U.S. patent application number 10/104449 was filed with the patent office on 2003-09-25 for apparatus and method for determining density, porosity and fluid saturation of formations penetrated by a borehole.
Invention is credited to Odom, Richard C., Tiller, Donald E., Wilson, Robert D..
Application Number | 20030178560 10/104449 |
Document ID | / |
Family ID | 28040598 |
Filed Date | 2003-09-25 |
United States Patent
Application |
20030178560 |
Kind Code |
A1 |
Odom, Richard C. ; et
al. |
September 25, 2003 |
Apparatus and method for determining density, porosity and fluid
saturation of formations penetrated by a borehole
Abstract
A borehole logging system for determining bulk density, porosity
and formation gas/liquid fluid saturation of formation penetrated
by a borehole. Measures of fast neutron radiation and inelastic
scatter gamma radiation, induced by a pulsed neutron source, are
combined with an iterative numerical solution of a two-group
diffusion model to obtain the formation parameters of interest.
Double-valued ambiguities in prior art measurements are removed by
using the iterative solution of the inverted two-group diffusion
model. The system requires two gamma ray detectors at different
axial spacings from the source, and a single neutron detector
axially spaced between the two gamma ray detectors. The system can
be embodied as a wireline system or as a logging-while-drilling
system.
Inventors: |
Odom, Richard C.; (Benbrook,
TX) ; Tiller, Donald E.; (Fort Worth, TX) ;
Wilson, Robert D.; (Fort Worth, TX) |
Correspondence
Address: |
Patric H. McCollum
PD Holdings (USA) Inc.
Suite 1700
363 N. Sam Houston Parkway East
Houston
TX
77060
US
|
Family ID: |
28040598 |
Appl. No.: |
10/104449 |
Filed: |
March 19, 2002 |
Current U.S.
Class: |
250/269.2 ;
250/269.4; 250/269.6; 250/269.7 |
Current CPC
Class: |
G01V 5/104 20130101 |
Class at
Publication: |
250/269.2 ;
250/269.4; 250/269.6; 250/269.7 |
International
Class: |
G01V 005/10 |
Claims
What is claimed is:
1. A method for determining at least one parameter of material,
comprising the steps of: (a) irradiating said material with a
source that emits neutrons; (b) measuring neutron radiation at a
first axial spacing from said radiation source; (c) measuring gamma
radiation at a second axial spacing and at a third axial spacing
from said radiation source; and (d) combining said measures of said
neutron radiation and said gamma radiation in an iterative solution
of at least a two-group radiation diffusion model to obtain a
single-valued measure of said at least one parameter.
2. The method of claim 1 wherein: (a) said material is earth
formation penetrated by a borehole; and (b) said source produces
fast neutrons; and (c) a plurality of said parameters are
determined.
3. The method of claim 1 wherein said neutron radiation is fast
neutron radiation and said gamma radiation is inelastic scatter
gamma radiation.
4. The method of claim 1 wherein said third axial spacing is
greater than said first axial spacing, and said first axial spacing
is greater than said second axial spacing.
5. The method of claim 2 wherein said plurality of parameters
comprise formation bulk density, formation porosity, and formation
fluid saturation.
6. The method of claim 1 wherein: (a) said neutron radiation is
measured with a first type of radiation detector; (b) said gamma
radiation is measured with a second type of radiation detector; and
(c) said first and said second types of radiation detectors and
said neutron source are disposed within a logging tool.
7. The method of claim 6 further comprising the step of conveying
said logging tool within said borehole with a wireline.
8. The method of claim 6 further comprising the step of conveying
said logging tool within said borehole with a drill string.
9. A method for determining a plurality of parameters of earth
formation penetrated by a borehole, comprising the steps of: (a)
irradiating said formation with fast neutrons produced by a fast
neutron source that is sequentially pulsed; (b) during a sequential
pulse, measuring fast neutron radiation at a first axial spacing
from said source; (c) during said sequential pulse, measuring
inelastic scatter gamma radiation at a second axial spacing and at
a third axial spacing from said source; (d) iterating a numerical
solution of a two-group radiation diffusion model to obtain a
linearized relationship between a fast neutron diffusion
coefficient and said measure of fast neutron radiation, wherein
said linearized relationship is independent of fluid saturation,
fluid type and pore space volume of said formation; and (e)
combining said fast neutron radiation measurement and said
inelastic scatter gamma ray measurements with said linearized
relationship to obtain a measure of bulk density of said
formation.
10. The method of claim 9 comprising the additional step of
combining said measure of bulk density with said fast neutron
radiation measurement to obtain a measure of liquid saturation of
said formation.
11. The method of claim 10 comprising the additional step of
combining said measure of bulk density with said measure of liquid
saturation to obtain a measure of porosity of said formation.
12. The method of claim 9 wherein said third axial spacing is
greater than said first axial spacing, and said first axial spacing
is greater than said second axial spacing.
13. The method of claim 9 wherein: (a) said fast neutron radiation
is measured with a first type of radiation detector; (b) said
inelastic scatter gamma radiation is measured with a second type of
radiation detector; and (c) said first and said second types of
radiation detectors and said neutron source are disposed within a
logging tool.
14. The method of claim 13 further comprising the step of conveying
said logging tool within said borehole with a wireline.
15. The method of claim 13 further comprising the step of conveying
said logging tool within said borehole with a drill string.
