GEORG Geothermal Workshop 2016 SESSION B3, Midstream. Presentation title: modelingStrmodelling of casings in high-temperature geothermal wells

1. Structural modeling of casings in high
temperature geothermal wells
12/20/2016
#GGW2016
Gunnar Skúlason Kaldal (gsk@isor.is)
PhD student at University of Iceland / Engineer at ÍSOR – Iceland GeoSurvey
Magnús Þór Jónsson, Halldór Pálsson and Sigrún Nanna Karlsdóttir
University of Iceland

2. Introduction
• PhD project at University of Iceland
• Structural models of casings for evaluating well integrity and casing
failure modes
• Examples FEM analyses and results
• Thermal expansion is one of the most severe structural concerns in
high temperature geothermal wells
• Cemented steel casings are constrained by the cement and high forces
generate plastic (permanent) deformations as the casings warm up
• Failure modes from thermal expansion (and contraction) include:
• Casing collapse (in the form of a bulge/pucker)
• Tensile rupture where the casing (pin) is teared out of the coupling (box) by the
threads or the pipe body
#GGW2016

3. High temperature
geothermal wells
• Typical casing program includes 3 casings (API grades)
• Conditions sometimes call for more casings
• Casings are cemented over their full length
• Perforated liner supports the wellbore in the
production section of the well
• Expansion spool is used to allow thermal expansion of
the production casing at the wellhead
Fig: Sigrún Nanna Karlsdóttir
#GGW2016

4. High temperature
geothermal wells
• The design of high temperature geothermal wells is
based on API materials and methods, and knowledge
gained over the past decades in the geothermal
industry
• The design procedure for typical high temperature
geothermal wells is good and failures are not very
common in conventional wells
• There are however several exceptions
• Many parameters influence success of wells
• “Structural success” is an important one, for usability
and overall reliability and safety
#GGW2016

5. Design challenges for the future
• Most casing failures that occur in wells are directly related to
large temperature changes and high annular pressure
• Typical wellhead temperatures in high temperature geothermal
wells is 200-300°C
• In IDDP-1, still the hottest recorded well to date, superheated
steam was produced at the wellhead with temperatures of
450°C
• Future aim is to produce from supercritical source where
temperatures could reach as high as 550°C – IDDP-2?
• This provides new challenges in casing design as design
standards do not account for these high temperatures
#GGW2016

6. Failure modes
• Collapse (bulge/pucker) – pressure difference/thermal expansion
• Tensile rupture – thermal expansion (axial)
#GGW2016
HS Orka / ÍSOR
Adopted from a diagram by Rahman & Chilingarian, 1995

7. FEM Modeling
• The nonlinear behavior of materials, displacements
and friction between contacting surfaces are solved
with numerical methods.
• The (Nonlinear) Finite Element Method is used.
• Thermal and structural models of the cased section of
the well.
• The models are used to evaluate the structural
integrity of the casings when subjected to transient
thermo-mechanical loads.
• Three models presented:
– Cased section of the well (2D axi-symmetric)
– Connection in concrete (2D axi-symmetric)
– Section of the well (3D collapse analysis)
i.
ii.
iii.
#GGW2016

8. FEM results 1 2
3 4
700m
100 m
• Production history modeled.
• T-P logs and wellhead data are used as load.
• Transient thermal analysis is performed and
the results used as load in the structural
analysis.
1. Cooling due to drilling.
2. Thermal recovery.
3. Discharge (12 min).
4. Discharge (3 months).
◦C
#GGW2016

9. • The wellhead rises as the casing suddenly
warms up during discharge.
• The production casing expands and slides
inside the wellhead (expansion spool). Wellhead displacement.
Temperature distribution
after 9 days of discharge.
Stress-strain curves are implemented for steel. Stress reduction at elevated
temperatures is accounted for by scaling E, σy and σu (acc. to Snyder). Friction is defined between
casings and concrete.
Karlsdottir, S.N. and
Thorbjornsson, I.O.,
2009
FEM results #GGW2016

