7649re information to your upre information to your up727.ppt

re information to your upre information to your up

1. Modelling Unconventional Wells
with Resolve and Reveal
Juan Biörklund (Gauloise Energía)
and
Georg Ziegler (Wintershall Holding)

2. Modelling Unconventional Wells
PVT Modelling
Reservoir Modelling
Well Modelling
Chemical EOR
Water Chemistry
Reveal Features
Adsorption
Diffusion
Rock Mechanics
Rock Failure
Fracturing
Designed to study specialized near well bore effects and
assess their impact on the production from the reservoir

3. Modelling Unconventional Wells
Quick Calculations
Tight reservoirs with complex well geometries
Systems with up to 20,000 wells with 100 wells being
drilled per month
Shale Oil / Gas Wells
Horizontal, inclined and fractured wells
Gas adsorption, diffusion, directional stress dependent
permeability variations etc...
Capabilities (Resolve & Reveal)

4. Modelling Unconventional Wells
Dimensionless transient reservoir response
capturing the well, fracture and reservoir geometry
Dimensionless pseudo-pressure pseudo-time (pD tD)
Match this response to production history
Modify reference parameters k and phi(r
w
2
)
Adjust PVT reference conditions for pD and tD
Concept

5. Modelling Unconventional Wells
Perform predictions
Use matched pD tD and super-position to predict Pwf
Transient IPR for rate estimation
Well scheduling
Automate the procedures using visual workflows
Model definition and workflowin Resolve
Indirectly accesssimulation capabilities of Reveal
Concept (cont.)

6. Modelling Unconventional Wells
6
Shale Gas Well

7. Modelling Unconventional Wells
7
Microseismic
FracXfis 3 -5 x longer
than seen in FracPro,
proved with Microseismic
Multi-Fracs does not interfear
near wells (approx. 300m)
(Sketch not same as seen)

8. Modelling Unconventional Wells
Fracture Input Parameters
8

9. Modelling Unconventional Wells
Fracture Finite Element Grid
9

10. Modelling Unconventional Wells
History Match
10
History match a shale well
-Additional Input Data -
Young’s modulus (248000 bar).
Poisson’s ratio (0.275).
Biot’scoefficient (0.75).
Langmuir adsorption curve (70 scc/g).
Fracture permeability variation (no data).
Compaction table (Palmer & Mansoori–SPE 52607).
Fracture geometry and properties (NRT Log & Gohfer)

11. Modelling Unconventional Wells
History Match
11
History matching : Reveal model

12. Modelling Unconventional Wells
Input Parameters

13. Modelling Unconventional Wells
History Match: Reveal Model
13
Calculated
Pressure
Measured PressureRate

14. Modelling Unconventional Wells
pDtDCurves -Resolve
14
pD tD curves
Why
Amethod that allows us to perform rapid calculations
with complex well geometries
What they are
Dimensionless transient reservoir response that
captures the well, fracture and reservoir geometry

15. Modelling Unconventional Wells
pD tD Curves -Resolve
15
pD tD curves (pseudo-pressure pseudo-time) 


 0
2103266.6
3
P
p
D
B
dp
Q
kh
p

 



t
Tw
D
c
dt
r
k
t
0
2
3
102637.024


16. Modelling Unconventional Wells
pD tD Curves -Resolve
16
Optimization
Resolve -Powerful Optimizing Tool
on Field Level Production
Why Resolve
Handles large number of wells
Easy to analyze well performance
Infinite scenarios to find best match

17. Modelling Unconventional Wells
pD tD Curves
17
Use GAP with Eclipse / Reveal / pD tD
One well
Reveal
Model
One well
pdtd
curve
Two well
Eclipse
Model

18. Modelling Unconventional Wells
Resolve: Modelled vs. History

19. Modelling Unconventional Wells
19
Shale Oil Well

20. Modelling Unconventional Wells
Issues
No gas break-through seen in the simulation
What is the fracgeometry causing this behaviour?
Tight Reservoir Module uses only analytical method (pD tD)
3 Fracssimulated as one, inflow problems
Solutions
Model the fracsexplicitly with Reveal fracgridding option
Evaluate resulting pD tD curves from analytical vs. numerical
Detailed Resolve model explains where GOR increase when
pressure is below Pbat certain fracs
Oil Models: Resolve & Reveal

21. Modelling Unconventional Wells
Resolve pD tD Match
Dimensionless parameters –no geometry info

22. Modelling Unconventional Wells
Resolve Workflow

23. Modelling Unconventional Wells
Setup Resolvemodel and ran several cases to test sensitivity
on different parameters and interaction between them such as:
Porosity, Stimulated K, Matrix K
Fracture half length, Fracture height
Stimulated Rock Volume (SRV)
Compressibility
Control Reveal with Resolve
RESOLVE Logic

24. Modelling Unconventional Wells
Resolve / Reveal Sensitivities
24Case_NamePorosityPermeabilityCompressibilityFracture Half LengthFracture HeightSRVMatrixK
Case_240.04 0.0008 1.00E-05 262 82 130.0001
Case_250.04 0.0008 1.00E-05 262 82 130.0002
Case_260.04 0.0008 1.00E-05 262 82 130.0003
Case_270.04 0.0008 1.00E-05 262 82 130.0005

25. Modelling Unconventional Wells
What is different between a Resolveanalytical and a
Revealnumerical model?
Transfer fracture model from Resolve tight module to
Reveal to back calculate geometry.
Oil Reservoir Model

26. Modelling Unconventional Wells
Pressure after
one month
Reveal Oil Reservoir Model
Model variable compaction table, stress dependent
perm and changing cleat compressibility

27. Modelling Unconventional Wells
Porosity Reduction with Pressure
27/11/2013 02/03/2014 30/06/2014
Effect of Cleat Compressibility
4.01 3.96 3.90 3.65 3.80 3.74

28. Modelling Unconventional Wells
Reduction of Oil Density over Time
12/12/2013 31/01/2014 30/06/2014
Changing Oil Density over Time
41.5 40.4 39.3 38.2 37.1 36.0

29. Modelling Unconventional Wells
Porosity and compressibility have similar effects on BHP
match
Changes on SRV affects the late match
Stress -Permeability function has an effect on the transient
behaviour (early decline / build up)
BHP calculation need careful considerations due to the
instability of low rate wells.
Stress dependent permeability needs to be included in the
Reveal models to obtain a match
Reservoir Model -Conclusions

30. Modelling Unconventional Wells
30
Thank you for your attention
Questions?