Enhanced Oil Recovery by Water Alternating Gas (WAG ......Water Alternating Gas (WAG) Injection Studies JIP Institute of Petroleum Engineering, Heriot-Watt University, Edinburgh Scotland,

Enhanced Oil Recovery by

Water Alternating Gas (WAG)

Injection: The Opportunity and

the Challenge

Centre for Enhanced Oil Recovery & CO2 Solutions

Characterisation of Three Phase Flow and

Water Alternating Gas (WAG) Injection

Studies JIP

Institute of Petroleum Engineering,

Heriot-Watt University, Edinburgh

Scotland, UK

Contact:

Prof. Mehran Sohrabi

Telephone: +44 (0)131 451 3568

Email: [email protected]

Sohrabi M., Fatemi M., Ireland S.

Presented by: Mobeen Fatemi

06 May 2014

mailto:[email protected]

North-Sea WAG Injection Potential

Distribution of EOR field applications by method in the North Sea (total of 19

projects). (SPE 99546; 2006)

2

WAG48%

MEOR5%

FAWAG11%

SWAG5%

MGI31%

WAG Injection

O

W

3

O

W

G

WAG Injection

4

O

W+ O

G+O

G+W+O

WAG involves major

complexity and

hysteresis, caused by

injection alternation

that happens during

process.

WAG Injection

5

WAG Injection

WAG involves major complexity and hysteresis, caused by

alternating injection that happens during process.Therefore,

numerical simulation of WAG injection becomes extremely

complex (involves the 2-phase and 3-phase Pc and kr and also

their hysteresis behavior).

Our current understanding of the physics (and sometime

chemistry) involved in three-phase flow is currently limited

and hence quantifying and predicting the outcome of these

processes are difficult.

6

7

Measuring 3-phase kr is very difficult and time consuming

hence many correlations have been proposed for calculating

3-phase kr from the more readily available 2-phase data

Note: These models are usually based on water-wet systems and high

IFT gas/oil.

WAG Injection

8

Formulation available in the existing reservoir simulators are

not capable of adequately account for the complex interplay

of hysteresis, capillary pressure, wettability, IFT, trapped

phase saturation and their impact on flow under three-phase

flow regime.

WAG Injection

Water Flood

Gas injection

Alternating slugs of gas and water (WAG)?

Sequence of gas and water injection?

Simultaneous injection of water and gas (SWAG)?

Wettability?

Interfacial Tension (gas type)?

Rock type?

Rock Permeability?

Miscibility?

Trapped phase saturation and hysteresis?

Reliability of Reservoir Performance Prediction

9

06/05/2014 10

06/05/2014 10

JIP at Heriot-Watt University: Research Approach

To experimentally study parameters and understand mechanisms

involved in GF, WAG and SWAG injections.

To generate 2-phase and 3-phase relative permeability curves using

the results of the coreflood experiments.

Evaluate existing 3-phase models

Develop improved methodologies for obtaining 3-phase kr and

hysteresis for reliable numerical simulation of WAG injection.

11

Experimental Condition

Different injection scenarios WAG, SWAGContinuous GasContinues Water

Different IFT valuesHigh IFT (immiscible)Low IFT (intermediate)Very low IFT (near-miscible)

Different wettability conditionsWater-Wet, Oil-Wet & Mixed-Wet

Different Core SamplesCarbonates, Sandstones

12

13

This research programme was first launched on November

1997 with main aim of understanding the process of WAG

injection. The scope of work was later extended to three-

Phase flow (including WAG).

Inlet Outlet

Cover plate

Two-dimensional etched pore structure

Micromodel Experiments

Core Length

/ cm

Diameter

/ cm

Porosity

/ frac.

Permeability

/ mD

Core 1 67.0 4.98 0.17 1000

Core 2 60.5 5.08 0.19 64

Core Flood Experiments

Pressure

/psia

Temperatur

e

/oC

Ρg/kgm-3

ρL/kgm-3

µg/mPa.s

µL/mPa.s

IFT

/mNm-1

1200 37.8 86.6 466 0.0141 0.0793 2.7

1790 37.8 184.8 345 0.0206 0.0474 0.15

1840 37.8 211.4 317 0.0249 0.0405 0.04

Core properties

Fluid properties

Injection Production

14

Effect of IFTo/g: (1000 mD, Gas Injection, Mixed-wet)

15

0.0

0.2

0.4

0.6

0.8

1.0

0 2 4 6 8 10

Oil

Re

cove

ry (

Co

re P

V)

Injected Gas (Core PV)

1000mD, GF, MW, 1825 psi

1000mD, GF, MW, 1200 psi

σ = 2.70

σ = 0.04

Effect of IFTo/g: (65 mD, Gas Injection, Mixed-wet)

16

σ = 2.70

σ = 0.04

65mD, Mixed-Wet 1000mD, Mixed-Wet

0

0.2

0.4

0.6

0.8

1

0 2 4 6 8 10

Oil R

ecovery

(fr

ac.