10. Wellhead displacement survey
Photographic series of the wellhead of HE-46 during discharge.
52 mm40 mm
Merged photographs of the wellhead of RN-32 after 9 days of discharge.
26
mm
#GGW2016

11. • Wellhead displacement measured during discharge.
Kaldal, G.S., Jonsson, M.T., Palsson, H., Karlsdottir, S.N.,
2015. Structural modeling of the casings in high
temperature geothermal wells. Geothermics 55, 126 –
137.
Wellhead displacement survey
#GGW2016

12. • Modeled wellhead displacement compared to
data.
Model results
Kaldal, G.S., Jonsson, M.T., Palsson, H., Karlsdottir, S.N.,
2015. Structural modeling of the casings in high
temperature geothermal wells. Geothermics 55, 126 –
137.
#GGW2016

13. • Anchoring of couplings in concrete.
• Large stresses are produced near couplings.
Concrete failure
Model i. Cased section of the well Model ii. Coupling in concrete
#GGW2016

14. • Connection displacement in concrete.
• Upward displacement of 5 mm.
Damaged concrete
Concrete failure

15. • Case study: structural analysis of IDDP-1.
• Operation history modeled.
• Initial conditions.
• Warm-up.
• Discharge.
• Shut-in and quenching.
Kaldal, G.S., Jonsson, M.T., Palsson, H., Karlsdottir, S.N.,
2015. Structural modeling of the casings in high
temperature geothermal wells. Geothermics 55, 126 –
137.
Discharge history of IDDP-1
FEM results #GGW2016
Ingason et.al, Geothermics 2013:

16. • Stress and strain analysis. Discharge phase V of IDDP-1
Anchor casingProduction casing
FEM results #GGW2016

17. • Permanent strain is generated in the casings
during the operation history.
Anchor casing
Production casing
Discharges: Cyclic stress-strain and temperature of
the production casing at 50 m depth.
FEM results #GGW2016

18. • Collapse analysis of the production casing.
• Some instability needs to be introduced.
Collapse analysis
0 5 10 15 20 25 30 35 40 45
0
10
20
30
40
50
60
70
80
90
100
D/t ratio
K55Collapsepressure[MPa]
Yield strength collapse
Plastic collapse
Transition collapse
Elastic collapse
9 5/8 (47.0 lb/ft)
13 3/8 (68.0 lb/ft)
ISO/TR 10400:
Eigenvalue buckling analysis (theoretical collapse strength).
1
5
2
6
3
7
4
8
Casing: OD = 13 3/8 in, t = 12.2 mm
API collapse resistance: 13.4 MPa
 Eigenvalue buckling analysis (theoretical collapse strength).
 Nonlinear buckling analysis (includes nonlinearities).
 Effect of initial geometry; mode shape perturbation, effect of
ovality and external geometric defect.
 Collapse shape with and without external concrete support.
#GGW2016

19. Nonlinear buckling analysis.
Other defects:
 Mode shape perturbation
 Ovality
 External defect
 Water pocket in concrete
Casing: OD = 13 3/8 in, t = 12.2 mm
API collapse resistance: 13.4 MPa
• Limit load for a perfectly round casing: 38.4 MPa
• Limit load using mode shape perturbation: 21.6 MPa
• API collapse resistance: 13.4 MPa
0 10 20 30 40 50 60 70 80
0
10
20
30
40
50
60
Load,externalpressure[MPa]
UX displacement [mm]
Perfectly round casing
1st mode shape perturbation (0.0005 scaling)
1st mode shape perturbation (0.001 scaling)
Collapse resistance, 13.4 MPa (API, ISO/TR)
Elastic collapse (Timoshenko 1961)
Mode shape perturbation
Dmax
Dmin
Effect of ovality
0 10 20 30 40 50 60 70 80
0
10
20
30
40
50
60
Load,externalpressure[MPa]
UX displacement [mm]
Perfectly round
Ovality (0.1%)
Ovality (0.5%)
Ovality (1.0%)
Ovality (2.0%)
Ovality (3.0%)
Collapse resistance
Elastic collapse
Von Mises stress at collapse: 440 MPa
Collapse at 300°C and 20 bar (wall pressure)
Water pocket in concrete
Collapse analysis

