NUMERICAL STUDY OF WATER CONING CONTROL WITH DOWNHOLE WATER SINK (DWS) WELL COMPLETIONS IN VERTICAL AND HORIZONTAL

WELLS

A Dissertation Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Doctor of Philosophy in The Department of Petroleum Engineering

by Solomon Ovueferaye Inikori B. Engineering, University of Nigeria, Nsukka-Nigeria, 1988 M. Engineering, University of Benin, Benin City-Nigeria, 1993 August 2002

ACKNOWLEDGEMENTS The author expresses special gratitude and appreciation to professor Andrew K. Wojtanowicz for his encouragement, patience and guidance as research advisor and the chairman of the examination committee. A deep appreciation is also extended to other members of the examination committee, Dr. Chris White, Dr. Zaki Bassiouini, Chair of the Craft and Hawkins Department of Petroleum Engineering, Dr. Dandina Rao and Dr. Dimitris Nikitopoulos for their valuable review, guidance and assistance throughout this research. The author would also like to acknowledge the members of the Joint Industry Panel (JIP) of the Downhole Water Sink Technology Initiative (DWSTI) for the financial support and the supply of relevant field data for this research. The author wishes to also acknowledge Dr. John McMullan, Dr. Ephim Shirman, Dr. Omowunmi Iledare of the Petroleum Technology Transfer Council at the Louisiana State University, Dr. Sam Ibekwe of Mechanical Engineering Department at the Southern University, Rev. Chibuike Azuoru, (CPA), of Southern University, Mrs. Mary Azuoru of the Louisiana State Department of Education, members of the African Christian Fellowship and Christ the King Christian Center, Baton Rouge, my colleague, Djuro Novakovic and the staff of the Petroleum Engineering Department for all the support and encouragement. Finally, the author is grateful to his wife, Evarista Omoyoma Inikori and three sons, Erosuo Godson Inikori, O’fejiro Hubert Inikori and Rukevwe Nicholas Inikori for all their support, patience, advise, prayers, understanding and encouragement throughout this doctoral program. To God be the glory. Great things He has done. ii

TABLE OF CONTENTS ACKNOWLEDGEMENTS ............................................................................................. ii LIST OF TABLES .......................................................................................................... vii LIST OF FIGURES ....................................................................................................... viii NOMENCLATURE........................................................................................................ xii ABBREVIATIONS ...................................................................................................... xviii ABSTRACT ……………………………………………………………………………xxi CHAPTER 1 INTRODUCTION......................................................................................1 1.1 Background and Purpose .....................................................................................1 1.2 Statement of Research Problem ...........................................................................3 1.3 Significance and Contribution of this Research...................................................5 1.4 Research Method and Approach..........................................................................6 1.5 Work Program Logic ...........................................................................................8 CHAPTER 2 LITERATURE REVIEW .......................................................................10 2.1 Water Coning in Vertical Wells.........................................................................10 2.2 Other Industry Solutions for Water Coning in Vertical Wells ..........................16 2.2.1 Perforation Squeeze-off and Re-completion..............................................17 2.2.2 Conformance Control Technology ............................................................17 2.2.3 Downhole Oil-Water Separation Technology ...........................................18 2.2.4 Total Penetration Approach .......................................................................19 2.3 Water Coning Control with Dual Completions in vertical wells .......................19 2.4 Capillary Pressure and Relative Permeability Hysteresis in DWS Water Coning Control ..................................................................................................22 2.5 Horizontal Well Technology..............................................................................23 2.6 Horizontal Well and Water Influx Control ........................................................26 2.7 Friction Pressure Loss Consideration in the Horizontal Wellbore ....................29 2.8 Water Cresting Control in Horizontal Wells .....................................................32 2.9 Flow Regimes of Oil-Water Flow in Horizontal Pipes/Wellbore......................33 2.10 Friction Pressure Loss of Oil-Water Flow in Horizontal Wells ........................35 2.11 Numerical Simulation of Water Coning in Wells..............................................36 2.12 Current Methods of Controlling Water Cresting in Horizontal Wells...............37 2.12.1 Horizontal Well Completion With Stinger ................................................37 2.12.2 Water Cresting Control With Variation of Perforation Density .......................................................................................................38 2.12.3 Water Cresting Control with High Angle Wells........................................40 2.13 Summary............................................................................................................41

iii

CHAPTER 3 DWS VERTICAL WELL PERFORMANCE WITH VARIABLE WATER SATURATION...................................................43 3.1 Pressure Stabilization and Steady-state Water Cone Simulation.......................45 3.2 Verification of Results with Analytical Model IPW..........................................52 3.3 Capillary Pressure Transition and Simulation of DWS Well ............................54 3.3.1 Capillary Pressures.....................................................................................54 3.4 Relative Permeability Hysteresis and the DWS Technology Modeling............62 3.4.1 Fractional Flow Analysis ...........................................................................65 3.4.2 Imbibition and Drainage Relative Permeabilities ......................................66 3.4.3 Simulation of Relative Permeability Hysteresis in Eclipse .......................69 3.4.4 Analysis of Results from the Relative Permeability Hysteresis and Capillary Pressure Studies.................................................71 3.5 DWS Well Productivity Performance Studies...................................................74 3.5.1 DWS Well Oil Recovery Performance Study............................................75 3.5.2 DWS Well Productivity Index Comparison ..............................................76 3.5.3 Watercut Performance Evaluation .............................................................81 3.6 DWS Well Design Recommendations for Old Wells........................................82 CHAPTER 4 DWS WELL DELIVERABILITY WITH OIL-FREE DRAINAGE WATER..............................................................................84 4.1 E & P Produced Water Regulations and Requirements.....................................85 4.2 Overview of DWS Environmental Performance ...............................................86 4.3 Capillary Transition and the IPW and Clean Water Drainage...........................88 4.4 Modeling Clean Water Performance of DWS in Old Fields. ............................89 4.4.1 Model Description. ....................................................................................90 4.4.2 Capillary Pressure Transition Evaluation. .................................................91 4.4.3 Relative Permeability Curves. ...................................................................92 4.5 Presentation and Discussion of Results .............................................................92 4.6 Design Considerations for DWS Wells With Oil-free Drainage Water ............94 4.7 Effect of Location of the Water Sink in the Capillary Transition Zone ............97 4.8 Recommendations for Designing DWS Well and Production Schedule ...........98 CHAPTER 5 EFFECT OF FLOW FRICTION ON WATER CRESTING IN HORIZONTAL WELLS.......................................................................100 5.1 Limitations of Water Coning Control With Horizontal Wells – Field Evidence ..........................................................................................................101 5.2 Model of Oil-Water Flow in Horizontal Wells................................................104 5.2.1 Flow Regimes and Conditions in Horizontal Wellbore...........................104 5.2.2 Properties of Oil-water Mixture in Horizontal Wellbore.........................107 5.2.3 Frictional Pressure Loss in Horizontal Wellbore.....................................114 5.2.4 Generalized Friction Factor Correlation for Horizontal Wells and Pipes........................................................................................120 5.2.5 Verification of “Generalized Friction Factor (GFF)” Relation ....................................................................................................121 5.3 Numerical Study of Water Cresting in Horizontal Wells With Compound Frictional Pressure Loss Model ....................................................125 iv

........................................................................................1....................1 Equivalent Length/Drainage Area of Horizontal Wells.............1..........1..................167 8..................1..........................................................................170 8...1.............................3.................................127 CHAPTER 6 WATER CRESTING CONTROL IN HORIZONTAL WELLS WITH DOWNHOLE WATER SINK TECHNOLOGY...................................2 Watercut Performance Evaluation ...............1 Effective Length of Bilateral Water Sink Horizontal Well .........................................................2...168 8........157 7.........................................................................................5..1 Feasibility of Water Cresting Control with Dual Completion Technology...............2 Method 2: Ellipse Shaped Drainage Area Method ..............3 Water Cresting in Horizontal Wells and DWS Technology.........161 7............................................................................1....................3.161 7.....................................................1...........................................2 Water Cresting Control Performance of the Tail-pipe and Bilateral Water Sink Technology..........................1 Drilling and Completion of the Tail-pipe Multi-lateral Well Option .............172 REFERENCES .........2 Two Vertical DWS Wells Vs Horizontal Well Watercut Analysis................140 6..5 Economic Evaluation of the DWS Technology Vs Horizontal Well ..................4 Comparison of Vertical Wells with DWS Technology and Horizontal Wells ..............153 7........................................159 7...173 v ......................................................................................................1 Oil Production Comparison Study.........................167 8...................................................137 6............153 7......1 Changes in Horizontal Wells Water Crest Development Due to ‘New Model’ .......2 Oil-Free Water Production Capacity of DWS with Capillary Pressure Zone.2 Drilling and Completion of the Bilateral Water Sink Well Option .............................................4...144 6....................152 7................................................................................2 Recommendations....1 Method 1: Two Half-circle and Rectangle Approach...................................................169 8............4.......147 CHAPTER 7 COMPARISON OF DWS VERTICAL WELLS AND HORIZONTAL WELLS IN BOTTOM WATER DRIVE RESERVOIR..............................................................................................................................................149 7......................................1 Conclusions....2 Simulation Grid Properties of Horizontal Well and Vertical Well........1 Two Vertical DWS Wells Vs Horizontal Well Oil Production Analysis ............164 CHAPTER 8 CONCLUSIONS AND RECOMMENDATIONS..................3...............1 Relative Permeability Hysteresis and Capillary Pressure Transition Zone on DWS Well Performance........................................................1....171 8....155 7.......163 7.............134 6....................3 Comparison of Watercut Performance of Vertical and Horizontal Wells ....156 7.......................134 6..............4 Vertical Wells with DWS and Conventional Horizontal Wells .......................................

...................198 APPENDIX D – DATA DECK FOR TAIL-PIPE WATER SINK (TWS) CONCEPT FOR WATER CRESTING CONTROL ................216 vi ..........................193 B....................183 APPENDIX B – WATER CONE REDUCTION STUDY ON WELL LVT ................................................................4 Approach to Study .......................194 B.............................212 APPENDIX F – SAMPLE DATA OF MODIFIED PIPE ROUGHNESS...........197 APPENDIX C – DATA DECK FOR BILATERAL WATER SINK (BWS) CONCEPT FOR WATER CRESTING CONTROL ...................................................................................196 B............ ...... ...................5......................................................................................................................188 B....5.2 Objective ........................7 Conclusion of Studies ...............APPENDIX A – CAPILLARY PRESSURE/RELATIVE PERMEABILITY HYSTERESIS DATA DECK ...................................................................................................................195 B.......................................................5..................................................3 Input Data for Eclipse Simulator .........................................214 VITA ……………..................................................1 Executive Summary...193 B...........1 Scenario A – Watercut Reduction Without the Effect of Capillary Pressure....................................189 B.........188 B....DWS WELL COST DATA SPREADSHEET..............................................................................................................5............ .....3 Scenario C – Watercut Reduction Performance with Leaking Cement....................190 B..............................................................4 Scenario D – Watercut Reduction Performance with high rate at top Completion...205 APPENDIX E ...........188 B...............2 Scenario B – Watercut Reduction Performance with Capillary Pressure Effect.................5 Discussion of Results..6 Appropriate Watercut Level for Deployment of DWS Technology..................195 B......................

..................................144 Table 6-2 Improved Oil Recovery in Horizontal Well with DWS Technology............. 90 Table 5-1 Experimental Values for the Constants............... a............. 87 Table 4-4 LVT Reservoir Input Data.................................................................. 145 Table 7-1 Reservoir/Rock Properties of Horizontal Well Vs Vertical Well Water Cresting/coning Performance Studies ............................................................................................. 64 Table 4-1 Requirements on Hydrocarbon Contamination of Produced Water.............. 85 Table 4-2 Hydrocarbon Contamination of Drainage/System Water....................................................................LIST OF TABLES Table 3-1 Craig's Rules-of-Thumb for Determining Rock Wettability..................... 87 Table 4-3 PAH Contamination of Drainage /System Waters ... 155 Table 7-2 Drilling and Completion Cost for 2-Vertical Wells and 1 Horizontal Well.......................................................... b and Cn........ 118 Table 6-1 Oil Recovery for Conventional Horizontal Well with Fanning Friction Factor.......... 165 vii ..............

....................................... 47 Figure 3-4 Unstable Wellbore Pressure Distribution with time........................ 39 Figure 2-7 Schematic of High-Angle Well with tip Close to the OWC ........................................... 62 Figure 3-12 Relative Permeability Curves for Drainage and Imbibition............................................................. 40 Figure 3-1 Typical Inflow Performance Window (IPW) and its Regions ...................................... 34 Figure 2-5 Horizontal Well Completion with a Stinger........................................................................................... . 70 Figure 3-14 Negligible Hysteresis Effect Due to use of Same End-point Saturations for both Imbibition and Drainage............................ 51 Figure 3-8 Comparison of IPW ................ 49 Figure 3-5 Creeping Watercut for Unstable Flowing Bottomhole Pressure........... 50 Figure 3-6 Overlap of Pressure at Different times Indicate a Stable Steady-state Flow Process .............. 21 Figure 2-4 Oil-Water Flow Regimes in Horizontal Pipes .................... 59 Figure 3-11 Effect of Capillary Transition Zone on DWS IPW............... 52 Figure 3-9 Typical Capillary Transition Pressure Zone ........................... 38 Figure 2-6 Schematic of Typical Variable Perforation Density ................................................................................................................ 50 Figure 3-7 Watercut Profile for Stabilized Bottomhole Pressure (Plateau)............................................ 46 Figure 3-3 Unstable Flowing Bottom-hole Pressure at Top Completion ................... 68 Figure 3-13 Relative Permeability Hysteresis Curve with Scanning Option .....LIST OF FIGURES Figure 2-1 Schematic Representation Water Coning in Vertical Wells ................................Numerical Simulator and Analytical Model.............................................................. 11 Figure 2-2 Cone Stability Analysis......................................... 72 viii ............................................................................ 44 Figure 3-2 Steady-state Numerical Simulation Models.................. 58 Figure 3-10 Capillary Pressure Hysteresis........................ 13 Figure 2-3 Schematic of the dual completion for the Nebo Hemphill project...................................

..................................................................... 80 Figure 3-21 Watercut and Cumulative Oil Recovery Performance.................................................................................................................................................................................................................. 75 Figure 3-17 Field liquid PI Comparison between DWS Well and Conventional Well .......... 102 Figure 5-2 Finite conductivity Model of Horizontal Well (Non-uniform Crest Development)........................................ 96 Figure 4-9 Optimization of Oil Production Rate for DWS Well With Oil-free Water Drainage..................................................................................... 91 Figure 4-4 Relative Permeability Curves from Cores............... 89 Figure 4-3 Capillary Pressure Plot from Leverett J-function........................................... 96 Figure 4-10 Oil Contamination in Water sink Completion placed within Capillary Transition Zone............................................................................ 102 ix .......................................................... 79 Figure 3-20 Improved Oil Productivity Index with Rate.................... 93 Figure 4-6 Transition Zone Enlargement around Dual-completed (DWS) Well ... 98 Figure 5-1 Uniform Flux Model of Horizontal Well shows Uniform Water Crest Development along the Length of the Well......................................................................................Figure 3-15 Combined Effects of Capillary Pressures and Relative Permeability Hysteresis................. 92 Figure 4-5 Transition Zone Enlargement around Conventional Well ............................................................... 77 Figure 3-18 Oil Productivity Index (OPI) Comparison ................ 81 Figure 4-1 Reduction in Size of IPW Due to Capillary Transition Zone ............... 78 Figure 3-19 Oil Recovery Plot for Same Drawdown Pressure ..............................................95 Figure 4-8 IPW for LVT Reservoir With WC Isolines........ 88 Figure 4-2 Shows Zone of Concurrent Production of Contaminated Fluid in both Completion................................ 93 Figure 4-7 IPW Comprising Transition Zone Effect (LVT reservoir)................................ 73 Figure 3-16 Oil Recovery Performance Study of DWS Well................................................................................................................. Critical Rate lines Intercept and Diverge at Higher Rates Enveloping Region 4 of Twophase Inflow at the Top and Bottom Completion........................................................

................................124 Figure 5-12 Friction factor comparison (Model Vs weighted average) in oil-water flow with phase inversion ........................................................................ 112 Figure 5-5 Effect of Viscosity Model on Fanning Friction Factor ...........................................N... Alaska (Courtesy. 137 Figure 6-4 Multi-lateral Well Project in Purdue Bay.................. 130 Figure 5-16 New Water Crest Development ................. 135 Figure 6-2 Bilateral Horizontal Well Water Sink Option................. A..................... 2000) .................................................... 131 Figure 6-1 Tail-Pipe Water Sink Completion Schematic ............131 Figure 5-17 Weighted Average Approach shows Symmetrical Water Crest .... 122 Figure 5-10 Performance of 'Generalized Equation' With in-situ pipe Roughness Vs Blasius Equation for Smooth Pipe..................................... 126 Figure 5-13 Early Water Breakthrough in Horizontal Wells.............................................. 116 Figure 5-7 Effect of Perforation and Fluid Influx on Wellbore Friction ................................................................................... 121 Figure 5-9 Friction Factor Plot from 'Generalized Equation' Vs Yuan's Experimentally Derived Correlations (10SPF).. 128 Figure 5-15 Friction Pressure Loss Effect on Horizontal Well Bottom Hole Pressure ................... Bakerhughes website) ................. 136 Figure 6-3 A level 5 Multi-lateral Schematic for Tail-pipe Water Sink......... 142 x .... 128 Figure 5-14 Early Water Breakthrough but gradual Development of Watercut.......................................... 123 Figure 5-11 Performance of 'Generalized Equation' Vs Blasius Equation for smooth pipe.............................. ................................................................... 110 Figure 5-4 Effect of Mixture Viscosity Model on Reynold's Number . 113 Figure 5-6 Effect of Perforation Without Fluid influx on Wellbore Friction Factor................................................................. 141 Figure 6-5 Dual Completion Multi-lateral Well in Saudi Arabia (Al-Umair................................ 119 Figure 5-8 Friction Factor Plot from 'Generalized Equation' Vs Yuan's Experimentally Derived Correlations (5SPF).Figure 5-3 Oil-Water mixture Viscosity Evaluation............Skewed toward the Heel of the Horizontal Well.................................................................................................................................................................

........................................ 160 Figure 7-7 Oil Production with DWST Vs Horizontal Well ...Figure 6-6 Cumulative Oil Recovery Comparison.......................................... 148 Figure 7-1 Productivity Comparison between Vertical and Horizontal Wells ..................................................... 193 Figure B-7 Watercut Performance with High Rate at Top Completion ................................ 157 Figure 7-5 Cumulative Oil Recovery Performance Analysis ............................................. 154 Figure 7-4 Oil Production Rate Performance Evaluation ................ 150 Figure 7-2 Horizontal Well Drainage Area .......................................... 192 Figure B-6 Watercut Reduction Forecast with and without Capillary Pressure..... 161 Figure 7-8 Cumulative Oil Recovery Performance of Horizontal Well Vs Two DWS Vertical Wells ............................... 189 Figure B-2 Relative Permeability Curve for LVT Well Studies............................................................................ 195 Figure B-8 Watercut Performance of DWS in a Developing Cone Scenario..................................................................................... 191 Figure B-4 Watercut Reduction Performance of DWS Technology ................................ 196 xi ............................... 190 Figure B-3 LVT Well Reservoir Grid Model ......................... 164 Figure B-1 Capillary Pressure Plot for LVT Well ........ Conventional Horizontal Well....................................... 191 Figure B-5 Watercut Forecast of LVT Well without Capillary Pressure ............................................................. 146 Figure 6-7 Evaluation of Effective Length of Bilateral Water Sink Well ............................................. Horizontal Well with Tail-pipe Water Sink and Horizontal Well with Bilateral Water Sink ............ 162 Figure 7-9 Watercut Performance of Horizontal Well and Two Vertical DWS Wells ........................... 153 Figure 7-3 Elliptical Drainage Area for Horizontal Well ................... 158 Figure 7-6 Watercut Performance Evaluation of Vertical and Horizontal Wells ............

lim exponent [dimensionless] cross-sectional area [L2] constant [dimensionless] Oil formation volume factor [L3/L3] water formation volume factor [L3/L3] constant [dimensionless] main flow pipe inner diameter [L] pipe diameter [L] perforation diameter (siz) [L] change in height of water cone.NOMENCLATURE a A B Bo Bw Cn D d dp dX dr F f f Ms f MRI fo ftp fT fw fw fw. [L] change in circumference of cone arc. [L] dimensionless function friction factor (Fanning) [fraction] Moody pipe static friction factor [fraction] Moody perforated pipe friction factor with influx [fraction] fractional flow of oil [fraction] two-phase flow friction factor [dimensionless] total friction factor [dimensionless] wall friction factor [dimensionless] fractional flow of water [fraction] fractional flow of water at the inversion point [fraction] limiting water fraction at which the relative viscosity is 100 xii .INV fw.

gc H h ht ho hcb hct hp J(Sw) K k ke kg kh ko (ko)wc kv kr krnw kro kow (kw)or L - acceleration due to gravity [L/T2] height of transition zone above the oil-water contact [L] formation thickness [L] height of total formation thickness. [L] Leverett capillary function [dimensionless] Richardson constant [dimensionless] absolute permeability. [L2] relative permeability [fraction] non-wetting phase relative permeability [fraction] relative permeability to oil [fraction] effective permeability to water [L2] effective permeability to water at residual oil saturation [L2] length of horizontal well [L] xiii . [L] height of completion from top of formation. [L2] effective permeability [L2] effective permeability to gas[L2] horizontal permeability. [L] height of completion bottom from top of formation. [L2] effective permeability to oil [L2] effective permeability to oil at connate water saturation [L2] vertical permeability. [L] perforation thickness. [L] average oil column thickness.

[L3] isotropic reservoir flow rate. Qc QcH Qcv QD qi Qo Q(o. [L3/T] oil or water flow rate [L3/T] xiv . Qiso. [L3/T] critical rate of vertical well. [L3/T] critical rate of horizontal well.NRe NRem P Pc Pct pe pw PHydrostatic Preservoir Pwellhead DP PIH PIV Q Qani. [L3] critical rate.w) - Reynolds number [dimensionless] mixture Reynolds number [dimensionless] pressure [M/LT2] capillary pressure [M/LT2] threshold capillary pressure [M/LT2] boundary pressure [M/LT2] wellbore pressure [M/LT2] hydrostatic pressure [M/LT2] reservoir pressure [M/LT2] wellhead Pressure [M/LT2] differential pressure [M/LT2] productivity index of horizontal well [L4T/M] productivity index of vertical well [L4T/M] main flow rate. [L3/T] dimensionless flow rate influx through perforation [L3/T] oil flow rate. [L3/T] anisotropic reservoir flow rate.

[L] radius of capillary (L) drainage radius of horizontal well. (L) well drainage radius.qw r’ rc reH reV re rw Rw rwH rwV (S*w)drainage (S*w)imb Sr Sw Swc Swc(drain) Swc(imb) Swc(imb) Swc(scan) Swi tBT tD tDBT - water flow rate [L3/T] effective radius. (L) drainage case water saturation [fraction] imbibition case water saturation [fraction] residual saturation of the non-wetting phase [fraction] water saturation [fraction] connate water saturation [fraction] drainage connate water saturation [fraction] imbibition connate water saturation [fraction] imbibition connate water saturation [fraction] scanning curve connate water saturation [fraction] irreducible water saturation [fraction] time-to-water break through. (L) wellbore radius of vertical well. [L] wellbore flow resistance [dimensionless] wellbore radius of horizontal well. [T] dimensionless time dimensionless time-to-water break through. [L] wellbore radius. (L) drainage radius of vertical well. [T] xv .

[L] change in cone height. [L] dimensionless cone height (t D )SC BT u ut vm vso vsw XA Z dz ZD Greek Letters: a adip a’ a” b D d e F friction factor exponent [dimensionless] dip angle [degree] dimensionless factor for water breakthrough transformed dimensionless factor constant [dimensionless] Difference or increment increment pipe roughness [L] Potential or datum corrected pressure [M/LT2] xvi . (Sobocinski & Cornelius) velocity [L/T] total velocity [L/T] mixture superficial velocity [L/T] oil superficial velocity [L/T] water superficial velocity [L/T] location of a constant pressure boundary (L) cone height.(t D )BJ BT - dimensionless breakthrough time. (Bournazel & Jeanson) dimensionless breakthrough time.

[M/LT2] porosity. [M/L3] water density.dF f j m mm mo mw mr p q ro rw s sm so swo sw - change in potential. [M/L3] surface tension [M/T2] mixture surface tension [M/T2] oil surface tension [M/T2] water-oil interfacial tension [M/T2] water surface tension [M/T2] xvii .142 contact angle [degree] oil density. [fraction] perforation density [L-1] viscosity [M/LT] mixture viscosity [M/LT] oil viscosity [M/LT] water viscosity [M/LT] relative viscosity [dimensionless] constant = 3.

: Avg: bbl: BLPD: BOPD: Bt.: DBD: DL: DVWCON: DWS: DWSVW: E&P: EPA: FWL: HW: HWCON: HWWFIP: Analytical Average barrel Barrel Liquid Per Day Barrel Oil Per Day Break Through bottom Bilateral Water Sink capillary Completion Conventional Dual Bore Deflector Detection Limit Dual Vertical Wells CONventional Downhole Water Sink Downhole Water Sink Vertical Well Exploration and Production Environmental Protection Agency Free Water Level Horizontal Well Horizontal Well CONventional Horizontal Well With Friction and Infinite Porosity xviii .ABBREVIATIONS Anal.: btm: BWS CAP: Compln: Conv.

: ND: NPDES: OC: OCS: OOIP: OPI: O/W OWC: PAH: PI: ppb: PV: RCRA: RF: Sim.IMPES: IPW: JIP: LSU: mD: Mod.: Stb: SVWCON: IMplicit Pressure Explicit Saturation Inflow Performance Window Joint Industry Panel Louisiana State University milliDarcy Modified Not Detectable National Pollutant Discharge Elimination System Oil Cut Outer Continental Shelf Original Oil In Place Oil Productivity Index Oil in Water Oil-Water Contact PolyAromatic Hydrocarbons Productivity Index pounds per barrel Pore Volume Resource Conservation and Recovery Act Recovery Factor Simulator stock tank barrel Single Vertical Well CONventional xix .

: Tailpipe Water Sink United Kingdom United States Versus Watercut Water Hydrocarbon Ratio Water Oil Ratio Water in Oil Weighted xx .TWS UK: US: Vs: WC: WHR WOR W/O Wt.

This study identified two factors that could aid in a better modeling of the technology in old vertical wells – inclusion of capillary transition pressures and relative permeability hysteresis. field reports indicates that water breakthrough into horizontal wells could be quite dramatic and tend to erode the merit of high deliverability. However. It also shows the results of the application of this relation in the modeling of water cresting in horizontal wells subject to bottom water drive. xxi .5 billion dollars is spent annually to solve the problem of produced water in oil and gas wells.ABSTRACT Approximately 2. This study analyzed the problem of water cresting in horizontal wells and developed a “generalized compound friction pressure loss relation” for horizontal wells and pipes. Downhole Water Sink (DWS) technology is one industry solution to control water coning in oil wells. DWS technology involves the segregated production of oil and water through separate completions with zonal isolation packer. Another widely recommended industry solution to the problem of produced water is horizontal well technology. However. These results reveal an asymmetrical distribution of water influx skewed toward the heel in line with field observations. The new relation includes factors such as perforations. and radial influx of fluid into the wellbore as well as phase inversion. It also identified a pressure enhanced capillary transition zone enlargement around the wellbore as responsible for the concurrent production of contaminated fluid from both completions. several problems have been experienced in the application of the technology in watered-out oil wells. oil-water emulsions.

The results of the initial studies shows that oil recovery could be improved by as much as 7 percent over conventional horizontal wells.Finally. xxii . the study presents two innovative dual-completion concepts for controlling water cresting in horizontal wells adapting the principles of the Downhole Water Sink technology.

“Produced water is a fact of life in Louisiana. is produced water. as well as nationally.1 Background and Purpose In 1998.halliburton.CHAPTER 1 INTRODUCTION 1. as well as the capital investment in surface facility construction and to address other environmental concerns.2 billion barrels of produced water was generated compared to less than 200 million barrels of oil and condensate and a little over 200 million BOE (barrel oil equivalent) of gas 1 . In 1993. In 1993.us/statistics. produced water from 40 counties in the state of Colorado was estimated as 220. while 2001 production stood at 360 million barrels of water against 25. dispose or re-inject produced waters. These costs according to Halliburton include the expense to lift.5 billion barrels of crude oil as against about 25 billion barrels of associated unwanted water for the same year (Wanty.oil-gas.com. 1998-2001). In fact Kimbrell (2001) asserted that. as well as in the Middle East.6 million barrels in comparison to only about 22.).1997.S. The amounts of produced water are overwhelming compared to the amounts of hydrocarbons produced. 2001).the U.co. over 1. Similar problem of produced waters exists in the North Sea and in the Niger Delta Basin.5 million barrels of crude oil (www. The largest volume of waste associated with oil and gas production operations in Louisiana. The extent to which produced water problem is a big nuisance in the oil and gas industry is reflected in the fact that unwanted production of water has been estimated to cost the petroleum industry about $45 billion a year (www. produced about 2.46 million barrel of oil produced over the same period.state.

produced in the same period. The production of disposable oil-free water at the water sink completion in these old vertical oil wells with severe water coning history is also addressed. Further. Thus. Several research efforts have been directed at understanding the mechanism of water coning in vertical oil wells or water cresting in horizontal wells. the statewide water-hydrocarbon ratio (WHR) averaged approximately 3. This need requires advancement of Petroleum Engineering practice from sheer prediction of what is happening in the reservoir to solving the problem of cash flow. and prediction of watercut behavior with time after water breakthrough. From 1990 to 1993. The need for a practical solution that allows operators to produce oil at reasonable economic rate and yet live pleasantly with water production issues is paramount.2 (Kimbrell. This research presents a solution to this need. the research reviews the problem of water cresting in horizontal wells and investigates the feasibility of extending the DWS principle to controlling water cresting in horizontal wells and to recover some of the bypassed oil. The purpose of this research is to evaluate the performance of the DWS technology in old watered out vertical wells. the time to water breakthrough at production rates above the critical rates. Unfortunately. which is traditionally tied to oil production rate and ultimate recovery. the proffered critical oil rates are usually uneconomic for any practical purposes. the research represents an advancement of Petroleum Engineering practice from merely predicting water influx parameters to controlling water influx into producing oil wells at economic production rates and at the same time recovers some of the bypassed 2 . These research efforts have led to the calculations of the critical oil rate to avoid water coning. 2001).

here. the encroachment of the water into the producing oil well caused by the pressure drawdown around the wellbore creates problems over time. 3 . It has been established that effective planning and execution of well completions and well management could potentially eliminate the irritation of sand production in the industry. At the same time. 1.2 Statement of Research Problem The two recurring problems of critical concerns that require remedial solution and re-completion in the oil and gas industry are: (1) water shut-off repairs in bottom or edge aquifer support reservoirs. The production of oil and/or gas from a well causes pressure around the wellbore to drop. A very significant development has been made in the area of sand production control in vertical and deviated wells.oil. recommended for the control of water cresting in horizontal wells. A counteracting gravitational force. Oil reservoirs with bottom water drive have the advantage of a high oil recovery due to energy support from the aquifer. the problems associated with produced water. The technology has been tested in vertical wells and the principles are. and the ensuing bypass of recoverable oil still remains a challenge to the industry. These problems include water handling and bypass oil. which include the surface-handling problem. It involves the concurrent production of the oil and water through separate completions with zonal isolation packer. This creates a pressure gradient around the wellbore vicinity. On the other hand. and (2) sand control repairs in unconsolidated deltaic deposits. However. this research work reviews the performance of this concurrent fluid production technology in watered-out vertical oil wells with severe water coning history. due to the difference between the oil density and water density.

Joshi (1991) presented several calculations of critical oil rate for vertical and horizontal wells using different correlations and obtained results ranging from 57 stb/day to 84 stb/day for vertical wells and from 190 stb/day to 470 stb/day for a 1640 ft long horizontal well. After the initial water breakthrough in the well. Consequently.causes the oil-water contact interface to remain stable. While vertical wells act like point source concentrating all the pressure drawdown around the bottom of the wellbore. However. The capital outlay of oil production operation requires high oil production rate to ensure early capital recovery. horizontal drilling technology has been widely recommended as a solution to the development of these reservoirs experiencing high water coning problems. well penetration as well as the viscous forces. 4 . water production increases rapidly and oil production decreases. this rate is generally too low for any economic significance. 1971. mobility ratio. Bournazel and Jeanson. Over the passed two decades. 1991. The shape and nature of the cone development depends on several factors ranging from production rate. 1975). at a certain production rate. However. Studies indicate that there is a critical rate of oil production that mitigates water coning into the well – usually designated Qcrit (Joshi. horizontal wells act more like a line sink and so distribute the pressure drawdown over the entire length of the wellbore. Thus operators have to come to grips with the problem of producing water and oil. Chappellar and Hirasaki. the contact rises up until water breakthrough into the producing well. the viscous forces due to the pressure gradients around the wellbore become greater than the gravitational forces making the oil-water interface unstable. horizontal and vertical permeability.

The facts. recent evidence indicates that horizontal wells are. not free the problem of water production. actually becomes a disadvantage when water breaks through into the wellbore. However. One such problem is that the increased contact with the reservoir. other technologies commonly employed to minimize water production problems have not solved the problem of contaminated water handling or addressed the problem of bypassed oil in the reservoir. horizontal well water cresting has increasingly become a concern to the industry. 1. themselves. Existing solutions to the problem of water coning in vertical wells or water cresting in horizontal wells have either recommended uneconomic oil production rates or have not adequately addressed the recovery of bypassed oil due to water influx. 5 . remain that the reduced pressure drawdown is not a total solution to water coning. by their nature and geometry.3 Significance and Contribution of this Research The significance of this research is five folds. successfully. Conversely. Thus. · First.The result is reduced pressure drawdown around the wellbore and delayed water influx (cresting) compared to vertical wells. which is an advantage in terms of fluid production rates. the Downhole Water Sink (DWS) technology have proved to be very successful at recovering some of the bypassed oil while encouraging high oil production rates with reduced water handling in vertical wells. however. the research presents an advancement of the modeling of the DWS in vertical wells using existing commercial numerical simulators with all the capabilities of modeling several complex reservoir phenomena such as capillary pressure transition zone and relative permeability hysteresis. have inherent problems. Further. Horizontal wells.

4 Research Method and Approach This research used existing commercial numerical simulator (Eclipse 100/200) and a radial grid model to study the introduction of complexities such as capillary pressure transition zone and relative permeability hysteresis in DWS technology in watered out vertical wells. Thus. · Finally. · Third. 1. · Fourth. Two options were found adequate for most of the 6 . the research presents an extension of the downhole water sink technology principle to the control of water cresting in horizontal wells evolving two variants with the acronyms: (1) Tail-pipe Water Sink Technology (TWS). as well as phase inversion from oil dominated flow to water dominated flow.· Second this research presents a design and optimization criteria for clean drainage water production from the water sink completion in old oil wells due the presence of the capillary transition zone. the research presents a more robust friction pressure loss relation for horizontal wells and pipes (generalized friction pressure loss relation) incorporating factors such as perforations and perforation density. It assumes a steady-state process with adequate water influx from the aquifer. two-phase oil-water flow effects and the formation of emulsion. fluid influx into the wellbore. the first step was to evaluate and adopt an adequate steady state model for the commercial simulator. and (2) Bilateral Water Sink (BWS) technology. the application of the ‘generalized relation for friction pressure loss in horizontal wells’ has led to a more practical modeling of the problem of water cresting in horizontal wells showing the severity of the water crest at the heel of the well and bypassed oil at the toe.

The research developed a ‘generalized compound friction pressure loss relation’ for horizontal wells incorporating all of the variables of perforations. The IPW is a plot of fluid withdrawal rate at the top oil zone completion on the x-axis and fluid withdrawal rates at the bottom water zone completion on the y-axis. a fluid flow re-distribution technique that focuses on reducing the water influx at the heel has been developed to counter the crest development in this region and reduce the severity of oil bypass at the toe. Thus. 7 . An ‘equivalent pipe roughness function’ is then calculated from a re-arranged Colebrook friction pressure loss relation. due to limitations of existing commercial simulators in modeling horizontal well production. and flow regimes. The application of the ‘generalized friction pressure loss relation’ showed an increase in the water cresting around the heel of the well and bypassed oil at the toe. This ‘equivalent pipe roughness function is input into the commercial numerical simulator to give the same pressure loss effect in the wellbore. However. For horizontal wells. a Cartesian grid model with finer grids around the wellbore was adopted in the numerical simulator to model water influx into the wellbore. two-phase flow. This new compound friction pressure loss relation is then applied to model the performance of water cresting in horizontal wells. the research focused. fluid re-injection and the introduction of an infinitely large pore space at the bottom of the reservoir to simulate a bottom-water-drive reservoir. the Inflow Performance Window (IPW) was used as an adequate evaluation tool. in part. For most of the studies on DWS performance in watered-out wells. on the development of an effective tool for evaluating friction pressure loss in horizontal wells.studies. water influx.

It also presents a comparative study of oil recovery with DWS vertical wells and 8 . Chapters five and six are ground breaking areas of this research presenting a more robust tool for evaluating compound friction pressure loss in horizontal wells considering two phase oil-water flow and the concept of phase inversion. Chapter six ends with the recommendation of two techniques for controlling water cresting in horizontal wells.5 Work Program Logic This study is divided into eight chapters. Chapter two presents a literature review of scientific research into the problem of water coning and water cresting and industry solutions to mitigating these dual phenomena. Chapter seven reviews current industry practice of comparing the performance of a single vertical well with a horizontal well concluding that it does not compare apple for apple. Chapter four then takes a look at the environmental attraction of the technology in these old wells in terms of production of disposable oil-free water at the sink. It also presents a concise statement of the origin of water influx and ends with a presentation of the relevance of this study and approach. Chapter three gives an overview of the application of downhole water sink technology to the control of water coning in vertical wells with specific focus on application in watered-out wells and reservoirs with capillary transition zone. The friction pressure loss model presented in chapter five is easily adaptable to three-phase oil-water-gas flow by use of appropriate liquid-gas viscosity. The introduction chapter presents an overview of the problem of water influx into producing oil wells and the enormity of the problems associated with this unwanted phenomenon.1.