IOIP

)

Injected Fluids (Core PV)

WAG, MW, 65mD

Water Injection, MW, 65mD

SWAG (Qg/Qw=0.25), MW, 65mD

Gas Injection, MW, 65mD

0

0.2

0.4

0.6

0.8

1

0 2 4 6 8O

il R

ecovery

(C

ore

PV

)

Injected Fluids (Core PV)

WAG Injection, MW, 1000mD

Water Injection, MW, 1000mD

SWAG (Qg/Qw = 0.25), MW, 1000mD

Gas Injection, MW, 1000mD

Mixed-Wet Rock (65mD vs. 1000 mD; σg/o = 0.04 mN.m-1)

Performance of Different

Injection Scenarios

WAG WAG

GF

GF

17

65mD, Mixed-Wet 1000mD, Mixed-Wet

0

20

40

60

80

100

0 2 4 6 8 10

Pre

ssure

Dro

p (

psi

)

Injected Fluids (Core PV)

SWAG (Qg/Qw=1), MW, 65mD

WAG, 65mD, MW, IDIDID

Water Injection, MW, 65mD

Gas Injection, MW, 65mD

0

5

10

15

20

25

0 2 4 6 8Pre

ssure

Dro

p (

psi

)

Injected Fluids (Core PV)

SWAG, MW, 1000mD

WAG, MW, 1000mD

Water Injection, MW, 1000mD

Gas Injection, MW, 1000mD

Mixed-Wet Rock (65mD Vs. 1000 mD; σg/o = 0.04 mN.m-1)

Effect of Injection Scenario on

Injectivity

18

19

G OW G WG WSw,im=18% , Soi=82%

W G

65 mD

WAG-IDIDIDID

Mixed-wet

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

0 2.5 5 7.5 10 12.5

Prod

uced

Oil

(fra

c. S

orw

)

Injected WAG (Core PV)

WAG-IDIDIDID, 65mD, MW, 1840 psia

WAG-IDIDIDID, 65mD, MW, 1790 psia

WAG-IDIDIDID, 65mD, MW, 1215 psia

σ = 2.70

σ = 0.04

σ = 0.15

Effect of Gas/Oil IFT on WAG

20

W1

W2

W3

W1

W3

W2

σ = 2.70σ = 0.04

0.1

1

10

100

0 0.5 1 1.5 2 2.5 3

Inje

ctiv

ity

(cc/

psi

)

Injected Brine (Core PV)

W1

W2

W3

W1

W2

W3

σ = 0.15

Effect of IFTOil/Gas on Injectivity

21

One of the major achievements of this research is a

growing set of core flood data covering a wide range

of pertinent parameters (IFT, Wettability, Hysteresis,

K, Rock Type, etc) investigating various injection

strategies.

As far as we know, no such comprehensive data is

available in published literature.

A unique set experimental data

22

Measuring 3-phase kr is very difficult and time

consuming hence many correlations have been

proposed for calculating 3-phase kr from the more

readily available 2-phase data.

+

0.0

0.0

0.2

0.4

0.6

0.8

1.0

0.0

0.2

0.4

0.6

0.8

1.0

k r

1.00.80.60.40.2

k rog

k rg

0.0

0.0

0.2

0.4

0.6

0.8

1.0

0.0

0.2

0.4

0.6

0.8

1.0

1.00.80.60.40.2

k row

k rw

2Ph Oil & Gas 2Ph Oil & Water 3Ph Oil k r

So

Sg=1

Sw

k ro0.01

0.80

Use 2Ph k rg in 3Ph

and 2Ph k rw in 3Ph

SwSg SwSg

Three-phase kr Determination

22

23

Fluid injection

Unsteady state 2-phase test

2-phase kr

Simulation of WAG

test using 3-phase krmodels

Swir =18%, k=65 mD

WAG experiment

Fluid injection Swir =18%, k=65 mD

Comparing experiment and simulation results e.g.

recovery and pressure representing the accuracy

of that particular 3-phase kr model.