16. A system for determining a plurality of parameters of earth
formation penetrated by a borehole, comprising: (a) a logging tool
comprising a pressure housing in which are disposed (i) a source of
neutrons, (ii) a neutron detector disposed at a first axial spacing
from said source, (iii) a first gamma ray detector disposed at a
second axial spacing from said source, and (iv) a second gamma ray
detector disposed at a third axial spacing from said source; and
(b) a processor which combines gamma radiation measured by said
first and said second gamma ray detectors and neutron radiation
measured by said neutron detector using an iterative numerical
solution of a two-group radiation diffusion model to obtain said
plurality of parameters.
17. The system of claim 16 further comprising: (a) a conveyance
unit located at the surface of the earth; and (b) a conveyance
member with a first end operationally connected to said logging
tool and a second end operationally connected to said conveyance
unit.
18. The system of claim 16 wherein: (a) said neutron source is
operated to emit sequential pulses of fast neutrons, (b) said
neutron detector measures fast neutron radiation during a pulse of
fast neutrons; and (c) said gamma ray detectors measure inelastic
scatter gamma radiation during a pulse of fast neutrons.
19. The system of claim 16 wherein: (a) said iterative solution
linearizes a relationship between said measure of fast neutron
radiation and a fast neutron diffusion coefficient; and (b) said
fast neutron radiation measurement and said inelastic scatter gamma
ray measurements are combined with said linearized relationship to
obtain a measure of bulk density of said formation.
20. The system of claim 16 wherein said third axial spacing is
greater than said first axial spacing, and said first axial spacing
is greater than said second axial spacing.
21. The system of claim 16 wherein said conveyance member comprises
a wireline.
22. The system of claim 16 wherein said conveyance member comprises
a drill string.
23. The system of claim 16 wherein said neutron detector measures
fast neutron radiation during a pulse of fast neutrons and said
gamma ray detectors measure inelastic scatter gamma radiation
during the same pulse of fast neutrons.
24. A method for determining a parameter of material from a measure
of neutron radiation at a first axial spacing from a neutron source
and two measures of gamma radiation at a second axial spacing and
at a third axial spacing from the source, said method comprising
combining said measures of said neutron radiation and said gamma
radiation in an iterative solution of a two-group radiation
diffusion model to obtain single-valued measure of said
parameter.
25 A method for determining a plurality of parameters of earth
formation from a measure of fast neutron radiation at a first axial
spacing from a neutron source and two measures of inelastic scatter
gamma radiation at a second axial spacing and at a third axial
spacing from the source, said method comprising: (a) iterating a
numerical solution of a two-group radiation diffusion model to
obtain a linearized relationship between a fast neutron diffusion
coefficient and said measure of fast neutron radiation, wherein
said linearized relationship is independent of fluid saturation,
fluid type and pore space volume of said formation; and (b)
combining said fast neutron radiation measurement and said
inelastic scatter gamma ray measurements with said linearized
relationship to obtain a measure of bulk density of said
formation.
26. A system for determining a plurality of parameters of earth
formation with a logging tool which induces and measures neutron
radiation and gamma radiation, said system comprising a processor
which combines two axially spaced measures of said gamma radiation
and a measure of said neutron radiation using an iterative
numerical solution of a two-group radiation diffusion model to
obtain said plurality of parameters.
Description
BACKGROUND OF THE INVENTION
[0001] 1. Field of the Invention
[0002] This invention is directed toward logging of earth
formations penetrated by a borehole, and more particularly directed
toward the determination of formation density, formation porosity
and formation fluid saturation from measures of fast neutron and
inelastic scatter gamma radiation induced by a pulsed source of
fast neutrons. A two-group diffusion model is employed which
defines a distributed source of gamma radiation within the
formation resulting from fast neutrons impinging upon and
inelastically scattering from nuclei within the formation.
Ambiguities in prior art measurements are removed by using
iterative solutions of the inverted two-group diffusion model.
[0003] 2. Background of the Art
[0004] Density logging systems, which are compensated somewhat for
the effects of the borehole, were introduced in the mid 1960s in
the paper "The Physical Foundation of Formation Density Logging
(Gamma-Gamma)", J. Tittman and J. S. Wahl, Geophysics, Vol. 30, p.
284, 1965. The system introduced by Tittman et al., commonly
referred to as a compensated gamma-gamma density logging system,
was designed to operate in boreholes which are "open" and contain
no steel casing. An instrument or "tool" is lowered into the well
borehole on a cable, and the depth of the tool is determined by the
amount of cable deployed at the surface of the earth. This type of
tool contains an intense gamma-ray point source and preferably two
gamma-ray detectors at differing distances from the source. The
gamma ray detectors measure gamma rays that are scattered from
electrons in the formation, and back into the borehole. Since for
most earth formations, the electron density is in constant
proportion to mass bulk density, the "backscatter" gamma ray
intensity at the detectors provides a measure of formation bulk
density. Two detectors at different axial spacings are preferably
employed to allow the measurement to be compensated for the effect
of mudcake that tends to accumulate on the borehole wall from
drilling fluid used in the drilling process. The gamma-gamma
density tool has a characteristic shallow depth of investigation
into the formation of about 4 inches ("Depth of Investigation of
Neutrons and Density Sondes for 35% Porosity Sand", H. Sherman and
S. Locke, Proc. 16th Annu. SPW1LA Symp., Paper Q, 1975) and
therefore is heavily influenced by the near borehole environment.
This tool cannot make quantitative density logs in boreholes that
have been cased, where the casing is typically steel and is
surrounded by a cement sheath.