Chapter eight provides conclusions from this research work. 9 .horizontal wells in bottom water drive reservoirs. including recommendations for future research.

2.1 Water Coning in Vertical Wells Production of oil through a well that partially penetrates an oil reservoir underlain by water creates a differential pressure between the wellbore and the reservoir. Mathematical models have been developed to predict the performance of oil wells with water coning problems after water breakthrough. The breakthrough occurs when the cone shaped profile becomes unstable due to the high-pressure drawdown around the wellbore. Several correlations have been developed as a result of these research efforts to predict the time for water breakthrough. the cone height above the initial oil/water contact (OWC) also increases until water breaks through into the wellbore. Several research efforts and solutions have also been developed to reduce the level of severity of the post water breakthrough performance of the well. As production rate is increased. the critical oil rate to avoid water breakthrough and the post breakthrough behavior of the water influx at supercritical rates of production. Muskat and Wyckoff (1935) and 10 . This differential pressure causes the oil/water interface to deform into a cone shape (Figure 21).CHAPTER 2 LITERATURE REVIEW A survey of literature shows that tremendous amount of research work has been done ranging from experimental studies to analytical and numerical simulation studies in order to understand and predict water coning and cresting in vertical and horizontal wells respectively. An overview of research findings on the subject of water coning/cresting will be presented in this chapter.

Muskat. Wheatly (1985) observed that Muskat and Wyckoff relation over predicts the 11 . (1949) observed that coning is a rate-sensitive phenomenon. The equilibrium condition necessary to avoid water breakthrough requires that the vertical pressure gradient in the oil zone caused by viscous forces be balanced by the differential-gravity gradient in the underlying water zone. the water cone is termed “stable”. oil W ater Figure 2-1 Schematic Representation Water Coning in Vertical Wells Muskat and Wyckoff (1935) also determined the critical oil rate analytically solving Laplace equation for single-phase flow and for a partially penetrating well. However. which develops only after certain equilibrium conditions are unbalanced by increasing the pressure differential beyond critical limits. When this occurs.

also observed that the value of the well radius does not significantly affect the critical rate of production. and the potential. In effect. F. (or datum corrected pressure). Wheatly. The difference in observation could be due to the fact that Wheatly studied a partially penetrating well while Schol studied a fully penetrating well. 12 . but at a wellbore penetration about one-third the total oil-zone thickness for an isotropic reservoir. This is contrary to Schol’s (1972) observation. “If the vertical pressure gradient is higher than the hydrostatic pressure gradient of the water. the upward dynamic force resulting from wellbore pressure drawdown causes the bottom of the oil zone to rise to a height where the dynamic force due to pressure difference is balanced by the weight of water beneath this point. An important result of Guo and Lee’s study is that the maximum water-free production rate (critical rate) does not occur at zero wellbore penetration as one might expect from intuition. z. That is.critical oil production rate because it did not account for the presence of the cone when calculating the pressure distribution in the reservoir. As the radial distance from the wellbore increases. they assumed that the pressure distribution in the reservoir is the same with or without water coning. pressure drawdown decreases and so does the cone height above the original oil-watercontact creating a locus of cone-shaped balance-points called the water cone as shown in Figure 2-1. Archer and Wall (1986) observed that the maximum oil potential gradient possible for a stable cone could be written in terms of the cone height. the unstable water cone can be observed and the associated critical oil rate exists”. Guo and Lee (1993) demonstrated that the existence of the unstable water cone depends on the vertical pressure gradient beneath the wellbore.

the following condition holds: dF dF = dr dz 2-1 Where. the length of the cone arc.2. dr. and dF/dz = dF/dr and the cone becomes unstable. z. is equal to the height of the cone. In Figure 2-2b. In Figure 2-2a.433( r w . Summarizing. the length of the cone arc is greater than the height of the cone above the OWC (dF/dz > dF/dr) and so the cone is stable (no water breakthrough). is in specific gravity. Archer and Wall concluded that the viscous to gravity balance equation for cone height is given by: z= DF 0. is the length of the cone arc along its locus. they concluded that at the critical oil rate. however. in field units. dz.r o ) 2-2 (where r. is the vertical height of the cone above the OWC and. z. is the cone height in ft. r. and F is in psi) (a) Low Rate: Stable F 1 F2 dr dz Water Oil OWC (b) Critical Rate F dz dr 1 F 2 Figure 2-2 Cone Stability Analysis 13 .Using the illustration in Figure 2.

For the purpose of completeness. Bw is in bbl/stb) Bournazel and Jeanson (1971). the correlations are listed below: 14 . rw are in ft. Q = stb/day. kh. hct. mo is in cP. hp = ft) For time-to-water breakthrough evaluation.hcb ) 887.* (kv/kh). tBT. hcb. Qc is stb/day. Dr = psi/ft. several researchers have also proposed different correlations. = Qiso. r is in psi/ft. kh = mD.h p / ho ) mo 2 Qc = 2-6 And for anisotropic reservoir. and Bo. are in field units. are in ft. proposed a fast water coning evaluation relation for the critical oil rate for isotropic reservoir as: 5. developed an expression for the critical oil production rate as follows: Qc = 2pht k h k ro Drg (ho . and ho. re. Sobocinski and Cornelius (1965) as well as Bournazel and Jeanson (1971) proposed three-step correlations using dimensional analysis.2m o ln r 'Bo ö ÷ ÷ ø 2-3 æ ho . however. ho. The correlations are almost the same except for the evaluation of the dimensionless time. kv are in mD.hcb r ' = 4ho k h / kv ç ç h -h è o ct 2-4 ln r ' = [ ln[re /(rw + r ' )] 1 2 2 2 1 .(rw + r ' ) / re ] 2-5 (where r’. mo = cP. they simply recommended that: Qani. g = ft/s2.14 E (-05) * k h ho Drg (1 .Chappellear and Hirasaki (1975). (Equations 2-6 to 2-11.

645 E . and 0. Several significant variations exist in the results of the numerous studies on water coning predictions. 1971) ë 3 .69 E . = 0. 15 . 1991).6 for 1 < M < 10.or three-phase flow in porous media. 1991). Variations in the magnitude of 90 percent were recorded for the critical rate and as high as 140 percent for the time-to-water breakthrough (Joshi.p ÷ ho ø è ZD = (in field units) m oQo Bo 2-7 (t ) SC D BT 2 ù Z é16 + 7 Z D .3Z D = ê ú (Sobocinski & Cornelius.7 Z D û 2-9 Thus the time to water breakthrough is evaluated with Equation 2.æ h ö 3.r o )kh ho2 ç1 . Joshi (1991) evaluated the results of the various correlations for critical oil rate and time to water breakthrough.r o )kv (1 + M a ' ) 2-10 a'.04 * ( r w .10 using any of the dimensionless time evaluated from either Equation 2.0 M = m o (k w ) or m w (k o ) wc 2-11 M is the water-oil mobility ratio.0. t BT = m ofh(t D ) 1. (kw)or is the effective permeability to water at the residual oil saturation and (ko)wc is the effective permeability to oil at the connate water saturation (Joshi.8 or 2.9. These variations arise mainly due to the approximating assumptions used to simplify the complex mathematical solutions of the two.2Z D û 2-8 é ZD ù (t D ) BJ BT = ê ú (Bournazel & Jeanson.0.04 * ( r w .5 for M < 1. 1965) 4ë 7 .

Some of the methods are briefly described below. the common points of agreement of all the studies on vertical well water coning could be summarized as follows: 1. High pressure gradient/high fluid velocity (production rate) stimulates water coning. A corollary is that more money is spent on infill drilling to recover bypassed oil reducing the economic gains of the field.2 Other Industry Solutions for Water Coning in Vertical Wells Several practical solutions have been developed to try to minimize the level of severity of this water coning in vertical wells. Completion at the topmost part of the oil zone (preferably the top 45 percent) away from the oil-water-contact delays the time to water breakthrough. The approach. 3. is to delay the time to water breakthrough such as increasing the distance between the bottom perforation and the original oil water contact or minimize the amount of water in the wellbore thus reducing the hydrostatic head. The development of the water cone creates bypass oil problem and consequently shortens the producing life of the well. After water breakthrough in the wellbore.However. basically. 2. 4. 6. water production increases rapidly and significant resources and efforts have to be put into water disposal and oil/water separation facilities. 16 . 2. Critical oil rates for water-free oil productions are typically uneconomical for all practical purposes. 5.

Several water production control chemicals have been developed for different applications.2.2. High permeable sand layers in contact with the water zone are often times responsible for the high water influx and these could be isolated by cement squeeze operation during workover to minimize the level of water production. In some cases.halliburton. In some cases. While several publications have appeared giving case histories of field applications of these technologies.2 Conformance Control Technology “Reservoir conformance is the process of applying various methods and technologies to a reservoir or wellbore to reduce or control unwanted water or gas production so that recovery efforts are efficiently enhanced and operator profitability improved” (www.1 Perforation Squeeze-off and Re-completion In cases where the reservoir deposition is such that shale barriers are inter-bedded with the sandstones as in laminated sands. the long17 . application of fluid mobility modifiers could be considered in the form of conformance technology. et. In such cases. the shale barriers could form effective seal/separation between the sand layers. Cement squeeze operation may not be feasible or effective if adequate zonal isolation is not possible due to absence of shale barrier streaks. Other applications include selective relative permeability modifier that is capable of reducing relative permeability to water without affecting relative permeability to the oil and/or gel treatment. matrix fracture sealants are injected into the formation. described next.al. if the production of unwanted water is traceable to fracture paths. Matthews. 2001. 2.com.2. 1996).. an entire perforation is completely squeezed off and the well re-completed higher up the structure away from the oil-water-contact (OWC).

term effect on reservoir properties and overall well performance remains a controversy to industry operators. Ehlig-Economides et. al (1996) observed gel treatment to build a coning barrier has been found to be ineffective and uneconomical in certain fields.
2.2.3 Downhole Oil-Water Separation Technology

Another industry solution to the problem of unwanted water production is the development of Downhole Oil-Water Separation technology. This technology uses hydrocyclone separators and special design downhole pumps installed in the completion (production) string to separate the oil/water mixture within the wellbore. The elimination or reduction of water hydrostatic column by this method means more energy is available to the reservoir in terms of drawdown pressure. The result is increase in oil production rate. To effectively understand the concept of downhole water separation technology, the basic nodal analysis equation is listed as follows:
Pwellbore = Preservoir - DP = PHydrostati c + Pwellhead

2-12

The above equation rearranged gives:
DP = Preservoir - Pwellhead - PHydrostati c

2-13

Assuming a constant reservoir pressure and constant wellhead pressure, Equation 2.13 indicates that a reduction in the hydrostatic pressure gradient in the wellbore translates to an increase in the drawdown pressure in the reservoir. Following Darcy equation (2.14), the increased pressure drawdown translates to an increase in oil production, assuming all other reservoir properties remains constant.
Q= 0.00708kh ( DP ) mB ln(re / rw )

2-14

18

Q is in stb/day; k is in mD; h is in ft; Dp is in psi; m is in cP; re, rw are in ft, and B is bbl/stb. The technology provides reduced surface water handling. The fundamental problem of water interference with oil production within the reservoir creating bypass oil still remains unresolved with this technology. In other words, the problem of bypassed oil by the water development has not been solved by this technology.
2.2.4 Total Penetration Approach

In this approach the perforation interval is extended to cover the entire oil zone and into the bottom water zone. The aim is to maintain radial flow of fluid and so avoid cone development and the attendant oil bypass. Production of water starts immediately and so requires increased water handling facilities. As production increases, over time, the tendency for cone development is inevitable (Ehlig-Economides, et. al., 1996). Also, the co-mingled production of high water volume and oil in one string could create unwanted environmental problem caused by the disposal of the contaminated water. All or most of the above solutions addresses only one aspect of the two-prong problem of water coning namely; (1) increased watercut and water handling problems and; (2) bypassed oil in the reservoir as a result of water coning/cresting around the wellbore.
2.3 Water Coning Control with Dual Completions in vertical wells

Pirson and Mehta (1967) presented the results of studies performed using numerical simulators. The studies investigated several solutions to the problem of water coning in vertical wells. One of the solutions is the re-injection of produced oil into the reservoir below the oil zone perforations to suppress the development of the cone. This 19

technique known as the “ Oil Doublet Model” was not attractive economically. Another option investigated by Pirson and Mehta is the selective production of oil and water from their respective zones using dual completions. They concluded that, though the approach reduces the growth of the cone, it gives the same overall produced water-oil ratio at all times. Widmyer (1955) patented a water coning control technique in which he proposed a dual completion string to separately produce the oil and water zones and thus counter the cone development. He suggested perforating the top and bottom of the oil zone and producing each perforation with a production string. Considering the cost of the dual completion string, Driscoll (1972) suggested a variant of the dual perforation technique. He suggested two perforations – one in the oil zone and one in the water zone below the original oil-water contact. Fluid from both perforations is to be co-mingled in one production string with the possible use of a packer and adjustable flow choke to control oil/water rates. The demerit of this approach is the reduction in oil rate as a result of the increased hydrostatic head of the co-mingled fluid. The problem of separation of the produced fluid and disposal of the contaminated water makes this technology not so different from the conventional completion. It, however, solves the problem of bypass oil in the reservoir. Ehlig-Economides et. al (1996) observed that the concept of critical rate is a misnomer as water is bound to be produced in any reservoir with strong bottom water drive. They also observed that total penetration and dual penetration method of completion yields the most of oil production and recovery but at a cost of handling high rate and volume of water production.

20

In the dual completion technology with downhole water sink (DWS), the top completion is placed as high as possible, within the top 20 percent of the oil zone and the second perforation placed at some depth below the oil water contact. As will be seen in chapter 4, the location of this bottom completion (the water sink) is a very important determinant of the environmental and production efficiency of the technology. Fisher et. al (1970) used a numerical simulator to study the application of the dual selective production completion in the Bellshill lake, Blairmore Pool in Canada and concluded that it can reduce the water cone development and even eliminate it completely in certain cases. Wojtanowicz, Xu and Bassiouni (1991) and Wojtanowicz and Shirman (1997, 1998), conducted theoretical studies on the hydrodynamics of water coning control using a production well with tailpipe water sink (DWS). The model used flow potential distribution generated by two constant-terminal-rate sinks located between the two linear boundaries of the oil-water contact (OWC) and no-flow top boundary in the oil zone and constant-pressure outer radial boundary representing a steady state process.

Water Oil PCP 18’ 3’ Oil Water 7’ 5’ 64’ OWC

Figure 2-3 Schematic of the dual completion for the Nebo Hemphill project

21

Shirman and Wojtanowicz (1998) reported that more oil could be produced with less water in the top completion while producing oil-free water in the lower completion. 2. in Louisiana (Swisher.4 Capillary Pressure and Relative Permeability Hysteresis in DWS Water Coning Control Researchers have largely ignored capillary pressure transition and relative permeability hysteresis effects on the phenomenon of water coning. incorporating these effects complicates the mathematical solutions. as the in-situ gravity segregation in the Kern County. Thus they could not effectively model the performance of old oil wells with severe water coning history. al (1997) reported the result of the installation of the technology by Texaco Inc. For most of the analytical studies. The first field application of the technology was performed in 1994 at the Nebo-Hemphill Field located in LaSalle Parish. A limitation of these studies is that they lack the ability to analytically represent the oil/water transition zone (capillary transition) and related effects of water saturation changes on oil contamination in the produced water.The studies indicated that oil recovery could be improved by efficient water production from the bottom completion. but was actually capable of reversing a water cone after breakthrough. It also shows that the bottom completion can produce oil-free water. The result of the field application indicated that the technology did not only prevent water coning. 1995). Letkan and Ridings (1970) and Mungun (1975) included the effect of capillary pressures in their numerical coning model but did not report on the effect of the inclusion or non22 . Bowlin et. The well performance showed a remarkable improvement in oil production rate and total recovery over conventional wells. California.

Effectively. While vertical wells act like point source concentrating all the pressure drawdown around the bottom of the wellbore. the average water saturation is governed by the amount of mobile water and hence the height of the transition zone. The result is reduced pressure drawdown around the wellbore. He concluded that the presence of a capillary transition zone increases the amount of produced water.5 Horizontal Well Technology Several researchers have recommended horizontal well technology as a solution for the development of reservoirs with water coning problems. to a great extent. They observed that the vertical permeability determines. horizontal wells act more like a line sink and so distribute the pressure drawdown over the entire length of the wellbore. Kurban (1999) evaluated the performance of DWS wells in the presence of capillary transition zone and relative permeability hysteresis and concluded that both concepts have significant effect on the performance of the technology. Kurban (1999) did not report on the mechanism of the effect of these parameters on DWS well performance.inclusion in the models. The nature of fluid influx into a horizontal well remains 23 . the time to water breakthrough and the rate of decline of oil production after water breakthrough. Reed (1984) investigated the mechanism of water production in a high permeability sandstone reservoir with a capillary transition zone comparable to the formation thickness for a conventional well. They also concluded that the shape of the relative permeability curve has no significant effect on the performance of the wells. Miller and Rogers (1973) studied the performance of oil wells in bottom water drive reservoirs. He concluded that watercut is principally dependent on the average fractional flow of water and thus the average water saturation in the drainage area of the well. 2.

The most controversial and least accepted of all three is the uniform flux assumption. Infinite conductivity assumes there is no pressure loss in the wellbore. Researchers have presented three models to explain the flow pattern in a horizontal wellbore. Gilman (1991). Uniform flux assumes that the influx of fluid into the wellbore from the reservoir in constant along the length of the horizontal well. Several authors. uniform flux and finite conductivity. While this may be true under certain conditions. They argued that the productivity calculated from the uniform flux model is an underestimation. Research efforts have led to the development of mathematical equations for the evaluation of the performance of horizontal wells.a challenge to analysts. the ratio of wellbore friction pressure loss to reservoir pressure drawdown could be quite significant in some other conditions that its effects cannot be neglected. Babu and Odeh (1988) developed a generalized expression for evaluating the productivity of horizontal wells assuming uniform influx of fluid along the wellbore. The finite conductivity assumption is more encompassing as it incorporates the effect of friction pressure loss in the wellbore and changes in the distribution of fluid flux along the wellbore. such as low permeability – large wellbore conditions. Suprunowicz (1992. They are: infinite conductivity. 1993) reacted to this paper questioning the validity of the uniform flux model. Geiger (1984) developed a correlation for evaluating the productivity index of single-phase oil production or the pre-water breakthrough oil production rate in an isotropic reservoir (Equation 2-15). Geiger (1984) 24 . Proponents of the infinite conductivity model argue that the effect of friction pressure loss along the wellbore is small compared to the reservoir pressure drawdown and so is negligible. Peaceman (1991).

ì ï ' ï k Lï 1 PI H = 0.0145* í m ï æ 1 + 1 .L ÷ ç 2reH ÷ h æ h lnç ÷ + lnç ç L ÷ L è 2prwH ç 2reH ÷ ç ø è 2-16 Consequently. Geiger (1984) developed a correlation to compare the performance 2-17 ö ÷ ÷ ø (where.also gave the productivity index of a vertical well (Equation 2-16) in the same paper in order to compare the performance of vertical and horizontal wells. Equation 2-17 has been widely accepted by reservoir engineers in evaluating the performance of a horizontal well and the need or lack of it in field development planning. rwH and rwV are the respective wellbore radius of the horizontal and vertical wells. in an isotropic reservoir as: æ reV ö lnç çr ÷ ÷ PI H wV ø è = 2 ö PIV æ ç1+ 1. 25 .(L / 2r ) 2 eH ç ï( L / h) lnç L /(2reH ) ï è î [ ]ö÷ + lnæç ÷ ø h ç 2p * r wH è ü ï ï ï ý öï ÷ ÷ï øï þ 2-15 æ ö ç 2pkh 1 ÷ PI v = ç ÷ m ç ln rev ÷ rwv ø è of vertical and horizontal wells of length. L is the length of the horizontal well and h is the reservoir net thickness). reH and reV are the respective drainage radius of a horizontal and vertical well. L.

2.89 E . Papatzacos et. Another advantage of horizontal wells frequently advanced by analysts is that the standoff from the bottom water can be maximized with horizontal well than for vertical wells.0616438(a”) – 0. for predicting cone evolution toward the horizontal well in the transient state.9624955 + 0. Chaperon generated a table of data for different scenarios. (1986) also presented useful correlations for evaluating the critical oil rate of horizontal wells in a steady-state process. with a maximum error of 44 percent for the function F as follows: F = 3.04 * æ kH h 2 ö L DPç çm B ÷ ÷F XA è o oø 2-18 Where 1 ≤ a” < 70 and 2XA < 4L a" = X A kV kH h 2-19 For the dimensionless function F. al. To develop a correlation for the time to water breakthrough. In studying the effect of coning toward vertical and horizontal wells in anisotropic formation. Geiger (1989) and Karcher et.6 Horizontal Well and Water Influx Control The reduced pressure loss in the wellbore causes a delay in water influx in horizontal wells compared to vertical wells.000540(a”)2 2-20 Like Muskat. Chaperon neglected the influence of the cone shape on the flow pattern. al. Joshi (1991) developed a correlation. they assumed a constant pressure on the 26 . Chaperon (1986) developed a correlation for the prediction of critical oil rate in horizontal wells as: QcH = 4. (1989) developed an analytical method. However. assuming a moving boundary approach.

they neglected the friction pressure loss in the wellbore. (1986). Yang and Wattenberger (1991) developed several correlations from the result of numerical simulation studies for predicting water-oil-ratio (WOR) as well as the critical rate and time to water breakthrough for horizontal wells considering a closed outer boundary reservoir in transient state. Broman et. In other words. To reduce the complexity of the mathematical solutions. QD.r o ) kV k H 2-22 Papatzacos (1991) later developed correlations for critical rate calculations in horizontal wells using the gravity equilibrium assumption on the moving boundary. (1990) and Cunningham et. al.moving boundary. (1992) reported on the success of the application of horizontal wells in the re-development of the Prudhoe Bay field in Alaska. Reviewing field performance of horizontal wells. Nevertheless.r o ) t BT hfm o 2-23 m oQo Bo hL( r w . they are useful tools in analyzing the performance of horizontal wells. al. al. Murphy 27 .2284kV (r w . The result of their analysis yielded an expression for the dimensionless time to water breakthrough as: tDBT = 1/6QD 2-21 where. Sherrad et. all the analytical equations presented in Equations 2-15 through 2-23 assumed an infinite conductivity concept.12 and t DBT = 0. the dimensionless flow rate is given by Equation 2-22 and the dimensionless time to water break through is given by Equation 2-23: QD = 39. Both solutions refer to an infinite acting reservoir.

although critical rates are typically higher for horizontal wells than for vertical wells and the time to water breakthrough is also longer for horizontal wells than for vertical wells. the typical high production rates expected of horizontal wells soon creates the problem of unwanted water influx. Thus typical production rates are usually higher than the critical rates for these wells. One such problem is that the increased contact with the reservoir. Several other countries. the typical critical oil production rates to avoid water influx to the wellbore are also too low for any economic purpose. Horizontal wells. have inherent problems. Like their vertical well counterpart. Recent field evidence shows that the fact of reduced pressure drawdown is not a total solution to water influx. 1991) on the Troll Field in Norway amongst others. Lien et. The uniform flux model assumes that bottom water rises uniformly and 28 . by their nature and geometry. Peng and Yeh (1995) also concluded that the use of horizontal wells is a proven technology for reducing coning problems and improving recovery in reservoirs underlain by water. actually becomes a disadvantage when water breaks through into the wellbore causing a very rapid increase in watercut. The nature of water influx into horizontal wellbore remains a challenge to researchers. Chen (1993) in Oman.(1988) also reported the success of horizontal well in producing the Helder Field in Netherlands. have used the horizontal well technology to improve oil recovery from reservoirs with strong bottom water drive. al (1990. In other words. The result is that high mobility bottom water invades the oil zone and ultimately moves toward the well. most of these reports have been written quite early in the life of these fields developed with horizontal wells. However. which is an advantage in terms of fluid production rates.

2000).. Chaperon evaluated the case of laminar flow in a pipe and considered the pipe 29 . Friction pressure loss in the wellbore have been adduced as a reason for the skewness of horizontal well flow profile toward the heel (Guo and Lee. several research projects conducted in the past 10 years have shown that approximating horizontal wells with horizontal pipe flow is grossly inadequate (Su and Gudmundsson. 2. 2000). The pipe roughness can be much greater in horizontal wells than for regular pipe.. al. 1993. 1999. al. This is quite understandable considering the effect of perforations and the fluid influx along the well length. Chaperon (1986) was the first to consider the effect of pressure drop along the wellbore. Yuan (1997) recommended that the regular pipe friction factor correlations should not be used to predict the pressure drop in a horizontal well.forms a symmetrical crest along the well length. Most reservoir simulation tools neglect the friction pressure loss component or at best approximate it with friction pressure loss in horizontal pipes using the Fanning friction factor or Moody friction factor (Eclipse Manual. 1997.7 Friction Pressure Loss Consideration in the Horizontal Wellbore In order to evaluate the effect of the friction pressure loss component. More of horizontal production (over 80 percent) occur within the first 50 percent of the well length indicating that water breakthrough will first occur at the heel and spread gradually toward the toe of the well (Lien et. VIP. 2000. Yula et. However. 1991). researchers and industry operators have been faced with another challenge of developing an adequate correlation for the friction pressure loss. Yuan. 1993). Physical evidence from production logs in horizontal wells with water cresting problem show that this is not the case.

p1.15 E10 * r 2 (k is in mD) 2-24 The pressure drop in the wellbore. However. however. For a well of length 1500 ft and ro = 0. therefore. The consequence is that critical rate predicted by the link sink approximation method will be overestimated. Joshi (1991) performed several calculations to try to show that friction pressure loss in the wellbore is small and. r. Guo et. L is in ft.45cP.inner bore. (1993) observed that the line sink theoretical approximation of the horizontal well is inadequate as it assumes the infinite conductivity model. Several parameters such as perforations 30 . On the other hand. negligible.125 bbl/stb.000 stb/day. Bo = 1. to behave as a medium of permeability where the pipe bore equivalent permeability is given by: k = 1. mathematical models used for evaluating friction pressure loss in pipe are inadequate in horizontal wells. used correlations developed for pressure loss in pipe flow that are considered inadequate for horizontal wellbore flow conditions. al. they did not validate their observation. They observed that the inclusion of wellbore friction pressure loss causes a pressure gradient in the pressure drawdown in the formation along the wellbore at the downstream end of the horizontal well. p2 are in psi and ro is in ft. (DP in psi) could be obtained by re-arranging the Poiseuille’s law for total flow rate through capillary as in Craft and Hawkins (1991): DP = QBo m o L 1. DP calculated from the above relation (Equation 2-25) is 0.417 ft. and mo = 1. For a flow rate of 5. These calculations.30(10)10 pro4 2-25 where Q is in stb/day.01 psi.

25 in smooth pipe flow. f = 0. In order to accommodate the increased friction pressure loss proposition. Dikken assumed that the effect of turbulence and increased pressure loss could be approximated by varying the factor.15 as against 0. The use of the Blasius type correlation for smooth pipe neglects the in-situ pipe roughness. They concluded that pressure loss increment in the magnitude of 15 percent or more is 31 . fluid influx into the wellbore and oil production rate affects the flow geometry of the produced liquid which in turn affects the level of pressure loss in horizontal wells. Typical commercial pipes used in horizontal wells have manufacturer recommended in-situ pipe roughness that poses a resistance to the flow of fluid. especially when the high production rates of horizontal wells is taken into consideration. a. in the typical Blasius equation for friction pressure loss in pipe from between 0. Su and Gudmundsson (1993) developed a correlation to evaluate the extent of increased pipe roughness due to perforation and consequent increase in pressure loss. The consequence is that pressure loss in horizontal wellbore is of a higher magnitude than in conventional pipe flow. al (1992) proposed a correlation for friction pressure loss in horizontal wellbore using a Blasius-type pressure loss equation. The Blasius equation for friction pressure loss in smooth pipes is presented in equation 2-22. Dikken (1990) observed that laminar flow assumptions for horizontal wells are inaccurate when the radial influx of fluid is considered.0791 a N Re 2-26 Asheim et. He observed that the flow geometry is rather turbulent in the wellbore.roughness. They incorporated an expression for the ratio of fluid influx to main wellbore flow.0 to 0.

They concluded that pressure loss in the wellbore of horizontal wells could be quite significant compared to that of pipe flow. due to the fact that the technology is relatively new. 2. Gonzalez-Guevara and Camacho-Velazquez (1996) and Yula (2000) performed experimental studies on the effect of pressure loss along the wellbore on the productivity of single-phase fluid in horizontal wells. (1989) observed that the euphoria of horizontal well technology for the development of thin oil columns in an unconsolidated sand deposit in a deltaic 32 . al (1993. Yuan (1994. However. This is.8 Water Cresting Control in Horizontal Wells The subject of control of water cresting in horizontal wells is yet to receive much attention as in vertical wells. the commercial numerical simulator employed in McMullan and Larson studies uses the Fanning friction pressure loss correlation for fluid flow in pipe. McMullan and Larson (2000) also performed numerical simulation of the horizontal well productivity incorporating friction pressure loss along the wellbore. Unfortunately. the experience of water cresting in the horizontal wells over the past few years have prompted some level of research into the development of possible solutions. 1999). (1998. Yuan et. using a 2-inch diameter pipe was quite significant. 1999). Chugbo et. The summary of the above literature review is that friction pressure loss in horizontal wellbore cannot be adequately represented by the correlations for pressure loss in conventional pipe flow and that friction pressure loss in horizontal wells could be quite significant. 1997). Ozkan et. which has been considered inadequate for effective evaluation of friction pressure loss in horizontal wells. perhaps.possible. al. The result. al.

they did not report the result of further studies to validate this claim. It is 33 .environment may be erroneous and dangerous. They concluded that inserting a stinger about 0. a better understanding of the flow regimes for oil-water flow in horizontal pipes is necessary. (1997) investigated the horizontal well completion with a stinger introduced into the perforated pipe to redistribute the crest profile along the wellbore. Renard et.. Asheim and Oudeman (1997) recommended an optimal perforation scheme that reduces perforations at the heel and increases perforation density toward the tail end and so redistribute the fluid influx model to possibly achieve uniform influx pattern. (1997) recommended drilling of multilateral wells instead of several horizontal wells to economically increase and accelerate the overall recovery of oil in reservoirs underlain by an active bottom aquifer.25 times the length of the horizontal well provides a better pressure drawdown distribution and therefore more uniform flux along the wellbore. the flowing fluid mixture changes matrix from water-in-oil (W/O) flow to oil-in-water (O/W) flow. al. Evaluations of the various flow patterns indicate that at a certain fractional flow of water. She did not present the results of research to validate the proposition. Permadi et. al. Six different flow regimes have been identified for oil-water flow mixtures in horizontal pipe (Trallero. The six regimes are summarized in Figure 2-4. Unfortunately. This point of flow regime transformation has been termed “the inversion point”. Ehlig-Economides (1996) proposed that a dual horizontal well completion.9 Flow Regimes of Oil-Water Flow in Horizontal Pipes/Wellbore In order to understand the properties of the evolving mixture fluid. one in the oil zone and one in the water zone could reduce water cresting problem in horizontal wells. et. al. 2. 1996).

In the latter case. Thus. Considering the radial influx of fluid along the length of the well. the flow regimes will be that of various level of dispersion of one phase in the other. the dominant forces become that of slippage between the flowing mixture components.characterized by a sudden change in the mixture viscosity that in turn affects the frictional pressure loss. 34 . It has also. Oil-Water Flow (a) Segregated Flow (a-i) Stratified Flow with some mixing (a-ii) Stratified Flow (b) Dispersed Flow (b-i) Water Dominated Flow (b-i-1) Dispersed oil in water (b-i-2) Emulsion of oil in water (b-ii) Oil Dominated Flow (b-ii-1) Emulsion of water in oil (b-ii-2) Dual dispersion Figure 2-4 Oil-Water Flow Regimes in Horizontal Pipes In the flow conditions prevalent in horizontal oil wells with the fluid influx models. oil-water mixture viscosity is more appropriately described by emulsion viscosity relations. however. been observed that friction pressure loss effect is predominant in the W/O flow regime than in the O/W regime. the flow process is rather dominated by unstable emulsion due to turbulence. it is reasonable to conclude that the presence of a stratified flow in horizontal wellbore is almost impossible but rather.

the viscosity relationship depends on the continuous phase. Martinez et. They observed. Most commercial numerical simulators have adopted weighted average approach for the mixture viscosity also. The density and fluid velocity are evaluated as weighted average of the flowing mixture as a function of the fractional flow. However. al. In fact their experiments indicates the mixture behavior follows that of a power-law fluid. that a flowing oil-water mixture behaves more like a non-Newtonian fluid. However. al. 2-phase oil-water flow has not received commensurate attention as in gas-liquid flow. Arirachakaran et. They observed two broad classification of oil.10 Friction Pressure Loss of Oil-Water Flow in Horizontal Wells Most pressure loss relations have been developed for single-phase flow or for 2phase gas-liquid flow conditions. The flow of oil in the presence of increasing watercut creates interfacial drag forces that tend to change the mixture viscosity. Studies indicate that the basic equation for the evaluation of the Reynolds number is all that is affected. Richardson (1950) also developed a correlation for the evaluation of emulsion viscosity in oil-water flow systems. Broughton and Squires (1937) presented several correlations for the evaluation of changing viscosity of oil in the presence of increasing watercut in a turbulent emulsionlike flow regime. (1988) studied the dispersion viscosity of oil-water mixture flow in pipes. (1989) explained that the relation for flowing oil-water mixture viscosity is a complex one that cannot be represented with the weighted average concept. They also proposed that oil-water mixture exhibits a nonNewtonian rheological behavior and so apparent viscosity of the mixture cannot be 35 .water flow systems namely oil dominated and water dominated flow with a phase inversion point. the mixture viscosity component is more complex and cannot be evaluated as the weighted average.2.

Schramm (ed. They concluded that fully implicit model accepts larger time increment sizes and is more efficient for problems involving high capillary forces but requires more computer time.) (1992) stated that the mixture of two immiscible liquids flowing under turbulence forms an emulsion when one of the phases becomes a collection of droplets dispersed in the other phase. Arirachakaran et al. 2. MacDonald and Coates (1970) performed water coning studies using implicit pressure-explicit saturation (IMPES) analysis with the production term treated explicitly and a fully implicit analysis. whereas slip consideration is critical in the water-dominated flow. (1989) developed a mathematical model for the evaluation of the phase inversion point.calculated using weighted mixing rules. Yang and Wattenbarger (1991) used radial model with logarithmic grid distribution for vertical wells and a 3-Dimensional Cartesian model for horizontal well studies with finer grid distribution around the wellbore and coarser grid away from the wellbore. RØnningsen (1995) studied eight different North Sea crude and developed a Richardsontype correlation for evaluating emulsion viscosity of oil-water flow in pipe. They also recommended radial model with fine grid around the wellbore for vertical well conceptual studies. They observed that in the oil-dominated flow regime. Several other authors including Wu et.11 Numerical Simulation of Water Coning in Wells Several researchers have employed numerical simulators to study the complex phenomenon of water coning/cresting in vertical and horizontal wells respectively. al (1995) and McMullan and Larson (2000) used a 3-Dimensional Cartesian model with finer grid in the oil zone and coarser 36 . friction pressure loss is critical. Chappelear and Hirasaki (1974) also used radial model for studies of water coning in vertical wells.

5)..12 Current Methods of Controlling Water Cresting in Horizontal Wells Various methods have recently been recommended by researchers to restore the productivity of horizontal wells. al. Permadi et. They also reported that the rate reduction due to the insertion of the un-perforated stinger is relatively small and is overcome by the much longer delay in the time to water breakthrough (Permadi.al. Thus the use of radial model with logarithmic grids for modeling vertical well conceptual water coning studies and a 3-dimensional Cartesian grid model with fine grid distribution at the oil zone for horizontal well is employed in an implicit function commercial numerical simulator for this study. 1997). Bowlin et. 2. et. They recommended a stinger outside diameter-to-liner inside diameter ratio of 0.667 as most appropriate for optimum 37 . 2.25 times the producing length of the horizontal wellbore is adequate to redistribute the pressure drawdown and have a more uniform flux along the wellbore.1 Horizontal Well Completion With Stinger This method involves redistributing the pressure losses along the wellbore by inserting a piece of pipe of a smaller diameter into the completion/production liner (Figure 2.grid in the water zone together with implicit type commercial numerical simulators for horizontal well studies. A brief overview of each of the current technology is presented in the following section. studied this approach both numerically and with a Hele-Shaw physical model and concluded that a stinger length of 0. (1997) and Kurban (1999) also employed radial grid systems to evaluate the performance of Downhole water sink technology in vertical wells.12. This smaller piece of pipe is typically called “the stinger”.al.

12. In other words. They recommended that the completion method be used in horizontal wells where frictional pressure losses along the perforated section of the well restrict the production performance (Brekke.2 Water Cresting Control With Variation of Perforation Density Another scheme commonly recommended for controlling water cresting in horizontal wells is variation of perforation-density to uniformly distribute the influx of fluid (Figure 2.combination of annulus pressure loss and production rate increase. Figure 2-5 Horizontal Well Completion with a Stinger Brekke and Lien reported that the stinger completion method. the perforation density is less at the heel than at the toe of the horizontal well. the completion of the horizontal well with stinger creates an improved sand face pressure profile by controlling the inflow of fluid along the wellbore. 38 . and Lien. could provide up to 25 percent increase in well productivity during the first part of the production life. they observed that inserting the stinger was capable of delaying the time to water breakthrough in horizontal wells by as much as four times. 1992).6). For most of the pipe length studied. 2. Thus. in combination with reduced perforation density. This process helps to redistribute the frictional pressure loss along the perforated part of the well.

the optimum distribution is when each segment has the same inflow. For the case of water cresting control. However.Asheim and Oudeman investigated the option of varying the perforation density in order to create a uniform production and injection profile along the wellbore. 1997). and Oudeman. the perforation density is least at the heel of the well and increases toward the toe. The perforation density in each segment is varied (either by varying size of perforation or by varying the number of perforation for the same hole sizes) until the optimum distribution is achieved. However. the well is partitioned into small number of uniformly perforated segments. the sweep efficiency is higher and there is attendant delay in water breakthrough time and improved ultimate recovery (Asheim. the perforation strategy of most interest is uniform fluid inflow from the reservoir into the well. To accomplish this type of optimization. 39 . They observed that the uniform inflow design may involve some marginal reduction in productivity compared to uniform perforation density method. Heel Figure 2-6 Schematic of Typical Variable Perforation Density Toe Marett and Landman (1993) observed that in cases where water or gas cresting is anticipated.