Evaluation of 3-phase kr Modells

23

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8

PVinj

Oil

Reco

very

, P

V

EXPERIMENT

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8

PVinj

Oil

Reco

very

, P

V

EXPERIMENT

BAKER2

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8

PVinj

Oil

Reco

very

, P

V

EXPERIMENT

SWI

BAKER2

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8

PVinj

Oil

Reco

very

, P

V

EXPERIMENT

STONE2

SWI

BAKER2

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8

PVinj

Oil

Reco

very

, P

V

EXPERIMENT

STONE2

SWI

BAKER1

BAKER2

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8

PVinj

Oil

Reco

very

, P

V

EXPERIMENT

STONE1

STONE2

SWI

BAKER1

BAKER2

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8

PVinj

Oil

Reco

very

, P

V

EXPERIMENT

STONE1

STONE2

SWI

BAKER1

BAKER2

LARSEN

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8

PVinj

Oil

Reco

very

, P

V

EXPERIMENT

STONE1

STONE2

SWI

BAKER1

BAKER2

LARSEN

STONE-EXPONENT

Numerical Simulation of WAG (Mixed-Wet)

24

25

Existing three-phase kr models lead to large errors

in prediction of WAG performance.

What is the actual 3-phase kr during WAG

?

25

26

Another major achievement of the project is development

of a software for obtaining three-phase kr and Pc.

3-phase kr can be obtained directly instead of indirectly from

2-phase.

kro =kro (Sw, Sg)

krw =krw (So, Sg)

krg =krg (Sw, So)

Gas

Oil

Water

OilWater Gas

Direct 3-phase kr - 3RPSim

27

injection core

using our in-house simulator

(3RPSim) to estimate 3-phase

kr values by history matching

experimental results e.g.

recovery and pressure

Determination of 3-phase kr by

history matching experimental

results:

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8

PVinj

Oil

Reco

very

, P

V

EXPERIMENT

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8

PVinj

Oil

Reco

very

, P

V

EXPERIMENT

BAKER2

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8

PVinj

Oil

Reco

very

, P

V

EXPERIMENT

SWI

BAKER2

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8

PVinj

Oil

Reco

very

, P

V

EXPERIMENT

STONE2

SWI

BAKER2

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8

PVinj

Oil

Reco

very

, P

V

EXPERIMENT

STONE2

SWI

BAKER1

BAKER2

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8

PVinj

Oil

Reco

very

, P

V

EXPERIMENT

STONE1

STONE2

SWI

BAKER1

BAKER2

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8

PVinj

Oil

Reco

very

, P

V

EXPERIMENT

STONE1

STONE2

SWI

BAKER1

BAKER2

LARSEN

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8

PVinj

Oil

Reco

very

, P

V

EXPERIMENT

STONE1

STONE2

SWI

BAKER1

BAKER2

LARSEN

STONE-EXPONENT

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8

PVinj

Oil

Reco

very

, P

V

EXPERIMENT

STONE1

STONE2

SWI

BAKER1

BAKER2

LARSEN

STONE-EXPONENT

Heriot-WattSimulator

Numerical Simulation of WAG (Mixed-Wet)

28

New Hysteresis model

Required two-phase data for running this model1. Oil/water : krow2. Oil/gas : krog3. Gas/water : krgw, krwg

three-phase pore occupancy

Saturation function

accounting cyclic

Hyst

Two-phase kr

3Phri rij rjk rik rkjk f k k k k (1 )(1 )

i

j k

Sf

S S

SPE #152218. Three-Phase Relative Permeabilityand Hysteresis Model for Simulation of WaterAlternating Gas (WAG) Injection. MehranSohrabi 29

Three-Phase Flow JIP at Heriot-Watt University

Micromodel Experiment Core-flood Experiment

Mechanisms of Flow Mechanisms of FlowExamining different

injection Scenario

Generating kr and Pcdata

Analyse experimental

data

Evaluate capability of existing simulators and models

Three-Phase kr Three-Phase Pc Hysteresis IFT scalingTrapped

saturationViscous

fingering

Modelling

New improved mathematical

model for calculating flow

parameters (kr , Pc , trap

phase, hysteresis..)

High quality measured data

for different rock and fluid

conditions (kr & Pc)

methodologies to correct the SCAL

data due to experimental artefact

(viscous fingering, end-Effects)

In-house Software

(3RPSim)

Deliverable

30

Project’s Sponsors

31

http://www2.petrobras.com.br/ingles/index.asphttp://www2.petrobras.com.br/ingles/index.asphttp://www.dongenergy.com/EN/index.htmhttp://www.dongenergy.com/EN/index.htmhttp://www.bhpbilliton.com/http://www.bhpbilliton.com/http://www.bp.com/home.do?categoryId=1&contentId=2006973http://www.bp.com/home.do?categoryId=1&contentId=2006973

32