Summary
There remains considerable debate about the impact of fracture-face-matrix
damage on cleanup and productivity of gas wells. The impact of the Cinco-Ley
fracture-face-skin factor has been studied extensively, normally with
single-phase-flow examples. Two-phase-flow examples are usually limited to the
impact on long-term production. This paper uses a new two-phase-flow simulator
to demonstrate the impact of fracture-face-matrix damage on both
fracture-treatment cleanup and fracture-face-skin evolution during cleanup and
subsequent production. The simulator was validated by comparison of calculated
skins with pressure-buildup simulations evaluated by use of a high-end,
third-party, pressure-transient-analysis package.
The study has demonstrated that fracture-face skin relative to gas flow can
be calculated continually throughout a simulation of fracture-treatment cleanup
and production. It was found that at lower matrix permeabilities and subsequent
higher capillary pressure curves, the impact of water saturation in the damage
zone becomes much more important. Specifically, the effective fracture-face
skin relative to gas can be several times higher than expected on the basis of
single-phase flow. Furthermore, the simulation results show that at lower
matrix permeabilities, the time required to achieve a reasonable fracture-face
skin relative to gas flow can require considerable production time, on the
order of several weeks, even for moderate damage factors. The results
demonstrate that in tight gas reservoirs (≈0.01 md and less), even a moderate
amount of matrix damage in a fracture face can result in high fracture-face
skins and exceedingly long times for treatment cleanup. As such, it becomes
very important to minimize fracture-face-matrix damage during tight gas
fracturing treatments.
Introduction
The issue of whether fracturing fluids damage the productivity of
propped-fracture wells is at least 50 years old (van Poollen 1957). Yet, even
today, our industry is concerned with the causes of and remedies for the slow
cleanup of fracturing treatments in low-permeability gas wells. Two of the main
areas of focus are the conductivity of the propped fracture and flow impairment
in the fracture-face matrix. Numerical modeling has been a mainstay of the
efforts to understand the processes that occur in the formation during and
after a fracturing treatment (Soliman and Hunt 1985; Iqbal and Civan 1993;
Neghaban et al. 1998; Yi 2004; Montogomery et al. 1990; Sherman and Holditch
1991). A recent analysis has demonstrated that the evaluation of and causes of
inadequate production of fracturing treatments are quite complex (Barree et al.
2005).
This paper will focus on issues within the fracture-face matrix, progressing
through two major sections. The first describes the new simulator and documents
the validation process. The process included production comparison with an
in-house simulator and pressure-transient analysis with a commercially
available software package. Pressure-transient analysis was quite effective for
validation of finite-conductivity-fracture properties.
The second section describes application of the new simulator to evaluate
single-phase and two-phase fracture-face damage, with results from multiple
studies of relative permeability and capillary pressure effects. Of particular
interest was the impact of two-phase flow on the observed fracture-face skins
and on cleanup of the matrix after a fracturing treatment. Numerical
simulations were conducted to demonstrate the potential impact of clay swelling
and clay dispersion as damage mechanisms in the fracture-face invaded zone.
These two damage mechanisms were chosen to potentially represent separate cases
in which damage increases the capillary pressure and in which damage does not
increase the capillary pressure in the invaded zone. These results lead to a
final discussion of fracture cleanup and its impact on productivity.
© 2009. Society of Petroleum Engineers
View full textPDF
(
1,432 KB
)
History
- Original manuscript received:
26 June 2006
- Meeting paper published:
24 September 2006
- Revised manuscript received:
22 July 2008
- Manuscript approved:
7 August 2008
- Published online:
2 March 2009
- Version of record:
26 February 2009
View Cart