3 Water Cresting Control with High Angle Wells Another technology that has been extensively used to develop bottom water drive reservoirs with severe water coning problems is the ‘high-angle well’ technology (Figure 2. In a bottom water drive reservoir. Oil OW C W ater Figure 2-7 Schematic of High-Angle Well with tip Close to the OWC Typically. Results of field experience “demonstrate that water breakthrough in high angle wells can occur at two points close to the downstream/heel end and at the toe of the well. this angle is usually higher than about 80 degrees.12. the tip end of the well tend to be placed closed to the plane of the oil-water contact while the heel end is placed closed to the top of oil sand far away from the oil water contact. slanted wells are classified by the angle of inclination of an axis through the wellbore and a vertical line intersecting the axis.7).2. In a Hele-Shaw model experiment on the performance of high-angle wells. but it always occur close to the heel end in horizontal wells”. Permadi reported that oil recovery tended to be higher 40 . For high-angle wells. Thinner oil leads to bottom water invading oil around the toe after breakthrough and results in increased displacement efficiency (Permadi. P. 1996).

An important observation is that each of the three methods enumerated above involves some form of rate restriction and specifically in the case of high-angle well. most of the analytical models developed to evaluate the performance of oil wells subject to water influx (coning) have made several simplifying assumptions to reduce the complexity of their solutions. However. However. Several solutions have been developed to minimize the impact of the concurrent production of unwanted water in oil wells. It involves segregated production of the oil and water with a dual completion string with zonal isolation. 41 . 1996). the extent of reach of the well is limited by the thickness of the oil-bearing reservoir.13 Summary Water influx into an oil well is caused by the high pressure gradient around the wellbore during oil production. 2. In order to avoid this problem of water influx. Consequently. researchers have developed correlations for evaluating the time to water breakthrough into the well and the performance of the well after water breakthrough. However. water went through the whole part of the well length of high-angle wells in a relatively shorter time while some part of horizontal well length in the upstream region was never invaded by water even at high watercut (Permadi. the oil well should be produced below the critical oil rates. Downhole Water Sink technology is one industry technique that allows for production rates above the critical oil rates at the oil zone with reduced contamination with water. these critical oil rates are usually too low for any economic reason.for thicker oil columns and higher well angles. Phenomena such as capillary pressure transition zones and relative permeability hysteresis have often been ignored in these models.

horizontal wells are themselves not free from water influx problems (termed water cresting). Also. 42 . typical critical oil rates to avoid water influx into horizontal wells are too low for any economic purpose.Another industry technique for developing reservoirs subject to water coning is horizontal wells. Finally. Most of the analytical solutions for horizontal wells neglect the friction pressure loss in the wellbore or assume that it is negligible compared to the reservoir pressure drawdown. Like the vertical well counterpart. the review of literature reveals that detailed studies of water coning in vertical wells using numerical simulators involves radial grid with a fully implicit models while studies on horizontal wells use 3-D Cartesian grids with local grid refinement around the wellbore also in a fully implicit model. However. several researchers have recommended various approach to redistribute the wellbore pressure of the horizontal well to minimize the level of water influx at the heel of the well.

et al. The consequences is that. 1991) and the result of the theoretical studies showed that the hydrodynamic isolation of the bottom and top completions is the key factor to effective control over the production performance of the dual completion. The technology enhances radial (horizontal) flow of the reservoir fluid and so involves a less drawdown pressure than in the conventional well with water coning. Typically. This fact becomes an advantage to the DWS technology.CHAPTER 3 DWS VERTICAL WELL PERFORMANCE WITH VARIABLE WATER SATURATION The application of Downhole Water Sink (DWS) completion in the control of water coning in vertical wells involve the segregated production of water and oil using a dual-type completion with a zonal isolation packer. The Department of Petroleum Engineering at the Louisiana State University conducted theoretical studies on the DWS technology (Wojtanowicz. pressure drawdown in the conventional well is greater than the DWS well. Subsequent studies led to the development of a speadsheet-based software (SCON) and an innovative technique (the Inflow Performance Window – IPW) to evaluate the performance of the 43 . The production of water from the water zone creates a “pressure sink” and counters the development or progression of the water cone toward the oil zone completion. The basic idea is to perforate both the oil zone and the water zone and produce each fluid with a separate completion string. for the same amount of fluid (oil and water). vertical reservoir permeabilities are lower that horizontal permeabilities by a magnitude of 2 to 10 in most cases. Hunt Petroleum performed the first successful field trial of the DWS completion technology in the Nebo Hemphill field in Louisiana in 1994.

oil breaks through into the bottom completion (the sink) and at the extreme end of the window is a line of high cone instability where production of segregated fluid is almost impossible. 120 Water Rate (bbl fluid/day) 100 80 60 40 20 0 0 ugh etion o r h akt ompl e r C B Oil ottom in B n egio r p -flo ity Flip nstabil of i te ega r g Se 10 n gio e wR Flo ugh o r h kt rea pletion B m ter Wa op Co in T 30 40 50 60 20 Oil Rate (bbl fluid/day) S:Oil Bt Line S:Water Bt Line Figure 3-1 Typical Inflow Performance Window (IPW) and its Regions Above the segregated fluid region.1.technology. Four zones are distinguishable from the window as shown in Figure 3. The IPW is a 2-dimensional plot of fluid production rate at the top completion on the x-axis and fluid production rate from the bottom completion (the water sink) on the y-axis. Inside the triangular-shaped window is the region of segregated fluid production where oil production at the top completion is water-free and water production from the bottom completion is oil-free. The cone could be a water cone 44 . Below. this window is the domain of clean water production from the sink (bottom completion) and watercut oil production from the top completion.

there exist a stabilized limiting watercut and a stabilized flowing bottom hole pressure. (1991).4 Ibm/ft3 and the oil density is about 53. and Shirman (1997. Consideration of the fluid flow mechanisms of capillary transition pressures and relative permeability hysteresis complicates the mathematical analysis and thus. The concept of steady-state flow process also pre-supposes that. this study assumes a steady-state flow process.0 psi can create a cone height of as much as 15. The numerical simulation of DWS performance employs a conceptual model. et al.0 ft for a water/oil model where the water density is about 62. which assumes that the bottom aquifer creates enough pressure support to achieve a steady state flow behavior. 3. a pressure differential as small as 1. The consequence of the above observation is that effective simulation of steady-state water cone development 45 .1 Pressure Stabilization and Steady-state Water Cone Simulation Numerical simulation of water cone development is very sensitive to little changes in pressure. the studies were performed with numerical simulators.04 Ibm/ft3 respectively. As in most coning studies on reservoirs with strong bottom water drive. for a particular production. The “analytical model” calculates steady state pressure distribution around the well with single or two-phase production at the top and bottom completions.breaking through at the top completion or change quickly to an inverse oil cone breaking through at the lower completion with very little variation in production rates. As an example. The steady state simulation approach was also necessary to effectively evaluate the results of the “analytical model” developed by Wojtanowicz. the pressure drawdown becomes stabilized after a very short period of time. 1998) while at the same time verifying the results of the simulation with known theoretical solutions. The consequence is that for a particular flow rate.

or at the lowest bottom grid in the aquifer zone to simulate a bottom water drive reservoir.in j e c t io n 3 . (2) Use of infinitely large porosity in last radial grid cell creating steady-state boundary pressures. However. I n fi n it e P o r o s it y m o d e l. L a r g e B o tt o m a q u if e r s u p p o r t to m a i n ta in P r e s su r e R e . L a r g e b o tto m w a t e r d r iv e 2 . P r o d u c e d F lu id R e . Researchers have recommended three approaches for the simulation of a steady-state production process with commercial simulators.in j e c ts P rodu ced F lu id s (m in im u m b o tto m a q u if e r ) L a s t w a te r c e ll h a s in f in it e p o r e volu m e Figure 3-2 Steady-state Numerical Simulation Models However. T H R E E B A S IC M O D E L S O F S T E A D Y S T A T E IN F L O W S IM U L A T IO N 1 . the first option requires effective material balance solution to establish appropriate enchroachable water.requires a high level of accuracy. This is time consuming considering the amount of simulation that is required to create enough data points to generate an inflow performance window (IPW). and (3) Fluid re-injection approach. Each of the above solutions is capable of generating adequate reservoir energy in the aquifer zone to simulate a steady-state process. it was found to be inadequate as 46 . The second option of infinitely large porosity at the radial boundary was employed in all previous work on this project. They are: (1) Creation of a large aquifer with adequate enchroachable water.

To effectively present the above argument.it stabilized pressure at the boundary but failed to stabilize pressure around the wellbore. This ensures that the energy supply from the aquifer is uniform through out the drainage area of the reservoir. This option creates a drawdown pressure profile at the near wellbore region different from that at the radial boundary for different time frames. The graph shows a different pressure within the first five years of simulation forecast and the remaining fifteen years. This equation is produced with a slight 47 . Figure 3-3 shows a typical pressure profile generated as a function of time using the radial boundary infinite porosity model. it is necessary to present the equation of Darcy for a steady-state fluid flow process. In order to effectively simulate a steady-state bottom water drive system with stable pressure at all points. IN F IN IT E P O R O S IT Y M O D E L A N D D E L A Y E D S T A B IL IZ A T IO N D e la y e d S ta b iliza tio n C r e a te s tw o P re s s u re P ro f ile s (In finite p o ro s ity m o d e l) 1988 1987 1986 1985 1984 1983 1982 1981 1980 1979 0 Top BHP (psi) A B 5 10 T im e (y e a r s ) 15 B 20 A Figure 3-3 Unstable Flowing Bottom-hole Pressure at Top Completion This development of differential pressure gradient was considered inadequate for water coning studies since the water cone development is a near wellbore phenomenon. the infinite porosity grid should be the lowest set of grid blocks in the water zone and not the last radial grid.

mo.350 ft away from the wellbore converges as expected for steady-state fluid flow process. To understand the cause of the unstable pressure distribution with time. 48 . pe.00708 * kkro h * ( pe . The pressure distribution at mid-perforation grid blocks is shown in Figure 3-4. k = mD. Qo = 0.00708 * kkrw h * ( pe . Bw = bbl/stb. While the pressure at +/. the wellbore pressure should be constant with time.pw ) æ re ö mo Bo lnç çr ÷ ÷ è wø 3-1 Qw = 0. Thus it was necessary to further investigate this abnormality. it diverges at the near wellbore region over time.modification for the 2-phase oil-water flow incorporating the relative permeability function to account for the component of oil and water in the fluid flow. This difference in near wellbore pressure profile over time was attributed to the time lag in the infinite porosity steady-state option. mw = cP. it was necessary to evaluate the radial pressure profile from the near wellbore region to the outer reservoir boundary.pw ) æ re ö m w Bw lnç ÷ çr ÷ è wø 3-2 The above equations show that for a steady-state process with constant pressure boundary and constant flow rate. Q= stb/day. pw = psi. assuming no slippage (field units. rw = ft). The results presented in Figure 3-3 shows a differential pressure profile with time. h = ft and re. Bo.

Figure 3-5 shows that the watercut never reaches the expected limiting value for this infinite porosity option but rather creeps.00 0 15 Years Section B-B 3 Years Section A-A 50 100 150 200 250 300 350 400 Pressure Radius (ft) Figure 3-4 Unstable Wellbore Pressure Distribution with time The conclusion from this study is that the infinitely large pore space option at the radial grid boundary is inadequate for the pressure sensitive studies of water cone development.50 1985.50 1984. It is expected that the watercut should reach a stable level called “limiting watercut” over time for a steady-state process. 49 .50 1983.00 1985.Well P1 BHP vs radius plot (Infinite Porosity model) 1987. 1999).50 1986. Consequently.00 Well P1 BHP (psia) 1986.00 1983. previous research efforts adopted two acronyms called “engineering approximation” and “asymptotic extrapolation” in order to define a cut off point for determining the critical rates needed to generate the IPW.50 1987. Kurban reported that the inability to stabilize the flowing bottom hole pressure means that limiting watercut could not be achieved (Kurban.00 1984. Another tool employed to evaluate the effectiveness of the radial boundary infinite porosity model is the watercut profile over time.

00 1 9 7 2.00 0 Well P1 BHP (psia) T im e 15.2 0.4 0.00 0.00 1 9 8 0.t h in lin e ) T im e 3 .0 0 0 1 00 0 .00 1 9 7 6.00 1 9 8 2.00 1 9 8 4.8 Water Cut.B s e c t io n .00 1 9 8 6.7 0.6 0.A s e c t io n .00 1 9 7 4. Fraction 0.1 0 0 200 400 600 800 Time.3 0. Days 1000 1200 1400 1600 Figure 3-5 Creeping Watercut for Unstable Flowing Bottomhole Pressure The last model of steady-state simulation in commercial numerical simulator is the fluid re-injection method. By this approach. Figure 3-6 is made with the fluid re-injection approach.Field WaterCut vs Time (Creeping case) 0.9 0. S T A B IL IZ E D P R E S S U R E A T B O U N D A R Y A N D A T W E L L B O R E W e ll P 1 BHP vs r a dius plot (Ne w m ode l .t hic k lin e ) 2 0 0 .m id-uppe r c om ple tion) 1 9 8 8.y e a r s ( A .00 1 9 7 8.00 1 9 7 0.0 0 0 P re ssu re R a d iu s (ft) Figure 3-6 Overlap of Pressure at Different times Indicate a Stable Steady-state Flow Process 50 .ye a r s ( B . the produced fluid (oil and water) is reinjected into the reservoir to maintain the reservoir energy and thus simulated a steadystate flow process.0 00 80 0 .5 0.0 0 0 6 0 0 .0 00 4 00 .

01 psi both at the wellbore and at the outer radial boundary as shown in Figure 3-6. a minimum of eight data points representing the critical water production rates of the bottom completion for several given oil rates at the top completions are needed. Thus.This approach appears to be effective in stabilizing flowing wellbore pressures with accuracy of ±0. On a typical IPW. Operator personnel need few hours of training to generate the Inflow Performance Window using in-house commercial simulators. The ability to determine a stable limiting watercut from the results of the simulator reduces the time and efforts needed to determine each of the critical rates of the two completions. a simplified approach to determining these points also increases its attraction to industry operators. Figure 3-7 Watercut Profile for Stabilized Bottomhole Pressure (Plateau) 51 . The foregoing exercise has been an attempt to minimize the time expended in generating a typical IPW using commercial numerical simulators.

3.Figure 3-7 shows the watercut profile for the fluid re-injection model generated from the graph software.Numerical Simulator and Analytical Model Water Rate (bbl fluid/day) 52 . it was determined that the bench-mark to verify the results of the commercial simulator would be the results from the analytical model. A disadvantage of this approach is that oil production follows a straight-line profile over time with no distinct decline since the oil volume in the reservoir remains constant with time.2 Verification of Results with Analytical Model IPW In the absence of field data in form of history match.) Figure 3-8 Comparison of IPW . 100 90 80 70 60 50 40 30 20 10 0 0 5 10 15 20 25 30 35 40 45 Oil Rate (bbl fluid/day) A:Oil Bt(Anal. Consequently.) A:Water Bt(Anal. The limiting watercut stabilizes over time as shown in the plateau. fluid re-injection method was adopted as adequate for simulations performed on vertical wells.) S:Water Bt(Sim. now possible to obtain the critical flow rates necessary to generate the IPW from the results of the reservoir simulation runs without applying the concept of “asymptotic extrapolation” or “engineering approximation”. thus. It is.) S:Oil Bt(Sim.

The plot shows a close agreement in the size and shape of the two windows. This is attributable to the fact that the analytical model did not incorporate the dynamics of relative permeability variations with water saturation changes. It therefore tends to overestimate the size of the domain of segregated fluid production. The analytical model uses only end-point mobility considerations and so neglects changes in relative fluid mobility due to increasing water saturation. The solid line plots represent the IPW generated from numerical simulator while the hatched line plots represent an IPW generated from the “analytical model” for the same reservoir and fluid data. the next step was to study the performance of the DWS wells under certain fluid flow mechanisms such as 53 . the stability of the oil-water contact (OWC) range decreases so that there is a maximum oil rate above which the drainage-production system is inherently unstable. neglecting the effect of capillary transition. The term. A primary difference between the result of the “analytical model” and that of the numerical simulator.The logical step was to compare the Inflow Performance Windows from both numerical simulation and analytical model for the same reservoir and fluid properties. The variation in the area of the two IPW is about 0.050 or 5 percent. Figure 3-8 is a superimposed plot of IPW from numerical simulator and a similar IPW from the “analytical model” previously developed for DWS studies. Swisher and Wojtanowicz (1995) indicated that as the oil rate increases. is that the IPW for the numerical simulator appears to close faster than that of the analytical model. “unstable” as used by the DWS technology refers primarily to the deformation of the cone profile causing concurrent production of water and oil from both completions. Having verified the performance of the numerical model.

In field units. The reservoir pressure gradient must be such as to overcome the effect of these attractive inter-molecular forces before fluid mobility can be achieved. solids and gases in the reservoir. The goal is to investigate how the DWS technology will perform in these old oil wells with severe water coning history as compared to the performance in a new oil well with distinct oil-water contact. when reservoir energy is not high enough to overcome these inter-molecular forces the oil is trapped as residual oil.1 Capillary Pressures Capillary pressure is simply the pressure exerted by inter-molecular attractive force between immiscible fluids. 3. This capillary pressure in the reservoir is a function of the chemical composition of the rock and fluids. the following definitions are needed: 3. Consequently.3 Capillary Pressure Transition and Simulation of DWS Well To clearly understand the effect of capillary forces on water coning and hence the DWS technology performance. capillary pressures are given by (Craft and Hawkins. 1991): Pc = 9.3.519(10) -7 s wo cosq rc 3-3 The radius of capillary is related to reservoir permeability by: 54 . the pore size distribution of the sand grains in the rock and the saturation of the fluids in the pores.capillary transition zone characterized by the presence of mobile oil and water as well as incorporate the directional properties of relative permeability termed “hysteresis effects”. Old oil wells with a history of severe water coning typically presents this watered out zone of co-mingled mobile oil and water in a manner similar to the presence of a capillary transition zone due to low reservoir permeability.

but rather occupy the inner space is called the ‘non-wetting phase’. This preferential spreading or adhering capacity is called the wettability of the rock pore wall and the fluid that spreads/adheres more on the surface of the wall is called the wetting phase. This angle is influenced by the tendency of one of the two immiscible fluids to preferentially spread on or adhere to the pore wall surface.5. For unconsolidated sands and limestones. Consequently. An increase in the saturation or content of the wetting phase is called an “imbibition process” while an increase in the concentration (saturation) of the non-wetting phase is called a “drainage process”. r. measured through the wetting phase fluid against the pore walls is known as the contact angle. The other fluid that does not spread on the pore wall. q is in degrees.15(10)13 rc 2 3-4 Where. 1944) for consolidated sands where Pc is proportional to 1/k0. k has no definite relationship but Pc is rather a function of k. and k is in mD.5 shows that capillary pressure effects are less severe in high permeability reservoirs than in low permeability reservoirs. particularly the asphaltene content of the oil. rc is in ft. The degree of wettability depends on the chemical compositions of the two fluids. swo is in dynes/cm. Pc is in psi. Combining the two equations yield an expression for capillary pressures as a function of reservoir permeability: Pc = 3. as well as the nature of the pore wall.228s wo cos q k 3-5 The above relation has been confirmed by experiments (Hassler and Brunner.k = 1. Equation 3. and Sw. Pc vs. for normal 55 . The angle. q.

A minimum threshold pressure. a further increase in the displacement pressure produces no further increase in the hydrocarbon saturation. At the point of irreducible wetting phase saturation. Capillary pressure is typically measured in the laboratory using air-mercury system. The threshold capillary pressure. the capillary transition relates to the distribution of water saturation as a function of height above the free water level (FWL) in a reservoir. Pct. is required to displace the fluid in the largest pore space. Thereafter.recovery. Thus. is thus. smaller pore spaces are filled up by the migrating hydrocarbon until an irreducible level of water (wetting phase) is reached called the connate water saturation or irreducible water saturation. the effect of capillary pressures is neglected if the reservoir permeability is high in the magnitude of thousands of millidarcies. the reservoir rock pore space is initially occupied by the wetting phase fluid (water for sandstone). the minimum pressure required to overcome the intermolecular forces between the surfaces of the two immiscible fluids in order to move the wetting phase fluid (water) from the pore space by the non-wetting phase (oil). 56 . The laboratory measurements are then up-scaled to reservoir condition using the Young-Laplace equation: Pc1 = Pc 2 * (s cos q )1 (s cosq ) 2 3-6 Typically. capillary threshold pressure. The region between the threshold capillary pressure and the capillary pressure due to the irreducible water saturation is known as the “capillary transition zone or region”. The migration of hydrocarbon into this pore space causes the water to be displaced.

7 to extrapolate the value of capillary pressure with height above the oil-water contact. The equation 3. 1999).9. 1986): Pc ( S W ) = H ( SW ) * ( r w .(Archer and Wall. if the OWC is known. Alternatively. 1986). where the capillary pressure information is not available. a drainage test can be conducted on a core sample from the reservoir to determine the threshold capillary pressure from which the depth of the FWL can be determined.This threshold capillary pressure found in reservoir rocks is proportional to the height above the free water level (FWL) datum. ro are in Ibm/ft3. 1986). The experimental measurement of capillary pressure transition is rarely a straight line. and (Yokoyama. The relationship between height above FWL and capillary pressure is derived from consideration of the gravity-capillary pressure force equilibrium and is expressed in field unit as(Archer and Wall. while an oil-water contact observed on a particular well in the reservoir will depend on the threshold pressure of the rock type present in the vicinity of the wellbore. where a region of 100 percent water saturation is found. The free water level is thus a property of the reservoir system.7 can be used to evaluate the threshold capillary pressure when the height of the oil-water contact above the field FWL is known together with the densities of the oil and water. However.r o ) 144 3-7 Where. H(Sw) is in feet. Rather. researchers have used the Equation 3. (Willhite. 1981). it is curved line as shown in Figure 3. 1986). and rw. (Weinstein. Pc is in psi. 57 . (Kurban.

This dimensionless value can in turn be used to evaluate the capillary pressures for the reservoir rock in question. capillary pressure data are available from adjacent wells or reservoirs with similar properties. Pc = 0 Swirr Wetting Phase Saturation. It expresses a relationship between capillary pressures and permeability as follows: J (SW ) = Pc ( S W ) k s cos q f 58 3-8 . a to Height above FWL Top Transition Zone Pct Observed Oil Water Contact Free Water Level. Where.‘clean oil’ Pc. Sw Figure 3-9 Typical Capillary Transition Pressure Zone This linear extrapolation approach assumes the OWC as the zone of 100 percent water saturation. the equation of Leverett can be used to evaluate a dimensionless parameter called the Leverett J-function. FWL.

1 Capillary Pressure Hysteresis The capillary pressure that must be overcome to displace the wetting phase from the pore space by the non-wetting phase differs from that required to displace the nonwetting phase from the pore space by the wetting phase.3. Pc. called the “suction pressure” does not coincide with the 100 percent wetting phase (water) saturation point. Capillary Pressure. Rather it stops short of the 100 percent point and defines the residual saturation of the non-wetting phase. When the wetting phase (water) saturation increases by displacing the non-wetting phase from the pore space.3.1. This difference creates separate paths for the capillary pressure effect as shown in Figure 3-10.0 Wetting Phase Saturation. the imbibition threshold pressure. psi Drainage Imbibition Pct 0 Swirr Figure 3-10 Capillary Pressure Hysteresis 1. Sw 59 .

imbibition capillary pressure data were not available. The prolonged production of water causes a zone of co-mingled variable fluid saturation to develop.1. 1986). as will be seen later. the experimental difficulties in determining the definition of the imbibition capillary pressure curve. combined with the difficulties of using the information in reservoir simulation models. a zone of varying water saturation) and this was effectively achieved with the inclusion of the drainage capillary pressure data.e. the linear extrapolation approach was adopted necessitating the use of only one table of capillary pressures in the simulation.3. drainage and imbibition is termed the capillary pressure hysteresis as shown in the Figure 3. especially in old oil wells with severe water coning history.2 Capillary Transition Pressure and the DWS Technology in Old Oil Wells The concept of capillary pressure transition. capillary pressure hysteresis does not exist (Archer and Wall. However. This zone of mobile comingled oil and water is termed “the transition zone”. the drainage capillary pressure data were available in some cases and in other cases. The presence of this transition zone of varying water saturation indicates that the intermediate relative permeabilities.The threshold pressure of the drainage phase almost always coincides with the 100 percent wetting phase saturation point. however. The primary objective was to incorporate a capillary transition zone in the DWS model (i. play prominent roles in fluid mobility and hence oil recovery. for practical purposes. has led to the assumption by many reservoir engineers that. This deviation in the path of the two processes.10. 60 . However. For the reservoirs studied. 3. is very significant in modeling the performance of the dual completion water sink performance.

when the both imbibition and drainage phenomenon are involved as in the DWS technology. it will underestimate the amount of water required to stabilize the oil-water contact or at least avoid water breaking through into the top completion. For an increased capillary pressure transition height. the Inflow Performance Window reduces drastically. however. almost zero. The results of inclusion of the capillary transition effect in modeling of DWS completion technology for old oil wells with severe water coning history are quite significant. Reed. This agrees with the result of previous research in this type of reservoirs (Kurban. However. these effects cannot be ignored. over 50-ft of oil sand. The critical water rate is essentially unchanged.the imbibition curve. Water breakthrough times for thin reservoirs showed little or no change with inclusion of capillary transition zone. the imbibition and drainage properties. 1999). The reason for this response is because water breakthrough times are usually very small. probably due to the fact that the studies did not incorporate the effect of the directional properties of permeability to the various fluid phases namely. the critical oil rate is greatly affected. the effect becomes very pronounce depending on the extent of the transition. However. for thin reservoirs and thus capillary pressure effects will not be easily noticeable.Most water cone studies have neglected the effect of capillary transition in water cone development. A few authors (Henley. Neglecting the capillary pressure transition effect will not only overestimate the amount of moveable oil. for extensive reservoirs. The studies applied only one set of relative permeability curves . The 61 . 1960. This conclusion is. 1984) have simulated the effect of capillary pressures in water production and concluded that the effect of capillary transition zone was small and could be ignored.

The relevance of the above observation is that the “analytical model” tends to overestimate the performance of the DWS technology for these old wells. the saturation of the non-wetting phase increases representing a drainage process.gradient of the water breakthrough line increases with increasing size of the capillary transition zone. 3. the wetting phase saturation increases representing an imbibition process. 62 . When the water sink is activated to counter the cone development.4 Relative Permeability Hysteresis and the DWS Technology Modeling Another fluid flow mechanism of interest to the modeling of DWS completions is that of relative permeability hysteresis. During the water coning process. Water Rate (bbl fluid/day) 350 300 250 200 150 100 50 0 0 50 100 150 200 250 Oil Rate (bbl fluid/day) 30' C AP :Oil B t 10' C AP :Wa te r B t 30' CAP :Wa te r B t 0'C AP : Wa te r B t" 10' CAP :Oil Bt "0'C AP :Oil B t" Figure 3-11 Effect of Capillary Transition Zone on DWS IPW The consequence is a reduction in the size of the domain of segregated fluid production. Figure 3.11 show the reduction in size of the IPW for different level of capillary transition region.

kr is the relative permeability of the phase. intermediate wet or mixed-wet.max. Relative permeability curves are dependent. is reached when the non-wetting phase. For two-phase flow in porous media. oil. This preference depends on the rock –fluid composition and interaction. The minimum possible level of the wetting phase saturation. i. to a great extent on the rock-fluid interactions or wettability. Wettability refers to the preference of reservoir rocks to be coated or wetted by a particular fluid type in multi-phase system. The maximum wetting phase saturation Sw. For petroleum engineering studies we observe that rocks could be oil wet. the relative permeability to each phase is expressed as a function of saturation.e oil mobility becomes zero either due to high capillary pressure trapping as earlier discussed or due to decreasing reservoir energy. k is the absolute permeability of the porous medium. ke =k * kr where: ke is the effective permeability of the phase. Wettability of a surface depends on the threshold capillary pressure term “scosq". reaches its residual saturation level. water in our studies. 63 . This term indicates that capillary pressure controls the sequence of pore saturation changes and hence relative permeability so long as viscous forces are not controlling the flow process. is usually designated as the “connate water saturation” or “Irreducible water saturation” (Swc or Swirr). water wet.The concept of relative permeability has been conveniently used in reservoir engineering to relate absolute permeability of the porous media (100% saturated with a single fluid) to the effective permeability of a particular fluid in the system when that fluid occupies a fraction of the total rock pore volume.

One such rules-of-thumb is that proposed by Craig (Craig. 1993). Greater than 50% and approaching 100% The table indicates the need to also establish the wettability of the reservoir rock system before deciding whether DWS technology is applicable to the reservoir of interest 64 . Relative permeability to water at the maximum water saturation based on effective oil permeability at reservoir connate water saturation Greater than 50% water Less than 50% water saturation Generally less than 30% saturation. It has been quite valuable for water flooding projects and as a benchmark for reservoir simulation experts in analyzing results of experimental/laboratory data prior to the development of reservoir simulation models. 1986). Frequently less than 10% Saturation at which oil and water relative permeabilities are equal. 1993). Table 3-1 Craig's Rules-of-Thumb for Determining Rock Wettability Water wet Oil Wet Irreducible or connate water saturation Usually greater than 20 to 25 % PV Generally less than 15% PV.Relative permeability curves are currently generated for either oil wet or water wet rocks using published correlations (Archer and Wall. Table 3-1 list Craig’s rules-of-thumb for determining wettability from twophase relative permeability curve (Craig. Some “rules-of-thumb” has been proposed by researchers in order to characterize the wettability of reservoir rocks.

1 Fractional Flow Analysis Relative permeability curves dominate the fractional flow equation applied in numerical simulators to determine the flow rate of each phase in the reservoir. The research results presented in this dissertation have assumed a water wet sandstone reservoir for all the models. However. (1941) first presented the concept of fractional flow of fluid for an oil-water system in terms of the relative permeabilities and the fluid properties: 1+ fw = kkro ut m o æ ¶Pc ö .or not. A review of the basic fluid flow equations in porous media as they apply to the effect of capillary pressures and relative permeabilities will be undertaken at this point.gDr sin a dip ÷ ç è ¶x ø m k 1+ w o mokw 3-9 Civan and Donaldson (1987). The first and most important flow equation is the Darcy equation. Leverett. Fluid permeability characteristics are frequently presented in terms of permeability ratio kw/ko. Thus. an understanding of fractional flow equations and how they depend on the fluids saturation is necessary. 3. developed a more detailed expression (Equations 310 and 3-11) for evaluating the relative permeabilities of the wetting phase and the nonwetting phase incorporating the effect of capillary pressures in the solution of the fractional flow equation as follows: 65 .4. This is because the fractional flow equations are expressed in terms of relative permeability ratios. the generalized fractional flow equation often neglects the effect of capillary pressure transition but rather assumes a sharp fluid interface.

relative permeability exhibits directional properties leading to the concept of hysteresis. of preferential wettabilities between fluids and rock surfaces.m q . and in fluid properties. kw/k against saturation. neglects the directional properties (hysteresis) of relative permeabilities. = . kg/k. normalized plots of relative permeability.2 Imbibition and Drainage Relative Permeabilities Like.Q ç kA k rw è dQ ÷ ø -1 sw -1 3-10 And k rw m o f w æ k rw kA ¶Pc ö ÷ 1. extending (or even dispensing with) experimental measurements of effective permeabilities” (Archer and 66 . ko/k. In effect. or purely empirical relationships.4. exhibit general similarities of form. capillary pressure. however. “ Inspite of the wide variety of pore structure in reservoir rocks. to assist in smoothing.ç ÷ m w k ro m ¶ fo ç q x w w ø è 3-11 Civan and Donaldson indicated the above two equations can be solved for the respective values of relative permeabilities as functions of water saturations. The expression. . ¶x k rw ÷ ÷ f o dS ç ø w w w w è è ø + Lqw m w 1 æ d Dp ö ÷=0 . -ç Dp . It is. 3. then. extrapolating. attractive to attempt to formulate theoretical semi-empirical.æ ö f w kA ¶Pc ¶S w dPc ç ÷ 1 . k rw ò ç m q ¶S ¶x ÷ f o dS dS w w w w w ø swx=0 è æ ö f w kA ¶Pc ¶S w dPc æ ¶S w dS w ö -ç ÷ ç ¶x L + Q dQ ÷ ç1 . ¶S . . the relative permeability curve for an imbibition process is different from that for a drainage process for the same fluids.

0 (Corey Model. where a = 4.4 (Archer and Wall.w ý î 1 .3 Non-wetting Phase drainage Relative Permeability * * k rnw = 1 . 3-14 And (S ) * w imb * = Sw ( ) drainage -1 (S ) 2 * 2 w drainage 3-15 3.2.S w 1 + 2S w ( )( 3 ) (where S * w is the drainage case) 3-16 3.S wi Swi = irreducible water saturation 3-13 3.4.Wall.2. used for this study) 3-12 (S ) * w drainage = S w .4.4.S wi .2.4. relative permeability data were not available and so published correlation presented in sections 3. 3.1 Wetting Phase Drainage Relative Permeability * k rw = S w ( ) a .2 Wetting Phase Imbibition Relative Permeability * k rw = S w ( ) a imb (a = 4. In other fields.S r þ 2 3-17 67 . 1986) were used. 1 .S wi ü = í1 .2.2.2. relative permeability and capillary pressure data from cores were supplied by Joint Industry Panel (JIP) members. 1986).S wi .0).4.4.4 Non-wetting Phase imbibition Relative Permeability k rnw ì S . For some of the fields studied.1 to 3.

80 1. if there is capillary transition zone as in old oil field with severe water coning history.0) 0. For most of the reservoirs studied.3000 0.0000 0. residual oil saturation and the respective end-point effective permeabilities at these saturations and the absolute permeability in order to generate complete relative permeability curves.20 0. 1.fraction) Figure 3-12 Relative Permeability Curves for Drainage and Imbibition The typical set of imbibition and drainage relative permeability curves generated for this study is shown in Figure 3-12.8000 0. only 68 .60 0.00 Water Saturation (Sw .9000 Relative Permeability(fraction) 0. that is. if there is no capillary transition zone.6000 0.5000 0.1000 0. all that is required is information on some basic properties such as irreducible water saturation. the position of the end-point saturations plays a significant role in the fractional flow of the two immiscible fluids especially in cases where there is a sharp interface between the water and the oil zones.0000 0.4000 0. However. As will be observed later.00 Blue = Imbibition curves Red = Drainage (Corey Exponent=4. not only the end-points but the entire relative permeabilities at different saturations within the hysteresis loop plays a paramount role in determining the fractional flow of each phase.40 0.2000 0.7000 0.With these normalized equations.

the keyword “HYSTER” is used to turn on the hysteresis function in the ‘RUNSPEC’ section.3 Simulation of Relative Permeability Hysteresis in Eclipse The application of the dual completion in DWS wells causes frequent reversal in the direction of saturation change creating a relative permeability hysteresis effect. 1989). thus the end-points of the drainage and imbibition curves are the same. and selects one choice of models for relative permeability hysteresis (Schlumberger Geoquest. 3. Two user-supplied tables of saturation Vs relative permeability are specified using the key words “SATNUM” and “IMBNUM” in the ‘REGIONS’ section of the data deck. A further keyword. “EHYSTR” in the ‘PROP’ section sets the values of the two parameters that determine the scanning curves for capillary pressure hysteresis and non-wetting phase relative permeability hysteresis. 1997). The option of relative permeability hysteresis is used to specify different tables of saturation functions for both the drainage process of increasing saturation of non-wetting phase and the imbibition process of increasing saturation of wetting phase. The Geoquest Eclipse 100/200 requires that the relative permeability curves be changed manually when there is saturation reversal. First. The scanning curves help to determine the path of flow should the process be reversed 69 . “Some simulators have the capability to sense saturation reversals and introduce hysteresis effects into the relative permeability functions automatically.one value of connate water (irreducible water) saturation and one of residual oil saturation were provided from the industry sponsors.4. Other simulators require that the relative permeability curves be changed manually when saturation reversals occur (Mattax and Dalton.

consider a drainage process (decreasing water saturation when the water sink is turned on) starting at point 2. Two methods are generally available for scanning curve generation. 1 Relative Permeability (Fraction) Drainage Curve 4 Imbibition Curve Swc 3 5 Swc(imb) Swc(scan) 2 Swc(drain. the Carlson method and the Killough method. Figure 3-13 will be used to illustrate the scanning curve option in numerical simulators and its effect.abruptly at some intermediate point.) Increasing wetting phase (water) saturation Figure 3-13 Relative Permeability Hysteresis Curve with Scanning Option The Killough method has been selected for this study to ensure that the end point saturation of the scanned curves always lie between the end-points of the critical drainage and imbibition curve saturations. if the full drainage process 70 . In Figure 3-13.

4. Appendix A is a sample data deck for DWS well simulation with permeability hysteresis and capillary pressure transition zone. the simulator can either retrace the curve back to point 2. neglecting the presence of a capillary transition zone. the end-point saturation of the scanned curve lies between the critical saturations of the drainage and imbibition processes specified in the two saturation tables.4 Analysis of Results from the Relative Permeability Hysteresis and Capillary Pressure Studies The result of simulation studies on the effect of relative permeability hysteresis function. Again. for most of the studies in this chapter. However. the retracing option has been adopted for this research. It is a function of the maximum oil saturation reached in the simulation run process. (when the production from the top completion is increased) such that water saturation begins to increase (representing an imbibition process). This is necessary to maintain equal amount of displaceable oil. the Killough option as described earlier ensures that. Also. should the scanning option be adopted. is presented in Figure 371 . 3. the critical saturation for the imbibition and drainage flow process have been set to coincide in order to effectively evaluate the improved oil production capability of DWS wells and to maintain the same value of residual oil saturation and connate water saturation. if the “RETRACE” option is specified or create a new scanning curve from path 4 to5 as shown above. if the drainage stops midway say at point 4. the end point will be at point 1. the hysteresis option for capillary pressures has been ignored. in order to simplify the results of this study. For the emphasis of reproducibility of results.is carried out. This critical saturation at the end-point of the scanned curve is the trapped critical saturation. Also.