[0005] One technique for measuring formation porosity utilizes a
porosity sensitive tool known in the industry as a
"neutron-neutron" porosity system ("Dual-Spaced Neutron Logging for
Porosity", L. Allen, C. Tittle, W. Mills and R. Caldwell,
Geophysics Vol. 32, pp. 60-68, 1967). A two-group neutron diffusion
model is used in developing the response of the logging system. The
first group is an epithermal neutron group, and the second group is
a thermal neutron group. The downhole tool portion of the system
contains a source of fast neutrons, which is typically an isotopic
source such as Americium-Beryllium (AmBe). At least two detectors
sensitive to thermal or epithermal neutrons are axially spaced from
the source at different distances. The detectors respond primarily
to thermal or epithermal neutrons backscattered into the borehole
by the formation. The measured backscatter flux is, in turn,
primarily a function of the hydrogen content of the formation. If
it is assumed that most hydrogen within the formation is contained
in water or hydrocarbon in the pore space, the detectors respond to
the porosity of the formation. As with the compensated density
tool, the at least two neutron detectors respond to events at
differing radial depths in the formation. The ratio of the detector
response is formed to minimize the effects of reactions within the
borehole, and porosity is determined from this ratio. The radial
depth of investigation is about 9 or 10 inches, and the system can
be calibrated to operate in both open and cased boreholes. It is
again emphasized that this prior art methodology requires at least
two detectors that are axially spaced from the neutron source.
[0006] U.S. Pat. No. 5,900,267 to Richard C. Odom et al. discloses
apparatus and methods for measuring the density of formations
penetrated by a cased borehole. The borehole apparatus comprises a
pulsed, fast neutron source and two axially spaced gamma ray
detectors. The gamma ray detectors are biased to measure inelastic
scatter gamma radiation produced by fast neutrons interacting with
nuclei within the formation. Inelastic diffusion length L.sub.RHO
is determined from the ratio of responses of the two detectors, and
the measure of L.sub.RHO is subsequently related to formation bulk
density. The physics of the L.sub.RHO bulk density logging
technique is discussed in detail in the publication "Applications
and Derivation of a New Cased Hole Density Porosity in Shaly Sand",
Richard C. Odom et al., Paper SPE 38699, Annu. Technical Conference
and Exhibition, 1997. Although operable in cased boreholes where
previously discussed gamma-gamma density systems are insensitive,
the L.sub.RHO system yields ambiguous results in formations
saturated with liquid and saturated with a gas/liquid fluid
mixture.
[0007] Prior art techniques for determining gas saturation of
formations penetrated by an open borehole involve the combination
of the responses of the conventional gamma-gamma type density tool
and porosity sensitive neutron-neutron tool. When the density and
porosity tools are calibrated for the water-saturated pore space
condition, and when they log formations that are water-saturated,
they will produce values for formation bulk density and formation
porosity that are consistent with each other, assuming the tools
are logging in a rock matrix that is the same as that used for
calibration. However, when a formation zone is encountered where
the pore water is replaced by gas, the porosity tool gives an
erroneously low porosity indication, while the density tool
correctly indicates a decrease in bulk density with corresponding
apparent increase in porosity. This results in a "cross-over" of
the log response curves from the two tools thereby indicating the
presence of gas within the logged formation. This method is
problematic in cased boreholes because of the more shallow
investigation depth of the density log and its resulting greater
sensitivity to variations in borehole conditions, such as
variations in the thickness of the cement sheath, immediately
behind the casing.
[0008] Logging for gas in cased, cemented boreholes has been
performed using a logging tool containing a pulsed source of fast
neutrons. Details of this method may be found in "Examples of Dual
Spacing Thermal Neutron Decay Time Logs in Texas Coast Oil and Gas
Reservoirs", Trans. SPWLA 15th Annu. Logging Symp., 1979, and "The
Use and Validation of Pulsed Neutron Surveys in Current Drilling
Tests" Trans. SPWLA 19th Annu. Logging Symp., 1978. This
"pulsed-neutron decay time" or "pulsed neutron" tool, as it is
known in the art, was designed to detect the presence of
hydrocarbon liquids (oil) in formations where the water that
otherwise fills the pore spaces is normally saline. This
sensitivity to fluid type is achieved by measuring the formation
thermal neutron cross section. However, the cross section is not
very sensitive to the presence of gas, and therefore the logging
tool is not very useful as a gas indicator. Another type of
measurement can be performed with the pulsed-neutron decay time
tool that is more sensitive to the presence of gas. This involves
measuring a ratio of the tool's typically two axially spaced gamma
detector responses. This ratio can, in turn, be interpreted in a
manner that is sensitive to the presence of gas within the logged
formation. Since the measured gamma radiation is produced by
thermal neutron capture reactions, this response is similar to that
of the neutron-neutron porosity log in that both are responding to
changes in the spatial distribution of thermal neutrons which, in
turn, is a function of hydrogen density. For this reason, the gamma
ratio response, like the neutron-neutron porosity tool response, is
not capable of distinguishing between low porosity formations and
formations with higher porosity whose pore spaces are gas filled.
The density/gas saturation logging system set forth in this
disclosure, on the other hand, responds to the change in atom
density and hence can distinguish between the gas saturated high
porosity and liquid saturated low porosity formations.
