.858 2.706 2.582 2.479 2.389 2.311 2.241 2.178 2.120 2.067 2.019 1.974 1.932 1.892 1.855 Bg (RB/MSCF) 1.040 1.048 1.056 1.064 1.072 1.081 1.089 1.099 1.108 1.118 1.128 1.139 1.149 1.161 1.173 1.185 1.197 1.211 1.224 1.239 1.244 1.251 1.265 1.280 1.297 1.315 1.335 1.356 1.378 1.401 1.426 1.452 1.480 1.509 1.539 1.571 Rs (SCF/STB) 4856.8 4856.8 4856.8 4856.8 4856.8 4856.8 4856.8 4856.8 4856.8 4856.8 4856.8 4856.8 4856.8 4856.8 4856.8 4856.8 4856.8 4856.8 4856.8 4856.8 4856.8 2067.5 1903.7 1768.5 1652.8 1551.4 1461.0 1379.3 1304.9 1236.4 1173.1 1114.1 1059.0 1007.3 958.5 912.5 Rv (STB/MMSCF) 205.9 205.9 205.9 205.9 205.9 205.9 205.9 205.9 205.9 205.9 205.9 205.9 205.9 205.9 205.9 205.9 205.9 205.9 205.9 205.9 205.9 195.4 181.4 170.0 160.3 151.9 144.5 137.9 131.9 126.4 121.4 116.8 112.6 108.6 105.0 101.5 1.147 1.147 1.147 1.147 1.147 1.147 1.147 1.147 1.147 1.147 1.147 1.147 1.147 1.147 1.147 1.147 1.147 1.147 1.147 1.147 1.147 1.128 1.102 1.081 1.062 1.046 1.032 1.019 1.007 0.996 0.986 0.977 0.969 0.961 0.954 0.947 SG Gas

167 Table A-7: Standard PVT Properties (Cont.)

Pressure (psia) 2250 2200 2150 2100 2050 2000 1950 1900 1850 1800 1750 1700 1650 1600 1550 1500 1450 1400 1350 1300 1250 1200 1150 1100 1050 1000 950 900 850 800 750 700 650 600 550 500 Bo (RB/STB) 1.821 1.788 1.757 1.728 1.700 1.673 1.648 1.623 1.600 1.578 1.556 1.536 1.516 1.497 1.479 1.461 1.445 1.428 1.412 1.397 1.382 1.368 1.354 1.341 1.328 1.315 1.303 1.291 1.280 1.269 1.257 1.246 1.236 1.226 1.216 1.206 Bg (RB/MSCF) 1.605 1.641 1.678 1.718 1.759 1.803 1.850 1.899 1.952 2.007 2.066 2.128 2.195 2.266 2.342 2.424 2.511 2.606 2.707 2.817 2.936 3.065 3.207 3.361 3.532 3.720 3.928 4.161 4.423 4.719 5.056 5.444 5.894 6.423 7.053 7.816 Rs (SCF/STB) 868.9 827.5 788.1 750.5 714.6 680.3 647.4 615.9 585.6 556.5 528.4 501.4 475.4 450.3 426.1 402.7 380.1 358.2 337.1 316.6 296.8 277.6 259.0 241.0 223.6 206.8 190.4 174.6 159.3 144.3 129.2 114.5 100.5 86.9 73.9 61.5 Rv (STB/MMSCF) 98.3 95.3 92.5 89.9 87.4 85.1 82.9 80.8 78.9 77.1 75.4 73.8 72.3 71.0 69.7 68.6 67.6 66.6 65.8 65.1 64.5 64.1 63.7 63.5 63.5 63.6 63.9 64.3 65.0 66.0 67.3 68.9 71.0 73.5 76.8 80.8 0.941 0.935 0.929 0.924 0.919 0.914 0.910 0.906 0.902 0.899 0.896 0.893 0.890 0.887 0.885 0.883 0.881 0.880 0.879 0.878 0.877 0.876 0.876 0.876 0.877 0.877 0.879 0.880 0.882 0.885 0.888 0.892 0.897 0.903 0.910 0.919 SG Gas

168 Table A-8: Tubing Input Data

Min Allow Wellhead Pressure (psia) Tubing Roughness (inch) Tubing Efficiency (Frac) Depth (MD - ft) 0 650 1300 1950 2600 3250 3900 4550 5200 5850 6500 Depth (TVD - ft) 0 650 1300 1950 2600 3250 3900 4550 5200 5850 6500 Tubing ID (inch) 4.5 4.5 4.5 4.5 4.5 4.5 4.5 4.5 4.5 4.5 550 0.0018 0.70 Temperature (F) 60.0 83.8 107.6 131.3 155.1 178.9 202.7 226.5 250.2 274.0 297.8

