Section 8 Recovery Optimization (1).pdf

TABLE OF CONTENTS SECTION 8

RECOVERY OPTIMIZATION .................................................................................... 8-1 PRODUCTION OPTIMIZATION ................................................................................. 8-1 Concepts ...................................................................................................... 8-1 Approaches .................................................................................................. 8-2 INFILL DRILLING.................................................................................................... 8-4 CASE STUDY NO. 14 ............................................................................................ 8-7 How Infill Drilling Can Improve Recovery ..................................................... 8-7 CASE STUDY NO. 15 ............................................................................................ 8-8 Infill Drilling and Pattern Modification Increase Oil Recovery ....................... 8-8 3. HORIZONTAL INJECTORS/PRODUCERS FOR WATERFLOODING ............................ 8-9 Steady-State Water Injection Rate in a 5-Spot Pattern ................................ 8-9 Approximate Linear Rate for a 5-Spot Pattern ........................................... 8-10 Comparison of Vertical and Horizontal Well Rates ..................................... 8-10 Areal Sweep Efficiency with Opposed Horizontal Injection/Producing Wells.. 8-11 4. HIGH ANGLE AND MULTI-LATERAL WELLS ....................................................... 8-12 Situations Where Gas Injection Can Enhance Oil Recovery in a Mature Waterflood Project .................................................................................. 8-13 CASE STUDY NO. 15 .......................................................................................... 8-15 A Noval Approach To Increasing Oil Recovery From Waterflood .............. 8-15 OIL RECOVERY ENHANCEMENT BEYOND WATERFLOOD....................................... 8-16 Approach No. 1........................................................................................... 8-17 Approach No. 2........................................................................................... 8-18 Approach No. 3........................................................................................... 8-19 CASE STUDY NO. 16 .......................................................................................... 8-20 Time Frame: 1955 - 1971 ........................................................................... 8-20 Time Frame: 1971 – 1979 .......................................................................... 8-22 Time Frame; 1980- 1988 ............................................................................ 8-23 Time Frame: 1988 - 1996 ........................................................................... 8-27 Opportunities for Recovery Enhancements ................................................ 8-27 CASE STUDY NO. 17 .......................................................................................... 8-28 Waterflood Optimization ............................................................................. 8-28 LESSONS LEARNED ............................................................................................ 8-33 Useful Tidbits for a Waterflood Project ....................................................... 8-33 Waterflood Challenges ............................................................................... 8-35 'Common Sense' In Waterflooding ............................................................. 8-35 Why do waterfloods fall below expectations? ............................................. 8-36

© PetroSkills, LLC., 2007. All rights reserved.

i

Waterflooding A to Z Section 8

PROBLEM NO. 28 ............................................................................................... 8-40

ii

© PetroSkills, LLC., 2007. All rights reserved.

Section 8

R ECOVERY O PTIMIZATION

PRODUCTION OPTIMIZATION CONCEPTS Production optimization is an integral part of sustaining production at the plateau rate as long as technically and economically possible. Since oil price is market controlled, the only option available to an operator is to streamline the operation and thereby reduce the operating expense. All disciplines play an important role. 

Production makes sure that all wells operate at their top capacity and all field production, injection, and monitoring data is collected accurately and promptly transmitted to the corporate databases.



Facilities and project management makes sure that: (1) distribution and gathering facilities do not suffer from bottlenecks, (2) there is no production loss due to down time, and (3) facilities are modified to accommodate the changing production profile.



Reservoir makes sure that: (1) additional producers are available to compensate for production loss due to pressure decline, water and gas encroachment, and occasional facilities/plants upsets (2) reservoir injection-production balance is maintained at the desired level by modifying injection water distribution or control of production targets, (3) reservoir pressure is maintained in excess of the saturation pressure, (4) the trapped and stagnant oil is identified and that all parts of the reservoir participate in the recovery process, (5) reservoir energies are not wastefully depleted by production of excessive gas and water from the reservoir, and (6) disposal of produced water is done in HSE compliance.



Geosciences make sure that reservoir characterization remains at its best at all times so that the reservoir geologic models depict the reservoir as accurately as possible with the available geosciences and engineering data. They also keep a vigilant eye on reservoir simulation models to make sure that they have not been force fitted to obtain a satisfactory history match.



Management makes sure that sufficient funds and motivation is made available to: (1) gather the required data (missing as well as monitoring) in a timely manner, (2) test the emerging and evolving new technology without the fear of failure by the engineers, and (3) implement new technology with an open mind to its outcome.

© PetroSkills, LLC., 2007. All rights reserved.

8-1

Waterflooding A to Z Section 8

APPROACHES A large number of optimization approaches are being applied and are being field evaluated. Some of these are described below.

1. Infill Drilling Placement of additional wells within an existing fully developed reservoir is termed infill drilling. The immediate benefit is higher production rate. However, it may be sustained for a while or suffer a decline rate much higher than the previous. Reservoir pay continuity is the single most important factor controlling the success of infill drilling. This concept is further expanded later on.

2. Pattern Rotation (Modification) Pattern rotation is a common practice in waterflooding. By crossflooding, the oil that is not initially contacted by the original well pattern is now contacted and displaced by the new well pattern. The figure below shows the areal sweep expected in an inverted 5-spot pattern in the Mean San Andres waterflood. Upon drilling of additional wells and pattern rotation, the remaining oil is contacted by the floodwater. Hence, the sweep efficiency is greatly improved.

Infilled Pattern

Original Pattern

Swept Region

Swept Region (New)

Unswept Region

Unswept Region Swept Region (Old) Figure 8-1

8-2

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

3. Horizontal & High Angle Drilling It has been the greatest boon to production optimization and recovery enhancement in the last 20 years. Drilling technology has advanced to such a level that it is now possible to drill high angle, horizontal and even inverted wells with tailored trajectories within a very small target window. Such wells are used for: 1. 2. 3. 4.