[0009] U.S. Pat. No. 5,804,820 to Michael L Evans et al. discloses
a density logging system that uses a source of fast neutrons, and a
single gamma ray detector. A measure of fast neutron attenuation is
required to obtain a meaningful density measurement from the
response of the single gamma ray detector. A neutron detector is
used to monitor source neutron attenuation. Alternately, it is
taught that fast neutron attenuation can be inferred by making use
of the fact that neutrons interacting with the various components
of the tool result in the production of both inelastic and
epithermal capture gamma rays. For example, fast neutrons
interacting with the iron of the tool yield inelastic gamma
radiation. A measure of intensity of this gamma radiation can be
used to infer fast neutron attenuation. Gamma radiations from iron
are quite prominent in the inelastic gamma-ray spectrum and as a
result, are easy to use without complicated spectral deconvolution
techniques. There is no teaching of methods to resolve ambiguities
in density measurements for formations saturated with liquid and
formations saturated with a significant fraction of gas.
[0010] U.S. Pat. No. 6,207,953 to Robert D. Wilson discloses a
logging system that utilizes a source of fast neutrons. The system
combines a measure of fast neutrons along with measures of
inelastic gamma radiation at two axial spacings from the source to
provide data used to compute formation porosity and liquid
saturation. The computation is based on an interpretation chart
that plots fast neutron count rate against an inelastic gamma ray
ratio measured at the two axial spacings. Formation density can be
computed from formation porosity if matrix and fluid densities are
known. Inelastic diffusion length L.sub.RHO is a function of the
measured ratio. A liquid scintillator is specified for fast neutron
detection, providing both fast neutron and inelastic gamma counts
by pulse-shape discrimination. An alternate plastic scintillator
and gamma detector combination is also taught in the event that the
liquid scintillator is not suitable for a particular application.
Fast neutron energies are distinguished by use of pulse-height
discrimination to provide borehole size compensation for air-filled
boreholes. Although the system does minimize ambiguity in liquid
and gas filled formations, there are still deficiencies in the
methodology. The axial spacing of the neutron detector from the
source is not optimized to minimize adverse borehole effects
because of count rate statistics limitations. Inelastic scatter
diffusion is determined from the ratio of inelastic scatter
radiation at two axial spacings from the source. This methodology
essentially assumes a point gamma radiation source, which
physically is not robust as will be discussed in subsequent
sections of this disclosure.
SUMMARY OF THE INVENTION
[0011] The invention set forth in this disclosure is embodied as a
logging system to measure density, porosity and gas/liquid fluid
saturation in earth formations penetrated by a cased or open
borehole. The gas/liquid fluid saturation is typically expressed as
the portion (S.sub.W) of the total fluid that is liquid, or
alternately as the portion (S.sub.G) of the total fluid that is
gas, where S.sub.G=(1-S.sub.W). The system is designed to minimize,
or even eliminate, adverse response properties in prior art logging
systems discussed previously. Although the invention is directed
toward measuring various properties of earth formations penetrated
by a borehole, it should be understood that the disclosed apparatus
and methods can be used to measure properties of other types of
material such as construction materials, road bed materials, ground
surface composition, and the like.
[0012] Embodied as a logging system, the well borehole
instrumentation consists of an elongated, pressure housed downhole
instrument or "tool" preferably containing an electronic source of
high energy or "fast" neutrons, commonly referred to as a neutron
generator. Fast neutrons are emitted in sequential bursts or
"pulses". The tool also contains preferably three axially spaced
radiation detectors that measure the intensity and energy of
radiations produced by various reactions of fast neutrons with
formation nuclei and electrons. The preferred embodiment comprises
a single neutron detector and two gamma ray detectors.
[0013] The methodology consists of acquiring count rates from the
preferred three detectors, individually and in combination, as the
apparatus is raised or lowered inside a borehole. More
specifically, the intensities of fast neutron and inelastic scatter
gamma radiations are measured at their respective detector axial
spacings. Measurements are made during the neutron pulses, and
preferably made during the same neutron pulses. Basic physics
studies have led to a better understanding of the relationships
between measured quantities and the formation parameters of
interest. These studies have established important nuclear
parameters along with their relationship to the desired formation
properties of density, porosity, and fluid saturation. A model
based on two-group diffusion theory has been successfully inverted
to provide a means for determining formation density, fluid
saturation and porosity from two measurements of inelastic gamma
radiation and a single measure of fast neutron radiation.
[0014] Prior to summarizing features of the model and the data
processing methodology, a brief discussion of concepts underlying
the measurements is appropriate. At least one of the measured count
rates is proportional to fast neutron flux present during the
sequential neutron bursts from the pulsed source of fast neutrons.
At least one additional measured count rate is proportional to
prompt gamma radiation due to the inelastic scattering of fast
neutrons off of formation and borehole nuclei during each neutron
burst. Both of these measurements derive from nuclear interaction
processes of neutrons in the million electron volt (MeV) range with
the nuclei and electrons of atoms comprising the earth formation
and surrounding borehole environs. Both of these measurements are
made during the neutron burst. The fast neutron measurement
responds primarily to changes in hydrogenous liquid (water or oil)
saturation of the earth formation pore space, although it also
responds to changes in bulk density of the formation. The inelastic
gamma measurement responds primarily to changes in formation bulk
density, though it is also somewhat sensitive to the liquid content
of the formation. By combining measured fast neutron and prompt
gamma ray count rates, density, porosity, and fluid saturation of
the formation can be obtained if the lithology of the formation is
known.
[0015] It is important to again emphasize that neutron and gamma
ray measurements respond not only to properties of the formation
penetrated by the borehole, but they also respond to properties of
the borehole region. Prior art systems have used ratios of
responses of at least two detectors, at different axial spacings
and with different radial depths of investigation, to minimize or
"cancel" borehole effects. As an example, responses of two neutron
detectors at two differing spacings are used in prior art systems
to determine formation porosity. If the spacings are too great,
then the measured count rate can be statistically insignificant. If
the spacings are two small, then the radial depths of investigation
of the detectors are small thereby rendering detector responses
relatively insensitive to formation properties. If the separation
between the two detectors is too small, then count rates vary
little between the two detectors and variations in the count rate
ratio used to compute porosity can be statistically insignificant.
