Heat Transfer in Subsea

1
Heat Transfer in Subsea Production of Oil and
Gas by O.I. Levik Lecture given in
partial fulfillment of the requirements for the
degree of Doktor Ingeniør of the Norwegian
University of Science and Technology 27.
November 2000
2

• Basic Concepts of Heat Transfer
• Subsea Production Systems
• Heat transfer in different units
• Insulation requirements
• Deep water issues
• Case studies

Topics
3
OLGA Models
Partially Buried Pipelines
u?
houter???
H?
Q?
Buried Pipelines
H
D
4

• Flowline Insulation Requirements for Deepwater
• Rubel et al., Texaco (1994), SPE 28481
• Finding U and design to maintain the produced
  fluids above the hydrate temp. using PIPEPHASE.
• Internal convective resistance assumed
  negligible.
• Many thermal insulators not compatible to deep
  water, thus Pipe-in-Pipe (PIP) is a solution.
• The categories of flowlines in this study

Double carrier Single carrier No carrier
5
Flowline Insulation Requirements for
Deepwater Rubel et al., Texaco (1994), SPE 28481

• XXXXXX SLØYFE DENNE PLANSJEN?
• Conclusions
• For an uninsulated, half buried flow line the
  fluid
• temperature drops below the hydrate
  temperature.
• In the early phase with partial production and
  low
• watercut, only PIP with vacuum annulus gives
• the required protection
• Urequired 5Uvacuum 0.1 BTU/hr-ft2-F.
• During full production with a high watercut the
• required insulation is nearly obtained using a
  single
• carrier with the annulus filled with high
  performance
• insulation, e.g. kerosene gel
• Urequired 0.77Ugel 1.0 BTU/hr-ft2-F.

6
Gullfaks Satellites Operational Experience T.W.
Knudsen, Statoil (1999), SPE 56909
Rimfaks and Gullveig 11 km PIP with
polypropene coatings and 25 mm Rockwool
insulation. Very high insulation requirement 2
W/Km2. Gullfaks South 4.3 6.4 km bundle.
Production and hot water lines in sleeve pipe.
Gas and methanol injectors in outer carrier
pipe. Sleeve pipe insulated with 20 mm melanine
and press- urized with nitrogen to 19 bara for
mechanic support. Carrier pipe filled with
water. I STEDET FOR DENNE SIDEN FIGURENE 2, 4
OG 5 MED KOMMENTARER SOM ANGITT OVER OG NOTERT
FOR HÅND I PAPERET
7
Gullfaks Satellites Operational Experience T.W.
Knudsen, Statoil (1999), SPE 56909

• Simulations Ueffective 1.5 W/Km2 for
  producers.
• with 20 mm melamine assuming k 0.035 W/Km.
• Measurements Imply that k 1.1 W/Km.
• Poor thermal bundle performance because...
• Significant free convection in N2 gives heat
  loss.
• High pressure drop in hot-water line gives low
  rate.
• Design failure or damages?
• Revamping alternatives
• Replacing nitrogen with argon to reduce U by 15.
• Replacing nitrogen with cement at 1 bara.
• Gravel and sand dumping.
• Upgrading the hot-water circulation system.
• Conclusion
• Hydrate may form due to poor thermal performance.
• Current (1999) heat transfer models for complex.
• bundles are not reliable.

8
Thermal performance of insulated heated
bundle Chin et al., Kvarner/Amerada-Hess (2000),
SPE 58971
9
Agbami Manifold Transient Thermal Analysis A.B.
Sandøy, ABB, 2000
Water depth Sea water temp 1355 m 4.3 -
4.9 C. Manifold components Pipe, bends, tees,
valves. Basic assumptions Only radial heat
transfer from pipe. Pipe insulation thickness is
sufficient for valves too. Requirement Given
the initial temperature (63.9C) and the shutdown
duration (8 h), the minimum allowed temperature
(22.8 C) must not be violated. One-dimentional
Transient Working Equation Ui and Ai from
empirical engineering relations.
10
Special Problems from Sub-zero Water Temperatures
in Ormen Lange Many questions - still no
answers T(600m) 0 C T(bottom) -1 to -2
C, constant around the year T(bottom)
T(freezing point) T(freezing point, sea water,
3.5 salt) -1.87 C T(freezing point, formation
water, 0.7 salt) gt -1.87C T(freezing point,
condensed water, 0 salt) 0 C Conclusion
Ice formation - perhaps. Experiments are
inconclusive. Hydrate pressure at seawater
temperatures 4-6 bara Liquid coloumn static
pressures gt 4-6 bara. Conclusion Hydrate
formation in static system - perhaps. Depends on
success of liquid removal (head
removal). Actions MEG injection in long
pipelines ? Insulation and heating of short
pipelines ?
Ice ?
Hydrate ?
11
Thermal Performance of Subsea Equipment G.J.
Zarbas et al., Shell (1998)
Method General 2D finite-element PDE
solver. Steady-state and transient
simulations. Assumptions P gt Pbubblepoint , 3D
effects ignored. PIP Joints Well-covered joints
cause 20 increase in Uoverall Poorly-covered
joints cause 70 increase in Uoverall Conclusion
The joint covers must well overlap the joint.
Xmas-trees Cooldown time to hydrate temperature
upon shutdown. Typical design criteria 8-12
hours. Intuitive common belief Several
hours. Simulation of uninsulated subsea tree 1-3
hours. Conclusion Insulation of subsea
Xmas-trees is imperative.
12
Acknowledgments Alphabetical order
Idar Grytdal, Statoil Torbjørg Heskestad,
Statoil Henning Holm, Norsk Hydro Jan
Ingebriktsen, ABB Trygve Kløv, NTNU Roar
Larsen, SINTEF Ole Jørgen Nydal, NTNU Anne-Beth
Sandøy, ABB Pål Skalle, NTNU Erik Skjetne,
Statoil Elling Sletfjerding, Statoil