20. Nonlinear buckling analysis
0 50 100 150 200 250 300
0
10
20
30
40
50
60
Displacement [mm]
Load,externalpressure[MPa]
Concrete support (linear MP)
Without concrete support (linear MP)
Concrete support (non-linear MP)
Without concrete support (non-linear MP)
Collapse resistance, 13.4 MPa (API, ISO/TR)
Elastic collapse (Timoshenko 1961)
Casing: OD = 13 3/8 in, t = 12.2 mm
API collapse resistance: 13.4 MPa
Effect of external defect and concrete support
Collapse analysis #GGW2016

21. Summary and conclusions
• Analyses of the casings in high temperature geothermal wells were presented.
• High temperature and pressure differences generate many challenges in geothermal wells.
• Thermal expansion is one of the major cause of casing failures.
• Three models were presented here:
• (i) the cased well, (ii) detailed coupling in concrete and (iii) 3D model of a section (collapse analysis).
• The models are used to evaluate the structural integrity of casings.
• Can be used to analyze various load scenarios and material selections.
• Conclusions…
#GGW2016

22. Conclusions - Collapse
• Caused by excessive net external pressure
• Impurities and geometry (ovality, eccentricity, material..)
• Cement integrity and casing roundness (and other defects) have great
effect on collapse resistance of casings
• Cement is very important for lateral and radial support
• Collapse resistance can be increased by selecting proper materials (HC)
under strict quality control (casing roundness important)
• Biaxial loads affect collapse resistance
HS Orka / ÍSOR
#GGW2016
Dall‘Acqua et.al 2012 Burst and Collapse Responses of
Production Casing in Thermal Applications
Wu et.al 2008 Casing Failures in Cyclic
Steam Injection Wells
• Collapse strength reduction due to axial
tension is incorporated into API standards
(compression not)
• Axial compression plus net external
pressure probably also leads to reduction in
collapse resistance (is not well known or
standardized)

23. Conclusions – Tensile rupture
• Caused by excessive axial tensile stress when casings cool down
after production, i.e. due to long period shut-in or killing
operations
• The phenomenon is well understood, but limits of cooling rate
or limit in temperature variations ∆T remains unresolved
• The failure mode has occurred in several wells in Iceland in
connection to fast cooling while pumping cold water into a hot
well
• FEM analyses indicate that:
• Failures are more likely to form near changes in outer casings, e.g. at
material grade changes (T95-K55) and near casing shoes
• Thermal gradient between casing layers leads to thermal expansion
mismatch which generates stress/strain
• Slow temperature changes have less consequences than fast ones
• The thermal load is more severe for the innermost casing which is in direct
contact to the geothermal fluid than external casings (provided that
cementing in between is good)
HS Orka / ÍSOR
#GGW2016
Adopted from a diagram by Rahman & Chilingarian, 1995

24. What further information is needed?
• Material integrity at high temperatures >350°C
• Strength reduction
• Corrosion
• Creep and stress relaxation
• Can wells for 300°C+ be designed within the elastic
region of materials?
• Is it possible to quench/cool a >550°C hot well
without causing casing failures?
• What can we do to mitigate thermal expansion?
A well known problem of thermal expansion (ΔT day/night)
Conclusions – Looking ahead #GGW2016

25. Acknowledgements
• The University of Iceland research fund
• The Technology Development Fund at RANNIS –The Icelandic Centre
for Research
• Landsvirkjun – Energy Research Fund
• GEORG – Geothermal Research Group
• Reykjavik Energy, ON, HS Orka, Landsvirkjun, Iceland Drilling, Iceland
GeoSurvey (ÍSOR), Mannvit and the Innovation Center Iceland.
#GGW2016

26. Thank you
#GGW2016