It is, therefore, often difficult to select two axial detector
spacings that meet the above count rate, depth of investigation,
and count rate differential criteria.
[0016] Attention is next directed toward the model that has been
developed to process measured data. It is first noted that measured
gamma radiation is generated within the borehole environs by fast
neutron reactions. The gamma ray source is therefore dispersed
rather than a point source assumed in many prior art systems. The
model takes into account the dispersed nature of the inelastic
gamma source, and is based on two-group diffusion theory. Group one
describes the fast neutron transport from the logging tool into the
borehole environs, and group two describes the inelastic gamma
radiation transport within the borehole environs wherein a portion
reaches the gamma ray detectors in the logging tool. The diffusion
solution for the fast neutron group one becomes the source of
inelastic scatter gamma radiation for group two. The diffusion
length L.sub.FN characterizes the spatial distribution of fast
neutrons. The diffusion solution for the inelastic gamma group
describes the inelastic scatter gamma ray distribution, and
contains terms dependent on the fast neutron diffusion length
L.sub.FN and an inelastic scatter gamma ray diffusion length
L.sub.IG. The solutions are analogous to the two-group neutron
solutions developed in the previously referenced paper by Allen et
al. The solutions are for a uniform medium and with spherical
geometry. In principle, L.sub.FN is computed from the model using
fast neutron count rates measured at two or more suitable spacings
from the neutron source. These count rates are proportional to the
fast neutron flux. The inelastic gamma radiation flux is obtained
from a convolution integral for a spherically symmetric source
distribution. L.sub.IG is determined using L.sub.FN and inelastic
scatter gamma radiation count rates measured at two different axial
spacings from the neutron source. An iterative method using trial
values of L.sub.IG quickly produces a solution in the form of an
"intrinsic" diffusion length value, L.sub.IG,I. The iterative
solution eliminates previously discussed ambiguities in liquid and
gas saturated formations by producing a unique L.sub.IG,I for each
formation bulk density value.
[0017] The two-group diffusion model for the coupled fast neutron
and inelastic gamma transport problem is one-dimensional and
homogeneous. It does not, therefore, specifically account for a
finite borehole region including the borehole, casing and cement
annulus. In reality, fast neutron and inelastic gamma radiation
fluxes close to the source are dominated by transport in the
borehole region. Monte Carlo simulations show that spacings of 40
cm or greater are required before the fast neutron flux has
substantial formation sensitivity. Therefore, in order to obtain a
fast neutron diffusion length L.sub.FN related to the formation,
measurements should be made at spacings greater than about 40
centimeters (cm). In practice, it is difficult to achieve adequate
fast neutron count rate at axial spacings greater than about 60 cm.
The logging tool uses a single fast neutron detector axially spaced
at about 55 cm from the neutron source. Although this detector has
good formation sensitivity, the single count rate alone cannot
directly yield a value of L.sub.FN. This single fast neutron count
rate is related to the proper L.sub.FN value for use in the
iterative solution within the structure of the model. This is
accomplished by use of Monte Carlo simulations in combination with
the equations of the two-group diffusion model. The L.sub.FN values
corresponding to each fast neutron response are determined by
deriving a set of values that are self consistent, smoothly related
to fast neutron response, and that produce identical values of
inelastic gamma radiation diffusion length L.sub.IG for a common
bulk density .rho..sub.b computed from the model. The ratio of
inelastic scatter gamma ray count rates measured with two axially
spaced detectors is used in the computation of L.sub.IG. The
resulting L.sub.IG values exhibit the required inverse
proportionality to bulk density .rho..sub.b, which is one of the
parameters of interest. Fast neutron count rate is then combined
with the ratio of inelastic scatter gamma radiation and a parameter
proportional to the inverse of .rho..sub.b to determine formation
porosity and formation fluid saturation.
[0018] The present invention exhibits many advantages over prior
art systems. Ambiguities in measurements in liquid and gas
saturated formations are eliminated. The two-group diffusion model
used as a basis for data analysis includes a distributed source of
gamma radiation, which physically more closely represents the gamma
ray measurements rather than a point source gamma ray source. Fast
neutron diffusion length is measured with a single neutron
detector, which is axially spaced a relatively large distance from
the neutron source thereby minimizing adverse effects of the near
borehole region, yet responding with statistically significant
count rates for a given neutron source strength. Formation density,
formation porosity, and formation fluid saturation are uniquely
determined. The measurement can be made in both open and cased
boreholes. The system can be embodied in a variety of borehole
systems including a wireline logging system, a logging while
drilling system, a coiled tubing logging system, and a slickline
logging system. The neutron generator can be disabled when the tool
is removed from the borehole thereby reducing the risk of radiation
exposure to personnel and property. Other advantages will become
apparent in the following detailed description of the preferred
embodiments of the invention.
BRIEF DESCRIPTION OF THE DRAWINGS
[0019] So that the manner in which the above recited features,
advantages and objects of the present invention are attained can be
understood in detail, a more particular description of the
invention, briefly summarized above, may be had by reference to the
embodiments thereof which are illustrated in the appended
drawings.