Developing low permeability and stratified laminated oil and gas reservoirs Targeting thin oil columns which are underlain by water or overlain by gas Producing oil layers sandwiched between water layers Depleting oil pockets trapped or stagnant due to uneven water or gas advance

4. Multi-Lateral Wells Drilling technology is enabling to place multi-lateral wells in a reservoir in all shape and size. These are gaining popularity for the following applications: 1. Developing lenticular and tight heterogeneous reservoirs 2. Correcting for uneven water advance issues 3. Depleting layered reservoirs more uniformly

5. Downhole Oil-Water Separation This technology is in its infancy but offers a great promise in reducing operating cost (production, disposal, scaling and corrosion prevention, etc.) and complying with HSE regulations. Hence, a great deal of research is underway and field trials are being made to develop this technology.

6. Multi-Phase Metering and Pumping This technology will greatly reduce the cost of facilities. It will replace the traditional method of fluid separation (surface separators) and separate measurements of rates. Many vendors are developing this technology, many proto-types are being field tested in various parts of the world, and some are being installed in fields currently under development.

7. Automated Well Data Acquisition & Performance Monitoring SCADA has been around for a while (particularly in offshore and remote fields) and has proven effective in automatically collecting and transmitting real time data to the end user. Many benefits are derived: improved well utilization, minimizing spill volume, better well performance monitoring. With remote actuation devices, unattended well testing is made possible. 'Smart' wells will greatly reduce data acquisition and well intervention costs.

© PetroSkills, LLC., 2007. All rights reserved.

8-3

Waterflooding A to Z Section 8

INFILL DRILLING Drilling of additional wells to reduce well spacing in a fully developed reservoir is one of the many means to increase oil production rate. Infill wells reduce well spacing as shown in the following table. Drainage Area Acre 40 80 100 320 640

½ side of a square

Sq ft 1,742,000 3,405,000 6,960,000 13,930,000 27,878,000

660 ft 933 1320 1867 2640

radius of circle 745 ft 1054 1490 2160 2978

This may result in increased production rate as well as in incremental oil recovery over the flood life.

Oil Rate (STB/D)

Incremental

Qe

Figure 8-2

This may result in increased production rate, acceleration, for a while only but without any incremental oil recovery over the shorter flood life.

Oil Rate (STB/D)

Acceleration

Qe D

Time Figure 8-3

8-4

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

If reservoir pore space (and the fluids therein) in the interwell area is unaware of the presence of the neighboring wells, the decision on infill drilling is a simple one.

Pi Pi

SPACING

SPACING Figure 8-4

Infill wells are suitable for areas: 

With lower-than-expected oil recovery



Higher volumetrics - large well spacing; large pay thickness; high oil saturation; low FVF



Heterogeneous character



Discontinuous pay



Compartmentalized



Low permeability

The need for infill drilling is justified by demonstrating its economic benefit. A. If sufficient performance data is available on the existing wells, the Decline Curve Analysis method may be employed to project oil recovery at the economic water cut. If this estimate is substantially lower than the expectation, infill drilling should be a viable option. B. If a history-matched simulation model study is available for the reservoir, it may be used to conduct 'what if’ studies for options related to infill well program and well spacing. The incremental recovery or the accelerated production profile provides the economic incentive. Many examples of fields where infill drilling has resulted insubstantial improvements in oil recovery are available in the literature. A. An empirical relationship between waterflood ultimate oil recovery and well spacing is shown below. It includes projects in sandstone and limestone reservoirs with a wide range of properties.

© PetroSkills, LLC., 2007. All rights reserved.

8-5

Waterflooding A to Z Section 8

□ Mid Continent Projects x S. Permian Projects

70

Ultimate Recovery - % OOIP

+ N. Permian Projects 60

∆ W. Permian Projects

50

40 5 AC

30

10 AC 20 AC

Well Spacing

20

40 AC

10 0

500

1000

1500

2000

Distance Between Wells - Feet Figure 8-5

B.

Sand Hills Production Behavior due to Infill Drilling 6000

80 Acre

40 Acre 5-Spot

Oil Rate, BBL / D

5000

20 Acre

4000 3000 2000

1000 Primary

Infill

Water Injection

0 0

5

10

15

20

25

30

35

40

45

Cumulative Oil, MM BBLS Figure 8-6

8-6

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

CASE STUDY NO. 14 HOW INFILL DRILLING CAN IMPROVE RECOVERY The figures shown below illustrate production performance and. water cut behavior due to infill drilling the Mean San Andres Unit located in west Texas. Volumetric sweep efficiency was increased from 59% with 40 acre patterns to 85% with 10 acre patterns. Spacing (Acres) 40 20 10

20000 Rate, BOPD

Volumetric Sweep Efficiency, % 69 73 85

Recovery (MBO) 97 119 138

15000 10000 20 Acre

40 Acre

5000

10 Acre

0 40

60

80 100 120 Cumulative Oil, MBO

140

100.0 20 Acre

WOR

Economic Limit = 50

10.0

10 Acre

40 Acre

40 Acre EUR = 97 MBO

1.0

0.1 40

20 Acre EUR = 119 MBO

10 Acre EUR = 138 MBO

60 80 100 120 Cumulative Oil, Millions of Barrels

140

Production Performance Due to Infill Drilling Figure 8-7

Flood performance was also improved by rotating flood patterns. Crossflooding of patterns by new injectors contacted unswept oil. 

Goodwin, J.M. “Infill Drilling and Pattern Modification in the Means San Andres Unit”

© PetroSkills, LLC., 2007. All rights reserved.