“IMB ZERO CAP” also represents an imbibition process neglecting the effects of capillary pressures and “HYS ZERO CAP” represents the hysteresis option also neglecting the effect of capillary pressures. This result indicates that the two windows were essentially the same. the IPW windows will practically over lay each other in reservoirs with distinct oil-water contact as in Figure 3-14. 72 . Analysis of this result showed that the end points relative permeabilities plays a very prominent role. In effect. “DNG ZERO CAP” represents a drainage process neglecting the effects of capillary pressures. Thus.14. As long as the connate water saturation is the same for both imbibition and drainage relative permeabilities and the residual oil saturations are also the same for both imbibition and drainage cases. Effect of Relative Permeability Hysteresis on DWS IPW Window for Extensive reservoir without capillary transition 400 350 300 Water Rate (bbl fluid/day) 250 200 150 100 50 0 0 50 100 Oil Rate (bbl fluid/day) DNG ZERO CAP:Oil Bt IMB ZERO CAP:Water Bt DNG ZERO CAP:Water Bt HYS ZERO CAP: Water Bt IMB ZERO CAP:Oil Bt HYS ZERO CAP: Oil Bt 150 200 250 Figure 3-14 Negligible Hysteresis Effect Due to use of Same End-point Saturations for both Imbibition and Drainage In Figure 3-14. only the end point relative permeabilities are necessary for a step-function fluid flow model. the Figure 314 results essentially tests the effects of relative permeability hysteresis in reservoirs without capillary transition pressures.

‘HYS 0 CAP’ shows the segregated flow domain of the IPW for a case of zero capillary transition while ‘HYS 30’ CAP’ shows the segregated flow domain for the same 50 feet oil pay zone reservoir with 30 ft or 60 percent of the oil zone occupied by the capillary transition zone. Figure 3-15 show the performance of the DWS technology in a field with water saturation gradient such as an old oil field with severe water coning history or low permeability reservoirs with a capillary transition zone. the effect of the intermediate points of the various relative permeability curves becomes important. For this case. They arguing that only the end-point saturations and relative permeabilities are useful. It also explains why some industry analysts’ (operators) recommends the use of straightline relative permeability curves for high permeability reservoirs with distinct oil-water contact. Bottom completion production rate (stb/day) 350 300 250 200 150 100 50 0 0 50 100 150 2 00 250 T o p C o m p letio n p ro d u c tio n rate (stb /d a y ) H Y S 0' CA P :O il B t H Y S 30' CA P :W a t e r Bt H YS 0' CA P :W a te r Bt H YS 30' C A P : O il B t Figure 3-15 Combined Effects of Capillary Pressures and Relative Permeability Hysteresis However.This result agrees with Miller and Rogers’s (1973) conclusion of a similar study. The mobility of fluid at different saturation is different for the case of increasing water saturation (imbibition) when the cone is developing and for decreasing 73 .

when the production of liquid from the top is high. water saturation increases and the cone begins to develop and the flow path is that of imbibition.5 DWS Well Productivity Performance Studies Following the results of effects of capillary pressure and relative permeability hysteresis on the performance of DWS technology in these old wells presented in the preceding sections. The result of Figure 3-15 shows that the inflow performance window is smaller in reservoirs with capillary pressure transition zone. it could be observed that the critical factor in modeling the performance of the DWS well completion technology in old oil wells with severe water coning history is the size of capillary pressure transition zone (Inikori. The cone reversal process of the water sink creates a temporal drainage displacement when oil displaces the water cone. therefore. it was necessary to evaluate the oil productivity performance of the 74 . not adequately account for the mobility of fluid in the saturation transition zone. 3. and Wojtanowicz. It is important that capillary pressure measurements from cores be performed in candidate wells and the effect incorporated in the design model. Straight-line approximation of relative permeability data will. The effective way to model the performance of such wells with a zone of variable saturation is to incorporate the effects of capillary pressures with relative permeability hysteresis.water saturation during cone reversal. On the other hand. This indicates that the region of segregated fluid production becomes smaller for such reservoirs. 2001(A)). Combining the results of the Figures 3-11 and 3-15. The result of the study also show that the concept of Inflow Performance Window presents a powerful tool for understanding and quantifying the effects of capillary pressure transition zone in oil-water flow in porous media.

0 0 0 2 .0 0 0 4 .50 0 . and watercut performance evaluation.1 DWS Well Oil Recovery Performance Study In a study of DWS well oil recovery performance using a fluid re-injection model.0 0 0 50 0 .0 0 0 2 .50 0 .0 0 0 .0 0 0 .0 0 0 Oil Recovery (Stb) 3 .50 0 .50 0 . The upward sloping nature of the various curves indicates that the technology is capable of improving recovery of oil even in old wells 75 . Figure 3-16 shows a plot of oil recovery over a period of twenty years for various oil production rates. namely. 5.0 0 0 3 . oil productivity index analysis. Several oil production rate s at the top completion were evaluated varying the fluid withdrawal rates at the water sink completion.technology incorporating these effects.0 0 0 1.0 0 0 1. b b l/ d a y 2 0 0 B OPD 10 0 B OPD 3 0 0 B OPD 4 0 0 B OP D 6 0 0 B OP D 400 50 0 600 Figure 3-16 Oil Recovery Performance Study of DWS Well The intercept of the various lines with the oil recovery axis represents the typical oil recovery of conventional wells.0 0 0 . oil recovery performance evaluation.0 0 0 4 .5.0 0 0 0 10 0 200 300 Sink Dra iang e R at e. 3. Three main aspects of oil production performance of the technology in these old oil wells were considered. it was observed that the technology demonstrates superiority in oil recovery over conventional completion as shown in Figure 3-16.0 0 0 .0 0 0 .

Another merit of the technology that could be observed from the plot in Figure 316 is the flexibility in oil production rate.with water coning history. This result agrees with the performance of the field trial of the technology in the Nebo-Hemphill well in La Salle Parish in Louisiana (Swicher and Wojtanowicz. evaluation of its performance compared to conventional single completion presents some difficulty. The upward slope of the recovery curves demonstrates that oil recovery can be improved by increasing the water rate from the sink. This shows that the technology counters the cone development making the bypass oil available to the well.5.2 DWS Well Productivity Index Comparison The DWS technology incorporates two completions in one well and the target is optimum oil production at the top completion. The improved recovery represents the level of bypassed oil due to water coning in typical conventional completions. 1984). The Figure shows that oil recovery increases with increasing drainage rate at the water sink completion. we compare the productivity index of the combined fluid production from the two completions in a DWS well with the PI of equal fluid production from a conventional well producing both oil and water. Consequently. In effect. This is accomplished by varying the drainage water rate at the sink. One approach compares the total fluid (Combined Oil and Water production) productivity index of the DWS well with an equivalent fluid productivity index of a conventional well completion. The counter-cone development around the wellbore means reduction in amount of bypassed oil and consequent improvement in oil recovery. This allows operators to improve oil recovery over a shorter time and accelerated investment recovery. 3. Figure 3-17 demonstrates the 76 .

The results show that the DWS completion does not necessarily deplete reservoir energy faster than conventional wells for the same liquid production rates. In this figure. This is compared with PI due to the liquid production and pressure drawdown in the conventional well.1065-935dws Phc. Fluid flow involving vertical component of permeability is enhanced in conventional wells than in DWS wells. the field total fluid withdrawal capacity of a DWS well is compared with that of a conventional well.results of this approach. The reason for the above result/observation could be due to the value of vertical permeability of the reservoir. Field Productivity Index (stb/day/psi) 8 7 6 5 4 3 2 1 0 0 1000 2000 3000 4000 Time (Days) 5000 6000 7000 8000 Phc.C2000 Figure 3-17 Field liquid PI Comparison between DWS Well and Conventional Well 77 . In order to obtain the field productivity index. In most reservoirs vertical permeabilities are much smaller that horizontal permeability due to overburden load and compaction effects. the total liquid production rate is divided by the field pressure drawdown caused by production from the dual completion.

In practice. the wellhead pressure is kept constant as well as the liquid rate. This causes increased dissipation of reservoir energy for the same fluid rate for the conventional wells as in DWS wells. 7 6 Oil Productivity Index (Stb/Day/psi) 5 4 3 2 1 0 0 1000 2000 3000 DW S Oil PI 4000 Tim e (Days) 5000 6000 7000 8000 Conv. Also. watercut increases rather more dramatically for the conventional well than for the DWS well. partially penetrating.The withdrawal of fluid in conventional. well causes the OWC to move up over time. The limiting watercut is also greater for conventional wells. it is possible to observe that the effect of increased water production translates to increased drawdown pressure in both cases. Thus. However. for the particular reservoir studied. for the simulator. increased watercut is accompanied by a reduction in fluid rate and reduced wellhead pressure. W ell Oil PI Figure 3-18 Oil Productivity Index (OPI) Comparison 78 . This upward movement combined with the typically low value of vertical permeability creates increased reservoir pressure drawdown and attendant low fluid production.

for the same pressure drawdown. Consequently only the valuable oil production should be considered. the DWS completion shows tremendous superiority over the conventional completion as shown in Figure 3-18. the productivity index of oil production (OPI) is compared (Figure 3-18). The corollary of the above results is that. It also assumes that reservoir energy is sufficiently strong. Total Oil Production (Million stb) 20 18 16 14 12 10 8 6 4 2 0 0 100 200 300 400 500 600 700 Pressure Drawdown.Oil DWS-Oil Figure 3-19 Oil Recovery Plot for Same Drawdown Pressure 79 .In another more pragmatic approach. the DWS well is capable of delivering more oil than the conventional well without necessarily depleting the reservoir energy much faster. reservoir has strong water drive. This approach assumes that the water produced from the sink is environmentally friendly and could be disposed overboard at little or no handling cost. That is. psi Conv. Thus the oil productivity index is the ratio of oil production rate and pressure drawdown from the top completion compared with the ratio of oil production rate and pressure drawdown for the conventional well. Again.

The heavy line shows the performance of the DWS well and the thin line shows that of the conventional completion. Stabilised PI (stb/day/psi) 7 6 5 4 3 2 1 0 0 500 1000 1500 2000 2500 Fluid Flow Rate. This result further emphasizes a more attractive aspect of the technology. Figure 3-20 shows a plot of stabilized oil productivity index as a function of liquid withdrawal rate at the top completion. indicate that total oil production over the period of twenty years (cumulative oil recovery) was consistently better for the DWS well for this particular field. which is the ability to improve oil productivity index by increasing the rate of liquid production at the top.In a similar study of cumulative oil recovery over a period of twenty years the DWS well proved to be capable of delivering more oil than the conventional well for the same pressure drawdown in the top completion as in the conventional well. (stb/day) Conv. The two lines also. The Figure 3-19 demonstrates the superior oil recovery capacity of the DWS well over conventional completion. Compln DWS Compln Figure 3-20 Improved Oil Productivity Index with Rate 80 . The diverging trend of the two lines with increasing pressure drawdown indicates that the DWS well performs even better at higher production rates.

the DWS technology demonstrated tremendous potential in reducing watercut in the oil zone completion by a significant level over time.5 1000000 0 .4 800000 0 .6 1200000 0 . 3. In this study performed on LVT Well for a JIP member (Joint Industry Panel).3 Watercut Performance Evaluation A final parameter used to measure the performance of the DWS water coning control technology is the evaluation of watercut in the top completion compared with a conventional well.7 1400000 0 . the Y-shaped curve represents the cumulative oil recovery with and without the application of the DWS technology and the other prongshaped curve show the reduction in watercut at the top completion due to the deployment of the DWS technology.3 600000 0 . the diverging trend of the DWS well PI curve over the conventional well curve demonstrates the superior performance of the technology. 4000 5000 6000 0 7000 Figure 3-21 Watercut and Cumulative Oil Recovery Performance 81 .9 1800000 0 . Again.1 200000 0 0 1000 2000 3000 T im e (D a y s ) D W S W e ll W a te r C o n v . D W S O il R e c .2 400000 0 .The flexibility of being able to increase the rate of oil production by increasing liquid rates from both the top completion and the water sink is shown to improve the stabilized oil productivity index of the DWS well as in Figure 3-20.8 1600000 0 . 1 2000000 0 . W e ll W a te r C o n v . O il R e c .5. In Figure 3-21.

surface water handling facilities have reached their designed capacity and watercut from the top oil zone completion must be reduced while producing clean disposable water at the bottom completion. Figure 3-21 represents the results of studies on the LVT well in La Victoria Oil field. relative permeability hysteresis. Other studies on the LVT-34 Well in La Victoria Field are summarized in Appendix B. 82 . The results show that by varying the water production rate at the sink completion. By varying the rate of water production from the sink. 2 Further. the technology is capable of increasing the cumulative oil recovery while decreasing the limiting watercut from the top completion to a manageable level for the surface water-handling facilities. dual completion with zonal isolation in addition to reservoir heterogeneities and anisotropy can be modeled more accurately using numerical simulators. requested by another of the Joint Industry Panel members. 3. The complexities of capillary pressure transition. This reduction in watercut oil reduces water separation costs and the problem of overload of water treatment/handling facilities.This parameter is most useful in cases where water handling capacity of surface facilities is limited. In a typical study.6 DWS Well Design Recommendations for Old Wells 1 The design and modeling of DWS well completions for old oil wells with severe water coning history should incorporate the dual effects of capillary pressure transition zone and relative permeability hysteresis. the cumulative oil recovery could almost double while the watercut could be reduced by over 20 percent. numerical reservoir performance simulators should be used to effectively evaluate the expected performance rather than relying on semi-analytic methods developed for new wells with sharp oil-water interface.

especially capillary pressure. To minimize reverse coning problems. the water sink should be located at a depth of +/.20 percent of the total thickness of the water zone below the OWC. core samples should be obtained to have representative data.3 Where possible. oil-water contact as well as rock and fluid compressibilities. 4 The location of the water sink below the oil water contact is also important. For a new well without capillary transition effects. 83 . relative permeabilities. 5 Finally. especially where the produced water must be oil-free for environmental reasons. employing the water sink from start of production does not necessarily give the best performance. porosity. The rate of drainage water at the sink should also be graduated in steps over time. The sink could be turned on when the water cone is still developing (around 50 to 60 percent watercut) for optimum performance. the water sink production rates should be optimized.

The objective of this chapter is to present guidelines for the design and operation of DWS wells in watered out reservoirs (reservoirs with saturation transition) to achieve production of oil-free drainage water. The concurrent production of oil and water in both completions defeats one of the objectives of the DWS technology of producing disposable oil-free drainage water.CHAPTER 4 DWS WELL DELIVERABILITY WITH OIL-FREE DRAINAGE WATER Several years of oil production in reservoirs underlain by water creates a zone of saturation transition between the original oil-water contact and the zone with connate water saturation as has been discussed in chapter 3. it minimizes the watercut in the top oil zone completion. However. Water quality regulations vary from one country to another and even from state to state. the goal of all environmental regulations is that only oil-free water should be disposed overboard or re-injected. It can allow operators to meet environmental regulatory requirements for re-injection and overboard disposal for offshore operations. 84 . This zone contains mobile oil and water that is capable of flowing co-mingled into either the oil well (top completion) or the water sink (bottom completion) or both completions concurrently. A brief overview of environmental regulations governing the quality of produced water in oil and gas industry is presented in section 4. At the same time. Production of oil-free water is attractive to the petroleum industry.1.

Table 4-1 gives an overview of the requirements for maximum concentration of hydrocarbons. removal of produced water from the hazardous waste classification does not preclude it from emission control. One of the E&P wastes exempted from the hazardous classification is produced water. Produced water has to conform to certain minimum requirements. the EPA issued a regulatory guideline stating that control of E&P wastes under the RCRA Subtitle C regulations of hazardous wastes is not warranted (EPA530K-95-003).1 E & P Produced Water Regulations and Requirements The Federal Environmental Protection Agency (EPA) believe that the large volume of “special waste” from E & P operations are lower in toxicity than other wastes regulated as hazardous wastes by the Resource Conservation and Recovery Act (RCRA). in 1988. Thus. These requirements regulate the level of oil and hydrocarbon contamination of produced water. However.4. Table 4-1 Requirements on Hydrocarbon Contamination of Produced Water Parameters Requirements as at 1980 Requirement as at 1993 (Maximum Concentration) (Maximum Concentration) 29 mg/liter Oil and Grease Concentration (monthly average) Oil and Grease Concentration (Daily average) 48 mg/liter 72 mg/liter 42 mg/liter The National Pollutant Discharge Elimination System (NPDES) regulates the requirements regarding maximum hydrocarbon contamination for offshore produced 85 .

ethylbenzene or toluene. Various governments around the world have discharge limits for residual oil in produced water that range from 23-40 mg/l (about 21 – 40 ppm).water disposal in the United States. In some cases. “Average oil content of produced water in Norway in 1998 was 23 mg/l (21 ppm). benzenes. which in practice. demands re-injection of produced water. Polyaromatic Hydrocarbons (PAH) and Alkylphenols cause most worry. In offshore environments where environmental regulations are most stringent. The produced water may be injected into the lower part of the producing reservoir for pressure maintenance. amounting up to 49 tonnes in 1998” (Wolff. giving an oil discharge of about 2100 tonnes from about 100 million tonnes of total annual produced water. Some have also imposed zero discharge limits. clean produced water is disposable overboard saving substantial costs. Other state regulations in the US target organic carbons. This produced water disposal requirements depend solely on the level of hydrocarbon contaminants as in the OCS (Outer Continental Shelf) of the Gulf of Mexico and in the Niger Delta Basin. The first field application of DWS technology was in the Nebo-Hemphill field located in LaSalle Parish in Louisiana. 2000). the produced water could be re-injected. 4.2 Overview of DWS Environmental Performance The major environmental merit of the DWS technology is the ability to produce oil-free water and water-free oil with the separate completion. This technology not only permits production of environmentally friendly fluids. The environmental performance of the technology 86 . Aromatic compounds like Benzene and solubles such as Phenols make up additional 1200 tonnes. it counters the cone height growth around the wellbore overcoming the problem of bypassed oil in the reservoir.

In land operations.48 0.0 Ethylbenzene mg/l 1.94 System water refers to water samples taken from the separator facilities from conventional wells in the field.38 34.0 Xylene mg/l 1. This ability to produce oil-free water makes DWS extremely attractive in offshore operations as no water processing is needed prior to overboard discharge. and Wojtanowicz. 2Endpoint system water Drainage Water 11.66 6. oil-free water may be injected into the same well 87 . Table 4-2 Hydrocarbon Contamination of Drainage/System Water Parameter Total Dissolved Solids Oil and Grease Unit mg/l mg/l Detection Limit (DL) 1.0 1 Average of two measurement.was outstanding as shown in tables 4-2 and 4-3 (Swisher.74 26.0 2.32 System Water Mid-point1 End-point2 536.33 592.32 ND ND ND ND ND ND 11. An analysis of produced water from the sink and water from conventional completions in the field after one year of production indicated that the level of contamination of the produced water from the sink was below detection limit.25 0.61 450.100 484 <DL <DL <DL <DL Bezene mg/l 1.61 492.0 Toluene mg/l 1.54 0. 1995).23 1.70 9. 2End point system water Table 4-3 PAH Contamination of Drainage /System Waters Component Naphtalene Phenanthrene Fluorene Dibenzothiophene Anthracene Pyrene Other PAH Total PAH ppb ppb ppb ppb ppb ppb ppb ppb Unit 1 Effluent from Free Water Knockout.11 12.300 <DL <DL <DL <DL <DL System Water1 69.0 Drainage Water1 63.17 ND 0.35 6.

The shaded region in Figure 4-1 represents the reduction in the size of the IPW due to capillary transition zone. NO H YS H Y S 3 0 ' C A P : O il B t ZERO CA P. However. A 400 300 Water Rate (bbl/day) 200 100 0 0 50 100 150 200 A 250 O il R a te (b b l/d a y ) H Y S 3 0 ' C A P :W a te r B t ZERO CAP. NO HY S Figure 4-1 Reduction in Size of IPW Due to Capillary Transition Zone 88 . 4.3 Capillary Transition and the IPW and Clean Water Drainage In the previous simulations performed using the linear approximation of capillary transition (Chapter 3). The phenomenon indicates that there is an enlargement of transition zone (with mobile oil and water) due to the dynamic production process to a level much larger than that explained by the static capillary pressure effect. the size of the IPW was found to be smaller for the reservoir with capillary transition zone.below the drainage completion or lifted to the surface and disposed into injection wells without processing. field tests and some studies have also shown that sustainable drainage of oil-free water becomes somewhat difficult as the two completions (top and bottom) may receive co-mingled inflows of the two fluids.

0000 0 200 400 600 Fluid Production at Sink (BLPD) 0. the inclusion of capillary transition zone to account for the zone of changing water saturation is necessary.0250 0.0500 0.2500 0.0000 800 Contaminated Fluid Production Belt WC OC Figure 4-2 Shows Zone of Concurrent Production of Contaminated Fluid in both Completion This observation elicited the need for a better understanding of the mechanism of this concurrent fluid production in both completions. A-A showed concurrent production of contaminated fluid in both completions (Inikori and Wojtanowicz. 0.0450 0.0050 0.A section across the apex of the shaded region. 2001(B)). 4.0400 0. JIP members supplied relevant field data to initiate a numerical study of this phenomenon and to recommend guidelines for oil-free drainage water production.1500 0.0100 0.1000 0. 89 Oil Cut in Bottom Compleetion (Fraction) Water Cut in top completion (Fraction) .0350 0.0300 0.2000 0.0200 0.4 Modeling Clean Water Performance of DWS in Old Fields In order to effectively model the performance of old field with long production history.0150 0.

4. was used for this study.9604 1475 220 1489 225 71. 90 . Table 4-4 shows reservoir data employed for this study.9552 9596 .27 1.25 0.1 Model Description A 40 x 1 x 23 radial grid model of an actual reservoir. In order to represent this shape in the radial grid model. The anticline reservoir structure is slanted on either side of the well creating a dome shaped profile. Reservoir thickness is 85ft (40 ft net oil thickness and 45 ft water zone thickness).11 32 1. Thus.20 7 Also.50 185 0.76 0.4. Capillary pressure transition was also represented in the numerical model to account for the severe water coning history. Table 4-4 LVT Reservoir Input Data INPUT DATA Reservoir pressure Thickness of oil/gas column Thickness of water column Depth of OWC Position and length of production perforations Position and length of water zone perforations Horizontal permeability in oil column Vertical permeability in oil/gas column Horizontal permeability in water column Vertical permeability in water column Water density at temperature Water viscosity at temperature Reservoir temperature Porosity in oil column Porosity in water column Oil formation volume factor Oil gravity / gas gravity Oil/gas viscosity at temperature (@BHP Completion diameter/hole size Unit psi ft ft ft ft ft mD mD mD mD Ib/ft2 cP O F Fraction Fraction Rb/stb API cP inches LVT Reservoir 4200 40 45 9589 9540 . LVT. the model assumed a steady-state flow by assigning an infinitely large pore space to the bottom radial grids. certain grids were assigned zero porosity and permeability (Figure B-3 in Appendix B). the reservoir has a strong bottom water drive and high water coning history.

50 0. It has been adjudged suitable by researchers for conceptual evaluation studies as enumerated in chapter 3.800 1. 1986.000 0.00 1. 1986.00 2.00 0. 91 .4.4. et al.41 employed the linear approximation approach to obtain various sizes of capillary pressure transition zone. Fraction) Figure 4-3 Capillary Pressure Plot from Leverett J-function An alternative approach could be to use linear approximation equations to generate different sizes of capillary transition in order to quantify the effect of varying sizes on water coning performance (Weinstein. 1981).600 0. Willhite.200 0. The capillary pressure data from cores presents a useful tool in model a more practical performance of the DWS well for the oil-free drainage water production.400 0..000 Water Saturation (Sw.00 0. The results presented in section 4. 3.2 Capillary Pressure Transition Evaluation Capillary pressure data from adjacent field with similar reservoir properties were supplied by the operator.50 1. Using the Levereth J-function. a capillary pressure transition profile was developed (Figure 4-3).50 2. and Yokoyama.

3 Relative Permeability Curves Measured relative permeability data from cores in adjacent wells were also provided by the operator (Figure 4-4).200 0.800 0. The data indicate a water wet sandstone reservoir with a connate water saturation of 27 percent and a residual oil saturation of 19.4. hysteresis effects were not included in this model. The (capillary) transition zone is smallest at the reservoir boundary but increases toward the wellbore as shown in Figure 4-5.200 1.200 0. 1.200 0.2 percent. The data supplied represent an imbibition relative permeability measurement from cores in adjacent reservoirs.000 -0.000 0.800 1.400 0. fluid inflow to the well expands the size of the capillary transition zone (or swept zone for old oil wells) around the wellbore.600 0. 92 .600 0.400 0.000 Water Saturation. Consequently.5 Presentation and Discussion of Results For conventional wells with one completion.000 0.4. (Sw Fraction) Figure 4-4 Relative Permeability Curves from Cores 4.

3

2

1

Figure 4-5 Transition Zone Enlargement around Conventional Well

The numbers 1 through 3 show an enlargement of the capillary transition zone size from the static OWC toward the wellbore. The expansion is pronounced around the wellbore where the size approximately doubles. Most of the expansion occurs between the wellbore and a few hundred feet.

3 2

1

Figure 4-6 Transition Zone Enlargement around Dual-completed (DWS) Well

The enlargement effect of the transition zone increases for the DWS completion due to the dual pressure drawdown and diffusion effects (Figure 4-6). Like in the physical model, this diffusion causes a cross-flow expansion of the two fluids to concurrently 93

break through into the two completions. The level of expansion depends on the wellbore drawdown pressure as well as the reservoir flow capacity. An important corollary from this observation is that piston-like displacement concepts cannot be applied to low permeability reservoirs with capillary transition zone nor to old oil wells with severe watered-out zone.
4.6 Design Considerations for DWS Wells With Oil-free Drainage Water

The primary goal of this study was to understand the principles governing the concurrent production of oil and water in both completions and to formulate a design procedure for operating the DWS wells with oil-free water production at the sink. The study addressed the DWS Inflow Performance Window (IPW) under the conditions of diffusive transition zone. For conventional wells without capillary transition zone, the IPW typically consists of four regions (Figure 3-1). A triangular-shaped envelope represents the domain of segregated fluid production (water- free oil production from the top completion and oil-free water production from the water sink completion). This is the target region for environmentally sensitive areas. Below the envelope is the domain of water breakthrough into the oil completion. Above the envelope is the domain of inverse oil coning with oil breakthrough into the water sink completion. A fourth region beyond the apex of the envelope displays a “flip-flop” line representing a region of instability with some level of contamination in fluid production on either side. The DWS production schedule could also be implemented in the water breakthrough zone. This permits maximum oil production at the top completion with low watercut oil and at the same time produce oilfree water at the sink. 94

For the reservoir with capillary transition zone, this fourth region or “the flipflop” line representing line of cone instability is changed into an envelope of concurrent fluid production (Region 4 in Figure 4-7). It is characterized by a cross over of the water breakthrough and oil breakthrough lines. Thus, beyond the apex of the segregated fluid production triangle, both oil and water flow concurrently into both completions. A cleanwater zone, however, still exists below the oil breakthrough line (Region 3 in Figure 4-7).
1000 900 800 700 600 500 400 300 200 100 0 0

Water Sink Liquid Rate (BLPD)

Wa ter B r

eak thr

oug h

line

Region 4 A A Region 3
200 300 400

Region 1
2 ion Reg
100

th Oil Break
500

rough line

600

700

Top Liquid Rate (BLPD)

Figure 4-7 IPW Comprising Transition Zone Effect (LVT reservoir); Critical Rate lines Intercept and Diverge at Higher Rates Enveloping Region 4 of Two-phase Inflow at the Top and Bottom Completion

Thus, the oil breakthrough line represents the maximum rate of fluid withdrawal at the water sink completion to ensure production of disposable oil-free water. The task is to identify, a condition for maximum oil rate at the top completion while operating along the oil breakthrough line, i.e., producing oil-free water. An additional plot of watercut isolines has been superimposed on IPW in Figure 4-8. The plot shows increasing watercut with increasing rate of liquid production at the 95

top completion along the oil breakthrough line. Thus, there should be an optimum liquid rate at the top that should maximize oil production rate at optimum watercut along the section A-A of the Oil breakthrough line in Figure 4-7.
1200 Water Sink Liquid Rate (BLPD) 1000 800 600 400 200 0 0 100 200 300 400 500 600 700 Top Liquid Rate (BLPD)

WC, % =

0,

1,

7,

21,

47,

57,

67

Figure 4-8 IPW for LVT Reservoir With WC Isolines

250 Oil Rate (BOPD 200 150 100 50 0 0 100 200 300 400 500 600 700 Top Liquid R ate (BLP D )

A A

Figure 4-9 Optimization of Oil Production Rate for DWS Well With Oil-free Water Drainage

96

Rather it arises due the location of the sink with a region of mobile oil and water. the initial oil-water contact position for the well was unknown. This is shown in Figure 4-10. 97 . The positive values of the y-ordinate represents location of the water sink below the initial oilwater contact. The task is to investigate the performance of the water sink location in terms of liquid production. The prolonged history of water coning implies a zone of co-existing mobile oil and water. The contamination of water with oil in this case does arise from inverse coning of oil downwards. at the same time. The studies showed that locating the water sink completion in the swept zone causes the two completions to concurrently produce contaminated fluid. The conclusion from this investigation is that the water sink location should be below the original oil-water contact if environmental concerns are paramount. The result in Figure 4-10 shows that concurrent fluid production is immediately experienced in the two completions.A plot of oil production rate Vs total liquid rate at the top completion is shown in Figure 4-9. The negative values in the y-ordinate represent the location of the water sink (Lower completion) within the zone of saturation transition above the supposed initial oil-water contact. deliver oil-free water at the water sink completion. It gives an optimum top liquid rate at the point of maximum oil rate. a top liquid rate of 275 BLPD and a bottom water sink completion rate of 285 BWPD would be recommended in order to maximize oil production with minimum watercut at the top completion and. 4. For this reservoir.7 Effect of Location of the Water Sink in the Capillary Transition Zone In another application of the technology by a JIP member.

Adequate field data and production logs should be run to understand the extent of water saturation transition development over time and the possible location of the original oil-water contact. 2. The following comments arose from the studies: 1. 98 . A good understanding of field history from start of production and location of original oil-water contact is necessary.8 Recommendations for Designing DWS Well and Production Schedule As explained above.-20 Location above or below the OWC (ft) -15 -10 -20000 -5 0 0 5 10 15 20 20000 40000 60000 80000 100000 Total oil production (stb) Figure 4-10 Oil Contamination in Water sink Completion placed within Capillary Transition Zone 4. The data constitute initial condition for the equilibrium represented by the IPW. the design of DWS well with oil-free water drainage is entirely controlled by capillary pressure data and the initial size of capillary pressure transition zone. Another control parameter is the placement of water sink completion associated with a risk of drainage water contamination with oil.

To avoid early inverse oil cone and oil breakthrough (initial inverse oil cone). the water sink location should be as deep as the limit of water handling capacity allows.3. 5. Capillary pressure data from core analysis within the field or correlation fields should be used to derive suitable capillary pressure data from the Leverett J-function correlation for the pre-installation studies. 99 . In the absence of core data. A height of about 20 percent of the total thickness of the oil or water zone (whichever is smaller) could be used as a “rule-of-thumb” to determine the depth of the water sink below the original oil-water contact. al. 6. 4. linear approximations could be used together with information from the production logs.. capillary pressure information could be obtained from electric resistivity log responses using a capillary profile match (Ibrahim et. The sink should not be installed a few feet below the oil-water contact or in the transition zone where mobile oil can easily flow into the water sink. In the last option. 1994).

1997. 1999.. al. Tehrani. and Verhyden. Yuan. However.. Most analytical work on horizontal wells has ignored the effect of pressure loss in the wellbore or at most treats it as insignificant (Papatzacos et. 2000. In some cases water cresting in horizontal wells erodes the merit of the technology (Heysel. al. et. as they tend to neglect the complex flow geometry of horizontal wells and its implication on fluid influx. These include: (1) effect of two phase oil-water flow in the horizontal well after water 100 . et. 2000). 1992). et. 1992. 1992. Weipeng.CHAPTER 5 EFFECT OF FLOW FRICTION ON WATER CRESTING IN HORIZONTAL WELLS Horizontal drilling technology has been widely recommended by several researchers as a solution for the development of bottom water drive reservoirs experiencing water coning problems (Nzekwu. Modeling of fluid flow behavior and pressure drop in horizontal wells in bottom water drive reservoirs requires the incorporation of several important factors. several field experiences indicate that horizontal wells do not eliminate the problem of water coning in such reservoirs (Nzekwu. Reservoir simulation tools and several theoretical evaluation techniques are largely inadequate.. Recent research efforts that incorporate the effects of pressure drop in the wellbore stopped at the single-phase flow regime or rather the prewater breakthrough period (Yuan. al. 1992). 1999. Hansen. al. 1991). Ozkan. 1993. 1991).. and Tang. Modeling of horizontal well inflow performance and fluid flow processes remains a challenge to the industry.

the analysis of the complex problem of water coning in horizontal wells. (3) the effect of emulsion viscosity on the pressure drop along the well. al (1992) observed that neglecting pressure loss in the wellbore due to friction as contained in most theoretical analysis is a major limitation of theory in field application. Reiss and Jourdan. Guo et. (2) understanding of the flow regimes of immiscible liquid-liquid flow in horizontal wells and the formation of emulsion. The frictional pressure loss causes a gradient in the pressure drawdown in the formation along the horizontal wellbore and the implication of this is that the water crest is higher at the down-stream (heel) end than at the upstream (toe) end. The implication is that the infinite conductivity model of 101 . In this chapter. The well Lacq 90. This report indicates that horizontal wells are not a complete solution to the problem of water cone development around producing oil wells. (also known as water cresting because of the development along the well length) will be presented along with all of the critical fluid and well length parameters that affect the phenomenon. (4) the effect of perforations/slots along the well and implication on pipe roughness. (5) the implication of axial influx of fluid into the wellbore. Thus. 5.breakthrough. water breaks through first at the heel and further oil production causes the water breakthrough to advance towards the toe. (in the Lacq Superieur field) produced at 19 BOPD in 4 years with a watercut of 99 percent. 1984).1 Limitations of Water Coning Control With Horizontal Wells – Field Evidence One of the earliest references to the problem of water cresting in horizontal wells underlain by water was the performance test by Elf Aquitaine (Geiger.

horizontal well analysis does not conform to practical evidence and the critical oil rate predicated from such analysis is usually over-estimated. Figure 5-1 Uniform Flux Model of Horizontal Well shows Uniform Water Crest Development along the Length of the Well Figure 5-2. shows another cartoon of a more realistic profile of the water crest in horizontal wells. Figure 5-2 Finite conductivity Model of Horizontal Well (Non-uniform Crest Development) 102 . Figure 5-1 is a cartoon of the profile of water crest when the effect of pressure loss in the wellbore is neglected and assuming a uniform flux model. It shows that water crest development is uniform along the well length. The water crest profile is skewed toward the heel and so water breaks through at the heel and then advances toward the toe.

Whole oil fields developed with horizontal wells are increasingly experiencing severe water cresting problems and mechanical intervention in these wells is currently very limited compared to vertical or deviated wells. and Kydland. Roux and Bosio. in the Norwegian sector of the North Sea. Lien et. al. Havig. Seines. 1991).In the first major test of application of horizontal well in the Troll field. believed. Mobil’s experience with a horizontal well to mitigate water coning in the Ness field. 1991). The Obagi field experience with laminated thin reservoirs indicates that horizontal wells may not be successful in such reservoirs with bottom water drive (Chugbo. The result of production logging test showed that 80 percent of the well was initially open to flow and that 75 percent of this inflow took place in the first half of the completed interval (Lien.000 mD. The productive section of the well length was 1640 ft in a reservoir with permeability range of between 3. UK sector. the completed horizontal well showed rapid rise in watercut averaging nearly 65 percent despite all careful well planning efforts”. however. attributed the performance to geological uncertainties (Koonsman and Purpich.15. was reported to be unsuccessful. 1989). Several other research efforts have shown that horizontal well productivity is highest at the heel than at the toe.000 – 10. The cleanup production test indicated a PI above 6000 stock-tank m3/day/bar (2600 stb/day/psi) and a kV/kH ratio of 0. In summarizing this section. it is important to state that the use of horizontal wells in the development of reservoirs with bottom water drive does not eliminate the problem 103 . however. that the “unproductive 20 percent was caused by insufficient cleanup in certain intervals”. “In a field where average watercut from conventional vertical wells was 40 percent. They. the producing horizontal section was completed with an 8-inch pre-packed slotted liner to reduce wellbore friction pressure loss.

identified six flow patterns for oil-water flow in pipes under the two broad categories 104 . These important observations form a critical basis for solving the problem of water crest development in horizontal wells. al.2 Model of Oil-Water Flow in Horizontal Wells Effective modeling of the performance of producing horizontal oil wells subject to water cresting requires evaluating the following concepts: (1) flow regimes and reservoir conditions of producing horizontal oil wells.2.1 Flow Regimes and Conditions in Horizontal Wellbore Presently. no characterization work has been done on the phase distribution of oilwater flow in horizontal wells. the water crest development in horizontal wells is rather skewed toward the heel of the well and not uniformly developed. Trallero et. (2) effect of the flow of oil-water mixture on the fluid properties under reservoir conditions. Horizontal wells have the added dimension of perforations and radial fluid influx creating increased turbulence to the main flow. All of the existing research on phase characterization relates to liquid-liquid flow in horizontal pipes. Mixture of Oil-water flow in horizontal pipes has been characterized into two broad categories. Another broad characterization of oil-water flow is the definition of the flow pattern as either oil-dominated or water-dominated. namely. segregated flow and dispersed flow. 5.of water influx and that at the same time. (3) effect of axial fluid influx into the horizontal well with fluid flow along the wellbore and the nature of wellbore friction pressure loss in horizontals wells. 5. Segregated flow shows a continuity of both phases in the axial direction while dispersed flow shows a disruption in the continuity of any of the phases.

