An Efficient and Optimized Approach to Multiple Fracturing of ...€¦ · Well Evaluation Production Optimization Best Practices 11400 11600 11800 12000 12200 12400 12600 12800 13000

Injectivity Frcaturing and Assurance

Advantek Waste Management Services

11000 Richmond Ave, Suite 190

Houston TX 77042

713.532.7627

[email protected]

www.advantekinternational.com

An Efficient and Optimized Approach to Multiple Fracturing of Horizontal Wells

Copyright 2015 Advantek Waste Management Services LLC. This material is private and confidential 2

Outline

• Introduction

• Geomechanics

• Multiple Fracturing of Horizontal Wells

• Injection Performance Test

• Monitoring, Assurance and Compliance

• SRV Assessment

• Concluding Remarks


Variety of Fracturing Configurations

Frac PackHigh Perm

Selective Stimulation

Longitudinal Fractures

Multiple Transverse Fractures

Selective/ Whole Matrix Treatment

Stimulation for Selective Injectivity

EPA Technical Workshop on H/F DC, March 10-11/2011


Integrated Stimulation Process Optimization

Integrated Approach for Lowest Cost Development

Design Models

Operational

Monitoring and

Intelligence

Well Evaluation

Production

Optimization

Best Practices

11400

11600

11800

12000

12200

12400

12600

12800

13000

5000 6000 7000 8000 9000 10000

Minimum Horizontal Stress (psi)

Tru

e V

ertic

al

De

pth

(ft

)

Minimum Horizontal Stress Reservoir Pore Pressure

Overburden Pressure

11400

11600

11800

12000

12200

12400

12600

12800

13000

0.4 0.45 0.5 0.55 0.6 0.65 0.7

Fracture Gradient (psi/ft)

Tru

e V

ertic

al

De

pth

(ft

)

Min. RI Min. RI Min. RII Max. RII Sh FG

Log Interpretations

Tight

Thick to

V. Thick

Medium

Thin

V. TightMediumHigh

Reservoir PermeabilityTight

Thick to

V. Thick

Medium

Thin

V. TightMediumHigh

Reservoir Permeability

Re

serv

oir

Th

ick

ne

ssR

ese

rvo

ir T

hic

kn

ess

Well Type Selection


Shale Gas Rock Successful Exploitation - Key Enabling Technologies

Horizontal Drilling

Hydraulic Fracturing+


Fracture Shape & Proppant Concentration DiagramFor Different Injection

Scenarios and Complex Models

1500

2000

2500

3000

3500

4000

4500

5000

5500

4752 ft Injection PointMid Sand Layers Modeled5390 bbls Injected0.81 psi/ft Shale Stress Gradient

3280 ft Injection PointMid Sand Layers Modeled14000 bbls Injected0.85 psi/ft Stress Gradient

4752 ft Injection PointMid Sand Layers Not Modeled13400 bbls Injected0.81 psi/ft Stress Gradient

1.

2. 3.

(2D)24Oct1998TerraFracOutPutdata

0 500 1000HalfFractureLength(ft)

-300

-200

-100

0

100

200

300

400

500

600

700

800

900

Fractu

reHeig

ht(ft)

PropConcentration0.6135940.5726880.5317810.4908750.4499690.4090630.3681560.327250.2863440.2454380.2045310.1636250.1227190.08181250.0409063

SolidParticleConcentrationafter5,390bblsInjected(4752ftInjectionPoint,MidLayersModeled,

0.81psi/ftGradient)


0 100 200 300 400 500 600 700HalfFractureLength(ft)

-200

-100

0

100

200

300

Fractu

reHe

ight(f

t)

0.6240.5930.5630.5320.5010.4710.4400.4090.3790.3480.3170.2870.2560.2250.195

Magnus:SolidVolumetricConcentration(V=14000bblsat4bpm&3ppgSolids)


0 500 1000 1500HalfFractureLength(ft)

-400

-300

-200

-100

0

100

200

300

400

500

600

700

800

900

1000

Fractu

reHeig

ht(ft)


SolidParticleConcentrationafter13,400bblsInjected(4752ftInjectionPoint,MidLayersnotModeled,SolidsConsidered)


3280 ftCasing Shoe

4752 ftCasing Shoe

1. 2. 3.