8-7

Waterflooding A to Z Section 8

CASE STUDY NO. 15

10

20

30

40

40-Acre

20-Acre 20-Acre Well Spacing

Unflooded on 20 Acre Well Spacing

SOURCE: SOUTH WASSON CLEARFORK

PATTERN MODIFICATION

INFILL DRILLING

INFILL DRILLING AND PATTERN MODIFICATION INCREASE OIL RECOVERY

Waterflood Recovery Millions of Barrels Oil 8-8

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

3. HORIZONTAL INJECTORS/PRODUCERS FOR WATERFLOODING Compared to 5-Spot pattern with vertical wells, horizontal wells can: 1. Increase injectivity by as much as a factor of 10, depending upon well spacing and formation thickness. 2. Areal sweep efficiency can be increased by 25 to 40%. 3. The pressure gradient in the bulk of the reservoir can average several times greater. Because of the faster flooding rate, higher sweep efficiency, and higher bottomhole flowing pressure, project life and economics are more favorable.

STEADY-STATE WATER INJECTION RATE IN A 5-SPOT PATTERN

qs =

1.54 kh p 

  log10 

 W - 0.420  rw 

L

w

Where: qs = rate of water injection, bbls/day k = permeability, darcies h = sand thickness, ft p = pressure difference between injection and producing wells, psi  = viscosity, cp W = distance between like wells, ft rw = radius of injection well, ft

© PetroSkills, LLC., 2007. All rights reserved.

8-9

Waterflooding A to Z Section 8

APPROXIMATE LINEAR RATE FOR A 5-SPOT PATTERN

qL(S) =

4.52 khp 

L

Injection Well

W

TW

h

W

L

Figure 8-8

COMPARISON OF VERTICAL AND HORIZONTAL WELL RATES   W - 0.420  qL ( S ) = 2.93 q s  log rw  

8-10

Spacing (Acres)

W (Feet)

qL/qS

10,000 20,000 40,000 80,000 160,000 320,000

660.0 935.0 1320.0 1859.0 2640.0 3734.0

8.4 8.9 9.3 9.7 10.2 10.6

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

AREAL SWEEP EFFICIENCY WITH OPPOSED HORIZONTAL INJECTION/PRODUCING WELLS Areal sweep efficiency at water breakthrough in a direct-drive line pattern is given by the following equation:

EA = 1 -

0.441 h 2L

This equation shows that sweep efficiency in this case decreases with increasing formation thickness and decreasing well spacing. The figure below is a comparison of the linear and 5-spot well pattern.

Sweep Efficiency, %

100

90

Horizontal wells spacing, acres:

5

20

80

320

80

70

60 10

Vertical wells

20

30

50

100

200

300

500

1,000

Formation thickness, ft Figure 8-9

© PetroSkills, LLC., 2007. All rights reserved.

8-11

Waterflooding A to Z Section 8

4. HIGH ANGLE AND MULTI-LATERAL WELLS The schematic below shows the effect of a High Angle Slanted Well on enhancing the sweep efficiency of waterflood. DEPTH 5,500’

5,600’ Pay Pay Pay

5,700’

1,866’

1,866’

Figure 8-10

The schematic below shows the effect of Stacked Multi-Lateral Wells on enhancing the sweep efficiency of waterflood. DEPTH 5,500’

5,600’ Pay Pay Pay

5,700’

1,866’

1,866’

Figure 8-11

8-12

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

SITUATIONS WHERE GAS INJECTION CAN ENHANCE OIL RECOVERY IN A MATURE WATERFLOOD PROJECT Injection of gas alone or along with water (WAG - water alternating gas) can enhance oil recovery in specific situations. Some of these are shown below:

1. Attic Oil

Oil in the local highs is trapped due to oil-water density contrast.

Oil from the highs may be displaced by gas due to gravitational segregation.

Figure 8-12

2. By-Passed Oil Due to Water Under-Running G W

Subsequent gas injection may contact and mobilize part of the trapped oil.

Oil is left trapped towards the top due to gravitational under-running of water. Figure 8-13

© PetroSkills, LLC., 2007. All rights reserved.

8-13

Waterflooding A to Z Section 8

3. Trapped-Oil Due to Reservoir Heterogeneity

Oil is trapped in some lenses due to water under-running and shielding

Some of the trapped oil is contacted and mobilized by injection gas Figure 8-14

4. Slow Moving Oil Due to High Perm Contrast

Subsequent gas injection can displace oil from the low perm layer due to density difference and higher gas injectivity. It is particularly so if selective injection/production is practiced.

Oil is left trapped in the lower perm layer due to increasing back pressure in the wellbore. It also results from increasing water cut due to gravitational under-running of water in the high perm layer.

Figure 8-15

5. By-Passed Oil Under Shale Lenses

Water moves rapidly on top of shale lenses, thereby shielding some trapped oil under the shale lenses.

8-14

Gas will displace trapped oil due to gas-oil density contrast.

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

CASE STUDY NO. 15 A NOVAL APPROACH TO INCREASING OIL RECOVERY FROM WATERFLOOD Based on its laboratory tests and field projects, BP has observed that injection of fresher water (of lower salinity) in some reservoirs containing more saline formation water, results in incremental oil recovery. Laboratory Results: Reservoir condition core tests conducted in some shaly sandstone cores have exhibited the following behavior.

0.8 0.7

• Sor reduced from 40% to 12%

0.6 0.5 0.4 0.3

4,000 ppm injected brine

0.2 80,000 ppm injected brine

0.1

PV Injected

0 0

5

10

15

Figure 8-16

Ph change is seen in produced brine as well. This process appears to require:   

Presence of clay (Kaolinite) or some other mineral (not chlorite) Specific crude oil Connate Water Salinity > 15,000 ppm

Field experience in Milne Point Field, Alaska tends to suggest that areas flooded with fresher water resulted in higher oil recovery compared to other areas where seawater of higher salinity had been injected. No definitive explanation is yet offered, but other field trials are being considered at the present time in this and other reservoirs. Additional laboratory investigations are being conducted to better understand the process.

© PetroSkills, LLC., 2007. All rights reserved.