For the specific experimental set up. et.. and (b) dual dispersion.. Another important aspect of the phase inversion phenomenon is that friction pressure loss in the pipe is critical in the oil-dominated flow period while slippage between the liquids is critical in the water-dominated flow regime. al. Dispersed flow was divided into two headings: (1) water dominated dispersed flow made up of. al. 1999). produced stratified flow and between 7. These flow velocities corresponds to the laminar and turbulent flow regimes for the experimental setup.12 ft/sec or higher produced dispersed flow (Trallero. 1996). S. They observed that segregated flow could be either stratified flow or stratified with some mixing at the interface.. Flores et al (1999) reported that no segregated flow exists in deviated pipes with inclination greater than 33 degrees. An inversion point is reached at a certain concentration of water where the flow pattern changes from oil-dominated to water-dominated regime. 1990). 1989): 105 .65 ft/sec.of segregated flow and dispersed flow (Trallero. al.. This point called “phase inversion” has been typically characterized by a change in mixture viscosity as will be discussed later. Dikken observed that the typical flow regime encountered in horizontal oil wells are more likely to be turbulent than laminar considering the typical production rates of such wells (Dikken. The experiment did not represent the actual profile of horizontal wells as fluid influx was neglected in the property measurement section. developed a correlation for fluid phase inversion from oil dominated flow to water dominated flow in the fully laminar region (Arirachakaran. an average mixture velocity of 0 – 7. et. and (b) emulsion of oil in water. et. Arirachakaran et. al.65 ft/sec and 13. and (2) oil dominated dispersed flow comprising of (a) emulsion of water in oil. (a) dispersion of oil in water.

either as a result of sudden pressure drop across a choke valve on the wellhead or as a result of turbulent mixing in the wellbore or in transport pipeline” (Ronningsen. Also.. However.0.. amongst others. INV = 0. The emulsion formed by the agitation process of the two immiscible fluids under horizontal well conditions (oil and water) are.500 . The combination of the reservoir conditions and the axial influx of fluid into the flowing wellbore create the conducive environment for the formation of unstable emulsion. Schramm observed as follows: “If two immiscible liquids are mixed together in a container and then shaken. and watercut. al. an emulsion has been formed”(Schramm. the author has assumed a dispersed flow model for oil-water flow in horizontal wells as more representative of field experiences. unstable and can separate into their various individual components when the emulsifying conditions are removed. fluid mixture velocity.1108 log m 5-1 The evaluation of the above relation agrees with the observation of Soleimani et. This basic principle has been 106 . 1999). examination will reveal that one of the two phases has become a collection of droplets that are dispersed in the other phase. In fact. A. the phase inversion and general characteristics of the oil-water flow regimes depend on several other factors such as oil-water mixture viscosity.f w.). that a phase inversion seems to occur at a watercut of between 35 and 46 percent from oil-dominated flow to water-dominated flow (Soleimani. however. typical reservoir flow conditions are elevated temperatures in the magnitude of over 175o F. Combining all of the above analysis with the additional agitation created by the radial fluid influx in typical reservoir conditions. 1992). al. “Stable crude oil emulsions may form in systems containing mixtures of crude oil and formation water. 1995). (ed. et.

Thus the mixture density. and the mixture velocity are calculated as a weighted average of the concentration of each phase.2. The flow of oil and water mixtures under reservoir conditions in horizontal wells with axial influx of fluid creates a high level of turbulence conducive for the formation of emulsions.widely acknowledged in the petroleum industry but has not been applied in the mathematical model developments. 5.f w ) 5-3 where superficial velocities are: Q(o. mixture surface tension.w)/A Oil-Water mixture interfacial tension: 107 . The turbulence created by the influx of fluid coupled with the typical high production rates of horizontal wells encourages the dispersion of water in oil below the inversion point and also reduces the droplet sizes of the dispersed phase – a condition necessary for increased mixture viscosity.f w ) 5-2 Oil-Water mixture velocity: vm = vsw * f w + vso * (1 .2 Properties of Oil-water Mixture in Horizontal Wellbore The fluid mixture properties have been typically evaluated using weighted averages of the individual phase properties. Consequently. This assumes. the results of this study have assumed that the oil-water mixture in horizontal wells behave like unstable emulsion. a no-slip condition. Thus mixture density is: r m = r w * f w + r o * (1 . Thus oilwater flow in horizontal wells is best modeled as unstable emulsion and is very different from the flow of oil and water in pipes at surface conditions.

2. the volume fraction of the dispersed phase (water)”. 1995) Other researchers in the chemical industry have performed extensive research on the viscosity of oil-water mixtures (Broughton. A series of experiments conducted in the present study (Ronningsen’s) in fact indicate a fairly small effect (of droplet size distribution) even at 50% watercut. 1952) is widely used in the chemical industry. et. 1992). the relationship for the mixture viscosity is not as simple. al. 1932.s m = s w * f w + s o * (1 . (ed. and Squires. They developed a rather specific relation for evaluating the mixture viscosities for the range of oils studied. “The viscosity of an emulsion generally depends on: 1 Volume fraction of the dispersed phase 2 Viscosity of the continuous phase 3 Shear rate (if non-Newtonian) 4 Temperature 5 Viscosity of the dispersed phase 6 Nature and concentration of the emulsifying agents 7 Average droplet size and size distribution 8 Presence of solids in addition to dispersed liquid phase 9 For water-in-crude oil emulsions the viscosity of the dispersed phase (formation water) can be regarded more or less constant. …. 5. m m = mo * e b + K * f w 108 5-5 . … The most important factor that influences the emulsion viscosity. 1988). …… The viscosity of an emulsion is directly proportional to the viscosity of the continuous (oil) phase. Richardson.f w ) 5-4 However.2. al. 1937. observed that accurate prediction of the apparent mixture viscosity of oil-water mixture is a difficult one because the mixture exhibits non-Newtonian rheological behavior and so their apparent viscosities can not be simply calculated by using standard weighted average rules (Martinez.). Schramm. (Ronningsen.1 Oil-Water Mixture Viscosity Martinez et.. The modified Richardson equation (Richardson.

m is viscosity.lim ø è mø è 2. is the dispersed phase concentration at which the relative viscosity becomes 100. ö f w / f w. Figure 5-3 presents an overview of the divergence between mixture viscosity predicted by the Pal and Rhode’s relation and the current industry approach of weighted average. 1995). The hatched line (tagged Ronningsen in the legend) is the correlation derived from the experimental data on a wide spectrum of North Sea crude. The Pal and Rhodes relation correlates both Newtonian and nonNewtonian emulsions and the normalization of the watercut with the limiting watercut takes into consideration the effects of other forces other the hydrodynamic forces. However.49 5-6 where the fw.).lim æ mo ö æ ç ÷ mr = ç çm ÷ ÷ = ç1 + 1.lim. 109 . (ed. 1992). the more generalized and more recent relation in the chemical industry for oil-water mixtures is the Pal and Rhodes relation (Schramm. RØnningsen performed extensive research on viscosity of oil/water emulsions for samples of North Sea crude and developed a Richardson-type relation for the crude samples studied (RØnningsen. b is also a constant. K is the Richardson constant. The thin line (Pal & Rhode in the legend) is the mixture viscosity from the Pal and Rhodes equation and the heavy line (tagged Wt Avg in the legend) is the mixture viscosity from the weighted average approach. which depends on the system and fw is the concentration of the dispersed phase in the continuous phase (water fraction for water in oil emulsion).187 .( f / f ÷ w w .

al.0 Mixture Viscosity (cP) 100.. As a check.00 mo = 1.0 1.80 1. Avg P al & Rhode Ronningsen Figure 5-3 Oil-Water Mixture Viscosity Evaluation The Figure show the dramatic increase in oil-water mixture viscosity with increasing watercut using the Pal and Rhodes relation compared to the current weighted average approach employed by most commercial numerical simulators and researchers (Schlumberger.0 0.5 cP 0.00 Watercut in Oil (fraction) Wt. 1993).1 0. et.1000.40 0. the performance of mixture viscosity with increasing watercut as measured by Rønningsen’s experiment for the North Sea oil with the same properties has been included for comparison. Geoquest. 1997. Seines. Thus.34 cP mw = 0. The results of the experimentally derived correlation compares more closely with that of the Pal and Rhodes relation than the current weighted average approach widely used by the industry. 110 . the results presented in this research have incorporated the Pal and Rhodes relation.60 0.0 10.20 0.

most commercial simulators and researchers have found the Haaland’s equation more useful and potable.5.al. Three equations have been widely used by the petroleum industry in evaluating friction factor in pipe flow as well as horizontal wells. 1996).9 æ e ö10 9 ö 1 = -1. (Jr.8 logç +ç ÷ ÷ ç N Re è 3.2. the Haaland’s equation and the Blasius equation for smooth pipes. Thus numerical simulators have found it more efficient (VIP-Executive Reference.7 D ø ÷ f è ø 5-8 111 . Most commercial pipes used in the petroleum industry have in-situ roughness. The Blasius equation is a one-step solution but it is developed for smooth pipes and for Newtonian fluids..2. Thus. The Haaland’s equation. They are the Colebrook relation.). æ 6.2 Effect of Mixture Viscosity on Reynold’s Number/Friction Factor The effect of the mixture viscosity will be presented in terms of the impact on Reynold’s number and the friction factor. The Colebrook equation involves iterative processes and requires more computer time. relations is given as (Bourgoyne. Neglecting this pipe roughness function reduces the accuracy of the calculated friction factor with increasing flow rate. 1991): 1 æ 1. The Haaland’s relation incorporates the in-situ pipe roughness function and is a one-step solution. et.255 = -4 logç 0. on the other hand does not involve iteration.269 e + D ç f N Re f è ö ÷ ÷ ø 5-7 The equation involves the friction factor on both sides and so requires an iterative process for its evaluation. The Colebrook.

00 1. 112 .20 5-9 Q = 5.20 0.The VIP executive reference manual also stated that the pipe absolute roughness. f = 0.00 0. e.5cP.000 stb/day.0791 0.60 0.67 ppg.40 0.34cP Water Cut (Fraction) Weighted Average Emulsion. using equation 5-9.58 ppg. The effect of Reynold’s number on friction factor is in turn presented for smooth horizontal pipe. The Blasius equation represents a straight-line approximation of the Colebrook function on a log-log plot (Bourgoyne. mw = 0. Pal & Rhodes Figure 5-4 Effect of Mixture Viscosity Model on Reynold's Number All of the above functions assume a turbulent flow model for the typical oil rates encountered in horizontal wells. “can represent the effect of turbulence created by inflow through perforations”. rw = 8. (Jr. ro = 6. 1991). As an example Figure 5-4 presents the effect of correct application of the mixture viscosity on the Reynold’s number. mo = 1. The Figures compare the results of the Pal and Rhodes relation for mixture viscosity with the weighted average approach as a function of increasing watercut.80 1.25 N Re 300000 Reynold's Number (Dimensionless) 250000 200000 150000 100000 50000 0 0.).

60 0.00800 0.90 1 . The diverging trend confirms current observation by researchers that increasing watercut increases the mixture viscosity until the inversion point where the continuous 113 .58 ppg. rw = 8.30 0.00600 0.40 0.01 000 0.20 0.80 0.34cP 0. mw = 0.00 0.01 200 FanningFriction Friction Factor Fanning 0. As expected the decreasing trend of the Reynold’s number in Figure 5-4 translates to an increase in the friction factor and consequently. an increase in the pressure drop along the wellbore as watercut increases.00400 0.70 0.00200 0.5cP. Pal &Rhodes Figure 5-5 Effect of Viscosity Model on Fanning Friction Factor The continuous line represents the friction plot with increasing watercut using the Pal & Rhodes equation and the hatch line represents the expected friction factor with increasing watercut using weighted average considerations.67 ppg.The results show how dramatic the friction pressure loss can be when the appropriate model of liquid-liquid mixture viscosity is used. ro = 6. mo = 1 .50 0.1 0 Q = 5.000 stb/day. 0.00 W ater Cut (Fraction) W eighted Average Emulsion.00000 0.

NRe. 1999). is a function of pipe roughness. pressure loss in pipes/wells consists of several components (Brill.. in a study on oil-water flow in horizontal pipes (Arirachakaran. 114 . al. fr v 2 æ dP ö .phase changes from oil to water. reduces the effect of the increasing friction factor since friction pressure loss gives way to slippage between the fluids in the water-dominated flow regime. et. represents the (wall) friction component.3 Frictional Pressure Loss in Horizontal Wellbore In general. This effect of increased slippage is typically not included in most computer models as these models assume homogeneous fluids. 5-11 This term representing the acceleration of flow due to the fluid influx has frequently been evaluated and incorporated in the friction pressure loss term. For two-phase flow. æ dP ö æ dP ö æ dP ö æ dP ö +ç +ç =ç ÷ ÷ ÷ ÷ ç è dL ø Total è dL ø accelerati on è dL ø elevation è dL ø friction 5-10 For horizontal pipes/wells. e. about the inadequacy of the weighted average concept of mixture viscosity is also confirmed by Arirachakaran et. f. the elevation term is usually neglected (equal to zero). 5. made here. other factors such as fluid interaction may become important beside pipe roughness and Reynold’s number. the friction factor. and the Reynold’s number. however. rvdv æ dP ö ç ÷ = è dL øacc g c DL is the kinetic energy term. ç ÷ = è dL ø fric 2 g c D 5-12 For single phase flow. al. The observation.2. The concept of phase inversion from oil-dominated flow regime to waterdominated flow at about 35 – 50 percent watercut. 1989).

These additional factors could increase the friction factor by over 500 percent as will be seen from the results of this research presented in the succeeding sections of this chapter. Su and Gudmundsson (1993) investigated the effect of perforation pressure loss in pipes for single-phase fluid. The reason for this conclusion is because the mathematical models developed for horizontal pipes do not consider other effects such as perforations and axial influx of fluid into the wellbore. 115 . introducing the subscript. However. The author employed the relation of Su and Gudmundsson to evaluate the effect of perforation on friction factor. 1993). for mixtures. the equation for the mixture friction pressure loss becomes: 2 f tp r mvm æ dP ö ç ÷ = 2 gc D è dL ø fric 5-13 In field units. and Gudmundsson. result from field experiences of horizontal well performance does not support the fact that friction pressure loss is negligible.Thus.units ) ç ÷ = è dL ø fric 25. 1991). analysts have maintained that friction pressure loss in horizontal well is very small compared to the reservoir pressure drawdown and could be ignored (Joshi. m. equation 5-13 is written as: fr v 2 æ dP ö ( field .8 D 5-14 Until recently. More importantly. They did not include the effect of fluid influx. their results indicated that perforation effect could increase the friction pressure loss by as much as 15 percent or more. Su and Gudmundsson developed a three-step mathematical model for evaluating the friction pressure loss in perforated pipe (Su.

D = 4-1 /2-inch. Dikken performed an extensive research on the effect of pressure drop along the wellbore on horizontal well performance.00800 0.59ppg.01 000 0. d = 0.316ç ÷ p 2 D5 ç 4r ÷ ø è a 5-15 Dikken analyzed the pressure drop along the wellbore as: æ pm D ö 8r æd ö 2 -a ÷ * 2 5 Q ( x) 2 ç ÷ Pw ( x) = RwQ( x) = 0.00600 0.0. Gudmunsson B lasius (Smoo th pipe) Figure 5-6 Effect of Perforation Without Fluid influx on Wellbore Friction Factor The result show an increase of about 35 percent compared with the Blasius relation for smooth pipe. j=1 .5 cP. m w = 0.34 cP . He developed a mathematical expression for flow resistance. This result is presented in Figure 5-6. Rw.316ç ÷ ç è dx ø è 4Qr ø p D a 5-16 Expressing the equation 5-16 in terms of the Blasius expression for Moody friction factor gives: 116 .00000 0 20000 40000 60000 80000 1 00000 1 20000 2 spf. mo = 1 r o = 6.00200 0.01 200 0.00400 0. r w = 8. which incorporates friction factor for turbulent flow: æ pm D ö 8 r Rw = 0. Fanning Friction Factor Reynold's Number (Dimensionless) Su.5-inch.67ppg.

and n is the number of perforation. Q is the flow rate. et.25 for smooth pipe (Dikken.316 8r 8r æd ö 2 -a * 2 5 Q ( x ) 2 = f M * 2 5 Q( x ) 2 ç ÷ Pw ( x) = RwQ( x) = a ( N Re ) p D p D è dx ø 5-17 fM is the Moody friction factor. Yuan. for friction pressure loss in horizontal wells. Recent research efforts have included the effect of fluid influx and perforation for single-phase fluid (Yuan. r is the density of the fluid mixture.. 1999. The result of Dikken’s research further corroborates the author’s observation that the equations of friction pressure loss in pipes are quite inadequate in evaluation of horizontal well performance. Ashiem and Kolnes developed a correlation. he observed that the Reynolds’ Number exponent. Q is the main flow rate. al. The correlation accounted for the influx of fluid into the wellbore. 117 . The expression assumes a fixed value for the ratio of fluid influx to the main flow rate (Ashiem. While Dikken did not incorporate an expression for the fluid influx. et. D is the pipe diameter and fT. This translates to an over three fold (three hundred percent) increase in friction factor for perforated pipe over smooth pipe. 1990). The expression for wall friction follows the Blasius-type relation for smooth pipe and so neglects the effect of the pipe in-situ roughness. a. 1997)..15 for perforated pipe to 0. from experimental studies.16 ì ï æ qi ö æ D öæ qi ö ü ï ç ÷ ç ÷ + + 4 D 2 í ç ÷ ç ÷ç ÷ ý 0. in the Blasius equation for friction factor could vary from 0. 1992). is the total friction factor. al. D is the pipe diameter. 2 0.19 Q n Q N Re ï î è ø è øè ø ï þ fT = 5-18 qi is the fluid influx rate.0.

f T = f w + Cn * 2 D * j * qi Q 5-19 The authors used a Blasius-type expression for the wall friction factor as: f w = a b N Re 5-20 Table 5-1 Experimental Values for the Constants. Evaluation of the results of the experiment indicates that pressure the friction factor could increase by as much as four fold or more depending on the ratio of fluid influx to main flow rate (Figure 5-7).332 2.300 1.694 # Data from Yuan’s correlation and * represents data from Yuan’s experiments Table 5-1 represents values for the constants a.622 0.287 0.200 2.901 2.341 0.339 0.368 0.al.873 0.590 4.297 0.344 2.532 1.271 0. 1 2 3* 3# 4 5* 5# 6 7* 7# 8 9 SPF 5 5 5 5 10 10 10 10 20 20 20 20 PHA 90 180 360 360 90 180 180 360 90 90 180 360 a 0. 118 .266 0. b and Cn No.159 0.308 0.421 0.078 b 0. b and Cn presented in equations 519 and 5-20 (Tang.312 0.346 Cn 2.590 0. a.363 0.169 2.155 0.. Tang and Yuan research work used a Blasius-type relation to evaluate the pipe wall pressure loss and does not incorporate an expression for pipe roughness function.489 2.755 1. et.028 2.300 2.505 0.312 0.200 2. 2000).300 3.641 0.

5 cP. One task of this research therefore.67ppg. the Blasius relation for friction pressure loss is applicable to Newtonian fluids only.030000 Fanning FrictionFactor Factor FanningFriction 0.01 5000 0.0. mw=0.025000 0. is to develop a “generalized compound friction pressure loss relation” that effectively models the performance of horizontal wells and reduces to the normal horizontal pipe flow relations for both commercial and smooth pipes.000000 0 20000 40000 60000 80000 1 00000 1 20000 Reynold's Number (Dimensionless) Y uan. Finally. the experimentally derived correlations are case specific and require that a specific set of constants be applied for different flow conditions.020000 0.01 0000 0.035000 0. did not quite produce the constants obtained from the experimental data (Yuan. 119 . al. Blasius (Smooth pipe) /2-inch. rw=8.(1999). ro =6.5-inch. 1997).005000 0. al. PHA= 3600 . mo =1 . qi/Q =1 /200 Figure 5-7 Effect of Perforation and Fluid Influx on Wellbore Friction Again.59ppg. et.34 j=5 spf. D =5-1 cP. d =0. Also. the correlations developed by Yuan et.

75ï ç = 8í2. The evolving relation requires a three-step solution as follows: 1. However.7.1. The influx to main flow ratio is actually not constant but varies along the length of the wellbore. Employing. evaluate the Moody friction factor for the perforated pipe with fluid influx using the following equation. ì ü ææ N qi f MRI ö ï Re m ö ÷ + A . from step 1 above calculate a constant.8 log +ç ÷ ÷ ç N 3 .2.75ï ý ÷ ï ø þ 5-22 3. in the universal velocity distribution law for smooth wall. The component on 120 .11 æ 6. 1. 7 D ø ÷ è Re è ø -2 f MS 5-21 2.5 ln ç ÷ ý + 4.0. ì æN ï 8 ç Re Av = í 2 . Finally.5. by iteration.3. the influx to main flow ratio at the ultimate production rate is typically used for evaluation. fMs. Av.0 Dj v ÷ ç Q D ï ï èè 2 ø 8 ø î þ -2 f MRI 5-23 The Fanning friction factor could be calculated by dividing the Moody friction factor from step 3 above by the factor of 4. determine the Moody friction factor due to the in-situ pipe roughness.0d p . (f (Fanning) = fMRI/4). Employing Haaland’s equation for friction factor. The development of an adequate mathematical expression that effectively models the ratio of influx to main wellbore flow remains a research area. This was achieved by introducing the influx to main flow ratio concept from the Asheim and Yuan experimental data as well as a numerical constant.4 Generalized Friction Factor Correlation for Horizontal Wells and Pipes The first step was to modify the Su-Gudmundsson relation to accommodate the effect of fluid influx. fMs.9 æ e ö ö ç = . 5 ln ms ç 2 f ï è î f Ms 8 ü ö ÷ + 3.

qi/Q = 1 Max.5 Verification of “Generalized Friction Factor (GFF)” Relation The next step involves verifying the accuracy of the “generalized correlation”. j= 5spf. (Table 5-1) presents the most recent and detailed analysis and were considered an excellent calibration tool to verify the performance of the correlation for producing horizontal wells.01 00 0.01 50 0.34 cP.0050 0.0250 0.3 % Figure 5-8 Friction Factor Plot from 'Generalized Equation' Vs Yuan's Experimentally Derived Correlations (5SPF) 121 . The experimentally derived correlations of Yuan and the constants from her experiment.0200 0.2.0350 0. /1 00 ro rw = 8.5-inch. PHA =360o = 6.0000 0 20000 40000 60000 80000 1 00000 1 20000 Reynolds Number (Dimensionless) Yuan. Factor Fanning FrictionFriction Fanning Factor (Dimensionles s) 0.0 percent. Two plots of the verification exercise are presented below in Figures 5-8 and 5-9. D = 5-1 = 1 . The results show a very good fit with a maximum error margin of +/7.0300 0.Sarica Mod.67ppg. 5. d = 0.Gudmunsson /2-inch. Su. mo mw = 0.59ppg.the right hand side of the plus sign in equation 5-23 has been added to the Su and Gudmundsson relation by the author. Deviation = + /.5 cP.

Su.0000 0 20000 40000 60000 80000 1 00000 1 20000 Reynolds Number (Dimensionless) Yuan.Gudmunsson 0spf. d = 0.025 for all turbulent flow rates for the particular combination of fluid properties and wellbore geometry.0200 0. qi/Q = 1 Max.01 00 0.3 % Figure 5-9 Friction Factor Plot from 'Generalized Equation' Vs Yuan's Experimentally Derived Correlations (10SPF) 122 .01 50 0. Figures 5-8 and 5-9 also show that beyond a certain level of turbulence the dependence of friction pressure loss of the flowing fluid on the Reynold’s number decreases. = 80o mw 0. Deviation = +/.0300 0. The Fanning friction factor plot becomes almost horizontal indicating that friction pressure loss depends less on the fluid rate and more on the pipe roughness. 1997) and the solid line represents the results of the “generalized friction pressure loss relation” for the same fluid properties and flow conditions. PHA =1 mo /200 ro = 6.5 cP.The hatched line represents the plot of friction factor from the experimentally derived correlation of Yuan (Yuan.0250 0.5-inch.0350 0.Sarica Mod. j= 1 = 1 .34 cP.0050 0.67ppg. D = 5-1 /2-inch. Both plots also show that the friction factor is greater than 0. Fanning Friction Factor (Dimensionless) 0. rw = 8.59ppg.

The first involves checking the generalized friction factor relation for commercial pipes with the Blasius relation for smooth pipes (The Blasius equation neglects the effect of the in-situ pipe roughness). Two steps were taken.0004 50000 100000 150000 Blasius eqn Reynold's Number (Dimensionless) Mod.3. The close fit with the experimentally derived correlations and the generalized nature of the relation developed in this research makes it more attractive.67 ppg.0020 0.Again.59 ppg. mo = 1.0 percent. The third step involves testing the ‘generalized equation’ with the Blasius equation for smooth pipe in the absence of perforations and fluid influx.0040 0. Gudmunsson (With In-situ pipe roughnes) Figure 5-10 Performance of 'Generalized Equation' With in-situ pipe Roughness Vs Blasius Equation for Smooth Pipe 123 . e/D = 0.0060 0. rw = 8.5 cP. the plot in Figure 5-9 show a good fit within a maximum error margin of +/.0100 0. Most of the error occurs in the low Reynolds number region indicating that there is more perfect fit with increasing turbulence. mw = 0. The second step involves testing the performance of the GFF for a smooth pipe with zero in-situ pipe roughness. Fanning Friction factor (Dimensionless) 0. ro = 6.0080 0.34 cP.0000 0 D = 5-1/2-inches.

34 cP 50000 100000 150000 Blasius eqn Reynold's Number (Dimensionless) Mod. The result shows a close agreement (within +/. mw = 0. The plot shows an excellent fit for both cases.The result in Figure 5-10 show the amount of error introduced in the evaluation of friction pressure loss in commercial pipes when the Blasius equation for smooth pipe is used (as much as 10 – 15 percent error).0040 0.3 percent error) for smooth pipe consideration. Thus the relation could be applied to both horizontal wells and pipe flow.0000 0 D = 5-1/2-inches.59 ppg. the performance of the ‘generalized equation’ (also known as ‘modified Gudmundsson equation’) is compared with the Blasius equation for smooth pipes. Fanning Friction factor (Dimensionless) 0. This is demonstrated by the diverging trend toward the right. Gudmunsson (No pipe roughnes) Figure 5-11 Performance of 'Generalized Equation' Vs Blasius Equation for smooth pipe 124 . mo = 1.5 cP. rw = 8.67 ppg. the result from the equation is compared with the Blasius equation for smooth pipe.0020 0. In the next Figure. In Figure 5-11.0060 0.0100 0.0080 0. ro = 6. This error increases as the fluid flow rate is increased.

Appendix F shows sample 125 . et. al. Therefore. incorporating all of the above effects. 5. This ‘new pipe roughness’ function is then input into the numerical simulator.004 to 0.015 to 0. This shows the magnitude of error introduced when friction factor relations for smooth pipes are adopted for the modeling of horizontal wells. (Jr.. Users cannot edit the codes of most commercial numerical simulators.025 to 0. the friction factor for smooth pipes range from 0.008 representing an increase of over 400 percent to 625 percent. the approach adopted was to back-calculate an “equivalent pipe roughness” that will give the same friction pressure loss effect as the compound friction loss model.255 ö ÷ N Re f ÷ ø f - 5-24 Typical “equivalent pipe roughness” of the order of 0.).000065 ft. 1991) is re-arranged to solve for the new value of pipe roughness for use in the simulator.72d ç10 -1 / 4 ç è 1.An important observation from the foregoing analysis is that while the friction factor in horizontal wells with perforations and fluid influx effects range from 0. the results of application of the new compound friction factor in modeling water cresting will be presented.032. The Colebrook equation (Bourgoyne.1525 ft were calculated compared with conventional values of 0.3 Numerical Study of Water Cresting in Horizontal Wells With Compound Frictional Pressure Loss Model The essence of the studies on effective evaluation of friction pressure loss in horizontal wells. In this section. æ e = 3. is to effectively model horizontal wells subject to bottom water drive.

to accommodate the dual concepts of phase inversion and mixture viscosity. Di = 4. The results in Figure 5-12 show the tremendous increase in friction factor between the ‘new approach’ and the conventional ‘weighted average’ used by the commercial numerical simulators.02000 0. Avg Modified Factor (qi/Q=1/100 Figure 5-12 Friction factor comparison (Model Vs weighted average) in oil-water flow with phase inversion 126 .00 1.40 0.000 stb/day.60 Water Cut (Fraction) 0. mo = 1 . qi/Q =1 /1 00 (2).92-in.20 Modified Factor (qi/Q=1/200) Eclipse Wt. rw = 8. the new compound friction factor is calculated with the ‘modified Reynold’s Number’ for mixture viscosity up to the point of inversion.5 cP. j = 1 0 spf.20 0.05000 Fanning friction factor 0.03000 0.00 Q=9.67 ppg.59 ppg.80 1. Also. The phase inversion concept is modeled in the numerical simulator by re-completing the well at the watercut corresponding to the phase inversion point with a new value of pipe roughness.06000 0. Beyond the inversion point (Water dominated flow regime) mixture viscosity is calculated with the conventional weighted average method.5-in Lw = 1 500 ft 0. qi/Q =1 /200. 0.04000 0.00000 0. d(perf.values of equivalent pipe roughness calculated from the compound friction pressure loss relation. ro = 6.Do=5-1 /2-in.) = 0.01000 0.34 cP. mw = 0.

therefore in Figure 5-12 between the Eclipse weighted average plot and the “compound friction” weighted average plot with perforations and fluid influx” beyond the point of phase inversion shows the components introduced by the dual concepts of perforation and axial fluid influx.1 Time to Water Breakthrough Previous analytical tools that ignored the effect of pressure loss along the wellbore or used the friction pressure loss relations for horizontal pipes seems to overestimate the time to water breakthrough. These dual effects should not be ignored. These effects will be presented in the subsequent sections. A fourth point. Also evident from the Figure is the fact that the concept of phase inversion effect is over-shadowed by the dual effects of perforation and axial fluid influx in horizontal wells. is the level of bypassed oil in horizontal wells due to water cresting. The application of the new model affected three primary aspects of water crest development (Inikori. As shown in Figures 5-13 and 5-14. Beyond the phase inversion point. 5.1 Changes in Horizontal Wells Water Crest Development Due to ‘New Model’ To present a clear picture of the effect of friction pressure loss on water cresting in horizontal wells. and Wojtanowicz.3. 127 .Also the Figure 5-12 shows that the inclusion of perforation effect and axial fluid influx introduces a tremendous increase in ‘equivalent friction pressure loss’ in the horizontal wells. the friction factor for both cases employed the ‘weighted average’ approach. 2002).1.3. 5. the time to water breakthrough for the reservoir studied reduced from 150 days to 70 days. The wide margin. The friction factor increase could be as much as six fold depending on the ratio of influx to main flow. the compound friction model was applied to the numerical simulation of water cresting. which is a corollary of the crest profile.

4 0.1 0 0 50 100 150 200 250 300 350 Tim e (Day s ) HW -0.4 0.3 0.00065-5000 800 1000 1200 Figure 5-13 Early Water Breakthrough in Horizontal Wells 150 days has been predicted using the Eclipse Fanning friction pressure loss relation for pipe flow as well as the weighted average option for fluid mixture viscosity. Figure 5-14 is a highlighted version of Figure 5-13.1100-5000 HW-0.6 Water Cut (Fraction) 0.1100-5000 HW -0.0.2 0.7 0.6 0. 0.5 0.1 0 0 200 400 600 Time (Days) HW-0.8 0.3 0.2 0.00065-5000 Figure 5-14 Early Water Breakthrough but gradual Development of Watercut Water Cut (Fraction) 128 .5 0.9 0.

It shows that water tends to breakthrough earlier when adequate friction pressure correlation is used.000 stb/day production rate. The result of evaluation of drawdown pressure with conventional Eclipse simulation and the “new equivalent pipe roughness” function shows an increase in drawdown pressure. Some researchers isolate the pressure loss in the wellbore from the pressure drawdown in the reservoir due to fluid withdrawal. the water development is rather gradual. However. The effect of increase pressure loss in the wellbore translates to an increase in reservoir pressure drawdown (Figure 5-15). This is because water first breaks through at the heel and spreads gradually toward the toe of the well. As is expected. The change in time to water breakthrough represents a 50 percent reduction at 5. 129 . for this particular reservoir.2 Effect on Flowing Bottomhole Pressure Another effect of the increased friction pressure loss on horizontal wells is in the reservoir performance. This concept is not supported by current field evidence.3. the ultimate watercut is a sole function of the reservoir properties and this is demonstrated by the convergence of the two curves in Figure in Figure 5-13. This causes a dramatic increase in watercut as soon as the water breaks through into the well. Thus. The consequence of this observation is that the period of oil plateau production typically evaluated with commercial numerical simulators will be overestimated. 5. The effect is even more pronounced at increased oil production rate as shown by the diverging trend. In the infinite conductivity and uniform flux models.1. the water crest height above the Oil-Water Contact (OWC) is the same at all points along the length of the well. the water crest rises symmetrically and uniformly toward the wellbore.

. (Ozkan.3 Friction Pressure Loss and Water Crest Development/Profile Figures 5-16 and 5-17 show two different profiles of water crest development in horizontal wells.al. Figure 5-16 was generated using the modified pipe roughness function while Figure 5-17 was made using Eclipse weighted average and Fanning friction factor relations.0 0 0 6 5 Mo d if ie d Fr ic tio n e = 0 . al. 5.1 5 0 0 Figure 5-15 Friction Pressure Loss Effect on Horizontal Well Bottom Hole Pressure .60 Pressure Drawdown (psi) 50 40 30 20 10 0 0 5000 10000 15000 20000 25000 L iq u id Ra te ( s tb /Da y ) Ec lip s e Fr ic tio n e =0 . This equation is capable of accommodating any friction pressure loss correlation and could be used with the relation presented in this research. 1999). et. 2000).. 130 .3. Such analysts use the infinite conductivity equations to calculate the reservoir pressure drawdown and the smooth pipe flow equations to calculate the friction pressure loss (Jiang. et. The obvious conclusion from such analysis is that the effect of friction pressure loss in horizontal wells is insignificant.1. Ozkan has developed a more complete equation for reservoir pressure drawdown.

5000stb/day.Skewed toward the Heel of the Horizontal Well H W W FIP2. 5-1/2”. When water breaks through at the heel and spread gradually toward the 131 .000065. Figure 5-16 New Water Crest Development . 5000stb/day. 5-1/2”. 0.15000. e = 0. the water crest development is skewed toward the heel of the horizontal well.HWWFIP2. Figure 5-17 Weighted Average Approach shows Symmetrical Water Crest In Figure 5-16.

toe.. Secondary mechanical intervention is not yet well developed for horizontal wells. S. horizontal wells experiencing up to 80 percent watercut are yet producing 100 percent oil at the toe (Lien. Consequently. In Figure 5-16. In some cases. et.1. 1995). The profile indicates that water crest develops symmetrically toward the heel and the toe of the well with the peak at the middle section. If horizontal well cone water at the toe. often times. 132 . a lot of oil is left unproduced at the toe (upstream end) of horizontal wells due to water cresting (Permadi. often times.4 Water Cresting and Bypassed Oil in Horizontal Wells Another important aspect of water cresting in horizontal wells is the problem of bypassed oil.3. it creates a poor sweep efficiency and bypassed oil at the toe of the well. However.. when water breaks through at the heel. Thus.C. Figure 5-17 shows a symmetrical profile of water crest development along the wellbore for the weighted average approach and in-situ pipe roughness. 5. 1990). the watercut increases rather rapidly requiring that the well be abandoned and in-fill drilling may be required to drain bypassed oil. This picture indicates an efficient sweep by horizontal wells even with water cresting. Industry reports of horizontal wells water cresting supports this result. Field evidence have shown that. the solution is an easy one – run mechanical bridge plugs and isolate the toe section and continue production. bypassed oil should be at both the toe and the heel end (if any). oil production at the toe is yet 100 percent while the well watercut is about 84 percent. Field reports of horizontal well water cresting from production logs do not seem to support this result. the diagram in Figure 5-16 shows a more practical representation of water crest development in horizontal wells. Figure 5-16 adequately represnts the profile described above.al.

the author recommends that adequate multidisciplinary field studies be performed prior to the development of whole reservoir subject bottom water drive with horizontal wells. 133 .In concluding this chapter.

Two variants of the multi-lateral technology have been recommended and extensively studied for mitigation of water crest development in horizontal wells. Re-entry of existing wells for the purpose of drilling multi-lateral wells is also growing.al. 6... et. Thus.al.1 Feasibility of Water Cresting Control with Dual Completion Technology The petroleum industry has made tremendous technological progress in the area of multi-lateral well drilling and completions. 1995). a considerable length of horizontal well upstream section is never contacted by bottom water at even very high watercut (Permadi. Permadi. The technology of drilling multi-lateral sections from existing wellbore is even more attractive in offshore operations where the cost of adding one more slots to the subsea template could be expensive. They are: (1) a vertical well extension into the water zone and an upper section horizontal well 134 . 1997).CHAPTER 6 WATER CRESTING CONTROL IN HORIZONTAL WELLS WITH DOWNHOLE WATER SINK TECHNOLOGY The previous chapter argues that pressure gradients in horizontal wellbore are varied and so create non-uniform drawdown and fluid influx (Brekke. The chapter ends with the presentation of an extension of the principle of the downhole water sink technology for water coning control in vertical wells to water cresting in horizontal wells. The consequence is that water break through occurs at the heel end of the well and gradually extends to the toe. In this chapter. the author presents an overview of several solutions to the problem of water cresting in horizontal wells. 1992. et. and Lien.

Water Oil PCP 18’ Oil OWC Water 64’ Figure 6-1 Tail-Pipe Water Sink Completion Schematic 135 . The recommendation of the two approaches are based on the field evidence of the profile of water crest development in horizontal wells which indicates that the water first breakthrough at the heel and then extends along the length of the wellbore. The concurrent production of water from the water zone using the vertical tail pipe well or the lower horizontal well (in the case of the bilateral option) performs a dual role: (1) redistibution of the pressure profile along the wellbore but more specifically around the heel.targeted at the top of the oil pay for optimum oil recovery. (2) Two horizontal wells drilled laterally on top of each other with the upper section targeted at the top of the oil zone and the lower section targeted a few feet below the original oil-water contact. This concept will be called “the Tail-pipe Water Sink (TWS) option”. This approach will be called “the Bilateral Water Sink (BWS) technology option”. and (2) reduction of water cresting in the upper oil zone horizontal well.