Depth(ft TVD)

4752 ft

4752 ft

3280 ft


Pressure during Sequential Pumping

0 1000 2000 3000 4000 5000 6000 7000 8000

Time (min)

0

500

1000

1500

2000

Bot

tom

hole

Pre

ssur

e (P

SI)

Atoka Injections

BHP

80

100

120

140

160

180

Tem

p (f

)

Temp

Stress along Well due to Concurrent Parallel Fractures

Pressure Increase due to Series Fractures

Concurrent Fractures

Injected Fluid

Plugged Tip Zone

New Fracture

EPA Technical Workshop on H/F DC, March 10-11/2011


Fracture Dimension (Height)


Stress around Fractues


Injection Rate at Fracture Entry


Stress Impact during multiple fracturing in a Horizontal Well

From ARMA/USRMS 05-672

Inner Fractures

Outer Fractures

Single IntervalMultiple Intervals

Schematic


Various Created Fracture Geometries in Shales

1500

2000

2500

3000

3500

4000

4500

5000

55004752 ft Injection PointMid Sand Layers Modeled5390 bbls Injected0.81 psi/ft Shale Stress Gradient

3280 ft Injection PointMid Sand Layers Modeled14000 bbls Injected0.85 psi/ft Stress Gradient

4752 ft Injection PointMid Sand Layers Not Modeled13400 bbls Injected0.81 psi/ft Stress Gradient

1. 2. 3.

(2D) 24 Oct 1998 TerraFrac OutPut data

0 500 1000Half Fracture Length (ft)

-300

-200

-100

0

100

200

300

400

500

600

700

800

900

Fra

ctur

eH

eigh

t(ft

)


Solid Particle Concentration after 5,390 bbls Injected(4752 ft Injection Point, Mid Layers Modeled,

0.81 psi/ft Gradient)


0 100 200 300 400 500 600 700Half Fracture Length (ft)

-200

-100

0

100

200

300

Fra

ctur

eH

eigh

t(ft

)

0.6240.5930.5630.5320.5010.4710.4400.4090.3790.3480.3170.2870.2560.2250.195

Magnus: Solid Volumetric Concentration(V = 14000 bbls at 4 bpm & 3 ppg Solids)


0 500 1000 1500Half Fracture Length (ft)

-400

-300

-200

-100

0

100

200

300

400

500

600

700

800

900

1000

Fra

ctur

eH

eigh

t(ft

)


Solid Particle Concentration after 13,400 bbls Injected(4752 ft Injection Point, Mid Layers not Modeled, Solids Considered)


3280 ftCasing Shoe

4752 ftCasing Shoe

1.

2.

3.

Depth(ft TVD)

4752 ft 4752 ft

3280 ft

X (ft)

Y(f

t)

0 500 1000 1500

5000

5200

5400

5600

5800

6000

6200

6400

6600

PropCon0.558830.523260.487690.452120.416550.380980.345410.309840.274270.23870.203130.167560.131990.096420.06085

Frame 001 08 Dec 2003 TerraFrac OutPut dataFrame 001 08 Dec 2003 TerraFrac OutPut data

Simulated fracture using @FRAC®


PTA-IFO Tool

Pressure Falloff Data

Pressure Transient Theory is used to determine Reservoir Parameters from the measured Pressure Response (Flow Regimes)


Conventional Pressure Transient Response

• Relies on pressure and derivative plots after shut-in to draw information about reservoir characteristics.

• Typical flow regimes such as wellbore storage, linear, and radial flow can be deduced from the plots.

• Different flow regimes can be observed by modification of the basic governing equations and the boundary conditions.

• Modifications include , for example, consideration of complications imposed by geometry, reservoir boundaries, porosity, mobility etc.

Pictures: Courtesy of EPA and Schlumberger


(Warren and Root, 1963)

sn

sn

ss

ss

1

2

Dual porosity m

odel

Conjugate Fractures


Problem Configuration

Elliptical coordinate of the water front ξ0 = 0.5

Xf fracture half-lengthY0

X0


Xf fracture half-length


Xf fracture half-lengthY0

X0

Olson, 2008


Regular fracture networkSRV Extent

Reservoir Volume >> SRV

PTA

-IFO ID

EALIZED

MO

DEL

After C. Economides


PTA-IFO Conventional Methods of Analysis

• Method 1: Using Storage Dominated Flow – Recognized by a straight line on the

pressure vs. time plot.– The slope of this linear part of pressure

curve is equal to q/Cf from which Fracture Storage Coefficient can be determined. From Cf, Fracture Length can be calculated, considering different fracture types (PKN, CGK, Elliptical).