8-15

Waterflooding A to Z Section 8

OIL RECOVERY ENHANCEMENT BEYOND WATERFLOOD 

On the average, about 50% of OOIP remains in the reservoir after a waterflood project reaches its economic limit.



Even with the best of the flood management, there are various natural reasons for the above. A. Inefficient Displacement of Oil 

Retention of residual oil due to capillary forces



Relative Permeability Effects



Viscosity and Mobility Contrast

B. Incomplete Volumetric Contact 

Gravitational Segregation



Reservoir Characterization a. Layering (zonation) with Permeability Contrast b. Continuity of Layers c. Directional Permeability d. Fractures & Joints



A large number of Enhanced Oil Recovery (EOR) methods have been investigated in the laboratory, field tested in pilot tests, and implemented in the field. Since many of these are much more expensive than a waterflood, EOR field applications run hot and cold, depending upon the price of the oil.



8-16

These EOR methods aim to recover incremental oil by one or more of the following approaches:

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

APPROACH NO. 1 Minimize or eliminate the retentive capillary force. It is easy to appreciate this approach by looking at the capillary equation for the oil-water system.

PC =

2  Cos  R

As   PC 

Potential methods for achieving this fall under: 

Injecting water containing chemical additives such as surfactants and Alkali



Injecting solvents such as LPG and Alcohols



Injecting gases that achieve miscibility with oil through mass exchange --CO2, N2, Natural Gas

Oil recovery is dependent upon the reduction in oil saturation (original minus the remaining) in the reservoir pore space at any point in time. When oil-containing pore space is processed ideally (sweep = 100%) by water encroachment (natural or man made), remaining oil saturation at the flood-out reaches the minimum value called the residual oil saturation to waterflood. Capillary Number, an approximate measure of the ratio of Viscous to Capillary forces, shows the effect of reservoir forces on the residual oil saturation.

N=

p q L = PC PC kA

The figure below shows that:

RESIDUAL OIL SATURATION, %PV

40

30

BEREA SANDSTONE SAMPLE 1

BEREA SANDSTONE SAMPLE 2 CORE PLUGS FROM A ROCKY MOUNTAIN RESERVOIR (EXTRACTED)

20

10

0 10-8

10-7

10-6

10-5

10-4

10-3

10-2

10-1

CAPILLARY NUMBER

Figure 8-17

© PetroSkills, LLC., 2007. All rights reserved.

8-17

Waterflooding A to Z Section 8

1. Residual oil saturation in normal waterfloods cannot change much since oil viscosity and oil-water interfacial tension are fixed and the achievable rate variation is rather small. 2. However, the residual oil saturation can be reduced significantly by making a drastic increase (by several orders of magnitude) in the capillary number. Since rate change in a reservoir is not very practical and viscosity change is realized only in high viscosity crude oils, the FT reduction is the only practical means to achieve this. Surfactant and Micellar flooding are examples of such processes. 3. Residual oil saturation can theoretically be reduced to zero if IFT is eliminated completely. Many approaches have been devised, considered, and investigated. Many have been implemented in the field with varying degrees of technical and economic success. Miscible flooding is example of such a process. 

First contact miscibility is achieved with C3, C4, LPG solvents, and many Alcohols.



Multi-contact miscibility is achieved with lighter hydrocarbons, CO2, N2, and flue gases.

APPROACH NO. 2 Lower Mobility Ratio between Oil and Water. It is easy to appreciate this approach by looking at the definition of Mobility Ratio.

M=

w Kw o = x o w Ko

Potential methods for achieving this fall under:

8-18



Injecting water containing viscosity builders such as Polymers or Gels



Injecting heat in the reservoir, thereby lowering oil viscosity



Generate in-situ heat in the reservoir by combustion

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

APPROACH NO. 3 Increase contact between Water and Oil. It is theoretically possible by (a) improving injection and production profiles, and (b) lowering permeability of the high capacity layers, and (c) increasing permeability of the low capacity layers. Potential methods for achieving this fall under: 

Inject water containing special polymers with characteristics that lower permeability to water without impairing permeability to oil.



Inject slugs of water alternately with the slugs of gas (WAG process).

General Remarks Gas injection projects (first contact miscible and multi-contact miscible) have been widely used in the past when market for the gas was limited. Such projects are no longer considered attractive because of the current market value of natural gas. Polymer floods have lost popularity because the polymers are expensive (derived from crude oil) and because they suffer from high adsorption losses to the rock surfaces while propagating through the reservoir. Steam floods or single-well steam stimulation treatments are still popular as they target reservoirs containing high viscosity oils at shallow depths. Carbon dioxide floods are quite attractive as they are targeting the reservoirs having proven success with waterfloods.

© PetroSkills, LLC., 2007. All rights reserved.

8-19

Waterflooding A to Z Section 8

CASE STUDY NO. 16 This San Andres reservoir in Permian Basin, Texas is typical of many limestone reservoirs where geology is complex and rock properties vary both areally and vertically.

0

2000

4000 ft

Structure map TIME FRAME: 1955 - 1971 It was initially developed and produced under natural depletion with wells on 40-Acre spacing.

WELLS 40 PRODUCERS (40 ACRE SPACING) 13 WATER INJECTORS 21 DRY HOLES

WATERFLOOD UNIT

8-20

0

2000

4000 ft

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

Rate-Time Performance 10,000

8,000 WATERFLOOD INITIATED 6,000

A

B

4,000

DECLINE PRIMARY PRODUCTION

2,000

0 1940

1950

1960

1970

1980

1990

2000

Observations: 1. Production rate increases rapidly as the wells are completed and perforated selectively as cased-hole producers. 2. The peak rate is maintained for 3-4 years only; rapid rate and pressure decline begin. 3. Very little water production but producing gas-oil ratio increases rapidly. Course of Action: 1. Unitize the field for a pressure maintenance project. 2. Begin pressure maintenance by peripheral water injection

© PetroSkills, LLC., 2007. All rights reserved.

