Understanding Waterflood Behavior and EOR - connectivity...Understanding Waterflood Behavior, ... –Examining the early water breakthrough plots and injectivity ... •How can we

Connectivity Analysis; Understanding Waterflood Behavior,

Connectivity for EOR Analysis

Richard Baker

September-2012

Jackson Hole Wyoming

Outline

• Executive Summary

• Background/What are we trying to do?

• Target Identification (communication analysis)

• Controlling “hot streak” flow

• Conclusions

Executive Summary I

• Examination of 12 waterflooded fields and over 2000

injector-producer pairs (1400 well pairs in sandstones)

– We studied the nature of heterogeneity between wells using

communication analysis (connectivity analysis)

• Flow in these waterflooded fields is controlled by two

components;

– Waterflood Induced Fractures

– Matrix controlled flow

3 © 2012 Baker Hughes Incorporated. All Rights Reserved.

Results of communication analysis work

(connectivity);

• Stress orientation western Canada

– controlling flow is typically NE-SW

– Examining the early water breakthrough plots and injectivity

analysis indicates the presence of fracture flow at early

times

• At late times, good reservoir management and shear

failure often mitigates the effect of on trend fractures

• Waterflood Induced Fracture (Geomechanics) is important

factor controlling flow


BACKGROUND/WHAT ARE WE

TRYING TO DO?


EOR selection Typically select on fluid type, lithology


Waterflood

Gas flood CO2

Chemical flooding

Understand how fluids flow

in reservoir

What are we trying to do? The need to

differentiate these two cases is critical


Fracture or small

volume hot streak

Khot streak>>

Kmatrix (50 times)

Matrix dominated flow +

Large volume in hot streak

Khot streak>

Kmatrix (2-10 times)

Why???


Fracture or small

volume hot streak

Khot streak>>

Kmatrix (50 times)

Matrix dominated flow +

Large volume in hot streak

Khot streak>

Kmatrix (2-10 times)


http://www.worldatlas.com/webimage/

countrys/na.htm

Figure 1: North American Stress Orientation Map18

(stars ae field locations annotated by authors)



WHAT ARE WE DOING?

• Study of 1400 well pairs in sandstone in Western Canada

(waterflooded) 2000 pairs in total

• Connectivity or communication analysis


• General experience in

North America is that

induced Waterflood

Fractures control flow

more than we think


Objectives of Work

• . A key question though is; what type of heterogeneity

controls flow? Is it matrix flow, induced hydraulic fracture

flow or both?

SPE 161177

Analysis of Flow and the Presence of Fractures and Hot Streaks in Waterflood Field Cases

Richard Baker, Tim Stephenson, Crystal Lok, Predrag Radovic, Robert Jobling, Cameron McBurney,

• How can we control injected fluid if we have a

combination of induced hydraulic fractures + matrix flow?

WHAT ARE WE DOING?

CONNECTIVITY/

COMMUNICATION ANALYSIS


What are we doing?

Figure 18: Types of Communication Strengths

Map view

1 2

3

What are we doing?

15

Light blue water injection rates

Dark blue water production rates

Green Oil rates

How do we use production data

to determine pathways?

Example of Watercut map used as

surveillance data from Beliveau SPE Time (days)

Short term data

long term data

Objective of Communication Analysis

Map of communicating well pairs

We want to identify key wells that communicate

What was done?

• We examine 1400 well pairs in sandstone reservoirs

under waterflood in WCSB using communication analysis

method

Figure 1: North American Stress Orientation Map18

(stars are field locations annotated by authors)

Five waterfloods in non

fractured high permeability

Sandstone reservoirs

Successful waterfloods

None of the waterfloods are

considered to be naturally fractured

Early Waterflood Behavior

• The stress orientation

controlling flow is typically

NE-SW in western

Canada. Examining the

early water breakthrough

plots and the hall plots

indicates the presence of

fracture flow.