Table A-9: Economic Input Data

General Net Hydrocarbon Fraction Price First Year Gas Price Gas Price Escalation First Year Oil Price Oil Price Escalation Discount Rate Discount Rate 12.00% 4.000 1.00% 80.000 2.00% $/STB $/MSCF 87.50%

169 Table A-9: Economic Input Data (Cont.)

Capital Expenses Capex Total - Drilling (per well) - Flowlines and Trunklines - Production Facilities - Platform (for Offshore) - Others Operating Expenses First Year Opex Opex Escalation Tax Gas Severance Tax Oil Severance Tax Ad Valorem Tax 8.00% 8.00% 0.00% 0.5 2.50% Million $/Mth 1,480.0 25.0 100.0 500.0 400.0 30.0 Million $ Million $ Million $ Million $ Million $ Million $

Table A-10: Field Performance Prediction Input

Target Recovery at End of Plateau 0.55 Total Number of Wells 18

Table A-11: Field Performance Optimization Input

Target Recovery at End of Plateau 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75 Total Number of Wells 3 6 9 12 15 18 21 24 27 30

170 Appendix B User Guide 1. Open Field Development Plan.xlsm

2. Simulate standard PVT properties

a. Simulate standard PVT propertied from Walsh-Tolwer algorithm i. Select worksheet PVT from CVD ii. Input data in Reservoir Input section iii. Input data in Pressure Volume Relation section iv. Input data in Constant Volume Depletion section v. Input data in Calculated Cumulative Recovery section vi. Input data in Z-Factor of Produced Wellstreams section vii. Input data in Composition of Produced Wellstreams section viii. Adjust formulas in Pre-Calculation section ix. Adjust formulas in Walsh-Towler Algorithm section x. Adjust formulas in Standard PVT Properties at Dew Point and Below section xi. Adjust formulas in Standard PVT Properties Above Dew Point section xii. Copy all calculated standard PVT properties to FDP_Input_PVT worksheet

171

172 b. Simulate standard PVT properties from compositional data i. In PVT_Input_Pres worksheet, input reservoir and surface separators pressure and temperature conditions ii. In PVT_Input_Comp worksheet, input composition (mole fraction) and properties of pure components iii. In PVT_Input_BiCo worksheet, input binary interaction coefficient iv. In PVT_Input_Si worksheet, input volume-translate coefficient data v. In PVT_Output_Prop worksheet, click Calculate Black Oil PVT Properties button vi. In PVT_Output_Envelope worksheet, click Create Phase Envelope button (optional) vii. In FDP_Input_PVT worksheet, click Import from PVT Calculation button

173

174

175

176 3. Perform field performance analysis

a. Run field performance prediction and economic analysis i. In FDP_Input_Tank worksheet, input reservoir data ii. In FDP_Input_RelPerm worksheet, input relative permeability data iii. In FDP_Input_TPR worksheet, input tubing data iv. In FDP_Economic worksheet, input economic data v. In FDP_Input_Main worksheet, input target recovery factor at end of plateau and total number of wells required for field development vi. Click Run Performance Prediction and Economic Analysis button vii. Obtain economic results from FDP_Economic worksheet viii. Obtain field performance data from FPD_Output_Perf worksheet ix. Obtain reservoir performance data from FDP_Output_SGCT worksheet x. Obtain decline trend analysis data from DCA worksheet xi. Obtain graphical display of flow rates, pressures and cumulative production from QgGp, QoNp, and QwPrPwfPwh worksheets

177

178

179

180

181

182

b. Run economic analysis from latest field performance data i. In FDP_Economic worksheet, input economic data ii. In FDP_Input_Main worksheet, click Run Economic Analysis from Latest Performance Data button iii. Obtain economic results from FDP_Economic worksheet

183

184 4. Perform field optimization a. Run optimization with new performance prediction and economic analysis i. Repeat 3.-a.-i ) to 3.-a.-iv) ii. In FDP_Input_Main worksheet, input target recovery factor at end of plateau and total number of wells required for field development iii. Click Optimization 1: Run Performance Prediction and Economic Analysis button iv. Obtain field optimization results from Optimization worksheet

185

b. Run optimization with new economic analysis only (utilize latest performance prediction data) i. In FDP_Economic worksheet, input economic data ii. In FDP_Input_Main worksheet, click Optimization 2: Run Economic Analysis only button iii. Obtain field optimization results from Optimization worksheet

186

187