[0020] FIG. 1 illustrates an instrument suspended in a borehole and
conveyable along the borehole by a conveyance means at the surface
of the earth;
[0021] FIG. 2 illustrates formation bulk density plotted as a
function of the inverse of inelastic scatter gamma radiation
diffusion length illustrating double-valued results obtained with
prior art point source models;
[0022] FIG. 3 illustrates linearized fast neutron diffusion length
plotted as a function of measured fast neutron count rate;
[0023] FIG. 4 illustrates formation bulk density plotted as a
function of the inverse of linearized inelastic scatter gamma ray
diffusion length;
[0024] FIG. 5 illustrates measured fast neutron count rate plotted
as a function of the inverse of atom density of formation
material;
[0025] FIG. 6 shows a graphical solution for formation porosity and
fluid saturation, expressed as fractional liquid saturation
S.sub.W, using a previously determined value of bulk density and
measured fast neutron count rate;
[0026] FIG. 7 is a flow chart showing steps in developing the
iterative solution of the data analysis model; and
[0027] FIG. 8 is a flow chart showing steps in determining
formation parameters of interest in a field logging system.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0028] The invention set forth in this disclosure is embodied as a
logging system to measure density, porosity and gas/liquid fluid
saturation in earth formations penetrated by a cased or open
borehole. As mentioned previously, gas/liquid fluid saturation is
typically expressed as liquid saturation S.sub.W, or alternately as
gas saturation S.sub.G, where S.sub.G=(1-S.sub.W). Prior to
disclosing features of the model and the data processing
methodology, a brief discussion of concepts underlying the
measurements are appropriate.
[0029] 3. Basic Concepts of the System
[0030] At least one of the measured count rates is proportional to
fast neutron flux present during the sequential neutron bursts.
Another measured count rate is proportional to prompt gamma
radiation due to the inelastic scattering of fast neutrons off of
formation and borehole nuclei during each neutron burst. Both of
these measurements derive from nuclear interaction processes of
neutrons in the MeV range with the nuclei and electrons of atoms
comprising the earth formation and surrounding borehole environs.
One measurement is of the high energy (3 to 10 MeV) neutrons which
are transported back into the borehole and arrive at a fast neutron
detector suitably axially spaced from a pulsed neutron source
emitting typically 14 MeV neutrons. A second measurement is of the
high energy (0.1 to 10 MeV) inelastic scatter gamma radiation,
which is produced promptly during the source neutron pulse, and
which is transported back into the borehole and subsequently
measured at preferably two axially spaced gamma ray detectors.
Neutron measurements and gamma ray measurements are made during the
neutron source pulse, and preferably made during the each of same
sequence of pulses.
[0031] The fast neutron measurement responds primarily to changes
in liquid or gas saturation of the earth formation pore space,
although it also responds to a lesser extent to changes in bulk
density of the formation. The inelastic gamma radiation measurement
responds primarily to changes in formation bulk density, although
it is also somewhat sensitive to the fluid content of the
formation. By combining measured fast neutron and the prompt
inelastic scatter gamma ray count rates, bulk density, gas/liquid
fluid saturation and porosity of the formation can be obtained if
the lithology of the formation is known.
[0032] 4. Apparatus
[0033] FIG. 1 illustrates a logging tool 10 comprising a radiation
source 12 which is preferably a neutron generator producing pulses
of neutrons at energy of about 14 MeV. The pulsed neutron generator
consists of a high energy neutron producing target with associated
gas reservoir and accelerated ion beam, high voltage supplies and
pulsing electronics, as is well known in the art. The logging tool
10 also comprises a neutron detector 15 axially spaced a distance
r.sub.NE from the source 12 as indicated at 32. The logging tool 10
further comprises a near gamma ray detector 14 and a long spaced
gamma ray detector 16 axially spaced distanced r.sub.N and r.sub.L,
respectively from the neutron source 12. The distances r.sub.N and
r.sub.L are denoted by the numerals 30 and 34, respectively. An
electronics package 17 provides power and control for the other
previously described elements of the logging tool 10. The
electronics package can also contain data processing equipment,
such as circuits to determine the intensity and energy of radiation
impinging upon the radiation detectors 14, 15 and 16. Optionally,
the electronics package 17 also contains a processor to transform
measured radiation energy and intensity into parameters of
interest. Alternately, a processor 25 can be located at the surface
of the earth 27 wherein the processor cooperates with the logging
tool 10 through a conveyance unit 26.
[0034] Still referring to FIG. 1, elements of the logging tool 10
are enclosed within a pressure housing 18 for protection from the
harsh environment of the borehole 20. The tool 10 is conveyed along
a borehole 20 penetrating an earth formation 22 by a conveyance
system that includes a member 24 with a first end operationally
connected to the tool 10 using a suitable connector 21. The member
24 extends to the surface of the earth 27 and has a second end
operationally connected the surface conveyance unit 26. If the
conveyance system is a wireline logging system, the member 24 is a
wireline logging cable, and the surface conveyance unit 26
comprises wireline draw works and surface equipment well known in
the art. If the conveyance system is a drilling rig, the member 24
is a drill pipe string and the surface conveyance 26 comprises a
drilling rig, which is also well known in the art. Other conveyance
systems, including a coiled tubing system and a slickline system,
can be used to convey the tool 10 along the borehole 20. The
surface conveyance unit 26 can also contain data processing
equipment, such as circuits to determine the intensity and energy
of radiation impinging upon and interacting with the radiation
detectors 14, 15 and 16. The surface conveyance unit 26 can further
contain a processor to transform measured radiation energy and
intensity into parameters of interest.