8-21

Waterflooding A to Z Section 8

TIME FRAME: 1971 – 1979 Rate-Time Performance: 10,000

20 ACRE 5 SPOT PILOT

8,000

WATERFLOOD INITIATED 6,000

A

4,000

C B

SECONDARY RECOVERY DECLINE

PRIMARY PRODUCTION 2,000

0 1940

1950

1960

1970

1980

1990

2000

Observations: 1. Oil rate increases in some wells and not the others. 2. Pressure increase in the flank wells-but the crestal wells show no response. 3. Water breakthrough does not conform to the reservoir zonation scheme based on the current geologic model. Zonal production is erratic. Course of Action: 1. In-field pattern injection 2. Drill infill wells at closer well spacing 3. Conduct a pilot to investigate flood-ability 4. Re-assess reservoir volume and zonation. 5. Establish flood-ability of tighter intervals

8-22

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

TIME FRAME; 1980- 1988 1. Conduct an in-field waterflood pilot on 20 Acre well spacing.

WELLS

20 ACRE 5 SPOT PILOT

44 PRODUCERS (40 ACRE SPACING) 61 WATER INJECTORS (20 ACRE SPACING) 12 DRY HOLES

WATERFLOOD UNIT

0

2000

4000 ft

Rapid response – higher oil rate, lower GOR, and higher inter-well pressure 2. Conduct a geological zonation study. A new correlation based on a depositional (tidal bar and channel) model shows that reservoir continuity is totally different from the original description of flow units and flow barriers. A. Original Interpretation: Alternating layers of Oolite beds and flow barriers and localized thinning and thickening of both. B. Revised Interpretation: Oolite beds are localized deposits separated from each other by continuous flow barriers.

© PetroSkills, LLC., 2007. All rights reserved.

8-23

Waterflooding A to Z Section 8

1

3

5

7 0 25 50 75 FT 0

1320 FT

“CONVENTIONAL” CORRELATION ANHYDRITE MARKER

1

3

5

7 0 25 50 75 FT

NEW CORRELATION BASED ON CONCEPTUAL MODEL AND CORES ANHYDRITE MARKER

1

2

3

4

5

6

7

8

TIDAL CHANNEL

0 25 50 75 FT 0

OOLITE RESERVOIR

1320 FT

FLOW BARRIERS

3. Conduct petrophysical studies to establish productivity criteria. The new porosity cut-off value from the revised O-K relationship is lower (4% for K = 0.1 md) than that based on the original relationship (10% for K= 1.0 md). The revised cut-off values are confirmed by selective production testing of low permeability intervals in newer wells and by conducting core flood tests in laboratory.

8-24

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

PERMEABILITY - MILLLIDARCIES

1000

III CRYSTAL SIZE > 0.06 mm II/III CRYSTAL SIZE 0.02 – 0.05 mm II CRYSTAL SIZE < 0.02 mm

100

10

1.0

0.1

0.01 0

5

10

15

20

25

30

4. Acquire newer logs to get-better interpretation of rock properties. Comparison of 1956 Gamma Ray/Resistivity logs with 1988 Gamma Ray/Neutron Porosity/Resistivity logs on two wells about 30-ft apart shows that: A. Either the net pay thickness varies rapidly. B. Or the data acquired earlier and its interpretation was not conclusive.

© PetroSkills, LLC., 2007. All rights reserved.

8-25

Waterflooding A to Z Section 8

1988 NEW WELL

ORIGINAL 1956 WELL

GAMMA RESISTIVITY RAY Ω-m 0

NET PAY

GAMMA POROSITY RESISTIVITY RAY Ω-m %

FEET 0

80 1 10 100

100 25

0 1 10 100

100

Sw %

0

TOP SAN ANDRES PAY

1st PAY MAIN PAY

50

WET? OR TIGHT? 100

PERFORATIONS CORE

Course of Action: 1. Revise estimate of OOIP 2. Perforate all zones in both injectors and producers 3. Investigate the technical/economic benefit of closer well spacing 4. Consider the technical merit of applicable EOR processes

8-26

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

TIME FRAME: 1988 - 1996 Expand the 20 Acre/well pilot program to cover the entire field. Also, make sure that all partial penetration wells are deepened and all zones in all wells are perforated. 10,000 20 ACRE INFILL DRILLING 20 ACRE, 5 SPOT INJECTION 20 ACRE 5 SPOT PILOT

8,000

WATERFLOOD INITIATED 6,000

A

DECLINE C

D

B

4,000

SECONDARY RECOVERY DECLINE

PRIMARY PRODUCTION 2,000 S/P = 1.74 0 1940

1950

1960

1970

1980

1990

2000

Flood pattern is modified by well conversions.

OPPORTUNITIES FOR RECOVERY ENHANCEMENTS The relative contributions of methods employed for recovery enhancement in this waterflood are shown in the pie-chart below.

INJECTION PROFILE CONTROL (7%)

MODIFY FLOOD PATTERN (36%)

© PetroSkills, LLC., 2007. All rights reserved.

WORKOVERS (3%)

BYPASSED PAYS/ RECOMPLETIONS (11%)

INFILL WELLS (43%)

8-27

Waterflooding A to Z Section 8

CASE STUDY NO. 17 WATERFLOOD OPTIMIZATION Prudhoe Bay, located on the north slope of Alaska, is the largest oil field in North America. The field produces from the Ivishak formation, primarily made up of braided river deposits of sandstone and conglomerate. The reservoir contains saturated 27° API oil underlying a large gas cap. Permeability varies from 10 to 2,000 md. The reservoir is being produced under a variety of recovery methods. 

Gas cap injection for vaporization of condensate and residual oil residing in the original and expanded gas cap areas in the main part of the reservoir. Gravity drainage provides supplemental recovery.



