Summary
Downhole fluid analysis (DFA) has emerged as a key technique for
characterizing the distribution of reservoir-fluid properties and determining
zonal connectivity across the reservoir. Information from profiling the
reservoir fluids enables sealing barriers to be proved and compositional
grading to be quantified; this information cannot be obtained from conventional
wireline logs. The DFA technique has been based largely on optical
spectroscopy, which can provide estimates of filtrate contamination, gas/oil
ratio (GOR), pH of formation water, and a hydrocarbon composition in four
groups: methane (C1), ethane to pentane (C2–5), hexane and heavier hydrocarbons
(C6+), and carbon dioxide (CO2). For single-phase assurance, it is
possible to detect gas liberation (bubblepoint) or liquid dropout (dewpoint)
while pumping reservoir fluid to the wellbore, before filling a sample
bottle.
In this paper, a new DFA tool is introduced that substantially increases the
accuracy of these measurements. The tool uses a grating spectrometer in
combination with a filter-array spectrometer. The range of compositional
information is extended from four groups to five groups: C1, ethane (C2),
propane to pentane (C3–5), C6+, and CO2. These spectrometers,
together with improved compositional algorithms, now make possible a
quantitative analysis of reservoir fluid with greater accuracy and
repeatability. This accuracy enables comparison of fluid properties between
wells for the first time, thus extending the application of fluid profiling
from a single-well to a multiwall basis. Field-based fluid characterization is
now possible.
In addition, a new measurement is introduced--in-situ density of reservoir
fluid. Measuring this property downhole at reservoir conditions of pressure and
temperature provides important advantages over surface measurements. The
density sensor is combined in a package that includes the optical spectrometers
and measurements of fluid resistivity, pressure, temperature, and fluorescence
that all play a vital role in determining the exact nature of the reservoir
fluid.
Extensive tests at a pressure/volume/temperature (PVT) laboratory are
presented to illustrate sensor response in a large number of live-fluid
samples. These tests of known fluid compositions were conducted under
pressurized and heated conditions to simulate reservoir conditions. In
addition, several field examples are presented to illustrate applicability in
different environments.
Introduction
Reservoir-fluid samples collected at the early stage of exploration and
development provide vital information for reservoir evaluation and management.
Reservoir-fluid properties, such as hydrocarbon composition, GOR,
CO2 content, pH, density, viscosity, and PVT behavior are key inputs
for surface-facility design and optimization of production strategies.
Formation-tester tools have proved to be an effective way to obtain
reservoir-fluid samples for PVT analysis. Conventional reservoir-fluid analysis
is conducted in a PVT laboratory, and it usually takes a long time (months)
before the results become available. Also, miscible contamination of a fluid
sample by drilling-mud filtrate reduces the utility of the sample for
subsequent fluid analyses. However, the amount of filtrate contamination can be
reduced substantially by use of focused-sampling cleanup introduced recently in
the next-generation wireline formation testers (O’Keefe et al. 2008).
DFA tools provide results in real time and at reservoir conditions. Current
DFA techniques use absorption spectroscopy of reservoir fluids in the
visible-to-near-infrared (NIR) range. The formation-fluid spectra are obtained
in real time, and fluid composition is derived from the spectra on the basis of
C1, C2–5, C6+, and CO2; then, GOR of the fluid is estimated from the
derived composition (Betancourt et al. 2004; Fujisawa et al. 2002; Dong et al.
2006; Elshahawi et al. 2004; Fujisawa et al. 2008; Mullins et al. 2001; Smits
et al. 1995). Additionally, from the differences in absorption spectrum between
reservoir fluid and filtrate of oil-based mud (OBM) or water-based mud (WBM),
fluid-sample contamination from the drilling fluid is estimated (Mullins et al.
2000; Fadnes et al. 2001).
With the DFA technique, reservoir-fluid samples are analyzed before they are
taken, and the quality of fluid samples is improved substantially. The sampling
process is optimized in terms of where and when to sample and how many samples
to take. Reservoir-fluid characterization from fluid-profiling methods often
reveals fluid compositional grading in different zones, and it also helps to
identify reservoir compartmentalization (Venkataramanan et al. 2008).
A next-generation tool has been developed to improve the DFA technique. This
DFA tool includes new hardware that provides more-accurate and -detailed
spectra, compared to the current DFA tools, and includes new methods of
deriving fluid composition and GOR from optical spectroscopy. Furthermore, the
new DFA tool includes a vibrating sensor for direct measurement of fluid
density and, in certain environments, viscosity. The new DFA tool provides
reservoir-fluid characterization that is significantly more accurate and
comprehensive compared to the current DFA technology.
© 2008. Society of Petroleum Engineers
View full textPDF
(
3,900 KB
)
History
- Original manuscript received:
4 June 2007
- Meeting paper published:
4 September 2007
- Revised manuscript received:
26 April 2008
- Manuscript approved:
23 May 2008
- Version of record:
29 December 2008
View Cart