The vertical tail-pipe is simply an extension of the original vertical hole section drilled to test the properties of the reservoir prior to kick-off to drill the horizontal section. 136 . Water Oil PCP 18’ Oil OWC Water 64’ Figure 6-2 Bilateral Horizontal Well Water Sink Option The segregated production of water from the lower horizontal section creates a redistributive pressure effect around the heel of the primary oil-zone horizontal well and so reduce the level of water influx around the heel and reduce the incidence of bypassed oil around the toe of the well.The segregated production of oil-free water at the lower horizontal well or the vertical tail-pipe (similar to its variant for vertical wells called the Downhole Water Sink Technology) essentially creates a pressure sink around the heel of the horizontal well. Figures 6-1 show schematic representation of the tail-pipe water sink completion. Figure 6-2 shows the tandem horizontal well option for controlling water production in the horizontal wells (the Bilateral Water Sink technology).

The level 5 multi-lateral well is defined “as a system offering hydraulic isolation of the junction through the use of additional ‘straddling’ completion equipment” – Figure 6-3 (Emerson. A level 5 multi-lateral well completion technology is recommended for the tail-pipe water sink. Figure 6-3 A level 5 Multi-lateral Schematic for Tail-pipe Water Sink A variant of level 4 multi-lateral well could also be used in cases where the oil zone production will be accomplished through the tubing-casing annulus while the water production is accomplished through the main-bore completion tubing. 1998).6. al. Level 4 multi137 .1 Drilling and Completion of the Tail-pipe Multi-lateral Well Option Perhaps the easier of the two technologies in terms of drilling and completion is the tail-pipe option. et.1..

The recent development of the expandable composite liner makes this option very attractive. A stiff drilling assembly is then run through the whipstock guide oriented away from the azimuth of the horizontal section. The completion involves using the downhole split packer and a multi-lateral junction expandable seal. The vertical section is drilled into the water zone and protected with a primary casing string.1 Project Planning Options for Tail-pipe Water Sink For a project plan that incorporates the implementation of the tail-pipe water sink option as a remedial solution. A whipstock packer set 138 .1. 6. The main bore completion tubing is used to produce the water zone while oil zone production is accomplished through the annulus. This permits re-drilling of the cement/mechanical plug of the original rat hole into the water zone. The primary casing string contains a pre-milled casing joint carefully spaced out across the desired kick-off depth to land the horizontal section at the top-most part of the oil zone. The variation involves replacing the top permanent production packer with a dual bore bottom hole oriented completion guide. In the case of pre-planned tail-pipe water sink well a premilled casing kick-off window option could also be used as in the first multi-lateral well in Saudi Arabia (AlUmair. An oriented whipstock guide is set just below the kick-off point of the original horizontal well for the case of remedial application. the extension of the original rat hole into the water zone is accomplished with a whip stock guide. 2000). The vertical section is usually drilled with sufficient rat hole – over 200 ft of rat hole to permit drop off of lost retrievable accessories since hole cleaning usually constitutes a major problem for multi-lateral wells.1.lateral well is defined “as having a cased and cemented main bore with a cased and cemented lateral liner in order to provide full mechanical support at the junction”.

and completed as required. the Scoophead Diverter. After completing the lateral. cased and cemented.2 Operation Sequence and Equipment for Pre-planned Tail-pipe Water Sink The mainbore is drilled.1. After drilling the lateral. The lateral bore can then be perforated. isolated fluid production is desired. the lateral production string is run through the Scoophead Diverter and sealed off in a previously run production packer set in the lateral bore. stimulated.1. Once the Scoophead Diverter is landed. Additional completion equipment is installed to create the hydraulic integrity required for a Level 5 multilateral system. Selective re-entry into either bore is possible with coil tubing or wireline tools. a washover assembly tool is used to wash over and retrieve both the portion of the lateral liner extending into the mainbore and the whipstock and anchor assembly. 6. and tied into. the lateral casing is run and cemented in place with the top of the liner extending back through the casing exit and into the mainbore portion of the well creating a level 4 system with good mechanical junction integrity. For the segregated production required of the water sink technology. a Scoophead Diverter and Anchor System are run.across the depth of the pre-milled window guides the drilling bit through it to drill the horizontal section. The anchor system latches into and orients against the Packer positioned below the window. The final step in the multilateral process is dependent upon the type of production desired. A Liner Hanger Packer or TorqueMaster Packer is run in conjunction with whipstock System to create the casing exit window. A standard Dual bore Packer is run directly above. First. 139 .

6. the old vertical well is extended into the water zone as the water sink completion. the old oil zone perforations are first squeezed off with cement. A level 5 multi-lateral well completion is also recommended for adequate zonal isolation and segregated fluid production. Both wells are completed with sufficient mechanical and hydraulic isolation as level 5 multi-lateral well completion and segregated fluid production could be implemented. For such applications.6.2 Drilling and Completion of the Bilateral Water Sink Well Option The bilateral water sink (BWS) concept requires that the two horizontal well be drilled laterally above each other. In a typical project in Alaska. 140 . Multi-lateral technology has advanced tremendously such that lateral wells could be drilled a few feet from each other with the help of geo-steering tools and anti-collision magnetic surveys. The top horizontal well targets the top of the oil zone while the lower horizontal section targets the water zone.3 Application of Tail-pipe Water Sink Technology in Old Vertical Wells The descriptions of the operation procedures outlined above indicates that the tailpipe water sink technology could be extended as a remedial option for fields developed with vertical wells where severe water coning has created substantial amount of recoverable but bypassed oil.1. a watered out horizontal well was re-entered and a top horizontal well drilled and landed just 2 feet away in true vertical depth (Figure 6-4). A whipstock milling and drilling operation is then targeted at the top section of the oil zone to produce the bypassed oil.1. The two completions should have zonal isolation integrity between them to avoid co-mingling of the oil and water in the wellbore.1. Then.

The window joint is made of composite wrap type material permitting easy drillout and minimal debris problems. At the same time.Figure 6-4 Multi-lateral Well Project in Purdue Bay. pre-milled window joint. A typical sequence of operation is as follows: · Drill to the production casing point at the top of the oil water contact or any shale point between the oil and water zone. casing annular packer and a port collar for two-stage cementing. This pressure sleeve is retrievable after the drilling and cementing operations. Alaska (Courtesy. Bakerhughes website) The pre-planned option could use the proven application of dual completion well in Saudi Arabia (Figure 6-5). · Retrieve an inner sleeve from the pre-milled window joint 141 . the pre-milled window contained an internal pressure sleeve to tolerate differential pressures during cementing operations. · Run and cement the production casing including a landing collar. The project involves the selection of a pre-milled casing window joint.

and tail pipe assembly run into the upper lateral (oil zone completion). Set a production packer in the mainbore lateral (also known as the water sink lateral) · Set a dual bore deflector (DBD) in the production casing latch coupling with a lower seal assembly stung into the lower packer.N.. Figure 6-5 Dual Completion Multi-lateral Well in Saudi Arabia (Al-Umair. 2000) · Set a retrievable hydraulic dual packer in the production casing with a seal assembly stung into the dual bore deflector. A. 142 .· · · · · Set a drilling whipstock at the window Drill the upper lateral out of the window joint Retrieve the drilling whipstock Drill out the lower lateral in the water zone.

A new cement plug is then set prior o kick off. A mechanical-set multi-lateral hanger/packer is installed as datum for deploying a window master retrievable whipstock. pre-packed slotted liners could be used in the multi-lateral sections to mitigate the problem of sand production. In cases where borehole stability is a major concern due to unconsolidated sandstone reservoir. a primary requirement for the segregated fluid production in the water sink technology. a dual string packer is used in the mainbore so that each completion (zone) could be produced independently. The lower water sink section could be completed with a permanent packer and screens. Finally. The multi-lateral sections could also be completed with slotted liners. 2001). This level 5 multi-lateral well posed very little problem from drilling to completion and the technology is quite adaptable for the bilateral water sink completion option to control water cresting in horizontal wells. The vertical rat hole is re-entered. The cement and mechanical bridge plug is drilled out with sufficient rat hole below the water zone. Similar application for two zone reservoirs has been implemented in the OB1 project offshore Brunei (Bakerhughes website. The mechanical hanger/packer also provides a reference for the final completion. The upper lateral (oil zone completion) could be completed with scoophead divertor system and the junction adequately isolated and the production tubing sealed into a permanent packer previously set in the lateral.Figure 6-5 shows the schematic of the dual completion multilateral for the Saudi Arabian project that could be perfectly adapted for the bilateral water sink technology option. 143 . The remedial option for horizontal wells with water cresting problems follows the same approach as the tail-pipe option.

6094 Percent Increase % 0 Conventional Horizontal Well Horizontal well with Tail-pipe DWS Horizontal well with Bilateral DWS 3. Oil Prod.000065 ft and weighted average properties for the fluid mixture as modeled by the Eclipse 200 reservoir simulator. The poor recovery performance is attributable to the fact that water crest development is nearly symmetrical along the well length as the effect of friction 144 . the application of tail-pipe or bilateral water sink contributed very little to improving cumulative oil recovery over time.59 Table 6-1 was generated for a conventional pipe roughness of 0. In the conventional case with regular Fanning Friction Factor.6244 0.980 0. Sample data deck for Eclipse models of the horizontal wells with DWS Technology are contained in Appendices C and D. Table 6-1 Oil Recovery for Conventional Horizontal Well with Fanning Friction Factor Well completion type Cum.071 3.5 0. the sweep efficiency for the symmetric crest is high and bypassed oil is small.449 0. Recovery (stb) (In 20 years) Factor (RF) 3.6153 1.2 Water Cresting Control Performance of the Tail-pipe and Bilateral Water Sink Technology The benefits of the application of multi-lateral well technology in controlling water cresting in horizontal wells was evaluated with the compound friction pressure loss model and compared with the conventional eclipse 100/200 Fanning friction factor and in-situ pipe roughness and weighted average mixture properties.456.6. Well curvature radius was less that 200 ft and so the well could be classified as a short radius well.542. As the Figure 5-17 showed.490.

Using the modified friction pressure loss model. Oil Prod. Recovery Factor) (stb) (In 15 years) Factor (RF) Conventional Horizontal Well 2.5 7. After initial water breakthrough. Table 6-2 and Figure 6-6 show cumulative oil recovery over a period of 15 years.872 4. Original Oil In Place (OOIP) is 3.844 0.917. Table 6-2 Improved Oil Recovery in Horizontal Well with DWS Technology Well completion type (Modified Friction Cum. watercut rises to over 60 percent in less than four months. the water crest development becomes rather asymmetrical with water breaking through first at the heel and gradually spreading to the toe of the well.3 The well has start of perforation about 400 feet away from the vertical axis of the surface location. For this model the water breakthrough is rather dramatic. The table showed an improvement in recovery with the concepts when the water crest is greatest at the heel.pressure loss is minimized. Thus any technology that can effectively redistribute the pressure around the heel should be able to control water production and ultimate oil recovery. Thus water table moves up symmetrically toward the horizontal well as oil production progresses.013.762. The water crest development showed an increase at the heel with a gradual spread toward the toe.122 3.089 0.457 million stock tank barrels.463 0. 145 .799 Percent Increase % 0 Horizontal well with Tail-pipe DWS Horizontal well with Bilateral DWS 2. In another horizontal well with similar reservoir properties the modified friction factor is applied.

Horizontal Well with Tail-pipe Water Sink and Horizontal Well with Bilateral Water Sink The reservoir studied in Figures 6-6 and 6-7 has 85 ft total thickness with 40 ft net oil sand and 45 ft water zone.000 stb/day (1k) and the bottom completion liquid rate is 12.000 stb/day liquid at the top completion and 12.000 stb/day and 12. The length of the water sink completion is 33 percent (one-third) the length of the oil zone completion. 146 .000 stb/day at the top and bottom completions respectively.000 stb/day liquid rate. In the legend. Conventional Horizontal Well.33BL-85ft-1k-12k+ represents a bilateral water sink completion delivering 1. .3200000 3000000 Cumm. (stb) 2800000 2600000 2400000 2200000 2000000 2000 2500 3000 85ft-1k-0 3500 4000 Time (Days) 85ft-1k-12k+tp 4500 5000 5500 .33BL-85ft-1k-12k+ Figure 6-6 Cumulative Oil Recovery Comparison.000 stb/day liquid at the water sink completion. 85ft-1k-12k+tp represents the tailpipe water sink completion with liquid rates of 1.000 stb/day for DWS well and zero for the conventional horizontal well. Oil Prod. The top completion liquid rate is 1. 85k-1k-0 represents a conventional horizontal well delivering 1.

This result could be attributed to the fact that the lateral drainage influence of the vertical tail pipe is limited compared with the bilateral water sink that is another horizontal well.2.1 Effective Length of Bilateral Water Sink Horizontal Well With a water crest development maximum at the heel.The graph shows an improvement of about 300. what effective length of the bilateral water sink well is adequate to control water cresting in the oil zone horizontal well? Various lengths of the lower lateral water sink were investigated.162 stb or a fraction of 0. However. The tail pipe option promises to be more effective in short radius as well drain hole type of horizontal wells. 147 . the pertinent question that follows is. 6.000 stock barrel of oil recovery when the bilateral water sink completion concept for water cresting control is implemented. A period of 15 years as was selected as in previous simulations. The Figure also shows that the bilateral water sink is superior to the tail pipe water sink in controlling water cresting and hence oil recovery. for drain holes. the well lengths are usually small compared to conventional horizontal wells and the effect of friction pressure loss is negligible. The parameter to evaluate the effect of different lengths of the water sink completion is the oil recovery over time. As a result of the superior performance of the bilateral water sink. the next section presents the results of the performance evaluation study of the BWS concept. The results shown in Figure 6-7 indicate that only a third of the length of the upper horizontal well is required to effectively redistribute the water crest along the length of the upper oil zone horizontal well.60 percent. The cumulative oil recovery in a 15 year period differ only by 18.

Oil Prod. (stb) 2800000 2600000 2400000 2200000 2000000 2000 2500 3000 3500 4000 4500 5000 5500 Time (Days) 85ft-1k-0 85ft-1k-12k+BL1 . 148 .33BL-85ft-1k-12k+ Figure 6-7 Evaluation of Effective Length of Bilateral Water Sink Well Thus this study recommends a well length of 33 percent (or one third) of the length of the oil zone horizontal well as adequate to redistribute the pressure along the wellbore of the upper horizontal well.3200000 3000000 Cumm.

two vertical wells are equivalent to one horizontal well of length 1. themselves. In this chapter. certain assumptions and observations have to be made as follows: · · Commercial considerations are not included in this evaluation Comparing the performance of one vertical well to one horizontal well is inappropriate from reservoir engineering viewpoint as each well drain different areas of the reservoir. · Drainage area consideration is used in the evaluation of the performance of the two type of wells · Thus. the research focuses on comparing the performance of vertical wells and horizontal wells in developing fields with severe water coning problems.CHAPTER 7 COMPARISON OF DWS VERTICAL WELLS AND HORIZONTAL WELLS IN BOTTOM WATER DRIVE RESERVOIR Several researchers and industry operators have recommended the application of horizontal wells as a suitable solution for developing reservoirs experiencing water coning problems.500 ft. The discussions in chapters 5 and 6 reveal that horizontal wells are. 149 . not free from water influx (water cresting) problems and that in some cases. To undertake this study. the influx of water into horizontal wells could be so dramatic that it tends to erode its merit.

Horizontal wells. a horizontal well usually has less drawdown pressure than a vertical well for equal rates of production. these specific advantages of horizontal well have been the basis of their intensive use in place of vertical wells in developing reservoirs subject to bottom aquifer or a gas cap. For over 15 years. are quite superior to vertical wells because of the high production rates at low drawdown pressure.000 5.000 6. it can be stated that horizontal wells have become the standard industry procedure to develop these reservoirs.000 1. Moreover.000 0 1000 2000 Length of Horizontal well (ft) 3000 4000 Figure 7-1 Productivity Comparison between Vertical and Horizontal Wells 150 .000 4.000 3. Infact. the standoff from the bottom water can be maximized with a horizontal well located high in the oil pay zone. Due to its extended exposure with the reservoir.000 2. 7.000 0. The major factor responsible for coning in conventional wells is the pressure drawdown around the wellbore. The low drawdown pressure is as a result of the extended reach nature of the horizontal well within the reservoir.

The discussion in chapters 5 and 6 indicates that these equations tend to overestimate the performance of horizontal wells due to the assumption of a frictionless flow. the ratio of PI of a 1. and. and XA is the location of a constant pressure boundary or abscissa where the interface level is known to be.500ft horizontal well and a vertical well has been estimated to be 3.5 when the effects of friction pressure loss in the wellbore is considered. However.500ft has a productivity index three and half times greater than that of a single vertical well in the pre-water breakthrough period. As the Figure shows. = 1. This research has shown that friction pressure loss in the wellbore could be quite significant. As an example. h. L. 1984).The plot in Fig. 7. Eq. 151 Another .(4. h is the initial oil thickness or interface elevation at XA. F = 4.0 .5 using the equation 2-17 and Figure 7-1. Chaperon developed a simplified relation as follows: Qch L F = * Qcv X A qc 7-1 Note that for.1 shows the superiority of horizontal well PI over that of a vertical well for various lengths of the horizontal wells for single phase flow and in an isotropic and homogeneous reservoir using equation 2-17 (Geiger. q*c ≈ 1.500 ft. Also comparing critical rates of vertical and horizontal wells.6) gives: Qch/Qcv = 4L/XA. Thus for XA = 750 ft and length. for steady-state flow. An inherent shortcoming of all of the above equations is that they were developed assuming friction pressure loss in the wellbore is negligible. these ratio could be reduced to as low as 1. the critical oil rate should be about 8 times higher for horizontal well than for the vertical well. a horizontal well of length 1.

1 Equivalent Length/Drainage Area of Horizontal Wells Researchers have agreed that horizontal wells drain larger area of the reservoir due to its extended reach. A more pragmatic evaluation of the performance of horizontal wells over vertical wells in reservoirs with water coning/cresting problems should be to consider the concept of equivalent drainage area for both types of wells. If the technology of horizontal wells is to be compared with that of vertical wells considering the overall reservoir performance. It is. thus. The next section presents an “equivalent drainage area” tool for comparing the performance of horizontal wells and vertical wells. then the more scientific comparison should be on the basis of equivalent drainage area. Two mathematical considerations have been advanced for estimating the drainage area of a horizontal well. One approach treats the drainage area of the horizontal well as a combination of two half-circles at each end of the well length and a rectangle along the length of the well. Consequently. It also failed to consider the concept of equivalent drainage area. 7. It implies that both type of wells do not drain the same area of the reservoir. The width of the rectangle is equal to twice the length of the drainage radius of the conventional vertical well and the length of the rectangular section is equal to the length of the horizontal reach (length 152 . several studies have been initiated to evaluate an effective drainage area for the horizontal well. This advantage needs a more scientific analysis. applicable to the pre-water breakthrough production period.shortcoming of the above PI relation is that it assumes single-phase flow. One frequently advanced advantage of horizontal wells over vertical wells in field development practice is that fewer wells are required to drain the same reservoir for horizontal wells than for vertical wells.

A(ft2) = (L/2 + re)*re 7-3 153 . The other method considers the drainage area of the horizontal well as an ellipse with minor radius equal to the drainage radius of conventional vertical well (Joshi. The drainage is. 1991). 7.1 Method 1: Two Half-circle and Rectangle Approach Drainage area in method 1 is calculated as.1.1. πre2 + (2re*L) = Area (Drainage) 7-2 L a b Figure 7-2 Horizontal Well Drainage Area 7.of the perforated section).2 Method 2: Ellipse Shaped Drainage Area Method This method approximates the drainage area of a horizontal well with an ellipse assuming the length of the minor axis b to be equivalent to the drainage radius of the vertical well and the major axis a equal to half the length of the well plus the drainage radius.

14 acres.57 acres. Joshi average-area approach is presented as follows: For a vertical well of drainage radius re = 750 ft. Consequently. Joshi (1991) takes the average of the two as the equivalent drainage area of the horizontal well. devoid of economic considerations will be performed on the basis of one horizontal well of length 1. 154 . affective comparison of the performance of vertical wells and horizontal wells in reservoirs with water coning problems. The area for two such vertical wells is 81. the drainage area is (π*(750)2)/43560 = ) 40. The length of a horizontal well that would drain the same area is given by the following sample calculations: ü 1 ìæ æ L ö ö * íç p *ç + a÷ * a÷ + p * a 2 + (L * 2 * a ) ý = 81.a b Figure 7-3 Elliptical Drainage Area for Horizontal Well It should be emphasized that the two methods give different solutions to the size of the drainage area of the horizontal well.500ft to two vertical wells.14 Acres ÷ ç 2 îè è 2 ø ø þ ( ) 7-4 Substituting values in the above equation where a = 750 ft and solving for L gives a horizontal well with an equivalent drainage area of two vertical wells. Joshi’s approach has been adopted for this research. The horizontal well length is L = 1320 ft.

31 0.50 165 0.34 3.0E-06 9-5/8-in csg psi ft ft ft ft ft mD mD mD mD Ib/ft2 cP O F fraction fraction rb/stb O API cP sips sips sips 155 . Table 7-1 Reservoir/Rock properties of Horizontal Well Vs Vertical Well Water Cresting/coning Performance Studies Input Data Properties Unit Reservoir Pressure Thickness of oil/gas column Thickness of water column Depth of OWC/GWC (static) Length of oil production perforations (Horizontal) Length of water perforations (Horizontal Bilateral) Position and length of Oil zone perforations (Vertical) Position and length of water perforations (Vertical) Horizontal permeability in Oil/Gas column Vertical Permeability in Oil/Gas column Horizontal permeability in water column Vertical Permeability in water column Water density at temperature Water viscosity at temperature Reservoir temperature Porosity in oil column Porosity in water column Oil/gas formation volume factor Oil gravity Oil viscosity at BHT Water compressibility Oil compressibility Rock compressibility Completion diameter/hole size 2500 40 45 6194 1500 500 8000 800 8000 800 64. The modeling is performed with the Geoquest Eclipse software.35 1.0E-06 1.The results presented in the subsequent sections are derived from numerical simulation of two vertical wells and one equivalent horizontal well in a reservoir with same grid sizes and properties.2 Simulation Grid Properties of Horizontal Well and Vertical Well The reservoir considered in this section has initial thickness of the oil zone 40 ft and the thickness of the water zone is 45 ft.15 35 1. 7.5E-05 4.2 0.

3 Comparison of Watercut Performance of Vertical and Horizontal Wells To effectively evaluate the effective performance of each of the technologies.The Eclipse model consists of a 75 x 20 x 12 Cartesian grid (about 18. three parameters of utmost importance to this study will be analyzed namely oil production rate performance of each well type.750 ft are considered. The modified equivalent pipe roughness function of 0.1 shows the reservoir and fluid properties employed for this study. The horizontal oil well completion is placed within the first 30 percent of the oil pay following good industry practice for reservoirs with bottom water drive.000 grid blocks) with block centered geometry and implicit formulation. Locations of both water sinks are varied to obtain optimum performance and minimize initial inverse oil coning. and (2) a bilateral well with water sink running parallel below the oil completion in the water zone. Comparison is performed using the Joshi equivalent drainage area and length method. Table 7.1100 ft. The last row of grid blocks have been made large compared to the others (30 ft thickness) and assigned infinitely large porosity to simulate a steady-state flow process. The water sink is modeled (1) as a vertical tail pipe originating from the heel of the horizontal well. The friction option of the Eclipse has been activated for this study to include the effect of pressure variation along the length of the wellbore necessary for the finite conductivity assumption. 7. The thickness of each grid block from top of sand to about 15 ft below Oil Water Contact (OWC) is 5 ft giving 11 blocks. Two vertical wells with drainage radii of +/. Equivalent length of a horizontal well in the same area is 1500 ft. is used for this study. cumulative oil production over time and watercut performance analyses. 156 .

3. The medium line (DVWCON in the legend) represents the performance of two single wells delivering a combined total of 3.000 stb/day. 157 .500 stb/day each) and the thin broken line (SVWCON in the legend) shows the performance of a single vertical well delivering 3. As could be seen in Figure 7-4.1 Oil Production Comparison Study Horizontal wells have undoubted superiority over vertical wells in producing oil reservoir subject to water coning problems.000 stb/day (1.000 stb/day liquid rate.7. 3500 3000 2500 2000 1500 1000 500 0 0 500 1000 1500 Time (Day s) HWCON DVWCON SVWCON Oil Rate (stb/day ) 2000 2500 3000 Figure 7-4 Oil Production Rate Performance Evaluation The thin continuous line (tagged HWCON in the legend) represents the performance of a horizontal well producing at 3. the low drawdown pressure capability of horizontal wells is responsible for the longer period of plateau production prior to water breakthrough.

the single vertical well experiences water breakthrough rather early and so the production of oil declines almost immediately. 5. 7-5).00E+06 4.50E+06 3.50E+06 2.00E+06 3. for the equivalent drainage area concept with two vertical wells the cumulative oil recovery for the same combined rates as the horizontal well show that the cumulative oil recoveries converges (Fig. a horizontal well would perform better than one vertical well.00E+05 0. arising from the long oil production plateau period of horizontal wells is eroded by the dramatic increase in 158 .As shown in the diagram.50E+06 Oil Recovery (Stock Tank Barrels.00E+06 5.00E+06 2.00E+00 0 1000 2000 3000 4000 Time (Days) 5000 6000 7000 8000 HWCON DVWCON SVWCON Figure 7-5 Cumulative Oil Recovery Performance Analysis In terms of cumulative oil recovery over a period of twenty years. stb) 4.50E+06 1. However. comparing the performance of two vertical wells with a combined total oil rate equal to the oil rate of the horizontal well shows the field wide performance of horizontal wells may be comparable to that of equivalent number of vertical wells.00E+06 1. The gain in oil recovery in the pre-water breakthrough period. However. This picture is made clearer by comparing the cumulative oil recovery of the three types of field drainage options over time as shown in Figure 7-5.

Also included in the plot is the watercut of two vertical wells draining thee same equivalent reservoir area. if long-term production of oil is a critical consideration. The medium line represents the watercut performance of two equivalent vertical wells and the thin broken line represents the performance of a single vertical well. After water breakthrough the watercut in horizontal well increases rapidly and approaches that of a single vertical well producing at the same rate. from the equivalent drainage area evaluation. especially in high permeability reservoirs.2 Watercut Performance Evaluation The superior performance of a horizontal well over a single vertical well is in the period of delayed water breakthrough. However. In the Figure 7-6. the watercut in the horizontal well rapidly increases and exceeds the combined watercut of the two equivalent vertical wells over a short time after water breakthrough. The result presented in the plot shows watercut for the horizontal well overtaking the two equivalent vertical wells.watercut in the horizontal well post water breakthrough whereas the low initial recovery by the dual vertical wells due to early water breakthrough is improved by the gradual rate of increase of watercut post breakthrough. the thin continuous line represents the watercut performance of one horizontal well.3. This result indicates that horizontal wells may not be the preferred optimum solution for developing bottom water drive reservoirs with high permeability. Figure 7-6 is a plot of watercut performance of single vertical well delivering same oil rate as the horizontal well. It also further 159 . 7. It underscores the need for industry operators to carefully analyze and evaluate the added advantage of horizontal wells over vertical wells in developing reservoirs subject to water coning problems.

This analysis of watercut performance of horizontal wells combined with the difficulty of secondary mechanical intervention in horizontal wells after water breakthrough compared with conventional vertical wells or even high angles wells makes horizontal wells less attractive as a solution to developing these reservoirs experiencing severe water coning.8 0.emphasizes the fact that water cresting in horizontal wells could be so dramatic that it may erode the merit of initial high oil rate.1 0 -0.3 0.1 0 500 1000 1500 2000 2500 3000 3500 Time (Days) SVWCON HWCON DVWCON Figure 7-6 Watercut Performance Evaluation of Vertical and Horizontal Wells The ultimate oil recovery from reservoirs with bottom water drive and high deliverability performance may be less if such fields are developed with horizontal wells than for vertical wells as the field tend to be abandoned pretty early at high watercut. 160 .9 0.7 0.4 0. 1 0.5 0.6 0.2 0.

3500 3000 2500 2000 1500 1000 500 0 0 1000 2000 3000 4000 Time (Days) HWCON DVWCON DWSVW 5000 6000 7000 8000 Figure 7-7 Oil Production with DWST Vs Horizontal Well 161 . cumulative oil recovery and finally watercut performance in oil zone completion. the performance of two vertical wells completed with the Downhole water Sink (DWS) technology will be compared with that of an equivalent horizontal well.7. 7. three parameters will be used to qualify the superiority of one technology over the other namely.4 Comparison of Vertical Wells with DWS Technology and Horizontal Wells In this section. it makes otherwise bypass oil accessible to the top completion. not only reduces the water production in the top completion.1 Two Vertical DWS Wells Vs Horizontal Well Oil Production Analysis The application of the Downhole water sink (DWS) technology in vertical wells. Figure 7-7 and 7-8 shows the improvement in oil production rate and cumulative oil recovery for the two vertical wells with DWS and the single equivalent horizontal well.4. oil production rate over time. More oil is produced using DWS technology in vertical wells. As in the previous section.

The first spike in the heavy line indicates the point where drainage water production from the sink was initiated and the second spike shows a point where the rate from the sink was increased to optimize oil recovery.00E+00 0 1000 2000 3000 4000 5000 6000 7000 Time (Days) HWCON DVWCON DWSVW Figure 7-8 Cumulative Oil Recovery Performance of Horizontal Well Vs Two DWS Vertical Wells 162 .00E+06 5. The thin broken line indicates the performance of the two vertical wells without the DWS technology and the thin continuous line shows the performance of the horizontal well.00E+06 4.00E+05 0.50E+06 1.50E+06 3. stb) 5. Oil Recovery (Stock Tank Barrels.00E+06 3.50E+06 4. The initial high oil production rate merit of the horizontal well is soon marred by the influx of water and the vertical wells with DWS soon out perform the horizontal well.50E+06 2.00E+06 2. It shows that the technology can recover more oil than the horizontal well.The heavy line graph shows the oil production rate performance of the two vertical wells completed with the DWS technology. The performance of the cumulative oil recovery presents a clearer picture of the superiority of DWS technology in vertical wells with water coning problems.00E+06 1.

This delayed water break through time improves oil recovery performance of the DWS completions. 7. The thin broken line (DVWCON in the legend) represents the performance of the two vertical wells without the DWS technology and the thin continuous line (HWCON in the legend) represents the performance of a single equivalent horizontal well. after water breakthrough watercut in the horizontal well increases rapidly and soon overtakes the vertical wells.4.In Figure 7-8. Initially. This is shown by the heavy line with a cumulative oil recovery about 300. 163 . This is due to the low drawdown pressure and improved period of plateau production of oil from the horizontal well. However.2 Two Vertical DWS Wells Vs Horizontal Well Watercut Analysis The application of the DWS technology in two vertical wells delays the time to water breakthrough (Figure 7-9). The horizontal well shows the longest time to water breakthrough due to the distribution pressure drawdown along the wellbore. the horizontal well outperforms the equivalent two vertical wells due to the delayed water breakthrough time. application of DWS in vertical wells also improves the time to water breakthrough. after water break through.000 stb more than the conventional horizontal well. On the other hand. However. The horizontal well initially performs better than the vertical wells even with the DWS (Figure 7-8). the heavy line (DWSVW in the legend) represents the performance of two vertical wells with the DWS completion. The result is that the duo-vertical wells with DWS completion produces less water in the oil zone than the horizontal well. the advantage of the horizontal well is eroded and the two vertical wells with DWS technology outperform the horizontal well.

table 7-1).500ft.500 ft.2 0. is the DWS technology economically justified? In this section. 7.9 0.1 0. This analysis uses estimates for the cost of installation of two new vertical wells each equipped with dual production strings for the oil and water (DWS type completion.000ft and are assumed to have high deliverability. The horizontal well producing length is 1.4 0. a simple economic evaluation will be presented. The wells have a true vertical depth of 8.6 0.5 0. DWS in two equivalent vertical wells outperforms one horizontal well of length 1.5 Economic Evaluation of the DWS Technology Vs Horizontal Well The foregoing analysis shows the superior technical performance of the Downhole Water Sink technology in reservoirs with water coning problems.7 Watercut (Fraction) 0.4 reveals that vertical wells with DWS installations could recover more oil when employed in field development than horizontal wells.8 0. However. 164 .1 0 -0.3 0.1 0 500 1000 1500 2000 2500 3000 3500 Time (Days) DWSVW HWCON DVWCON Figure 7-9 Watercut Performance of Horizontal Well and Two Vertical DWS Wells The preceding analysis in section 7.

500 120.000 1.547.400 3.105.800 $7. The estimated time for installing the completion strings is 10 days for the dual-string vertical wells and 7 days for the single string horizontal wells.the drainage area equivalent of two vertical wells.600 1.000 to 300.400 $5. the ultimate oil recovery (Figure 7-8) indicates that the two DWS wells outperform the single horizontal well after water breakthrough (about three years). However.000 80. The estimated drilling time for one vertical well is 20 days and 27 days for the horizontal well. Table 7-2 Drilling and Completion Cost for 2-Vertical wells and 1 Horizontal Well Description 2 Vertical Wells 1 Horizontal Well Drilling Operation and Associated Services Completion Operation and Associated Services Casing Strings (Drilling) Completion strings Wellhead Equipment Subsurface Equipment Miscellaneous/Contingencies Total Costs 4.950 825. The well project was planned and supervised by a team of engineers and they supplied the spreadsheet for this cost evaluation.000 250.000 40.200 It costs less to drill one horizontal well than two DWS vertical wells.000 90. Cost data are obtained from a shallow offshore well drilled in the Niger Delta Basin.000 125.725. The detailed spreadsheet for the drilling and completion cost is attached as appendix E.242.000 180.070. This improved recovery may justify the extra cost of installation besides several other merits 165 .000 320.312.350 556. An additional 150.000 stb of crude oil is recovered above that of horizontal well.

Finally.of the technology enumerated in chapters 4 through 6. the rapid increase in watercut level of horizontal wells with water cresting problems may cause rapid overload of water separation facilities leading to break down and high maintenance costs. In cases where water breaks through at the toe (very rare cases). the toe section of the well could be isolated. These factors include reduced water handling costs. However. Another disadvantage of horizontal well technology is the difficulty of secondary mechanical intervention for the purpose of well repairs and maintenance. Thus. 166 . water breakthrough occurs at the heel making it difficult for any meaningful intervention. the economic justification of the application of DWS technology in oil field development should not be based only on the initial sunk cost of drilling and completing the wells but on other critical factors during the producing life of the field. in most cases.

The study ague that relative permeability curves for water and oil differ during the water cone development process (imbibition process) and during the cone reversal process (when the DWS water sink is implemented).CHAPTER 8 CONCLUSIONS AND RECOMMENDATIONS 8. 167 . This hysteresis. It presented a ‘generalized compound friction pressure loss relation for horizontal wells and horizontal pipes. The application of the technology in controlling water coning creates a hysteresis effect. coupled with the development of a saturation transition region for wells with severe water coning history have been studied in this research. The following summarizes the findings of this research. The results of application of the compound friction pressure loss relation indicate that friction factor could increase by the order of magnitude of over 400 percent. Finally the study presented to concepts of the Downhole Water Sink Technology for controlling water cresting in horizontal wells. The inclusion of relative permeability hysteresis and capillary transition zone in the design of DWS completions in watered out wells offer the necessary solution to the desire to produce oil-free disposable drainage water while reducing watercut in the oil zone and recovering bypassed oil with the technology This study also reviewed friction pressure loss in horizontal wells and agued that adequate representation of wellbore friction pressure loss provides a tool for effective understanding and modeling of water cresting in horizontal wells.1 Conclusions The control of water coning/cresting with Downhole Water Sink (DWS) technology in both vertical and horizontal wells has been analyzed in this study.

recompletion of conventional wells with severe water coning history with DWS technology would not be as beneficial as DWS completions of new wells. 168 . · Consequently. · Relative permeability hysteresis without significant capillary pressure transition zone has no effect on DWS performance if the end-point-residual saturations are the same for both imbibition and drainage curves.1. This pressureenhanced capillary transition zone enlargement is further made worse by the presence of the two pressure sinks in DWS wells causing expansion of the saturation transition zone until both fluids break through into the two completions. · The capillary pressure transition reduces the size of the IPW and thus reduces the domain of the water-free oil production. The practical implication of this effect is increased water pumping rate from the sink completion and increased pressure drawdown is needed for a required production of oil. · Drawdown pressure around the wellbore creates an enhancing effect that causes enlargement of the capillary transition zone around the wellbore.1 Relative Permeability Hysteresis and Capillary Pressure Transition Zone on DWS Well Performance The conclusions derived from this study on effect of relative permeability hysteresis and capillary transition pressure on water coning control performance of the DWS technology are: · Cone reversal process by DWS involves capillary pressure and relative permeability hysteresis effects and these concepts should be included in the model development.8.

the effect is not only more pronounced but it also alters the IPW plots – a basic tool for design. it might also lead to lesser recovery with DWS – potential subject of future studies. · The study of IPW shows isolines of increasing watercut to the right along the oil breakthrough line indicating that for oil-free water drainage design. This could be obtained from a graph of oil rate against liquid rate at the top completion.1. however. though the presence of capillary transition could result in concurrent production of contaminated fluid in both completions. Not only would it result in smaller oil rate. · The study also shows that liquid rate at the bottom water sink completion is limited by the oil breakthrough line. oil-free water production at the water sink can be achieved by incorporating a capillary transition zone in the model and design procedure.2 Oil-Free Water Production Capacity of DWS with Capillary Pressure Zone With respect to the environmental merit of the technology in old wells the study has the following conclusions: · The results presented in chapter 4 show that. · Finally. · Transition zone enlargement effect occurs in conventional wells due to diffusion resulting from pressure distribution around the well. The optimum rate of liquid withdrawal at the top (oil zone) completion is determined by the point of maximum oil rate. maximizing liquid production rate at the top completion – typical for DWS design – would not be effective.8. Thus. locating the water sink in the swept zone or the zone of saturation transition could cause concurrent production of contaminated fluids in both completions. 169 . There is a commingled inflow envelope (Region 4 in Figure 4-8) in addition to the envelope of segregated inflow (Region 2). For DWS wells.