Peripheral and in-field pattern water injection and a miscible gas injection (WAG) scheme are used in-down-dip areas where no gas cap existed. Zone /Lith

4 NWFB Gravity Drainage GDWF

WF/EOR

Gas Cap

0 Gamma 100

Permeabili Porosity ty

Fine grained

Coarse 3 Conglomerate

2 Medium Conglomerate 1

Deltaic

The recovery optimization program deals with the waterflood in the Northwest Fault Block (NWFB) that originally contained 1.75 Billion Barrels of oil. It is structurally a complex area that is bounded by faults on three sides. The eastern edge is under the influence of gas cap where gravity drainage potential exists. This area was initially produced under primary depletion (solution gas drive, minor water influx, and some gravity drainage) from 1977 through 1984. A water flood with inverted 9-spot pattern on 320-Acre spacing was initiated in 1984. It was later modified into a line drive by pattern size reduction, pattern reorientation, and infill well drilling. The flood is now considered mature (the lower high-quality pay zones 2 & 3 are approaching flood-out while the low-quality zone 4 above contains majority of the un-swept oil) with substantial potential for recovery optimization and subsequent EOR flooding.



Source: SPE 63152 - A case Study of the Northwest Fault Block Area of Prudhoe Bay, Alaska, Using Streamline and Traditional Waterflood Analysis

8-28

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

Problem Performance prediction using decline cure analysis, shown in figure, indicates that estimated ultimate recovery of 40-45% OOIP would fall short of the EUR of 55-60% (based upon theoretical analysis and projections of recovery from other flank areas).

Oil Rate MSTB/Day

1000

NWFB Area Oil Production Production Decline after Injection Management

100

Decline Rate 22% per Year Waterflood 10 Apr 03

Apr 01

Apr 99

Apr 97

Apr 95

Apr 93

Apr 91

Apr 89

Apr 87

Apr 85

Apr 83

Apr 81

Apr 79

Apr 77

Time Decline curve shows a very steep decline following the Hydraulic fracturing program which occurred in 1991 and 1992. Recovery optimization process deals with the following questions: 

What are the possible reasons for this shortfall?



How would you assess the real cause(s)?



What methods would you recommend for recovery improvement?

Methods Utilized by BP-Amoco Re-Interpretation of Pressure Profile Surveys RFT surveys in 1994 on several producers show that Zone 4 pressures are lower than Zone 3 pressures indicating lesser cross flow (lower vertical permeability) than earlier estimated.

© PetroSkills, LLC., 2007. All rights reserved.

8-29

Waterflooding A to Z Section 8

Pressure 3350

3400

3450

3500

3550

3600

3650

Depth (SS)

8600

3700

T24

Gradient = 0.5 psi/ft

8700

T23 8800 T22 8900 HOT 9000

Subsequent detailed core description analysis and holistic regional stratigraphy indicated that Zone 4 consists of a series of fine and coarse-grained sands with occasional tight mudstone layers. Analysis of Injection Wells Performance 1. Flowmeter surveys - these had previously been used to estimate the injection water distribution by zone. There was concern that these surveys only indicate near-wellbore distribution and can be very misleading if appreciable cross-flow between the zones occurs in the reservoir. The pressure fall-off surveys had already indicated that Kh value was significantly higher from the well test than the core data. This had been attributed to thermal fracturing that had occurred due to the injection of 80F seawater into the 200°F reservoir.

Σ (Pressure & Time)

2. Injection well performance was critically evaluated utilizing Hall Plot annotated with well history information. One such plot is shown below.

Avg. I I, STB/Day/psi

Phase

Re-Perforation

22.6

22.6

35.0

13.1

1

2

3

4

Cum. Water Injection

8-30

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

The plot clearly exhibits the following: 

The initial phase (Phase 1) was short lived.



The rapid increase observed in Phase 2 is due to thermal fracturing in the near wellbore vicinity.



Re-perforation to improve injectivity had no effect indicating that all zones were taken water according to their capacity.



The gradual transition from Phase 2 to Phase 3 is due to continued thermal fracturing as the colder seawater enters deeper into the reservoir.

The above behavior was interpreted as a fracture system that had connected Zone 4 with the higher permeability Zones 3 and 2. Hence, it was concluded that water injected into Zone 4 perforations moved into Zones 3 and 2 due to the combined effect of fractures and gravitational segregation. Streamline Flood Front Tracking Simulation Streamline simulations have been very successful in tracking fluid movement in the reservoir. They have grown from simple 2-D, 2-phase models to today's 3-D models accounting for gravity. The classical (simplified and idealized) methods are of no use as they treat the reservoir as a single layer system. The finite difference simulators provide meaningful results but they are slower, more cumbersome, and expensive. The streamline models provided detailed history match of oil and water production over the 23years project life for each well. Also, they directly quantify volumes between injectors and producers to provide a 3-D dynamic prediction of injection support. The results showed that the near-wellbore fracturing was dominating injection well conformance; seawater was transferring from Zone 4 into Zones 2 and 3 in many wells. Also, some injected water was leaving the waterflood area by leaking off into the gas-cap area.

Optimization Approaches Understanding of the vertical conformance issues lead to making the following changes in the waterflood management. Drilling of Injection Wells Many new injection wells were drilled (or side tracked) as high angle or near horizontal wells to place injection into the unswept upper Zone 4. Injection was conducted at high rates and at several injection lengths along the wellbore in hopes to contain the thermal fracture system in Zone 4 alone. Many of these wells encountered 25 to 75 feet of oil column remaining in locations just 500 feet away from the original wells.

© PetroSkills, LLC., 2007. All rights reserved.

