Summary
This article presents the detailed formulation for each of the three steps
of a horizontal gravel-pack displacement operation, including sand injection
and alpha/beta waves propagation. The main core of the model, aiming to define
alpha wave height, is based on a well known two-layer model. Initially
developed for hydrotransport applications, this kind of model has been adapted
by several authors for drilled cuttings transport analysis. Additionally, a
comparison between theoretical predictions and pumping charts from a field
operation performed in Campos basin is presented.
Introduction
Gravel packing is today the most frequently applied sand control technique
in Campos Basin, offshore Brazil. Because of the critical conditions, such as
the deep and ultradeep waters and low fracture gradients, great precision is
required to assure gravel-packing success. Several models available in the
industry for horizontal gravel pack design are essentially empirical, resulting
in imprecise predictions for extrapolated conditions.
These aspects were the primary motivators for the development of a
mechanistic model to describe the whole operation. It is a consensus among
design and operation engineers that a physically based software is a necessary
rigsite tool for determining operational parameters, especially when
last-minute data have to be considered. Reliable and fast results are required
to enhance the chances of a successful operation.
Several authors present experimental results of horizontal gravel packing
performed in test facilities: Forrest1 presents a correlation to estimate pack
length limits in highly inclined and horizontal wells based on a full-scale
model wellbore tests with viscous fluids and water.
Nguyen et al.2 developed a 3D numerical simulator based on the
finite-volumes method, which can monitor the transport process of the slurry in
both axial and angular directions. Conservation of mass and momentum is
considered in each sector element (finite-volume cell) to evaluate its fluid
transport process. Each sector element is assumed to process homogeneous
properties within its control volume. The model considers rheological
properties of fluid, effect of gravel settling, and friction pressure
calculation.
Penberthy et al.3 present several field tests in a 1,500-ft-long simulator
to identify the main variables that govern the phenomenon. Extensive
field-scale testing has aided in the development of procedures and operational
guidelines that are still today relevant. Software has also been developed that
is based on correlations to determine gravel transport velocity and mechanistic
models to determine pressure drop and friction factor.
Sanders et al.4 present a numerical model based on a pseudo-3D approach
aiming to simulate of an alternative flow path concept during the horizontal
gravel-pack placement. The model solves the equations of volume and
momentum conservation for the incompressible slurry in the wellbore. In
order to validate the flow-path concept both small-scale and large-scale
experimental tests using models ranging from 5 to 1,000 ft in length were
performed.
The dynamics of data acquisition to run gravel-packing simulations requires
continuous updating, and part of the information is accurately available only a
few hours before pumping starts. Other data, such as detailed caliper
information, remain unavailable in several cases. Considering the several input
uncertainties for the process, the necessity to run simulations in a short
time, and the limited processing capacity of portable computers, a major
premise for the development was to consider simplified models that could
fullfil the operational requirements. Of course, such models should be able to
capture the major phenomena governing the process and predict pressures
properly.
© 2005. Society of Petroleum Engineers
View full textPDF
(
1,784 KB
)
History
- Original manuscript received:
12 August 2003
- Revised manuscript received:
30 May 2005
- Manuscript approved:
30 June 2005
- Version of record:
15 September 2005
View Cart