Early Waterflood Behavior

• Water

breakthrough

usually occurred in

a on trend

direction (NE-SW)

Water cut

Time

On trend NE-SW

Off trend NE-SW

NW-SE

INJECTOR ANALYSIS


Monthly Injection Pressures vs. Injection

Rates over Two Year Period; VRR~1

Reservoir Pressure constant Theoretical line

Based on Darcy’s flow constant res. pressure Wellhead

Injection

Pressure

Injection

Rate

Injection Pressures vs. Injection Rates over Two

Year Period; VRR~1 Reservoir Pressure constant

Theoretical line

Based on Darcy’s flow

Banding behavior

@ same ~pressures

(3000-4000 Kpa)

Fracture

pressure

~ 4 300 KPA

(0.75-0.43) psi/ft

x600 m KB

Injection

Rate

Hall Plot; Evidence of Induced Fractures

Injection pressure Injection rate

Injection rate drops by ½ but

injection pressure is constant

Hall integral

Entire Field Late Stage Analysis

1 mile

Early stage waterflood (NE/SW)

Zoom in Portion of Field

1 mile

What has happened?

Partial conversion to line drive

but also

Communication is now East- West

Rose Diagram of Entire Fields Connectivity

The rose plots generated by the communication analysis tool

indicate no preferential flow direction in late stage waterfloods,

this indicates; 1) matrix flow and/or 2) shear failure fractures.

What is happening???? Early Stages; Map

View

Off trend well

On trend well

NE-SW Good initial

response

Water cut rise

first

What is happening???? Early Stages; Map

View

Off trend well

On trend well

NE-SW

• Watered out

Or

• injector

Water movement initially

NE-SW

What is happening???? Late Stage Waterflood;

Map View

Off trend well

On trend well

NE-SW

• Watered out

Or

• injector

On trend well

NE-SW

• Watered out

Or

• injector

Growth in

fractures

Mental Model Check

Injection pressure constant (2-3 times injection rates)

Initial communication in stress direction; NE-SW (on

trend)

Late stage communication in off trend direction

Either simple matrix flow

Or

– Shear fractures (constant wellhead injection pressure)

Probably both

Recap Targeting Study

• In 10-50% of the wells we see strong communication

between individual injectors and producer probably via

fractures

• Flow in these waterflooded fields is controlled by two

components;

– Fracture and/or high permeability matrix flow (hot streaks)

– Matrix controlled flow

So what can we do with this

knowledge?

Decrease injection pressures

Change producer into injectors

(streamline patterns)

Gel treatment/Polymer

SIMULATION STUDY

SASKATCHEWAN

Sandstone

thickness=3-4 M

μoil~350cP

μ polymer=7cP

API ~13

HISTORY MATCHED ~90 WELLS ON WATERFLOOD

Field BC

Perm=15 D

Perm=1 mD

Waterflood Polymer 20% Slug Polymer Gel

Oil Production non rate control

Method Incremental

RF

Ultimate RF

Waterflood 27%

Polymer Flood 5 32%

Gel Treatment 3 30%

Gel Treatment w/

Polymer Flood

11 38%

0

5

10

15

20

25

30

35

40

waterfloodalone

polymer floodalone

gel alone polymer flood+gel

Series 1

Series 1

Comparison of

Communication

Analysis with

Waterflood Case

(Blind test)

Conclusions

• Flow in these waterflooded fields is controlled by two components;

– Waterflood Hydraulic Induced Fractures (not NFR)

– Matrix

• Examining the early water breakthrough plots and injection pressures

and rates indicates the presence of fracture flow at early times

– Pinj≈ Pfrac extension

• At late times, good reservoir management and shear failure fracture

often mitigates the effect of on trend fractures

c

Background references

• Richard Baker, Tim Stephenson, Crystal Lok, Predrag Radovic, Robert

Jobling, Cameron McBurney; “Analysis of Flow and the Presence of Fractures

and Hot Streaks in Waterflood Field Cases”

• Heffer, K., Zhang, X., Koutsabeloulis, N., Main, I., Li, L., “Identification of

Activated (Therefore Potentially Conductive) Faults and Fractures Through

Statistical Correlations in Production and Injection Rates and Coupled Flow –

Geomechanical Modelling,” presented at SPE Europe Annual Conference

and Exhibition held in London, United Kingdom, 11-14 June 2007.

• Heffer, K., Greenhough, J., Main, I.G., Zhang, X., Hussein, A.M.,

Koutsabeloulis, N., “Low-cost Monitoring of Inter-well Reservoir

Communication Paths Through Correlations in Well Rate Fluctuations: Case

Studies From Mature Fields in the North Sea,” presented at SPE

Europec/EAGE Annual Conference and Exhibition held in Barcelona, Spain,

14-17 June 2010


Understanding Waterflood Behavior and EOR - connectivity...Understanding Waterflood Behavior, ... –Examining the early water breakthrough plots and injectivity ... •How can we

Documents