[0035] 5. System Response Model
[0036] The model used to develop data processing methodology is
based on two-group diffusion theory. Group one describes the fast
neutron transport and group two describes the inelastic gamma
transport. The model accounts for the previously discussed
dispersed nature of the inelastic gamma source. The model is
represented mathematically by the following equations:
.PHI..sub.FN=Q/(4.pi.D.sub.FN)(1/r)(exp(-r/L.sub.FN)) (1)
and
.PHI..sub.IG=(QL.sub.IG.sup.2/(4.pi.D.sub.IG(L.sub.FN.sup.2-L.sub.IG.sup.2-
))(1/r)(exp(-r/L.sub.FN)-exp(-r/L.sub.IG)) (2).
[0037] where
[0038] .PHI..sub.FN=the fast neutron flux at spatial position
r;
[0039] .PHI..sub.IG=the inelastic scatter gamma ray flux at spatial
position r;
[0040] D.sub.FN=the diffusion coefficient for fast neutrons;
[0041] D.sub.IG=the diffusion coefficient for inelastic scattered
gamma rays; and
[0042] Q=the strength of the fast neutron source.
[0043] The respective diffusion coefficients, D.sub.i, are related
to diffusion length L.sub.i by:
L.sub.i.sup.2=D.sub.i/.SIGMA..sub.r,i
[0044] where
[0045] i=FN, IG; and
[0046] .SIGMA..sub.r,i=the probability per cm for removal from the
respective energy group.
[0047] Equation (1) is the diffusion solution for the fast neutron
group. Equation (1) describes the fast neutron distribution, and
becomes the source term of dispersed inelastic gamma radiation for
group two. The diffusion length L.sub.FN characterizes the spatial
distribution of fast neutrons. Equation (2) is the diffusion
solution for the inelastic gamma group with a source term given by
equation (1). It describes the inelastic gamma distribution and
contains terms dependent on the fast neutron diffusion length
L.sub.FN and inelastic gamma diffusion length L.sub.IG. The
solutions are analogous to the previously referenced two-group
neutron model developed by Allen et al. These solutions are for a
uniform medium and with spherical geometry.
[0048] For the fast neutron group, .SIGMA..sub.r,FN is the
probability per cm for a neutron to be removed from the energy
range of about 3 MeV to 14 MeV. For the inelastic gamma group,
.SIGMA..sub.r,IG is the probability per cm for an inelastic gamma
to be removed from the 100 kiloelectron volts (keV) to 10 MeV
energy range.
[0049] In the prior art, L.sub.FN is typically computed from
equation (1) using a combination of fast neutron count rates
measured at two or more suitable spacings from the source. These
count rates are proportional to the fast neutron flux. This
combination process eliminates some terms from equation (1) which
are not measurable and are not generally known. The process,
however, requires two or more axially spaced fast neutron detector
systems which (a) adds complexity to the downhole logging
equipment, and (b) creates depth of investigation and count rate
statistics problems discussed previously.
[0050] Equation (2) for the inelastic gamma flux is derived from
the convolution integral for a spherically symmetric source
distribution: 1 IG = 0 .infin. S ( r ' ) G SS ( r , r ' ) 4 r '2 r
' ( 3 )
[0051] where the diffusion kernel G.sub.ss(r,r') is given by the
equation
G.sub.ss(r,r')=(L.sub.IG/4.pi.D.sub.IG)(1/rr')
{exp(-.vertline.r-r'.vertli-
ne./L.sub.IG)-exp(-.vertline.r+r'.vertline./L.sub.IG)} (4).
[0052] at spatial positions r and r'. The spherically symmetric
source distribution S(r') is given by the fast neutron distribution
of equation (1).
[0053] L.sub.IG is computed from equation (2) using L.sub.FN as
computed from equation (1) and inelastic gamma count rates measured
at spacings r.sub.N and r.sub.L (see FIG. 1). Preferably, r.sub.N
is about 35 cm and r.sub.L is about 95 cm. Again, the inelastic
count rates at r.sub.N=35 cm and r.sub.L=95 cm are proportional to
the corresponding fluxes .PHI..sub.IG in equation (2). The constant
coefficients in equation (2) cancel if a flux ratio is formed for
the two spacings:
(r.sub.N/r.sub.L)(.PHI..sub.IG(r.sub.N)/.PHI..sub.IG(r.sub.L))=(exp(-r.sub-
.N/L.sub.FN)-exp(-r.sub.N/L.sub.IG))/(exp(-r.sub.L/L.sub.FN)-exp(-r.sub.L/-
L.sub.IG)) (5).
[0054] The resulting expression cannot be solved directly for
L.sub.IG, but an iterative method using trial values of L.sub.IG
quickly produces a solution. The iterated value of L.sub.IG is free
of the ambiguity inherent in certain prior art double-valued
solutions discussed previously. More specifically, if a point
source model is used to compute inelastic scatter gamma ray
diffusion length, ambiguous double-valued results are obtained for
constant density pairs, depending upon whether the formation is
liquid or gas saturated. This is because of differing fast neutron
diffusion lengths resulting from differences in the two saturating
fluids. Examples of double-valued results obtained with prior art,
point source models will be presented in a subsequent section of
this disclosure. Inelastic gamma ray diffusion lengths computed
from the point source model will be designated as "L.sub.RHO" to
distinguish this parameter from inelastic gamma ray diffusion
length L.sub.IG computed using the two-group distributed source
model. The iterative solution of equation (5) eliminates this
ambiguity, producing a unique L.sub.IG for each density. It should
be noted that equation (5) will produce indeterminate results for
the case where L.sub.FN and L.sub.IG are the same. However, in
practice, L.sub.FN is usually smaller than L.sub.IG so this
difficulty is avoided. For gas saturated porosity, L.sub.FN and
L.sub.IG can have similar values. For the case of equal or
near-equal values, another form of the two-group solution would be
used to avoid the indeterminacy of equation (5).