• Method 2 : Using Linear Formation Flow – Recognized by a straight line with a half

slope on the log-log plot– From semi-log analysis and using fluid and

relative permeability data, the ratio {viscosity/(permeability*porosity*total compressibility)} can be calculated.

– From a plot of Pf vs. square root of time, the above ratio is related to flow and Fracture Dimensions (h*L) so that Fracture Length. Can be calculated.

• Method 3: Using late time Pseudo-Radial Flow– Permeability of Inner and Outer Zones plus

Mobility discontinuity can be determined


Conventional Pressure Transient Analyses for Fractured Wells

• Two typical models for a single vertical fracture:

• Infinite-Conductivity Vertical Fracture (ICVF)

• Finite Conductivity Vertical Fracture (FCVF)

• Typical type curves does not show dramatic and sharp change in pressure and derivative plots.

• Slope of the type curve indicate different flow regimes (linear or bi-linear flow) and fracture conductivity.

Picture: Courtesy of Schlumberger

0.1

1.0

10.0

100.0

1000.0

0.001 0.01 0.1 1 10 100 1000 10000

Time (hr)

Dp

, d

p/d

lnt

(ps

i)

pressure drop

Pressure derivative

Fixed fracture permeability

Stress-dependent fracture permeability

(After Economidis and Ghassimi,)


Pressure Transient Response in Multi-Fractured Vertical Wells

Cashiriari DCI Well Batch

10-3 10-2 10-1 100 101

10-4

10-3

10-2

10-1

Delta-T (hr)

DP

& D

ER

IVA

TIV

E (

PS

I/ST

B/D

)

ENDWBS

2010/03/26-0729 : OIL

Radial Composite Homogeneous Reservoir** Simulation Data ** well. storage = 0.39474 BBLS/PSI Skin(mech.) = -4.1053 permeability = 2.4799 MD Perm.(inner) = 38.500 MD Stor.rto+x o/i = 0.034984 Inner Radius = 116.87 FEET Skin(Global) = -5.0419 Mobility+x o/i = 0.064412 Perm-Thickness = 148.79 MD-FEET Initial Press. = 6554.01 PSI Smoothing Coef = 0.,0.

Static-Data and ConstantsVolume-Factor = 1.000 vol/volThickness = 60.00 FEETViscosity = 1.000 CPTotal Compress = .1827E-04 1/PSIRate = -4219. STB/DStorivity = 0.0002192 FEET/PSIDiffusivity = 179.0 FEET^2/HRGauge Depth = N/A FEETPerf. Depth = N/A FEETDatum Depth = N/A FEETAnalysis-Data ID: GAU001Based on Gauge ID: GAU001PFA Starts: 2010-03-26 00:00:15PFA Ends : 2010-03-29 23:33:15

• The multiple fractures create a fracture network system that appears to mimic the signature of radial composite flow with more complexities.

• The inner zone has what appears to be an increased permeability that is brought about due to the elevated conductivity of the fracture network in comparison to the native formation permeability and mobility.

• Figure shows analyses of a multi-fractured injection well using a radial composite mode. The reasonable match, however, does not accurately reflect the system physics and fails to correctly provide the understanding of the fracture network makeup.

Technology Premise


Current Limitations of Fracture Diagnostic Techniques

Parameter Technique LimitationFr

actu

re H

eigh

tTracer logs Shallow depth of investigation:

shows height only near the wellbore

Temperature logs

Difficult to interpret: shallow depth of investigation shows

height only near wellbore

Stress profiling

Does not measure fracture directly: Must be calibrated

with in-situ stress tests

P3D models Does not measure fracture directly: estimates vary

depending on which model is used

Microseismic Optimally requires nearby offset well: difficult to conduct

in the field

Tilt meters Difficult to interpret: expensive and difficult to conduct in the

field

Parameter Technique Limitation

Frac

ture

Len

gth

P3D Models Length inferred, not measured: estimate s vary greatly

depending on which model is used

Well testing Large uncertainties depending upon assumptions and lack of

pre-fracture well test data

Microseismic Optimally requires nearby offset well; difficult to interpret;

expensive

Tiltmeters Difficult to interpret; expensive and difficult to conduct in the

field

Frac

ture

Azi

mu

th

Core techniques

Expensive to cut core and run tests; multiple tests must be

run to assure accuracy

Log Techniques

Requires open hole logs to be run; does not work if natural

fractures are not present

Microsesmic Analysis intensive; expensive for determination of azimuth

Tilt meters Useful only to a depth of 5000 ft; requires access to large area;

expensive

(EPA 816-R-04-003 )


Pressure Transient Response for a Closing Fracture (Ideal Theoretical Type Curves, Single Vertical Fracture)

• Fracture closure is characterized by a sudden rapid change in pressure (determines the fracture closure pressure).