8-31

Waterflooding A to Z Section 8

Injection-Production Targets with Dynamic Allocations With information provided by the streamline simulation, the injection-production allocation became routine and the practice of continually maintaining voidage balance in every pattern was introduced. Also, flux out of the waterflood area was eliminated. Reduced Water Cycling Zones in all injectors that identified inefficient water injection were worked over for water shut-off treatments. The streamline simulation was used to reduce uncertainty so that workovers could be performed without impairing long-term effectiveness. Fluid Flux Management Streamline simulation identifies dead spots as well as watered-out zones. Changing injection intervals and rates yielded immediate benefits and stopped flux out of the waterflood area. The results of the optimization program are shown in the performance plot below. The decline rate was cut from 22 to 11% per year. In addition, the seawater injection rate was decreased by 50% and the resulting water production rate decreased appreciably.

Oil Rate MSTB/Day

1000

NWFB Area Oil Production Production Decline after Injection Management

100

Decline Rate 11% per Year

10 Apr 03

Apr 01

Apr 99

Apr 97

Apr 95

Apr 93

Apr 91

Apr 89

Apr 87

Apr 85

Apr 83

Apr 81

Apr 79

Apr 77

Time

8-32

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

LESSONS LEARNED USEFUL TIDBITS FOR A WATERFLOOD PROJECT Design Aspects 1. WF should be part of the initial development plan, especially offshore. 2. Reservoir with high porosity, permeability, high initial oil saturation at WF initiation, and relatively uniform distribution make the best WF candidates. 3. A successful waterflood requires reservoir continuity from injectors to producers in almost all layers. Additionally, the higher the interlayer cross-flow the higher the recovery. 4. Response to waterflood begins when 2/3 of fill-up is achieved. 5. Peak oil rate commences at the time of fill-up. 6. Artificial Lift should be considered early in development planning. 7. Total water requirement for most WF projects is around 1.5 to 2 times the reservoir pore volume. 8. A rule of thumb suggests water injection rate of 5 to 10 B/D/Ft for infield water Injectors, and 10 to 20 B/D/Ft for peripheral injectors. 9. Water injection rate should exceed the reservoir withdrawal rate by at least 10% (preferably 20%) to compensate for the unknown loss out of the reservoir (outside the pattern or into the aquifer). 10. Average reservoir pressure during a WF is a judicious balance between various considerations, such as well injectivity, well productivity, and reservoir oil/water flow behavior. 11. A 5-spot pattern generates higher pressure-gradient from injector to producer than a peripheral pattern, which helps to better sweep the low permeability zones. 12. Quality of injection water is a major issue. The specification should insure a trouble-free project.

Recovery Aspects 13. Displacement efficiency ranges between 50 to 80%. The influencing factors are: water/oil mobility ratio, injection volume, and production/injection strategy and controls. 14. For a vertical well development plan, areal sweep efficiency ranges between 60 to 90%. The influencing factors are: water/oil mobility ratio, well patterns, reservoir anisotropy, and injection/production strategy and controls. Higher efficiency may be realized in a horizontal well and multi-lateral well development plan.

© PetroSkills, LLC., 2007. All rights reserved.

8-33

Waterflooding A to Z Section 8

15. For a vertical well development plan, vertical sweep efficiency ranges between 40 to 80%. The influencing factors are: water/oil mobility ratio, reservoir stratification, reservoir anisotropy, well completion, and production/injection strategy. Higher efficiency may be realized in a horizontal well or a multi-lateral well development plan. 16. Incremental oil recovery for 30° or higher API gravity oils is about the same as the primary oil recovery. 17. Incremental oil recovery for 25° or lower API gravity oil is about one-half of the primary oil recovery. 18. Residual oil saturation in a WF is roughly 1/3 of the initial oil saturation. 19. Oil recovery from a WF in a limestone reservoir is around 35%. 20. Oil recovery from a WF in a sandstone reservoir is around 50%.

Operational Aspects 21. Casing/tubing integrity should be assessed before injection begins. 22. Tubing size should be adequate to allow injection and production at the desired rates. Re-sizing is often required after water breakthrough and increasing water-cut production. 23. Poor cement bond may result in water loss to the adjoining lower pressure reservoirs or aquifers. Not only loss of pressure in the reservoir, it creates lots of problems during history matching of a reservoir simulation model. 24. Facilities constraints should be minimized. Facilities modifications, upgrades, and expansions must be done in a timely manner. 25. Pressure loss from the pump discharge to the wellhead is the higher of the 100 psi or 10% of the discharge pressure. 26. Producing wells should be kept pumped down as much as possible to minimize backpressure on the wellface. 27. Incompatibility of the injected water with formation rock, water, oil, and metallurgy could result in major production problems (scaling, emulsions, corrosion, and asphaltenes). Achieving practical compatibility requires judicious chemical treatment and continuous quality control. 28. Injection at a pressure below and above formation parting (fracturing) pressure is applicable under the right conditions. 29. Intentional fracturing is often considered as this enables significant cost reduction due to: lower well count, lesser facilities, and less rigid water quality specifications. 30. Addition of nitrate to seawater to reduce corrosion due to SRB activities has been found to be an effective solution.

8-34

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

WATERFLOOD CHALLENGES 1. Forecasting water cut for projects where water cut is lower than 30 to 35%. 2. Establishing water quality specifications. 3. Water shut-off control techniques using chemicals are often reported to be successful by service providers. However, operators and producers don't have access to an impartial and independent information source (based on field applications). 4. Improving diagnostic tools for cost-effective identification of produced water source in horizontal wells. 5. Improving remedial techniques for cost-effective profile control in horizontal wells. 6. RPM polymers for high temperature reservoirs are needed. 7. The technology/practice of fracture growth containment and optimization of fracture size is almost like a black magic. More geo-mechanical understanding of the process and better simulation models are needed. 8. Zero discharge issues - targets and cost consideration.

'COMMON SENSE' IN WATERFLOODING 

Process the entire oil column. Water displaces oil mainly by physically pushing it towards the producers; maximize contact in all oil-bearing pore spaces.



Make sure that the injected water stays within the reservoir. If water leaves the reservoir because of fractures, natural or induced, it is a lost cause. If water is injected in the aquifer, consider some loss to aquifer.



