Ge Geomechanical omechanical Model del Pengc gcheng heng Fu, - - PowerPoint PPT Presentation

▶
ge geomechanical omechanical model del
SMART_READER_LITE
LIVE PREVIEW

Ge Geomechanical omechanical Model del Pengc gcheng heng Fu, - - PowerPoint PPT Presentation

Dec 04, 2022 154 likes •474 views

45 th US Rock Mechanics/Geomechanics Symposium American Rock Mechanics Association San Francisco, CA, June 29, 2011 Simulating mulating Complex mplex Frac acture ture Syste stems ms in Geoth othermal ermal Reservoirs servoirs Using ing

slide-1
SLIDE 1

Simulating mulating Complex mplex Frac acture ture Syste stems ms in Geoth

  • thermal

ermal Reservoirs servoirs Using ing an Expli licitly citly Coupled upled Hy Hydro ro- Ge Geomechanical

  • mechanical Model

del

45th US Rock Mechanics/Geomechanics Symposium American Rock Mechanics Association San Francisco, CA, June 29, 2011

Pengc gcheng heng Fu, Scott

  • tt M. Johnson,

hnson, and d Char arles les R. Carrig rigan an

Atmosphe spheri ric, c, Earth th, , and Energy Division Lawren ence ce Livermo more e Nati tional nal Laborato atory

slide-2
SLIDE 2
  • Hydraulic fracturing is an effective

method for enhancing permeability of geological formations.

Background

slide-3
SLIDE 3

How real fracture system looks like

(Large Block Test, Yucca Mountain. Wagoner, 2000) (Warpinski and Teufel, 1987)

slide-4
SLIDE 4

State of the art

PKN model PL3D model

(Adachi et al. 2000)

slide-5
SLIDE 5
  • Physical processes need to be covered:

– Fluid flow due to pressure gradient; – Rock deformation; – Variation of aperture width; and – Rock fracturing.

What do we need to simulate hydrofrac?

slide-6
SLIDE 6

Modules and their coupling

slide-7
SLIDE 7

Important Components

  • Flow solver – Finite volume method

Two mechanisms: ― Flow in fractures due to pressure gradient. ― Mass conservation with varying total fracture volume.

t w l q

h

q l P ) ( 6

3 j i h ij ij

L L w

i h j j h i j i h j h i h ij

L w L w L L w w w

3 3 3 3 3

) (

( )

ij ij i j

V P P

ref i i vap ref i i ref i i i

V m P V m V m K P / if / if 1

slide-8
SLIDE 8

Important Components

  • Fracturing criterion

– Estimates stress intensity factors using a generalized displacement correlation method – Handles mixed mode fractures

  • Adaptive remeshing
slide-9
SLIDE 9

Injection well Expected fracture path Core simulation domain Extended simulation domain (partial)

20 40 60 80 100 10 20 30 40 Fracture length l (m) Injection time t (second) Simulation Results KGD, closed-form solution

Predicted fracture growth rate

3 2 6 1 3

) 1 ( 679 . ) ( t Gq t l

Mesh of the numerical model

Model verification: classical KGD model

slide-10
SLIDE 10

Model validation: lab test results

Blanton, 1982

slide-11
SLIDE 11

Interaction Between Propagating and Existing Fractures

σxx=-20MPa σyy=-10MPa σyy=-10MPa

Pumping 18MPa

slide-12
SLIDE 12

Interaction Between Propagating and Existing Fractures

slide-13
SLIDE 13

Interaction Between Propagating and Existing Fractures

slide-14
SLIDE 14

Interaction Between Propagating and Existing Fractures

slide-15
SLIDE 15

Interaction Between Propagating and Existing Fractures

slide-16
SLIDE 16

Interaction Between Propagating and Existing Fractures

slide-17
SLIDE 17

Interaction Between Propagating and Existing Fractures

slide-18
SLIDE 18

Application to more complex fracture networks

Zero-pressure flow boundary Injection well Core simulation domain

Extended simulation domain (partially shown)

slide-19
SLIDE 19

Far field stress

Application to more complex fracture networks

15 MPa 10 MPa 12 MPa 10 MPa 11 MPa 10 MPa 10 MPa 10 MPa

Stress rotation Less anisotropy Less anisotropy Less anisotropy Stress rotation

slide-20
SLIDE 20

Application to more complex fracture networks

slide-21
SLIDE 21
  • Challenges:

– The coupling of multiple modules. – High computational cost.

  • Benefits:

– Explicit simulation of fracture-fracture and fracture-fluid interaction. – Capable of handling complex fracture networks. – Simple and physically meaningful input parameters. – Induced seismicity naturally emerges in the simulation.

  • Further development, enhancement, and validation

Concluding Remarks

slide-22
SLIDE 22
  • This work was performed under the auspices of the U.S.

Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.

Acknowledgements

Release number: LLNL-PRES-489801

slide-23
SLIDE 23

Application to more complex fracture networks

slide-24
SLIDE 24

3 2 6 1 3

) 1 ( 679 . ) ( t Gq t l

Model verification: classical KGD model

100 200 300 400 500 600 700 800 50 100

  • Pres. along frac. P(kPa)

l=25 m l=50 m l=75 m l=100 m 1 2 3 4 5 6

  • Apert. size w (mm)

2 4 6 8 10 12 50 100 Flow rate , q Distance from injection well (m)

slide-25
SLIDE 25

Application to more complex fracture networks

Pre-stimulation fracture network Post-stimulation fracture network

slide-26
SLIDE 26

Application to more complex fracture networks

σxx σyy

Results published in ARMA 2011 Symposium

slide-27
SLIDE 27

Application to more complex fracture networks

Connected to Well A Connected to Well B

slide-28
SLIDE 28

Application to more complex fracture networks

(a) Case D-1, left pumping only

0.011 Darcy

(b) Case D-2, right pumping only

0.013 Darcy

(c) Case D-3, left-then-right pumping

0.025 Darcy

(d) Case D-4, right-then-left pumping

0.026 Darcy

slide-29
SLIDE 29

Application to more complex fracture networks

13 MPa 10 MPa In situ stress

Injection well

Production well

Stimulated with 14 MPa pumping pressure Stimulated with 16 MPa pumping pressure Flow in unstimulated fracture network Preexisting fracture network

slide-30
SLIDE 30

Application to more complex fracture networks

Before stimulation After stimulation

P 1.50E+07 1.43E+07 1.36E+07 1.29E+07 1.21E+07 1.14E+07 1.07E+07 1.00E+07

slide-31
SLIDE 31

T 300 266 231 197 163 129 94 60

5 years

T 300 266 231 197 163 129 94 60

5 years

Application to more complex fracture networks

Before stimulation After stimulation