• Mixture of fracture storage flow and linear formation flow before closure.

• After closure, flow from the fracture into the formation will show a transition from linear formation flow to pseudo-radial flow.

• Fracture closure is characterized by a sharp peak

Pictures: Courtesy of van den Hoek (SPE 77946)


HEIGHT SHRINKAGE LENGTH SHRINKAGE


PTA-IFO Fracture Characteristics Determination during DFIT

• Method 1: Using Storage Dominated Flow – Recognized by a straight line on the pressure vs. time

plot.

– The slope of this linear part of pressure curve is equal to q/Cf from which fracture storage coefficient can be determined. From Cf, fracture length can be calculated, considering different fracture types (PKN, CGK, elliptical).

• Method 2: Using Linear Formation Flow Before Fracture Closure– Recognized by a straight line with a half slope on the log-

log plot.

– Occurs for longer closure times where a combination of storage flow and linear formation flow takes place before fracture closure.

– Type curve matching and the use of an existing analytical expression gives the fracture storage coefficient from which fracture length can be calculated.

• Method 3 : Using Linear Formation Flow After Fracture Closure – Recognized by a straight line with a half slope on the log-

log plot.

– Occurs for short closure times.

– From semi-log analysis and using fluid and relative permeability data, the ratio {viscosity/(permeability*porosity*total compressibility)} can be calculated.

– From a plot of Pf vs. square root of time, the above ratio is related to flow and fracture dimensions (h*L).

• Method 4: Using late time Pseudo-Radial Flow– Mobility discontinuity can be determined

0.0000001

0.000001

0.00001

0.0001

0.001

0.01

0.1

1

10

100

0.0001 0.001 0.01 0.1 1 10 100 1000 10000Dimensionless Time

Dim

ensi

on

less

Pre

ssu

re

x0=0.5 x0=1 x0=2 x0=3 x0=4 x0=5 Test Data 1 Test Data 2


@IPT and @IPTSH Data Input and Estimated Parameters

All Data InputParameters Estimated

Using Conventional Analyses Methods

Parameters Estimated Using Type-Curve Matching

Injection Rate Permeability of the Inner zone Permeability of the Inner zone

Volume Injected Fracture Storage Constant Fracture Storage Constant

Fluid Compressibility Fracture Half Length Fracture Half Length

Injection Fluid Viscosity Mobility Ratio Frcature Skin

Reservoir Porosity Frcature Conductivity

Formation Volume Factor Rate of Fracture Length Shrinkage

Formation Height (Thickness) Injection Layer Stress

Reservoir Poisson's Ratio Containment layer Stress

Reservoir Young's Modulus Diffusivity Ratio

Total Compressibility Mobility Ratio

Mobility Ratio Permeability of the Outer zone

Mobility Front, Elliptical

Diffusivity Ratio

Injection Layer Stress

Fracture Storage Coefficient is related to Formation Elastic properties, and fracture Length and Height.Mobility is defined as the Ratio of Permeability to Fluid Viscosity, Mobility Ratio = Inner Zone Mobility/Outer Zone MobilityDiffusivity is defined as the Ratio of Mobility to (Porosity x Compressibility), Diffusivity Ratio = Inner/Outer Zone Diffusivity


Example Case – Comparison between PIE and @IPT Plots

Actual Field Record of a fractured Injector during Fall Off


PTA-IFO Type-Curve Matching

• Model Selection– Infinite Conductivity-Dual Mobility– Finite Conductivity-Single Mobility– Finite Conductivity-Dual Mobility

• Fracture Types– KGD– PKN– Elliptical

• Fracture Shrinkage Modes– Height Shrinkage Only– Length Shrinkage Only– Combined Height and Length Shrinkage


Type-Curve Generation before Matching

• Method 1: Using Storage Dominated Flow – Fracture storage constant has been determined

and Fracture Half Length has been estimated for three types of fracture geometry.