Promote water distribution to contact as much reservoir as possible. If severe stratification with large flow capacity contrasts, consider sequential flooding or SIP/SPP. Well placements to provide the maximum areal contact.



Minimize water circulation via the flooded zones. All water cycling through the reservoir (without physically pushing oil), costs money to inject, produce, process, and dispose water without any return.

© PetroSkills, LLC., 2007. All rights reserved.

8-35

Waterflooding A to Z Section 8



Produce oil wherever present; minimize movement within the reservoir. Displacement energy is wasted without any production.



Well tubular and gathering facilities must not impose flow restrictions within the reservoir/wellbore/flowline network.



Design well, gathering, injection and production processing- facilities with an eye towards the life cycle of the flood project.



Minimize physical and chemical interactions between the injection water and the reservoir rock, oil, and water.



Collapse gas saturation in the reservoir as fast as possible. Never allow gas evolution during waterflood. ■

WHY DO WATERFLOODS FALL BELOW EXPECTATIONS? Waterfloods often do not meet rate-recovery expectations for many reasons. These reasons and their causes are described below: 

Delayed or Inadequate Oil response 1. The fill-up time is under-estimated due to inaccurate material balancing 2. The initial water saturation is higher than estimated 3. Inappropriate well locations



Oil target rate falls below expectation 1. Lack of inter-well connectivity/continuity/floodability 2. Reduced well productivity due to chemical/mechanical problems 3. Unrealized voidage balancing for the oil-bearing layers due to water short-circuiting through thief zones 4. Higher water saturation at the start of flood



Water injection falls below design rate 1. Wellbore plugging due to solids deposition 2. Injectivity impairment around the wellbore due to positive skin resulting from scales/emulsions/corrosion products/oil 3. Inter-well connectivity/continuity/floodability 4. Reservoir (kh) degradation away from the wellbore

8-36

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization



Water breakthrough is earlier than expected 1. Reservoir stratification with large kh contrast 2. Preferential water movement through layers with high initial water and gas saturation (relative permeability effect) 3. Presence of natural or induced fractures 4. Directional permeability in relation to well patterns



Water quantity and quality issues 1. More than anticipated water volume requirement resulting from out-of-zone losses, channeling, coning, and cusping 2. High operational costs due to water handling and disposal 3. Extended project life 4. Impairment of injectivity and productivity due to scales, emulsions, asphaltenes, suspended solids, and bacteria



Facility Issues 1. Under-sizing or over-sizing of facilities 2. Excessive downtime due to inadequate maintenance 3. Bottlenecks 4. Inappropriate designs

A review of historical waterflood projects that did not meet recovery expectations or failed outright shows that: 1. 45% suffer from poor sweep efficiency 

Presence of fractures and super-permeability streaks



High degree of reservoir anisotropy



Large permeability contrast between layers



High Kv/Kh ratio in thick reservoirs (water under-running)



Escape of oil into gas cap or aquifer



Viscous fingering due to unfavorable mobility ratio



Non-capture of oil due to reservoir drift

© PetroSkills, LLC., 2007. All rights reserved.

8-37

Waterflooding A to Z Section 8

2. 45% face higher than expected expenses 

Production problems due to scales, emulsion, corrosion



Injection problems due to plugging/channeling



Extensive well workovers



Equipment failure due to corrosion



Under-design of injection/production facilities

3. 10% fail due to poor initial design

8-38



Insufficient oil saturation for oil banking



Insufficient water injection due to OOZ loss



Reservoir Compartmentalization



High oil viscosity



Water incompatibility

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

© PetroSkills, LLC., 2007. All rights reserved.

8-39

Waterflooding A to Z Section 8

PROBLEM NO. 28 After reviewing the field in Problem 22, some development schemes for waterflooding were most likely developed. (If not, then develop some now.) With a simplified representation of that reservoir as shown below, place the injection and production wells from those development schemes in the simulation model provided. Now lets find out which schemes a) produce the most oil from a waterflood, b) provide the most optimum recovery on a per well basis (roughly related to Capital money invested,) and c) provide the most optimum recovery on a per barrel of water injected basis (roughly related to Operating Expenses. CONSTRAINTS CAN NOT - change the reservoir, grid, or rock and fluid properties No well additions or changes before Dec 1, 2001 CAN - Add production and injection wells and vary producing rules and completion layers, but with No more than 10 injectors and No more than 10 producers.

RECOVERY - will be compared for optimization after 15 years - Dec 31, 2015 WELLS Wells are completed uniformly through all 10 layers Production Constraints are Max Production Rate of 10000 BPD (all fluids - reservoir Bbls) Limiting BHP of 200 PSI Injection Constraints are Max Injection Rate of 10000 BPD (water - STB) (Bumps up to 20000 BPD when 26 is drilled) Max BHP of 9000 PSI

8-40

© PetroSkills, LLC., 2007. All rights reserved.

Waterflooding A to Z Recovery Optimization

GRID LAYOUT Areal View (7 layers)

J=

I = 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

29

30 26

27

NOT TO SCALE

© PetroSkills, LLC., 2007. All rights reserved.

8-41

Waterflooding A to Z Section 8

Direction

Grid

Blocks

Dimensions

Total

I

20

220'

4400'

North J

40

280'

11200'

Down K

10

9'

90'

East

FAULT BLOCK Boundaries

I1

I2

J1

J2

Left Fault Block

1

7

20

40

N (1)

1

10

1

16

C (2)

1

12

17

19

S (3)

8

16

20

40

N (1)

11

20

1

16

C (2)

13

20

17

19

S (3)

17

20

20

40

Center Fault Block

Right Fault Block

There is an impermeable shale break between the Etive and Rannoch Sands. Therefore the permeability between layers K = 8 and K=9 is zero.

All rock and fluid properties are provided in the simulation model.

8-42

© PetroSkills, LLC., 2007. All rights reserved.