This “generalized equation” is a modification of the equation earlier developed by Su and Gudmundsson. 170 . a new “generalized relation” for evaluating compound friction pressure loss in the wellbore of horizontal wells and in pipes have been developed. a new “equivalent pipe roughness’” function has been developed that creates the same friction pressure loss effect in the wellbore using the conventional Fanning friction factor approach. Using this approach. As part of establishing an adequate representation of water cresting performance of horizontal wells. To accommodate the limitations of existing commercial numerical simulators. Water breakthrough occurs first at the heel end of the well and spreads gradually toward the toe (upstream end of the well). 8.it is important that the water sink completion be located below the original oil-water contact. a more practical numerical modeling of water cresting performance of horizontal wells have been presented and the following conclusions made: · The water crest profile in horizontal wells is skewed toward the heel. The study also recommended the adoption of the Pal and Rhodes equation for evaluating the flow viscosity of oil-water mixture as opposed to the ‘weighted average’ approach.3 Water Cresting in Horizontal Wells and DWS Technology The study also evaluated the problem of water cresting in horizontal wells. · The time to water breakthrough evaluated with most of the current numerical or analytical models are optimistic and not representative of current field results.1.

·

According to the viscosity and friction pressure loss model used in this study, the pressure drawdown in the wellbore is actually higher than is currently assumed and the difference between current analytical models and the new approach diverges with increasing rate of production.

·

The application of a simple Fanning friction pressure loss relation for fluid flow in pipes to horizontal wells is probably inadequate. In recommending water cresting control option with the DWS technology, the

study evaluated the application of two innovative concepts namely “tail-pipe water sink (TWS) well” and “bilateral water sink (BWS) well” for redistributing the pressure along the wellbore. The results show that both the vertical tail pipe and the bilateral well concepts can redistribute the pressure profile along the wellbore and reduce the bypassed oil at the toe of the well. However, for long radius wells, the bilateral well option shows more promise in improving oil recovery. The study showed that only a third of the length of the oil-zone horizontal well is required for the water sink to adequately redistribute the pressure profile along the wellbore. Improvements in oil recovery (as much as 7 percent) have been predicted with the bilateral well technology.
8.1.4 Vertical Wells with DWS and Conventional Horizontal Wells

Finally, the study also compared the performance of vertical wells with DWS technology and horizontal wells in the development of reservoirs underlain by water. The results indicate that two vertical wells with DWS completion are capable of improving ultimate oil recovery more than horizontal wells in reservoir where water coning/cresting is critical. It also shows that water contamination of the produced oil could be less with the DWS completions than with horizontal wells. The result is reduced water handling 171

costs. Thus, the application of DWS in vertical wells may improve the ultimate oil recovery of reservoirs with strong water drive compared with the performance of horizontal wells in such fields.
8.2 · Recommendations

Studies of oil recovery performance of DWS wells with capillary pressure transition zone with oil-free drainage only would it result in smaller oil rate, it might also lead to lesser recovery with DWS – potential subject of future studies.

·

Further research work should be initiated to develop a relation for the ratio of fluid influx to the main wellbore flow as a function of the length of the horizontal well.
·

A study to evaluate the extension of the pressure loss relation developed in this dissertation to three-phase (oil-water-gas) flow should be performed.

·

Analytical evaluation of the effect of the increased pressure loss in the wellbore on reservoir pressure drawdown should be investigated.

·

A Rule-Of-Thumb relation should be developed to help industry operators in determining when to use or not to use horizontal wells in developing reservoirs subject to bottom water drive.

·

Field application of the bi-lateral well technology and the tail-pipe option in watered out reservoirs with substantial bypassed oil should be vigorously pursued.

·

Finally, the performance of the bi-later and tail pipe techniques for water cresting control in reservoirs with capillary transition zone should be investigated. 172

REFERENCES AL-Umair, N. A., 2000, “The First Multi-Lateral/Dual Completion Well in Saudi Arabia”, IADC/SPE Paper 62771, IADC/SPE Asia Pacific Drilling Technology Conference, Kuala Lumpur, Malaysia, September 11-13.

Archer, J.S. and Wall, C.G., 1988, “Petroleum Engineering – principles and practice”, Graham and Trotman, London, pp. 92-119 Arirachakaran, S., Oglesby, K.D., Malinowsky, M. S., Shoham, O., and Brill, J. P., 1989, “An Analysis of Oil/Water Flow Phenomena in Horizontal Pipes”, Paper SPE 18836, SPE Production Operations Symposium, Oklahoma City, OK, March 1314 Asheim, H., Kolnes, J., and Oudeman, P., 1992, “A flow resistance correlation for completed wellbore”, Journal of Petroleum Science and Engineering, 8, Elsevier Science Publishers, Amsterdam, pp 97-104 Asheim, H., and Oudeman, P., 1997, “Determination of Perforation Schemes To Control Production and Injection Profiles Along Horizontal Wells”, Paper SPE 29275, SPE Drilling and Completion, March pp 13-17. Babu, D.K., and Odeh, A.S., 1988, “Productivity of Horizontal Well”, Paper SPE 18334, 63rd Annual Technical Conference and Exhibition, Houston, TX, October 2-5. BaherHughes, 2001, “Multilateral level definitions and multilateral case histories” www.bakerhughes.com/bot/Multilateral.htm November. Bourgoyne, A.T. (Jr.), Millheim, K.K., Chenevert, M.E., and Young, F.S. (Jr.), 1991, “Applied Drilling Engineering”, Society of Petroleum Engineers, Richardson, TX., pp144-147. Bournazel, C. and Jeanson, B., 1971, “Fast Water-Coning Evaluation Method”, Paper SPE 3628, 46th Annual Fall Meeting of Society of Petroleum Engineers of AIME, New Orleans, LA, October 3-6. Bowlin, K.R., Chea, C.K., Wheeler, S.S., and Waldo, L.A., 1997, “Field Application of In-situ Gravity Segregation to Remediate Prior Water Coning”, Paper SPE 38296, SPE Western regional Meeting, Long Beach, CA, June 25-27. Brekke, K, and Lien, S.C., 1992, “New and Simple Completion Methods for Horizontal Wells Improve the Production Performance in High-Permeability, Thin Oil Zones”, Paper SPE 24762, 67th Annual Technical Conference and Exhibition, Washington DC, October 4-7. Brill, J.P.(ed.), 1999, “Multi-phase flow in Pipes”, SPE Monograph Volume Richardson, TX, pp 5 –20. 173 ,

Broman, W.H., Stagg, T.O., and Rosenzweig, J.J.,1990, “Horizontal Well Performance Evaluation at Prudhoe Bay”, Paper CIM/SPE 90-124, Joint International Technical Meeting of the Petroleum Society of the CIM and the SPE, Calgary, Canada, June 10 – 13. Broughton, G., and Squires, L., 1937, “The Viscosity of Oil-Water Emulsion”, Journal of Colloid Science 5, Academy Press, New York, NY, pp 253-263. Chappelear, J.E., and Hirasaki, G.J., 1974, “A Model of Oil-Water Coning for TwoDimensional, Areal Reservoir Simulation”, Paper SPE 4980, 49th Annual Fall Meeting, Houston, October 6-9. Chen, H., 1993, “Performance of Horizontal Wells, Safah Field, Oman”, Paper SPE 25568, SPE Middle East Oil Technical Conference and Exhibition, Bahrain, April 3-6. Chaperon, I., 1986, “Theoretical Study of Coning Toward Horizontal and Vertical Well in Anisotropic Formations: Subcritical and Critical Rates”; paper SPE 15377, 61st Annual Technical Conference and exhibition, New Orleans, LA, October. 5-8 Chugbo, A.I., Roux, G.D., and Bosio, J.C., 1989, “ Thin Oil Columns: Most People Think Horizontal Wells, Obagi Field Case Suggests the Contrary”, Paper SPE 19599, SPE 64th Annual Technical Conference and Exhibition, San Antonio, TX, October 8-11. Civan, F and Donaldson, E.C.,1987, “Relative Permeability from Unsteady State Displacements With Capillary Pressure Included” SPE 16200, SPE Production Operations Symposium, Oklahoma, March 8-10. Colorado state, 2001, “Production and Sales by County – Monthly”, www.oilgas.state.co.us/statistics.html Craft, B.C. and Hawkins, M.F., Terry , R.E., 1991,“Applied Petroleum Reservoir Engineering” Second Edition, Prentice Hall, Englewood Cliffs, New Jersey. Craig, F.F., (Jnr.), 1993, “Reservoir Engineering Aspects of Waterflooding” , Monograph Vol. 3, SPE Textbook Series, Richardson, TX, p76. Cunningham, A.B., Jay, K.L., and Opstad, E.,1992, “Unconventional Prudhoe Bay Wells Benefit From MWD Use”, World Oil, May, p43-52. Dikken, B.J., 1990, “Pressure Drop in Horizontal Wells and Its Effect on Production Performance”, Paper SPE 19824, Journal of Petroleum Technology, November., pp1426-1433. 174

“Multiple Producing Intervals to Suppress Coning..H. p249-255. J... – Dec. p33-39. p409-416. and Brill.B. 1970. J. Fourth Latin American and Caribbean Petroleum Engineering Conference.. S. Ehlig-Economides.P. K. Reiss. Lee. Jones. C. and Jourdan. L. Sarica. V.A. Solid Waste Emergency Response 5305. B. B. and Spath. Paper SPE 47810. R. Colorado. A. Emerson.T. “A Horizontal Well Model Considering Multiphase Flow and The Presence of Perforations”. Feb.L.H. and Tetreau. Gonzalez-Guevara. J. Denver.M. Paper SPE 56108. Jakarta. Paper SPE 21610 SPE Reservoir Engineering Journal. September 7-9. Chan.. November. 1999..and. 1993. 175 .Driscoll.P.J.R. Chen.1996. Port Spain. J. X. J.. SPE Reservoir Engineering Journal. A.G. EPA530-K-95-003. 1998.P. “Characterization of Oil-Water Flow Patterns in Vertical and Deviated Wells”. 1. Indonesia. TX. and Lim.. SPE Reservoir Engineer Journal.. Schlumberger Technology Corporation Giger. p102-109. April 23-26. C. and Camacho-Velazquez. B. E. R.. Paper SPE 15378.. 1972.. Paper SPE 36073.. Paper SPE 23994. IADC/SPE Asia Pacific Drilling Conference.. Paper SPE 36613. Houston. 1989. 1995.. SPE 59th Annual Technical Conference and Exhibition. “Multi-Lateral Case Histories in the Thailand. “Eclipse 100/200 Technical Description Manual”.. Geoquest Schlumberger. Flores. F. Trinidad and Tobago. “The Application of Numerical Coning Model to Optimize Completion and Production Methods To Increase Oil Productivity in Bellshill Lake Blairmore Pool” Journal of Canadian Petroleum Technology. “Analytical Two-Dimensional Models of Water Cresting Before Breakthrough for Horizontal Wells”.731. 1997.S. November. Giger... “A Simple Approach to Optimization of Completion Interval in Oil/Water Coning Systems”.M.. W. F. Oct. J. 1984. May Fisher. p147-148. October 6-9. R. 1996. 1991. February..638. Annual Technical Conference and Exhibition. “Crude Oil and Natural Gas Exploration and Production Wastes: Exemption from RCRA Subtitle C Regulation”. Letkeman. May. “Discussion of Productivity of a Horizontal Well”. September 16-19. EPA (United States Environmental Protection Agency).C..A. “Production Enhancement Strategies for Strong Bottom Water Drive Reservoirs”. “The Reservoir Engineering Aspects of Horizontal Drilling” Paper SPE 13024.M.” US Patent No. 3. SPE Journal of Production and Facilities. Guo. Malaysia and Brunei Area”. G. Gilman..

K. Molinard. “Derived Capillary Pressure From Well Logs”. A. 176 . October. Canada. Alberta.. Los Angeles Meeting. and Bassiouni. France.K. 35th Annual Meeting Heysel.F. A.O. 2001(A)... 1994. Energy Sources Technology Conference and Exhibition. and Craig. SPE/EPA/DOE Exploration and Production Environmental Conference.. “Contaminated Water Production in Old Oil Fields With Downhole Water separation: Effects of Capillary Pressures and Relative Permeability Hysteresis” Paper SPE 66536. “Assessment and Inclusion of Capillary Pressure/Relative Permeability Hysteresis Effects in Downhole Water Sink (DWS) Well Technology for Water Coning Control”. S. W.. TX. 1992. New Orleans. and Verhyden.L. pp 38-41 Inikori. C. and Kelkar. “Measurement of Capillary Pressures in Small Core Samples” AIME. Houston. “Horizontal Well Performance in the Dina Sandstone in the Provost Area of Alberta”.. Calgary. Lee. E. Jiang. Owens. November 16-18. E. Houston. H01027.L. M. R. 1992. Cannes. Desbrandes. and Wojtanowicz.. 25th ASME Annual Engineering Technology Conference on Energy.. SPE European Petroleum Conference.. and Brunner.. Paper CIM 92-93.. June 7-10. F.. S.L. 23rd ASME Energy Sources Technology Conference and Exposition. February 14-17. D. Inikori. (2001). ETCE2000/Prod-10061.halliburton. September 16. (www. Petroleum Engineer International. Jnr. 2000.. Paper SPE 25050. R.H. February 4-5. Annual Technical Conference of Petroleum Society of the CIM. Halliburton. and Wojtanowicz.. ETCE2001-17100.O. M. M.O. February 5-7. A. TX.. Ozkan. S. Petroleum Engineer International.. K. “Horizontal Well Solves Water Coning”. Document No. ETCE2002/Prod-29165.W. Z. 2001(B).. J. “A Scale-Model Study of Bottom-Water Drives” Paper SPE . Henley. TX. “Reservoir Conformance Technology – Maximizing Your Reservoir Value Through The Management of Unwanted Water and Gas”. New York. “Effect of 2-Phase Oil-Water Flow and Friction Pressure Loss in The Wellbore on Water Cresting in Horizontal Wells”..com) .. Houston. 1991.1944. Hansen.K. “A General Solution of Gas/Water Coning Problems for Horizontal Wells”. G. 2002. “Investigation of the Effects of Completion Geometry on Single Phase Liquid Flow Behavior in Horizontal Wells”. Inikori. A. February 26-28. W. and Wojtanowicz. San Antonio.Guo. Sarica. July. Ibrahim. B. 1960. Hassler..E..

1970. Dallas. Seines. Dallas.. and Larson.C. J. R.. “VIP. TX. Giger.. Paper SPE 2796. Thesis. Paper SPE 23096. Calgary. R.. p914-974)..C and Dalton.D. 1991. McMullan. LA. M. LA. 1991. S.169. A. New Orleans. February 5-6. 1986. 1989. and Purpich. 1996. 1941. LA.. Leverett. “Some Practical Formulas To Predict Horizontal Well Behavior”. 61st Annual Technical Conference and Exhibition of the Society of Petroleum Engineers.L. September 36. C.C. October 5-8. Aberdeen. W. 1999.Joshi. and Combe. and Havig. C. Paper SPE 15430. pp 389-420 Karcher. pp152 .. Louisiana State University and A & M College. Baton Rouge. SPE/Petroleum Society of CIM International Conference on Horizontal Well Technology. Kurban.. S. 1970. 2001. J. J. and Ridings..M. “Horizontal Well Technology”. H... SPE Annual Technical Conference and Exhibition (JPT. “Impact of Skin Damage on Horizontal Well Flow Distribution”. T. February 5-6. December Landmark. Pennwell Publishing Company. R. “Numerical Simulation of Downhole Water Sink Production System Performance”. Lien...). “The First Long-Term HorizontalWell Test in the Troll Thin Oil Zone”. Mattax. F.J. Monograph Vol. SPE Offshore Europe Conference. Houston. Paper 20715.H.A.2000.. “Methods for Numerical Simulation of Water and Gas Coning”. Proceedings of PTTC Symposium on Water Production. LA. Koonsman. 13. TX. Scotland.L. 1991. SPE Textbook Series. April 5. Second Symposium on Numerical Simulation of Reservoir Performance. Paper SPE/Petroleum Society of CIM 65507.. and Coats.P. OK. “Ness Horizontal Well Case Study”. 1990. “A Numerical Coning Model”. MacDonald. p135. “Reservoir Simulation”. (ed. Second Symposium on Numerical Simulation of Reservoir Performance. Tulsa. S. “Produced Water in Louisiana: Analyzing the Magnitude of the Problem”. Canada. B.H.. Kimbrell.O. Paper SPE 2812. November 6-8. K. Transactions of AIME 142. K. M.C.. TX Letkeman.J. September.Executive Reference” Landmark Corporation.. 177 . TX. “Capillary behavior in Porous solids”..Sc. August 1991. Baton Rouge. T. New Orleans. Richardson.

TX. Beijing. M. September 30.TX.R.T. Paper SPE 21611. “Prediction of Dispersed Viscosity of Oil/Water Mixture Flow in Horizontal Pipes”.J. Muskat. November 14-17. Alberta. S. SPE European Petroleum Conference. Houston.. B. “Physical Principles of Oil Production”. T. “An Approximate Theory of Water-Coning in Oil Production” AIME Transaction. Richardson. Mungan. Vol. “Further Discussion of Productivity of a Horizontal Well”. Paper SPE 19822. London. M.. P. “Reservoir Engineering Aspects of Horizontal Wells Application to Oil Reservoirs with Gas or Water Coning Problems”. “Influence of Pressure Drop Along the Wellbore on Horizontal Well Productivity”. Martinez. E. and Skjaeveland. p149 Peng. SPE 63rd Annual Technical Conference and Exhibition. R. and Rogers. Paper SPE 18221. “Optimal Perforation Design for Horizontal Wells in Reservoirs With Boundaries”. 114. Arirachakaran.. “A Theoretical and Experimental Coning Study”. 1991. and Wyckoff. A.. Singapore. February 8-10.E. O. Paper SPE 25502 and 57687. October 6-9. Herring. Las Vegas. Papatzacos. 1999. W.Marett.J. 1992. 1935. J.. 178 ... September. and Brill.. New York City. M. 30-October. and Landman. TX. Houston. Paper SPE 29958. Paper SPE 18340.P. 3 Muskat. NV. 1991. Shoham. 1993. R. Miller. N. N. Ozkan. August 1991).. TX. Peaceman. and Yeh. Asia Pacific Oil and Gas Conference and Exhibition... 13th IEA Collaborative Project on Enhanced Oil Recovery Symposium on Abatement of Water and Gas Coning. Inc. SPE Annual Technical Conference and Exhibition (Reservoir Engineering Journal. 1989. San Antonio..P. C.P. Banff. Sarica.. 1975.. September.. SPE Reservoir Engineering Journal.. 1993.. U. and Haci.. 226-236.. D.M. Martinsen. PR China. M. p. S. Paper SPE 25366. “Case Histories of Worldwide Application of Horizontal Wells For The Abatement of Water/Gas Coning”. 1973. October 2-5.W. SPE-AIME 49th Annual Fall Meeting. October 16-19.. International Meeting on Petroleum Engineering. “Cone Breakthrough Time for Horizontal Wells”. “Performance of Horizontal Wells in the Helder Field”. C. B. October 8-11. 48th Annual Meeting. McGraw-Hill Book Co.L. 1988. 1995. 1988. R.K. P.. 1949. Murphy. Paper SPE 4982.. “Performance of Oil Wells in Bottom Water Drive Reservoirs” Paper SPE 4633. p144-163.. SPE Journal 4 (3). Nzekwu.D.

141-159.. American Chemical Society.1 . Academy Press. pp 404-413. H. M. October 22-25.. 1972. M. May 22-24. I. “Behavior of Water Cresting Under Horizontal Wells”. Brazil. Kooloid –Z. (ed. No. Aavatsmark. Schols. Annual Technical Conference and Exhibition.. Permadi. Paper SPE 28968. S. “Correlation for predicting Viscosity of W/O-Emulsions based on North Sea Crude Oils”. Paper SPE 30743. S. and Wheatley.C.T. J. and Pratomo. TX.. Fifth Latin American and Caribbean Petroleum Engineering Conference and Exhibition.. and Kartoatmodjo. Vol. March 9-11. September. “Water Cresting Behavior Under High Angle Wells: An Experimental Investigation”. Y. “Potential of Multilateral Wells in Water Coning Situations”. August 30 – September 3. 1992.J.G. “Considering Wellbore Friction Effects in Planning Horizontal Wells”. W. . “Emulsions Fundamentals and Applications in the Petroleum Industry” Advances in Chemistry Series 231. R. 1997. 1967..W. Pirson. 1993.. Western Regional Meeting. K. S. R. 1984.S. Journal of Colloid Science 5. SPE International Symposium on Oilfield Chemistry.Z. and Alfonso. Journal of Petroleum Technology (JPT).. 1996. Paper SPE 12066. G. “The Formation and Flow of Emulsion”.M. TX. Paper SPE 39071.. Washington. Rio De Janeiro. P.. Gadelle. 1995. P. 1997.M. Houston.. L. (65) 1933 p.). Anchorage.. Lee. Dupuy. 42nd Annual Fall Meeting of the Society of Petroleum Engineers of AIME. pp 1-62. E. DC. R. and Rushworth.L. R.. Seines. TX.. Alamsyah.32 Richardson.. p1559-1566 Renard. San Antonio.N. January. 1932.. SPE Production Operations Symposium. E.G. New York. Dalls.L. H.. and Mehta. Ronningsen. OK. Wibowo. Reed. 1959. P.Permadi. NY.S. Oklahoma City.P. Journal of Petroleum Technology. Alaska. p6-11 Schramm.. “A Study of Remedial Measures for Water-Coning By Means of a Two-Dimensional Simulator”. “Horizontal Well Completion With Stinger for Reducing Water Coning Problems”.J. October pp 994 – 1000 179 . 1995. Richardson. Permadi.P.. “Oil and Water Production in a Reservoir With Significant Capillary Transition Zone”. Lien.. 88. Erdoel Erdgas. “An Empirical Formula for the Critical Oil Production Rate”.. Paper SPE 1808.. February 14-17. Paper SPE 35715. October 14... C.

. J. p160 Swisher. 68th Annual Technical Conference and Exhibition. R.. 1993.K. Paper SPE 25295.J.. Paper SPE 49052. D. B. E. “Performance of Horizontal Wells Completed with Slotted Liners and Perforations”. and Butler.M. R. “A Correlation for Predicting Water Coning Time”.H.. SPE Annual Technical Conference and Exhibition. Tehrani. October 27-30. Paper SPE 29693. Calgary. A. New Orleans.. 61st Annual Technical Conference an Exhibition. Lawrence. 1986. Paper SPE 30697. March 24-27. Dallas. SPE Annual Technical Conference & Exhibition.P. SPE/EPA Exploration and Production Environmental Conference..G. A.... E. “Application of Horizontal Wells at Prudhoe Bay”. and Wojtanowicz. Y. Paper SPE 26521. C.S.W. “Discussion of Productivity of a Horizontal Well”...D. and Butler. October 3-6. SPE Reservoir Engineering Journal. November 6-8. SPE International Meeting on Petroleum Engineering. November. “In-situ Segregated Production of Oil and water – A Production Method With Environmental Merit: Field Application”. and Yildiz. TX. “Spatial Distribution of Oil and Water in Horizontal Pipe Flow”. C. M. March 27-29.CIM 65516. Houston. R. 180 . p453454 Suprunowicz. and Wojtanowicz. A. Su.I. p594-600. M. Paper SPE 22390. G. “An Overview of Horizontal Well Targets Recently Drilled in Europe”..K.. LA. A. 2000. R.M.F. and Cornelius. Paper SPE 56524.D.. and MacDonald. Brice. “Friction Factor of Perforation Roughness in Pipes”. October 3-6. October 5-8. 1992... Z. A. T.K. SPE/Petroleum Society of Cim International Conference on Horizontal Well Technology.Sherrard.W. and Gudmundsson.. A. M. D.J. TX. Suprunowicz. Journal of Petroleum Technology. 1998. Sobocinski. LA.. Soleimani.. Kelkar. Houston. Sarica.Canada. Houston. Shirman. Swisher. 1995.. May. Beijing. October 22-25. D. May. 1993. 1965.Alberta.. “New Dual Completion Eliminates Bottom Water Coning”. New Orleans. Paper SPE 26262. Paper SPE/PS .D. SPE Reservoir Engineering Journal. Tang. China. and Hewitt. “Further Discussion of Productivity of a Horizontal Well”. and Wojtanowicz.. 1999.. Ozkan. Paper SPE 15376. TX. 1992. Annual Technical Conference and Exhibition. 1995. TX.. “More Oil with Less Water Using Downhole Water Sink Technology: A Feasibility Study”.

047. CA.cr.G..A.J. Wojtanowicz... International Meeting on Petroleum Engineering.L.: “Producing Petroleum from Underground Formations. June 26-28..gov/energy/factshts/003-97/FS-003-97. September 22-25.. M. 60th Annual Technical Conference and Exhibition. Int. Lawrence Berkeley Laboratory. 1985. Oklahoma City.P.usgs. 2000.... Widmyer. November. E. Richardson. J.. Paper SPE 36609. SPE paper. Las Vegas. TX. Paper SPE 21654. SPE Production Operation Symposium. Paper SPE 61182.. “An In-situ Method for Downhole Drainage – injection of Formation Brine in a Single Oil-Producing Well.K. “A field Study of Horizontal Well Design in Reducing Water Coning”. C. “Waterflooding”. Z. Symposium on Scientific and Engineering Aspects of Deep Injection Disposal of Hazardous and Industrial Wastes. Patent No.. 6-9 181 . Third Printing. B. May 10 –13. 1997. “Reduction of Emissions to Sea by Improved Produced Water Treatment and Subsea Separation Systems”. 1996.. G. Beijing. TX. 3.K. Berkeley.” Proc. and Wattenbarger. 2 & 5. and Environment in Oil and Gas Production. OK. Wu. Norway. pp. October. and Shirman. http://greenwood.. Hui Xu. Paper SPE 22931. Wheatley.html Weinstein. G. Fact Sheet FS-003-97.Trallero. E.S. Paper SPE 30016. Reynolds. SPE Annual Technical Conference and Exhibition. B. R. 66th Annual Technical Conference and exhibition. R. 1995. H. A. J.S. A. R. “USGS Research on Saline Waters Co-Produced With Energy Resources”. October 6-9 Wanty. Sarica. Denver. 1955 Willhite..” U. and Bassiouni. 1991..I. 1986. “Water Coning Calculations for Vertical and Horizontal Wells”.855. 1994.. NV... K. W. Stavanger. Wolff. Journal of Petroleum Technology. April 7-9. and Markitell. 2. and Brill. Oct. Dallas. CO.H. Yang. Safety. March. 1986... “ A Study of Oil-Water Flow Patterns in Horizontal Pipes”. Chappelear. PR China.E. J.A. “An Approximate Theory of Oil/Water Coning”. and Nolen. 14-17. “Second Comparative Solution Project: A Three Phase Coning Study”. Paper SPE 14210. J. SPE Textbook Series. “Oilwell Coning Control Using Dual Completion With Tailpipe Water Sink”. Wojtanowicz. SPE International Conference on Health.P.

pp 203-209. M.. “Effect of Perforation density on SinglePhase Liquid Flow Behavior in Horizontal Wells”. Paper SPE 57395. 1994.. 56th Annual Fall Technical Conference and Exhibition of Society of Petroleum engineers of AIME. Thesis. Sarica. 1981. Richardson... PhD. and Lake. 1997. C. H. “Investigation of Single Phase Liquid Flow Behavior in Horizontal Wells”. Y. San Antonio. 1999. SPE Journal of Production and Facilities. Tulsa. OK. H. Yuan. Dissertation. The university of Tulsa. August. “The Effect of Capillary Pressure on Immiscible Displacements in Stratified Porous Media”. October 5-7. Yuan. The university of Tulsa. Larry W. TX. 182 . and Brill. TX.. Tulsa. Yuan. “Investigation of Single Phase Liquid Flow Behavior in a Single Perforation Horizontal Well”. OK. J.Yokoyama. J. H.. Paper SPE 10109. Sc.P..

APPENDIX A – CAPILLARY PRESSURE/RELATIVE PERMEABILITY HYSTERESIS DATA DECK -.PERMEABILITY HYSTERESIS STUDY ON DWS IPW WINDOW (30 FT CAPILLARY) -.292 / -.436 4.628 95.============================================================== -.283 0.AUGUST 1999 -.843 126.670 200 / OUTRAD 850 EQUALS / 183 .602 -41.drv -0.035 72.704 55.909 220.gas -.133 1.093 0.================================================================ RUNSPEC TITLE DWS oil/water coning test run -.163 0.651 0.973 2.DONWHOLE WATER SINK TECHONOLOGY SIMULATION RUN ON ECLIPSE 100 -.948 31.420 13.984 7.215 0.263 290.859 -1.896 10.SOLOMON O INIKORI DIMENS 31 1 20 / RADIAL NONNC OIL WATER -.495 1.751 18.534 -5.604 3.IPW WINDOW VERIFICATION ON HILAL'S PROJECT DATA .disgas FIELD WELLDIMS 4 20 1 4 / AQUDIMS 1 1 1 1 1 40 / SATOPTS 'HYSTER' / TABDIMS 2 / START 1 'AUG' 1999 / NSTACK 100 / MESSAGES 8* 5000 2*800 / UNIFOUT GRID -====================================================================== -SPECIFY INNER AND OUTER RADIUS OF IST AND LAST GRID BLOCK IN THE RADIAL DIRECTION INRAD 0.493 0.480 166.123 0.374 0.147 23.

4444 0.30 0.6944 0.20 0.55 0.'DTHETA' 'PERMR' 'PERMZ' 'PERMR' 'PERMZ' 'PORO' 'PORO' 'DZ' / 'DZ' / 360 / 1000 500 1000 500 0.1736 0.30 0.0605 0.357 0.536 0.0 31 31 31 31 31 31 1 1 1 1 1 1 1 1 1 1 1 1 1 1 31 31 1 1 11 11 10 10 20 20 1 11 10 20 1 1 / / / / / / 1 11 10 20 1 1 'TOPS' 4720 1 31 1 1 1 1 / / RPTGRID 'DX' / INIT GRIDFILE 1 / PROPS ===================================================================== -NO RELATIVE PERMEABILITY DATA SUPPLIED.1898 0.0e-6 0.000 / TABLE 2 EHYSTR 1* 4 1.8374 1.0029 0.20 0. FACTOR.70 0.40 0.80 0.55 0.536 0.0000 1.000 / TABLE 1 -Sw 0.803 0. PRESSURE.46 0 / PVCDO 184 .428 1.0000 1.30 0.00 Krw 0.02 3.3337 0.0278 0.0000 Pc -1.Sw Krw Kro Pc -0. EMPIRICAL DATA FROM -COREY'S CORRELATION USED SWOF -.0000 0.0128 0.0000 0.6328 1.70 0.80 1.428 0.3164 0.0039 0.250 1.0000 Kro 1.0391 0.0000 0.40 0.1187 0. ETC -Pref Bw(ref) Cw Visw Viscosibility 1788 1.0 0.0002 0.37 1 1 1 1 1 1 5.0000 0.1526 0.00 1.0 5.1 'KR' 'RETR' / PVTW -WATER PVT DATA (REF.071 0.0000 0.803 0.071 0.250 0.357 1. VISCOSITY -COMPRESSIBILITY.3547 1.0366 0. WATER FM VOL.

-OIL PVT DATA FOR DEAD OIL WITH CONSTANT SOLUTION GAS (Bo AND Co NOT GIVEN-ASSUMED) -Pref Bo Co Viso Viscosibility 1788 1.26 1.5e-5 1.25 0 / RSCONST -DATA FOR THIS SECTION ALSO NOT SUPPLIED. -Rs Pb 0.379 1000 / ROCK -DATA ALSO NOT SUPPLIED. -Pref Cf 1788 4.0e-6 / DENSITY -OIL WATER GAS 53.9 62.47 0.0702 / REGIONS =================================================================== SATNUM 620*1 / IMBNUM 620*2 /

SOLUTION ================================================================== -DATUM DATUM OWC OWC GOC GOC RSVD RVVD SOLN -DEPTH PRESS DEPTH PCOW DEPTH PCOG TABLE TABLE METH EQUIL 4770 1788 4770 0 1000 0 / AQUFET -- 1 2 3 4 5 6 7 8 9 10 11 12 13 -- ITEM 14 - INITIAL SALT CONCN. WILL BE RUN AS A SENSITIVITY LATER 4770 1788 10E06 7.0E-6 20 1 1 31 1 1 10 20 'K+' / RPTSOL 'RESTART=2' 'FIP=2' / SUMMARY ==================================== WOPR 'P1' WWPR 'P1' WGPR 'P1' WWCT 'P1' FOPR FOPT FWPR FWPT FLPR 'P2' 'P2' 'P2' 'P2' / / / /

185

-- WELL BOTTOM HOLE PRESSURES WBHP 'P1' 'P2' / FWCT FPR FOE WOPT 'P1' 'P2' / WWPT 'P1' 'P2' / -RPTONLY -PRESENT DATA IN TABULAR FORM FOR GRAF PURPOSE IN PRINT FILE -REQUEST RUNSUM OUTPUT TO GO TO A SEPARATE RSM FILE RUNSUM SEPARATE SCHEDULE ======================================= -- SET MAXIMUM NUMBER OF LINEAR ITERATIONS IN THE NEWTON ITERATION TUNING / / LITMAX 2* 100 / RPTSCHED 'RESTART=2' 'FIP=2' / RPTRST 'BASIC=4' / -- INTRODUCING THE WELLS -- WELL GROUP -- NAME NAME PHASE WELSPECS 'P1' 'G' 1 1 4720 'P2' 'G' 1 1 4780 'i1' 'G' 31 1 4720 'i2' 'G' 31 1 4780 LOCATION I 'LIQ' 'LIQ' 'oil' 'water' J / / / / BHP PREF DEPTH

/ -SPECIFYING COMPLETION DATA -- WELL - LOCATIONOPEN WELL -NAME I J K1 K2 COMPDAT 'P1' 1 1 1 1 'P2' 1 1 13 14 'i1' 31 1 1 1 'i2' 31 1 13 14 / COMPIMB -WELL NAME 'P1' I-LOCN 4* J-LOCN 2 /

SAT SHUT 'OPEN' 'OPEN' 'OPEN' 'OPEN' TAB 1* 1* 1* 1*

CONN FACT 1* 1* 1* 1* DIA 0.585 0.585 0.585 0.585 / / / /

K-UPR

K-LWR

IMB TABLE #

186

'P2' /

4*

2

/

-- PROD. WELL CONTROLS WCONPROD -1 2 3 -- WELL OPEN/ CNTL -- NAME SHUT MODE 'P1' 'OPEN' 'P2' 'OPEN' /

DRAINAGE RATE IS VARIED AT FIXED OIL RATE. 4 5 OIL WATER RATE RATE 'LRAT' 'LRAT' 6 7 GAS LIQUID RATE RATE 3* 110 1* 3* 180 1* 8 RES RATE 1* 1* 9 BHP / /

wconinje -- WELL INJT/ OPEN/ CNTL/ 5 6 7 -- NAME TYPE SHUT MODE 'I1' 'OIL' 'OPEN' 'GRUP' 3* / 'I2' 'WATER' 'OPEN' 'GRUP' 3* / / GCONINJE 'FIELD' 'OIL' 'REIN' 2* 1.00 / 'FIELD' 'WATER' 'REIN' 2* 1.00 / / -SPECIFY WELL ECONOMIC LIMIT AS 98 PERCENT WATERCUT -- WECON -WELL MIN MIN MAX MAX MAX WKVR -NAME OIL GAS WATER GOR GWR PRCDR -RATE RATE CUT -- 'P1' 1* 1* 0.98 1* 1* 'WELL' 'Y' / -- SPECIFY REPORT AT FOUR YEAR INTERVAL FOR 20 YEARS TSTEP 48*30.4 TSTEP 48*30.4 TSTEP 48*30.4 TSTEP 48*30.4 TSTEP 48*30.4 / / / / /

END RUN

END

187

APPENDIX B – WATER CONE REDUCTION STUDY ON WELL LVT B.1 Executive Summary

A watered-out well LVT had been selected as a candidate for re-completion, installation, and field implementation of the Downhole Water Sink (DWS) technology for water coning control. Prior to watering out, the well had been produced for several years. As a result of water coning a sizable zone of high water saturation has developed around the well. The estimated watercut of the well before it was shut down was between 94 and 98 percent. The LVT well has 37 ft of oil sand underlain by 87 ft of water. The reservoir structure represents an edge water drive system with strong aquifer support. Initially, prior to the movement of the Oil-Water Contact (OWC), the well had 45-foot long perforated section covering the entire length of the oil bearing sand. Later, these perforations were squeezed off and the top 15 ft re-perforated (9540 – 9555 ft). A bottom section of water sink completion had been recently added within the interval: 9596 – 9606 ft. Also, an isolation packer has been set around the OWC to isolate fluids produced from the top and bottom completion through separate tubing strings.
B.2 Objective

Objective of this study was to determine the approximate length of time required to bring watercut in the top production from current level of 94-98 percent to a manageable level. Also, the study shall verify the effect of prolonged process of watering out wells on reinstating their productivity to oil with DWS. In addition, the study shall evaluate the effect of capillary pressure and leaking cement on the process of water saturation drainage around the well with DWS. 188

189 .00 0.00 3.00 0. a slight linear interpolation of the relative permeability data from M2 was used to generate estimates for LVT well calculations (Figure B-2).B.00 Water Saturation (Sw . The value of contact angle of 140 degrees was ignored in the generation of the Leverett function. in turn. 8.00 7.Fraction) Figure B-1 Capillary Pressure Plot for LVT Well Also. was used to generate an estimated capillary pressure correlation for LVT well.20 0. The correlation/curve is as shown in Figure B-1.00 2.40 0.00 Capillary Pressure (psi) 0.3 Input Data for Eclipse Simulator Capillary pressure data was supplied by the Joint Industry Panel (JIP) member.00 4.00 5. direction of measurement was not specified so we could not use the complementary angle.80 1.60 0. as it does not represent a water-wet reservoirs.00 6. Also. The data had been collected from another well in the M2 reservoir with permeability much lower than that in LVT well. The data was then used to construct the Leverett’s Jfunction which.00 1.

190 .800 0.000 0.500 0. The approach was necessary to determine the extent of water invasion and to create a more realistic picture of the current status of this well.fraction) Figure B-2 Relative Permeability Curve for LVT Well Studies Other modification of the provided input data was the increase of the rock compressibility value to solve simulator’s instability problem resulting from rapid changes in production rates.600 0.000 0.400 0. The nature of the reservoir employed for this study is shown in Figure B-3. certain grids in the radial model were assigned zero porosity and permeability representing shaled out areas.700 0.400 0.Relative Permeabilities (Fraction) 1. B.200 0.000 0. Prior to forecasting watercut reduction in LVT well efforts were made to simulate the actual behavior of the well from the beginning of initial production.300 0.100 0. in order to obtain a good history match of the well oil and water production profile. the vertical permeability was reduced to 1/100th of the horizontal permeability.900 0. All other data for this study were same as for previous work by the LSU DWS team.600 0.000 Water Saturation (Sw .800 1. Also.4 Approach to Study To create a conceptualized model of the reservoir with edge water drive.200 0.