• Method 2 : Using Linear Formation Flow – Data here does not clearly show linear flow;

however, Fracture Half Length has been estimated for three types of fracture geometry.

• Method 3 : Using Radial Formation Flow – Permeability of Inner Zone has been determined.

Method 1

Method 2 Method 3


Results of Shrinking Fracture by Type-Curve Matching

Best Fit

Limit Type-Curves


Results of Type-Curve Matching (Single Mobility)

Limit Type-Curves

Best Fit


Results by Type-Curve Matching (Dual Mobility)

Limit Type-Curves

Best Fit


Results of Type-Curve Matching (Shrinking Fracture in Single Mobility)

Limit Type-Curves

Best Fit


Limit Type-Curves

Best Fit

Results of Type-Curve Matching (Shrinking Fracture in Dual Mobility)


Transient analysis of post injection pressure decline is used to develop a picture of the stimulated reservoir volume, and system structure and dynamics.Transient analysis will help define fracture geometry and the dominant system characteristics.

Do boundaries exist?Are stimulated zones being created?Does the analyzed fracture penetrate into or through the stimulated zone and what does this mean in analyzing and

describing the system?Once analysis has defined a model, pressure responses characteristic of different generated fracture

lengths can be predicted and cross correlated during performance reviews.

1E-5 1E-4 1E-3 0.01 0.1 1 10 100 10000.01

0.1

1

10

100

1000

10000

Log-Log plot: p-p@dt=0 and derivative [psi] vs dt [hr]

10

100

1000

10000

100000

1.00E-03 1.00E-02 1.00E-01 1.00E+00 1.00E+01 1.00E+02

Delta t (hrs)

Del

ta P

, Del

ta P

' (p

si)

Xf = 50ft Xf =100ft

Xf =200ft Xf =500ft

Xf =1000ft

10

100

1000

10000

100000

1.00E-03 1.00E-02 1.00E-01 1.00E+00 1.00E+01 1.00E+02

Delta t (hrs)

Del

ta P

, Del

ta P

' (p

si)

Xf = 50ft Xf =100ft

Xf =200ft Xf =500ft

Xf =1000ft

Fall-off ModelInjectivity ModelPressure and Pressure Derivative Plot

Using PTA during IFO to Estimate SRV



Model Service Provision Microsoft Azure

@SSURERemote Sensing

SurveillanceField Performance

Site-to-Site VPN

Injection Assurance Platform (@SSURE)

@SSURE provides secure cloud access between clients and field ops for surveillance

Field-to-Tower/Satellite Secure Tunnel

Field-to-SiteSecure VPN

Secure LoginWeb Access


Concluding Remarks• Cloud Computations are fast and inexpensive as well as connective

• Engineers can have effective real-time monitoring and simulation

• Fracture models have been advanced to provide efficient and realistic assessment of multiple concurrent fracturing of horizontal wells

• Pressure transient analysis of fall off data following fracture treatments or injection operations have been utilized

• The case of fractured injectors with closing and shrinkage fractures shows that significant geomechanical details may be obtained from the data

• A more direct methodology is proposed for the determination of stimulated Reservoir volume (SRV), if it exist.

• Water hammer effects provide potential for closer fracture assessment and would require further analysis which is underway. (HIT link)


• Both the extent and the permeability elevation of the SRV are easily assessed from the PFO results.

• A measure of fracture length, height and containment stress contrast may be estimated closely, which helps is assessing fracture migration outside the target zone.

• Breaching, loss of containment and fluid migration must always be significant factors in job design and implementation.

• Assurance is a primary factor in stimulation via complete data collection, sophisticated modeling and live monitoring.

• Pressure transient tests have advanced and are currently successful for better identification of fractures.

• Multiple fractures in single wells must be designed with sufficient certainty and complexities and need close monitoring

Concluding Remarks (continue)


Thank YouAny Questions?

Advantek International11000 Richmond Avenue, Suite 190Houston, Texas, 77063713.532.7627admin@advantekinternational.comwww.advantekinternational.com

Documents

An Efficient and Optimized Approach to Multiple Fracturing of ...€¦ · Well Evaluation Production Optimization Best Practices 11400 11600 11800 12000 12200 12400 12600 12800 13000