|
|
Annular displacement flows of non-Newtonian fluids in Primary
Cementing
|
|
|
|
These flows are at the heart
of the primary cementing process. In this process a sequence of non-Newtonian
fluids are pumped in sequence along a narrow eccentric annulus. Typically the
cement slurry is preceded by a spacer fluid and/or wash, the purpose of which
is to effectively displace the drilling mud from the well and to act as a
buffer between drilling mud and cement. The fluid stages pumped are
typically very long, relative to the annular radii, and variations in the
annular geometry occur relatively slowly along the length of the well.
Therefore, the basic flow to understand is the displacement of one non-Newtonian
fluid by another, along a uniform inclined eccentric annulus. |
Figure 1: Snapshots from a typical displacement experiment in the
6m UBC annulus. With poor choice of fluids and operating conditions there is
a tendency to finger on the wide side.
Contributors: -
M.
Carrasco-Teja -
I.
Frigaard -
M.
Martinez -
M.
Moyers-Gonzalez -
S.
Pelipenko -
S.
Sood -
S.
Storey |
|
|
|
|
|
|
In
an eccentric annular geometry fluid velocities on the wide side of the annulus
are faster than those on the narrow side. Unless the fluid properties are
carefully chosen there is a tendency for the displacing fluid to finger
through the displaced fluid on the wide side of the annulus, leaving behind
fluid on the narrow side, (see Figure 1, right). In
the worst case, the fluid on the narrow side of the annulus may even remain
static, due to having a relatively large yield stress. This situation
corresponds industrially to the formation of a static mud channel. After the displacement
is complete these mud channels remain in place while the cement slurry
hardens and eventually form porous conduits connecting subsurface strata.
These channels may often be observed on logs of poorly cemented wells. Via
careful modeling and experimental study, it is possible to understand
quantitatively the situations when unsteady displacements (such as in Figure 1) occur, when static mud
channels form and when we have a steady displacement. This latter case is the
ideal situation for primary cementing, in which the entire displacement front
advances along the annulus at the same speed at each azimuthal
position. The
basis of our modeling work on laminar displacements is the recognition that
the annulus is relatively narrow and may consequently be treated as a Hele-Shaw cell, which reduces the problem from 3 spatial
directions to 2D, see Figure
2. |
|
|
|
|
|
|
|
Figure 2: Schematic of the Hele-Shaw
approach when we “unwrap” the eccentric annulus and
average across the annular gap. |
|
|
Our 2D modeling work allows
fast simulation of complete primary cementing jobs and evaluation of the job
design. It is also possible to evaluate job designs without simulation.
Extensions of this work include: -
Study of
interfacial instabilities during displacement -
Special
treatments necessary for horizontal displacements -
Effects of
cement pulsation -
Simulation of
the effects of casing reciprocation and rotation |
||
Relevant publications:1.
2.
S. Pelipenko, MSc thesis, 3.
S. Pelipenko and I.A.Frigaard, “On steady
state displacements in primary cementing of an oil well.” Journal of
Engineering Mathematics, 48(1), pp. 1-26, (2004). 4.
S. Pelipenko and I.A.Frigaard, “Effective
and Ineffective Strategies for Mud Removal and Cement Slurry Design.” Society
of Petroleum Engineers paper number: SPE 80999, April (2003). 5.
S. Sood, MASc
thesis, 6.
S. Pelipenko and I.A.Frigaard,
“Two-dimensional computational simulation of eccentric annular cementing
displacements.” IMA Journal of Applied Mathematics, 69: pp. 557-583,
(2004). 7.
S. Pelipenko and I.A.Frigaard, "Visco-plastic fluid displacements in near-vertical
eccentric annuli: lubrication modelling." J.
Fluid Mech. 520, pp. 343-377, (2004). 8.
M. Moyers-Gonzalez, PhD thesis, |
||
|
Contact: Ian Frigaard for more details |