0.Cap Figure B-4 Watercut Reduction Performance of DWS Technology 191 .-No Cap DWS wit h Channel Wit h Sink Prod.20 7000 st b/ day 100 st b/ day 1.20 0.FNC W/ Sink&Mod.00 Nat ural Cone Collapse at Low rat e 100BOPD Watercut (Fraction) 0.00 0 1000 2000 Channel ef fect 3000 4000 5000 6000 7000 Time (Days) No Sink Prod.60 Cone Collapse w/DWS / Cap.40 0.Figure B-3 LVT Well Reservoir Grid Model 1.80 0.-No Cap DWS. transition Channel ef fect Cone Collapse wit h DWS No Cap.

000 B LP D) No DWS No Cap(to p 1 00 B LP D) Watercut w/DWS No Capillary Figure B-5 Watercut Forecast of LVT Well without Capillary Pressure 192 .The next step was to simulate reduction of watercut due to the activation of DWS drainage under several scenarios (Figure B-4).00 0.000 B LP D btm) DWS w/Cap (1 00 -1 0. 1 0.” It shows that fully developed water cone in an oil well may not retract quickly even if the well is shut-in for a considerable length of time. The study also investigated the rate of natural collapse of water cone around the wellbore due to a reduction in the drawdown pressure simply by a rate reduction. “Natural Cone Collapse” indicates that it will take a ‘life time’ for the cone to naturally reduce to an appreciable level even at the very low oil production rate once the cone is fully developed around the wellbore.Low rate . The top completion was put at 100 stb/day production without the DWS.00 0 1000 2000 3000 4000 Time (Days) from start of production with/out DWS DWS No Cap(1 00 B LP D to p.20 Watercut Performance (Fraction) 1. The performance shown in Figure B-5 as a plot marked “watercut without DWS – Low rate and no capillary Transition” or.40 0.No Capillary Transition Watercut with DWS and Capillary Transition 0.60 Watercut without DWS .80 0.20 0. This is a typical industry practice as a solution to reducing the adverse effect of water cone development. 1.

water cone had developed over several years creating and extensive radius of watered out zone around the wellbore. The water sink production was initiated at a rate of 10.20 1. 1 0. They include watercut performance with or without capillary pressures using a top rate of 100 stb/day and a bottom drainage rate of 10.40 0.1 Scenario A – Watercut Reduction Without the Effect of Capillary Pressure The top completion was put on production at a rate of 100 stb/day.000 B LP D) No DWS No Cap (to p 1 00 B LP D) Figure B-6 Watercut Reduction Forecast with and without Capillary Pressure 193 .B.5 Discussion of Results Several scenarios have been investigated.80 0. Watercut Performance (Fraction) 1. This is because. At the initial stage. B.000 stb/day.60 0.20 0.000 B LP D btm) DWS w/Cap (1 00-1 0. The results showed several stages of watercut decline corresponding to the extent of cone development.No Capillary Transition Watercut w /DWS No Capillary Transition 200 400 600 800 Time (Days) from start of production with/out DWS DWS No Cap(1 00 B LP D to p.Low rate .00 0 Watercut w ith DWS and Capillary Transition Watercut w ithout DWs .000 stb/day.5. the well had been shut-off for 30 days representing the period of workover when the well was not producing. Prior to production.00 0. water production from the sink did not affect the watercut at the top completion appreciably.

it took about 270 days of continuous pumping water from the bottom completion to reduce the watercut to 80 percent. Well production history matching shows an increased limiting watercut of about 98 percent which corresponds better to its current value in LVT Well. Also.” in Figure B-6. a dramatic reduction in watercut occurs after the cone size has been reduced to a few feet around the wellbore. Figure B-6 is a highlighted plot of Figure B-5.5 percent during the first fifty days of drainage. The rate of watercut reduction could increase if the vertical permeability is greater than that used in the simulation.5. The results indicate that reduction in watercut would not be noticeable in a couple of days or perhaps a few weeks. Similar to the previous case. “with capillary transition.To reverse such a large cone the radial spread of the cone has to be minimized to few feet around the well before any appreciable decrease in watercut could be noticed. the minimum watercut achievable with capillary transition is about 27 194 . To achieve 37 ft of oil sand the OWC was moved to the original position.2 Scenario B – Watercut Reduction Performance with Capillary Pressure Effect This case is marked.02 percent/day. They reported steady reduction of watercut of 0. Modified capillary transition effect was incorporated in the study. Also. Inclusion of capillary effects delays any noticeable change in watercut from the top completion. Over three years of water pumping is required to reduce the watercut at the top completion from 98 percent to 80 percent. The Figure B-6 show watercut reduction of only about 1. B. Results of this study correspond qualitatively to those obtained by another JIP member in their field test of DWS.

This could be seen in Figure B-4. Beyond this period.5. this is for a period of about three years.90 0.3 Scenario C – Watercut Reduction Performance with Leaking Cement This case is marked.4 Scenario D – Watercut Reduction Performance with High Rate at Top Completion 1. The results showed that cement channeling would actually enhance watercut reduction. This scenario was included to predict the behavior of the cone/watercut at top completion in a case where there is no cement integrity behind the casing. The simulation run was made for 100 stb/day top rate and 10. the effect of the DWS water sink completion becomes impressive in reducing watercut. “channel effect.40 0.” in plots in Figure B-4.60 0. B.20 0.00 0. watercut would reduce to about 80 percent within three weeks of pumping.00 0 1000 2000 3000 4000 Time (Days from start of DWS) Figure B-7 Watercut Performance with High Rate at Top Completion Watercut (Fraction) 195 . It might seem that the effect of DWS technology is negligible in the case with capillary transition zone looking at Figure B-6.30 0.percent as compared to 10 percent without the capillary transition effect. In the presence of cement channel.5. B. However.10 0.50 0.000 stb/day water sink rate.70 0.80 0.

20 0. the lateral extent of the water cone around the wellbore is limited to a few tens of feet. Watercut reduction was simulated with 800 stb/day from top and 10. In effect.This case is depicted in B-7. B.6 Appropriate Watercut Level for Deployment of DWS Technology In this section the study evaluated the performance of the DWS technology in reducing watercut in the oil zone completion in a case where the water cone is at the development stage.70 Watercut (Fraction) 0. the minimum achievable value of watercut is about 27 percent – the value similar to the case with low top rate and capillary pressure effect.40 0. Also.10 0 1000 2000 3000 4000 5000 6000 7000 Time (Days) Figure B-8 Watercut Performance of DWS in a Developing Cone Scenario The result of the study shown in Figure B-8 indicated that the initial delay experienced when the Downhole Water Sink technology is implemented to collapse the 196 . The results show that the watercut reduction is very gradual.30 0.90 0.80 0.10 0.50 0. Watercut reduces to about 80 percent after about one year of continuous pumping.000 stb/day from bottom water sink completion for the case without the effect of capillary pressure. 0.00 -0.60 0.

the results show that the DWS technology does have the capability of reducing the watercut over time but the time frame may be slow depending on the extent of coning severity. Also. B. The watercut in top completion shows immediate reduction as soon as the water sink in activated. This confirms the initial preposition that the initial delay is due to the full development of the water cone over time.7 Conclusion of Studies The study demonstrates the effect of well production history on watercut control with DWS. 197 .cone is absent. As the well LVT had been seriously watered-out over number of years it might be necessary to allow much longer time for pumping to reduce watercut and restore oil productivity of this well.

31 0.gas -.APPENDIX C – DATA DECK FOR BILATERAL WATER SINK (BWS) CONCEPT FOR WATER CRESTING CONTROL -.============================================================== RUNSPEC TITLE DWS OIL/WATER CONING STUDIES ON HORIZONTAL WELL -.SOLOMON O INIKORI DIMENS -.DONWHOLE WATER SINK TECHONOLOGY SIMULATION RUN ON ECLIPSE 200 -. 2002 -.============================================================== -.SIMULATION OF HORIZONTAL WELL APPLICATION OF DWS .JAN.DISGAS FIELD WELLDIMS 4 150 1 4 / START 1 'JAN' 2002 / FRICTION 2 1 / NSTACK 100 / MESSAGES 8* 5000 2*200 / UNIFOUT GRID -====================================================================== DXV 40 DYV 60 EQUALS 'DZ' 'DZ' 'PERMX' 'PERMZ' 'PERMX' 'PERMZ' 'PORO' 'PORO' 'PORO' 5 30 1 1 75 75 1 1 1 1 1 1 1 1 1 75 75 75 75 75 75 75 20 20 1 1 1 1 1 1 1 1 12 20 20 20 20 20 20 20 11 12 1 1 9 9 1 9 12 / / TOTAL 85' 8 8 12 12 8 11 12 / / / / / / / STEADY STATE 60 60 60 60 10*60 60 60 60 60 60 / total 1200' 40 40 40 40 65*40 40 40 40 40 40 / total 3000' 8000 800 8000 800 0.35 1e10 198 .NX NY NZ 75 20 12 / NONNC OIL WATER -.

00 0.93 0 0.40 0.99 0.15 1.5e-5 1. WATER FM VOL. VISCOSITY -COMPRESSIBILITY.05 0.06 0 0. FACTOR.18 0.35 0 0.00 0. -Pref Cf 2500 4.379 1000 / ROCK -DATA ALSO NOT SUPPLIED.0e-6 / DENSITY -OIL WATER GAS 50.00 0 1.10 0. ETC -Pref Bw(ref) Cw Visw Viscosibility 2500 1.0 64.98 0 0.00 0 / PVTW -WATER PVT DATA (REF.34 0 / RSCONST -DATA FOR THIS SECTION ALSO NOT SUPPLIED.20 0.85 0.60 0.'TOPS' / COPY 6154 1 75 1 20 1 1 / 'PERMX' 'PERMY' / / INIT GRIDFILE 1 / PROPS ===================================================================== -NO RELATIVE PERMEABILITY DATA SUPPLIED.35 0.10 0.18 0 0. USED EMPIRICAL CORRELATION DATA SWOF -. -Rs Pb 0.0702 / SOLUTION ================================================================== -DATUM DATUM OWC OWC GOC GOC RSVD RVVD SOLN -DEPTH PRESS DEPTH PCOW DEPTH PCOG TABLE TABLE METH EQUIL 6154 2500 6194 0 1000 0 / RPTSOL 199 .2 0.60 0 0.02 3.Sw Krw Kro Pc -0.55 0.50 0 / PVCDO -OIL PVT DATA FOR DEAD OIL WITH CONSTANT SOLUTION GAS -Pref Bo Co Viso Viscosibility 2500 1.00 0.0e-6 0.70 0. PRESSURE.30 0.

INTRODUCING THE WELLS -.WELL GROUP LOCATION BHP -.SET MAXIMUM NUMBER OF LINEAR ITERATIONS IN THE NEWTON ITERATION TUNING / / LITMAX 2* 100 1* 50 / RPTSCHED 'RESTART=2' RPTRST 'BASIC=4' 'FIP=2' / PREF PHASE 'LIQ' 'LIQ' / / / -.LOCATIONOPEN/ SAT CONN WELL 200 .'RESTART=2' 'FIP=2' 'PRES' 'SWAT' / SUMMARY ==================================== WOPR 'P1' 'P2' / WWPR 'P1' 'P2' / WGPR 'P1' 'P2' / WWCT 'P1' 'P2' / WOPT 'P1' 'P2' / WWPT 'P1' 'P2' / WBHP 'P1' 'P2' / FOPR FOPT FWPR FWPT FLPR FWCT FPR FOE / -RPTONLY -PRESENT DATA IN TABULAR FORM FOR GRAF PURPOSE IN PRINT FILE -REQUEST RUNSUM OUTPUT TO GO TO A SEPARATE RSM FILE RUNSUM SEPARATE SCHEDULE ======================================= -.NAME NAME I J DEPTH WELSPECS 'P1' 'G' 20 11 6157 'P2' 'G' 20 11 6201 / --SPECIFYING COMPLETION DATA WELL .

41667 0.41667 0.41667 0.41667 FACT 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* DIA 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / 201 .41667 0.WPIMULT -.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.'P1' 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.5 K2 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' SHUT 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* TAB 0.-NAME I J K1 COMPDAT 'P1' 22 11 1 1 'P1' 23 11 1 1 'P1' 24 11 1 1 'P1' 25 11 1 1 'P1' 26 11 1 1 'P1' 27 11 1 1 'P1' 28 11 1 1 'P1' 29 11 1 1 'P1' 30 11 1 1 'P1' 31 11 1 1 'P1' 32 11 1 1 'P1' 33 11 1 1 'P1' 34 11 1 1 'P1' 35 11 1 1 'P1' 36 11 1 1 'P1' 37 11 1 1 'P1' 38 11 1 1 'P1' 39 11 1 1 'P1' 40 11 1 1 'P1' 41 11 1 1 'P1' 42 11 1 1 'P1' 43 11 1 1 'P1' 44 11 1 1 'P1' 45 11 1 1 'P1' 46 11 1 1 'P1' 47 11 1 1 'P1' 48 11 1 1 'P1' 49 11 1 1 'P1' 50 11 1 1 'P1' 51 11 1 1 'P1' 52 11 1 1 'P1' 53 11 1 1 'P1' 54 11 1 1 'P1' 55 11 1 1 'P1' 56 11 1 1 'P1' 57 11 1 1 'P1' 58 11 1 1 'P1' 59 11 1 1 'P2' 22 11 10 10 'P2' 23 11 10 10 'P2' 24 11 10 10 'P2' 25 11 10 10 'P2' 26 11 10 10 'P2' 27 11 10 10 'P2' 28 11 10 10 'P2' 29 11 10 10 'P2' 30 11 10 10 'P2' 31 11 10 10 'P2' 32 11 10 10 'P2' 33 11 10 10 / -.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.

1100 1 / -.2 ECONOMIC LIMIT AS 98 PERCENT WATERCUT MIN MAX MAX MAX WKVR GAS WATER GOR GWR PRCDR RATE CUT 1* 0.1100 1 / 22 11 10 1* 1* 'I' 33 1* / / -.WELL OPEN/ CNTL OIL WATER GAS LIQUID RES BHP -.2 / TSTEP 24*15.5 / REPRESENTS UNDAMAGED WELL SINCE SYSTEM IS 1/2 OF WHOLE -.SPECIFY REPORT AT ONE YEAR INTERVAL FOR XX YEARS / TSTEP 24*15.98 1* 1* 'WELL' AT ONE YEAR INTERVAL FOR XX YEARS / / 6 7 8 GAS LIQUID RES RATE RATE RATE 3* 3000 1* 1* 3* 0 1* 1* 9 BHP / / END RUN 'Y' / WCONPROD -1 2 3 4 5 -.NAME SHUT MODE RATE RATE RATE RATE RATE 'P1' 'OPEN' 'LRAT' 3* 3000 1* 1* / 'P2' 'OPEN' 'LRAT' 3* 0 1* 1* / / -SPECIFY WELL WECON -.98 1* 1* 'WELL' 'Y' / -.2 / END RUN 202 .PRODUCTION WELL CONTROLS ./ WFRICTN -.'P2' 0.SPECIFY REPORT / TSTEP 24*15.WELL MIN MIN MAX MAX MAX WKVR -NAME OIL GAS WATER GOR GWR PRCDR -RATE RATE CUT 'P1' 1* 1* 0.NAME SHUT MODE RATE RATE 'P1' 'OPEN' 'LRAT' 'P2' 'OPEN' 'LRAT' / -SPECIFY WELL ECONOMIC LIMIT AS 98 PERCENT WATERCUT WECON -.I J K Tlen1 Tlen2 Dirn RangeEnd Diam 22 11 1 1* 1* 'I' 59 1* / / WFRICTN 'P2' 0.2 TSTEP 24*15.DRAINAGE RATE IS VARIED AT FIXED OIL rATE WCONPROD -1 2 3 4 5 6 7 8 9 -.NAME OIL -RATE 'P1' 1* -.41667 0.-.WELL MIN -.WELL OPEN/ CNTL OIL WATER -.WELL DIAM ROUGH FLOW SCALE 'P1' 0.41667 0.2 / TSTEP 24*15.

SPECIFY REPORT AT ONE YEAR INTERVAL FOR XX YEARS / TSTEP 24*15.WELL MIN MIN MAX MAX MAX WKVR -.1 2 3 4 -.WELL MIN MIN MAX MAX MAX WKVR END -.SPECIFY REPORT AT ONE YEAR INTERVAL FOR XX YEARS / TSTEP 24*15.2 TSTEP 24*15.WELL OPEN/ CNTL OIL WATER GAS LIQUID RES BHP -.NAME OIL GAS WATER GOR GWR PRCDR -RATE RATE CUT 'P1' 1* 1* 0.2 / TSTEP 24*15.2 TSTEP 24*15.NAME SHUT MODE RATE RATE RATE RATE RATE 'P1' 'OPEN' 'LRAT' 3* 3000 1* 1* / 203 .NAME OIL GAS WATER GOR GWR PRCDR RUN -RATE RATE CUT 'P1' 1* 1* 0.2 / TSTEP 24*15.2 / 5 6 OIL WATER RATE RATE 'LRAT' 'LRAT' 7 8 9 GAS LIQUID RES RATE RATE RATE 3* 3000 1* 1* 3* 0 1* 1* WCONPROD -.98 1* 1* 'WELL' 'Y' / -.2 / WCONPROD -1 2 3 4 5 6 7 8 -.2 / TSTEP 24*15.2 / WCONPROD -1 2 3 4 5 6 7 8 9 -.NAME SHUT MODE RATE RATE RATE RATE RATE 'P1' 'OPEN' 'LRAT' 3* 3000 1* 1* 'P2' 'OPEN' 'LRAT' 3* 0 1* 1* / END RUN 9 BHP / / -SPECIFY WELL ECONOMIC LIMIT AS 98 PERCENT WATERCUT WECON -.WELL OPEN/ CNTL -.WELL OPEN/ CNTL OIL WATER GAS LIQUID RES -.TSTEP 24*15.NAME SHUT MODE 'P1' 'OPEN' 'P2' 'OPEN' / / / BHP / / -SPECIFY WELL ECONOMIC LIMIT AS 98 PERCENT WATERCUT WECON -.98 1* 1* 'WELL' 'Y' / -.2 / TSTEP 24*15.

NAME OIL GAS WATER GOR GWR -RATE RATE CUT 'P1' 1* 1* 0.2 / TSTEP 24*15.2 / END WKVR PRCDR END RUN 204 .SPECIFY REPORT AT ONE YEAR INTERVAL FOR XX YEARS / TSTEP 24*15.98 1* 1* 'WELL' 'Y' / -.'P2' / 'OPEN' 'LRAT' 3* 0 1* 1* / -SPECIFY WELL ECONOMIC LIMIT AS 98 PERCENT WATERCUT WECON -.2 / TSTEP 24*15.2 / TSTEP 24*15.2 / TSTEP 24*15.2 / TSTEP 24*15.WELL MIN MIN MAX MAX MAX -.

NX NY NZ 75 20 12 / NONNC OIL WATER -.35 1e10 6154 205 .JAN.SIMULATION OF HORIZONTAL WELL APPLICATION OF DWS .gas -.SOLOMON O INIKORI DIMENS -.============================================================== RUNSPEC TITLE DWS OIL/WATER CONING STUDIES ON HORIZONTAL WELL (TAIL PIPE STUDY) -.DONWHOLE WATER SINK TECHONOLOGY SIMULATION RUN ON ECLIPSE 200 -. 2002 -.DISGAS FIELD WELLDIMS 4 150 1 4 / START 1 'JAN' 2002 / FRICTION 2 1 / NSTACK 100 / MESSAGES 8* 5000 2*200 / UNIFOUT GRID -====================================================================== DXV 40 DYV 60 EQUALS 'DZ' 'DZ' 'PERMX' 'PERMZ' 'PERMX' 'PERMZ' 'PORO' 'PORO' 'PORO' 'TOPS' / 5 30 1 1 75 75 1 1 1 1 1 1 1 1 1 1 75 75 75 75 75 75 75 75 20 20 1 1 1 1 1 1 1 1 1 12 20 20 20 20 20 20 20 20 11 12 1 1 9 9 1 9 12 1 / / TOTAL 85' 8 8 12 12 8 11 12 1 / / / / / / / STEADY STATE / 60 60 60 60 10*60 60 60 60 60 60 / total 1200' 40 40 40 40 65*40 40 40 40 40 40 / total 3000' 8000 800 8000 800 0.============================================================== -.31 0.APPENDIX D – DATA DECK FOR TAIL-PIPE WATER SINK (TWS) CONCEPT FOR WATER CRESTING CONTROL -.

USED EMPIRICAL CORRELATION DATA SWOF -.00 0.93 0 0. VISCOSITY -COMPRESSIBILITY.35 0 0.40 0.00 0.379 1000 / ROCK -DATA ALSO NOT SUPPLIED.18 0 0.55 0.2 0.10 0.00 0 / PVTW -WATER PVT DATA (REF. -Pref Cf 2500 4.50 0 / PVCDO -OIL PVT DATA FOR DEAD OIL WITH CONSTANT SOLUTION GAS -Pref Bo Co Viso Viscosibility 2500 1.COPY 'PERMX' 'PERMY' / / INIT GRIDFILE 1 / PROPS ===================================================================== -NO RELATIVE PERMEABILITY DATA SUPPLIED. PRESSURE.85 0.98 0 0.0702 / SOLUTION ================================================================== -DATUM DATUM OWC OWC GOC GOC RSVD RVVD SOLN -DEPTH PRESS DEPTH PCOW DEPTH PCOG TABLE TABLE METH EQUIL 6154 2500 6194 0 1000 0 / RPTSOL 'RESTART=2' 'FIP=2' 'PRES' 'SWAT' / 206 .60 0.15 1.Sw Krw Kro Pc -0.0 64.5e-5 1. ETC -Pref Bw(ref) Cw Visw Viscosibility 2500 1.70 0.18 0.00 0 1.06 0 0.00 0. FACTOR.99 0.0e-6 0. -Rs Pb 0.05 0.02 3.20 0.30 0.0e-6 / DENSITY -OIL WATER GAS 50.34 0 / RSCONST -DATA FOR THIS SECTION ALSO NOT SUPPLIED. WATER FM VOL.35 0.60 0 0.10 0.

SET MAXIMUM NUMBER OF LINEAR ITERATIONS IN THE NEWTON ITERATION TUNING / / LITMAX 2* 100 1* 50 / RPTSCHED 'RESTART=2' RPTRST 'BASIC=4' 'FIP=2' / PREF PHASE 'LIQ' 'LIQ' / / / -.LOCATIONOPEN/ -NAME I J K1 K2 SHUT COMPDAT SAT TAB CONN FACT WELL DIA 207 .WELL GROUP LOCATION BHP -.INTRODUCING THE WELLS -.SUMMARY ==================================== WOPR 'P1' 'P2' / WWPR 'P1' 'P2' / WGPR 'P1' 'P2' / WWCT 'P1' 'P2' / WOPT 'P1' 'P2' / WWPT 'P1' 'P2' / WBHP 'P1' 'P2' / FOPR FOPT FWPR FWPT FLPR FWCT FPR FOE / -RPTONLY -PRESENT DATA IN TABULAR FORM FOR GRAF PURPOSE IN PRINT FILE -REQUEST RUNSUM OUTPUT TO GO TO A SEPARATE RSM FILE RUNSUM SEPARATE SCHEDULE ======================================= -.NAME NAME I J DEPTH WELSPECS 'P1' 'G' 20 11 6157 'P2' 'G' 20 11 6201 / -SPECIFYING COMPLETION DATA -WELL .

41667 0.WELL OPEN/ CNTL OIL WATER GAS LIQUID RES BHP -.PRODUCTION WELL CONTROLS .41667 0.DRAINAGE RATE IS VARIED AT FIXED OIL rATE WCONPROD -1 2 3 4 5 6 7 8 9 -.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.5 / -.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P1' 'P2' 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 22 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 1 11 10 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 10 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 'OPEN' 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 0.41667 0.41667 0.41667 0.1100 1 / -.41667 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' 'X' / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / -.41667 0.41667 0./ WFRICTN -.'P1' 0.41667 0.WPIMULT -.41667 0.41667 0.NAME SHUT MODE RATE RATE RATE RATE RATE 'P1' 'OPEN' 'LRAT' 3* 3000 1* 1* / 208 .41667 0.41667 0.41667 0.41667 0.41667 0.I J K Tlen1 Tlen2 Dirn RangeEnd Diam 22 11 1 1* 1* 'I' 59 1* / / -.41667 0.41667 0.WELL DIAM ROUGH FLOW SCALE 'P1' 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.41667 0.

2 / TSTEP 24*15.NAME SHUT MODE RATE RATE 'P1' 'OPEN' 'LRAT' 'P2' 'OPEN' 'LRAT' / -SPECIFY WELL ECONOMIC LIMIT AS 98 PERCENT WATERCUT WECON -.98 1* 1* 'WELL' 'Y' / -.SPECIFY REPORT / TSTEP 24*15.WELL MIN MIN MAX MAX MAX WKVR -NAME OIL GAS WATER GOR GWR PRCDR -RATE RATE CUT 'P1' 1* 1* 0.2 / TSTEP 24*15.2 / TSTEP 24*15.2 TSTEP 24*15.'P2' / 'OPEN' 'LRAT' 3* 0 1* 1* / -SPECIFY WELL WECON -.WELL MIN -.WELL MIN MIN MAX MAX MAX WKVR END 209 .NAME SHUT MODE 'P1' 'OPEN' 'P2' 'OPEN' / 5 6 OIL WATER RATE RATE 'LRAT' 'LRAT' 7 8 9 GAS LIQUID RES RATE RATE RATE 3* 3000 1* 1* 3* 0 1* 1* END RUN BHP / / -SPECIFY WELL ECONOMIC LIMIT AS 98 PERCENT WATERCUT WECON -.2 / TSTEP 24*15.2 / WCONPROD -.WELL OPEN/ CNTL -.SPECIFY REPORT AT ONE YEAR INTERVAL FOR XX YEARS / TSTEP 24*15.WELL OPEN/ CNTL OIL WATER -.NAME OIL -RATE 'P1' 1* -.1 2 3 4 -.2 / TSTEP 24*15.98 1* 1* 'WELL' AT ONE YEAR INTERVAL FOR XX YEARS / / 6 7 8 GAS LIQUID RES RATE RATE RATE 3* 3000 1* 1* 3* 0 1* 1* 9 BHP / / END RUN 'Y' / WCONPROD -1 2 3 4 5 -.2 ECONOMIC LIMIT AS 98 PERCENT WATERCUT MIN MAX MAX MAX WKVR GAS WATER GOR GWR PRCDR RATE CUT 1* 0.

WELL MIN MIN MAX MAX MAX WKVR END -.SPECIFY REPORT AT ONE YEAR INTERVAL FOR XX YEARS / TSTEP 24*15.2 / TSTEP WKVR PRCDR END RUN 210 .NAME SHUT MODE RATE RATE RATE RATE RATE 'P1' 'OPEN' 'LRAT' 3* 3000 1* 1* / 'P2' 'OPEN' 'LRAT' 3* 0 1* 1* / / -SPECIFY WELL ECONOMIC LIMIT AS 98 PERCENT WATERCUT WECON -.WELL MIN MIN MAX MAX MAX -.SPECIFY REPORT AT ONE YEAR INTERVAL FOR XX YEARS / TSTEP 24*15.2 / TSTEP 24*15.98 1* 1* 'WELL' 'Y' / -.98 1* 1* 'WELL' 'Y' / -.WELL OPEN/ CNTL OIL WATER GAS LIQUID RES BHP -.2 / TSTEP 24*15.2 / WCONPROD -1 2 3 4 5 6 7 8 9 -.2 / TSTEP 24*15.2 / WCONPROD -1 2 3 4 5 6 7 8 -.NAME OIL GAS WATER GOR GWR -RATE RATE CUT 'P1' 1* 1* 0.NAME SHUT MODE RATE RATE RATE RATE RATE 'P1' 'OPEN' 'LRAT' 3* 3000 1* 1* 'P2' 'OPEN' 'LRAT' 3* 0 1* 1* / RUN 9 BHP / / -SPECIFY WELL ECONOMIC LIMIT AS 98 PERCENT WATERCUT WECON -.98 1* 1* 'WELL' 'Y' / -.2 / TSTEP 24*15.-.NAME OIL GAS WATER GOR GWR PRCDR RUN -RATE RATE CUT 'P1' 1* 1* 0.2 / TSTEP 24*15.SPECIFY REPORT AT ONE YEAR INTERVAL FOR XX YEARS / TSTEP 24*15.WELL OPEN/ CNTL OIL WATER GAS LIQUID RES -.NAME OIL GAS WATER GOR GWR PRCDR -RATE RATE CUT 'P1' 1* 1* 0.

2 END / / / / 211 .2 TSTEP 24*15.24*15.2 TSTEP 24*15.2 TSTEP 24*15.

FIELD COUNTY/STATE X DESCRIPTION OF WORK: DWS Vertical Well MOB/DEMOB DRYHOLE COMPLETION TOTAL Development Drilling $0 2.000 57.000 70.000 20.000 280.000 4.000 28. REGULATORY & SAFETY MISCELLANEOUS (CONTINGENCIES) SHORE BASED FACILITIES CONTINGENCY TOTAL INTANGIBLES DAYRATE 35000 8500 8400 5000 3500 1000 4000 500 1000 28500 12500 1000 15000 12500 3000 1200 30000 10000 2750 300 9000 600 1000 1500 2500 1500 DAYS/QTY 20 8 0 0 20 20 20 20 20 20 4 20 20 5 0 15 20 5 0 20 20 25 20 20 20 20 30 DRILLING COMPLETION 700.700 8000 'MD 8000 'TVD 8000 'MD 8000 'TVD DRILL "S" SHAPED PILOT HOLE TO TEST GREEN & RED HORIZONS.000 100.000 4.000 8.000 24.000 32.MARINE CLOSED LOOP COMMUNICATIONS P&A EXPENSE ENVIRONMENT.000 22.000 98.000 16.AIR TRUCKING & HAULING TRANSPORTATION .000 20.800 8 8 8 8 8 0 0 8 3 8 8 8 8 8 8 8 8 212 .000 14.937. COMPLETE W/ 7" PERFORATED LINER DAYWORK DIRECTIONAL SERVICES & SURVEYS MOBILIZATION/DEMOBILIZATION MUD MOTOR INSURANCE RIG MANAGEMENT COMPANY SUPERVISION .800 TOTAL COST 980.600 150.APPENDIX E .000 2.000 10.000 72. FUEL.000 12.862.400 297.000 30.000 8.000 28.000 2.000 552.000 8.000 30.000 100. SET SIDETRACK PLUG.000 8.400 925.300 $3. SCRAPER & REAMERS MUD & CHEMICALS TESTING.000 12.800 28.000 80.000 75.000 114.000 42. DRILL HIGH ANGLE SIDETRACK TO GREEN HORIZON @ 85°.000 45.000 45.000 28.000 2.121.000 6.ONSITE QUARTERING AND CATERING POWER.600 150.000 30.000 20.000 75.000 20. CORING & WIRELINE MUD LOGGING OPEN HOLE LOGGING MWD/LWD CASING CREW & TOOLS RENTAL EQUIPMENT PERFORATING GRAVEL PACK PUMPING CASED HOLE LOGGING TRANSPORTATION . WATER & LUBE RIG SUPPLIES & SERVICES CEMENTING COSTS BITS.000 250.000 20.000 45.000 112.000 55.DWS WELL COST DATA SPREADSHEET Louisiana State University (Petroleum Engineering Department) Drilling Cost Estimate For Downhole Water Sink Well (Vertical) PROSPECT/FIELD NAME LEASE/WELL NO.673.000 9.000 33.000 70.000 50.000 250.000 114.800 20.000 12.000 77.400 225.

000 18.000 4.000 125.500 202. (CONTINGENCIES) TOTAL TANGIBLES TOTAL COSTS DESCRIPTIO DEPTH Unit Cost Length (ft) 30" x 1" wa 450 350 20" x 133 Ib 105 1500 13 3/8" x 68 Ib 45 4500 9 5/8" x 47 Ib 32 8000 7" 26 0 3.000 125.700 TOTAL ESTIMATED COST FOR TWO VERTICAL DWS WELLS $7.000 25.862.937.500 925.400 213 .500 202.5 10 16000 157.500 157.300 157.400 160.000 33.400 1.000 15.900 3.500 256.000 90.500 157.000 65.400 2.725.000 160.000 7.500 372.000 2.188.DRIVE PIPE CONDUCTOR CASING SURFACE CASING INTERMEDIATE CASING & DRILLING LINER PRODUCTION LINER TUBING WELLHEAD EQUIPMENT FLOAT SHOE AND CENTRALIZERS SUBSURFACE EQUIPMENT LINER HANGER FLOWLINE AND HOOKUP MISCELLANEOUS EQUIP.500 256.900 816.

01818 0.00355 214 .00298 0.00300 0.00935 0.70 0.01729 0.80 0.01527 0.01605 0.70 0.01753 0.01575 0.01641 0.70 0.01134 0.00 0.01778 0.90 1.70 0.20 0.00946 0.01541 0.34 5 0.01744 0.00320 0.10 0.00957 0.00535 0.30 0.02092 0.30 0.59 6.01537 0.92 8.01728 0.00928 0.00 0.00306 0.00335 0.20 0.40 0.00416 0.34 5 0.67 0.01815 0.01871 0.80 0.00381 0.01800 0.01671 0.01540 0.67 0.00300 0.00963 0.01523 0.0050 1000 3.40 0.00316 0.00 0.00 0.34 5 0.00919 0.00297 0.30 0.00328 0.01731 0.01228 0.50 0.00324 0.5 1.80 0.00 0.5 1.0200 5000 3.00938 0.60 0.67 0.01008 0.00 0.00968 0.01019 0.90 1.01738 0.00928 0.01586 0.67 0.00 0.00308 0.60 0.01549 0.01752 0.02037 0.34 5 0.10 0.30 0.30 0.00311 0.00931 0.00594 0.01533 0.80 0.90 1.50 0.20 0.00393 0.00299 0.01657 0.80 0.60 0.40 0.01772 0.40 0.5 1.01737 0.01753 0.59 6.01169 0.00 0.01979 0.00990 0.01200 0.02161 0.01777 0.00980 0.90 1.60 0.01678 0.00942 0.01801 e (ft) 0.01005 0.00 f 0.00328 0.01741 0.0200 1000 3.50 0.00 0.40 0.01809 0.00650 0.34 5 0.90 1.5 1.00297 0.00368 0.01842 0.00342 0.92 8.01829 0.00 0.01703 0.01568 0.00514 0.00405 0.00999 0.00312 0.01949 0.01760 0.10 0.01652 Q (stb/d) d -in r w (ppg) r o (ppg) m w (cP) m o (cP) j (spf) q i/Q 9000 3.10 0.92 8.01605 0.02002 0.01789 0.70 0.02010 0.59 6.59 6.10 0.40 0.59 6.01843 0.00349 0.00921 0.01748 0.70 0.01530 0.92 8.5 1.20 0.00943 0.01061 0.92 8.30 0.01740 0.00 f 0.01561 0.01252 0.00923 0.50 0.00926 0.34 5 0.90 1.80 0.00968 0.00950 0.01767 0.00302 0.0200 WC 0.60 0.00930 0.01534 0.01047 0.00313 0.10 0.0050 5000 3.01595 0.APPENDIX F – SAMPLE DATA OF MODIFIED PIPE ROUGHNESS Q (stb/d) d -in r w (ppg) r o (ppg) m w (cP) m o (cP) j (spf) q i/Q 9000 3.00348 0.00354 0.60 0.50 0.59 6.02062 0.01624 0.20 0.50 0.5 1.01735 0.01563 0.01756 0.92 8.67 0.67 0.01778 0.01944 0.01856 0.0050 WC 0.01033 0.20 0.00991 e (ft) 0.01733 0.01555 0.00303 0.

001289 0.00714 0.60 0.001834 0.60 0.5 1.00774 0.Q (stb/d) d -in rw (ppg) ro (ppg) mw (cP) mo (cP) j (spf) qi/Q 9000 3.001229 0.001262 0.01036 0.00715 0.34 5 0.001271 0.00719 0.001449 0.00775 e (ft) 0.00726 0.00792 0.30 0.00710 0.67 0.00 0.10 0.00845 0.0010 1000 3.00747 0.92 8.70 0.001372 0.34 5 0.00789 0.001329 0.50 0.80 0.001768 0.30 0.002866 0.00817 0.0010 5000 3.00741 0.00752 0.59 6.001694 0.0010 WC 0.00783 0.70 0.67 0.90 1.00064 0.001233 0.001343 0.00953 0.5 1.50 0.003098 0.00728 0.20 0.001253 0.20 0.92 8.00803 0.001616 0.001242 0.00703 0.00 0.00730 0.80 0.00918 0.5 1.001207 0.00 f 0.01012 0.001536 215 .001283 0.70 0.001884 0.60 0.59 6.92 8.10 0.67 0.00708 0.001425 0.30 0.001192 0.002339 0.00984 0.001299 0.40 0.001238 0.20 0.00 0.50 0.001247 0.00 0.00764 0.00705 0.90 1.00752 0.80 0.00712 0.59 6.003608 0.00722 0.001315 0.00734 0.001343 0.10 0.40 0.00 0.00128 0.001399 0.00831 0.90 1.00712 0.34 5 0.40 0.

VITA Solomon Ovueferaye Inikori. The doctoral degree will be conferred on him at the August 2002 commencement. Nsukka in September 1988. native of Ohrerhe . Nigeria. He received his bachelor of engineering degree in mechanical engineering from the University of Nigeria. and Madam Regina Okpalefe. Son of chief Johnson E. he received his master of engineering degree in mechanical engineering from the University of Benin. 216 . 1964 in Warri.Agbarho. His job experiences range from drilling and completions engineering to reservoir engineering practice. He enrolled in the doctoral degree program in Petroleum Engineering at the Craft and Hawkins Department of Petroleum Engineering in August 1998. He has eight years of full-time and over three years of part-time industry experiences in petroleum engineering practice with both independent and major oil companies. Nigeria. Delta State. Inikori (Late) native of Eku. was born on March 27. Benin-City. In April 1993.

Sign up to vote on this title
UsefulNot useful