[0055] 6. System Response
[0056] Count rate data used in the following discussions are in
arbitrary units. The fast neutron count rate C.sub.FN is for a
spacing r.sub.NE=50 cm. The inelastic scatter gamma ray count rates
C.sub.G,N and C.sub.G,L are measured at gamma ray detector spacings
r.sub.N=30 cm and r.sub.L=90 cm, respectively. The formation matrix
is sandstone unless otherwise noted. Count rate data used to
develop interpretation models are computed with the Monte Carlo
simulation method using the MCNP code. Other suitable numerical
simulation and transport computation models can be used to obtain
the required count rates.
[0057] FIG. 2 illustrates double-valued results in .rho..sub.b
obtained from the inelastic scatter gamma ray diffusion length
L.sub.RHO computed from the prior art point source model using a
ratio of inelastic scattered gamma radiation C.sub.G,N/C.sub.G,L
measured at r.sub.N and r.sub.L. Recall that C.sub.G,N/C.sub.G,L is
proportional to .PHI..sub.IG(r.sub.N)/.PHI..sub.IG(r.sub.L) as
defined above. Formation bulk density .rho..sub.b is plotted as a
function of the inverse of L.sub.RHO for sandstone formations
ranging in density from 1.86 gm/cm.sup.3 to 2.65 gm/cm.sup.3. The
points 42 represent gas saturated formations with a gas density of
0.00 gm/cm.sup.3. A smooth curve 40 has been fitted to the data
points 42 and the point 43. The points 44 represent salt water
saturated formations with a liquid density of 1.07 gm/cm.sup.-3. A
smooth curve 46 has been fitted to the data points 44 and the point
43. The curves 40 and 46 converge at the point 43, which represents
a sandstone formation with zero porosity or 0.0 porosity units
(p.u.). The ambiguity in .rho..sub.b can be clearly illustrated by
example. Assume that L.sub.RHO is determined to be 14.25 cm, or
1/L.sub.RHO=0.0701 cm.sup.-1 as illustrated by point 52. A line
projected vertically from the point 52 intersects the curve 46 at a
point 57, and intersects the curve 40 at a point 55. Lines
projected horizontally from the intersection points 57 and 55 yield
values of .rho..sub.b of 2.16 gm/cm.sup.3 (point 50) and 2.53
gm/cm.sup.3 (point 59), respectively. Two ambiguous values of
.rho..sub.b are, therefore, obtained for the same value of
L.sub.RHO depending upon the fluid saturating the formation
matrix.
[0058] It should be noted that the two-group diffusion model for
the coupled fast neutron and inelastic gamma transport problem is
one-dimensional and homogeneous. It does not include the
cylindrical arrangement of the various material in the near
borehole environs such as the borehole and fluid therein, casing,
and cement sheath filling the casing-formation annulus. The model
does not account for the logging tool being eccentered within the
casing. Fast neutron and inelastic scatter gamma ray fluxes close
to the source are dominated by transport within the near-borehole
region. Monte Carlo simulations show that spacings of r.sub.NE=40
cm or greater are required before the fast neutron flux has
substantial formation sensitivity. Therefore, in order to infer a
formation related L.sub.FN from equation (1), measurements should
be made at spacings greater than about 40 cm. In practice, it is
difficult to achieve adequate fast neutron count rate at spacings
greater than about 60 cm due to statistical limitations. Because of
the above factors, the value of r.sub.NE in the logging tool 10
(see FIG. 1) is selected to be about 50 cm. Again because of the
above factors, it is not practical to use two axially spaced
neutron detectors to obtain an indication of fast neutron diffusion
length. If a second detector were spaced closer to the neutron
source, formation sensitivity in this detector would be
significantly degraded. If a second detector were axially spaced
further from the neutron source, the measurement would be
significantly degraded by poor counting statistics. Although the
axial spacing r.sub.NE of the single neutron detector 15 in the
tool 10 (see FIG. 1) has good formation sensitivity, a single count
rate alone measured with this detector cannot produce a value of
L.sub.FN from equation (1). However, for constant neutron source
strength, it is possible to relate the single fast neutron count
rate C.sub.FN to the proper L.sub.FN value for use in the iterative
solution of equation (5). This is accomplished by use of Monte
Carlo simulations in combination with the equations of the
two-group diffusion model. The process is illustrated by the
following examples.
[0059] The analysis using the two-group diffusion model of the
present invention will be illustrated again using examples of
sandstone formations at several bulk densities .rho..sub.b. At each
density there are again two formations with one having gas
saturated porosity and another having salt water saturated
porosity. The porosities are different for each pair in order to
maintain constant density. The count rates C.sub.FN, C.sub.G,N and
C.sub.G,L are then input to the two-group diffusion analysis model.
The L.sub.FN values corresponding to each fast neutron response are
determined by deriving a set of values that are self-consistent,
smoothly related to fast neutron response, and that produce
identical values of inelastic gamma ray diffusion length L.sub.IG
for a common bulk density .rho..sub.b when used to solve equation
(5). The resulting L.sub.IG values are constrained in that they
must exhibit the required inverse proportionality to bulk density
.rho..sub.b. As a starting point in this procedure, L.sub.FN values
are computed by using equation (1) to analyze curves of fast
neutron flux .PHI..sub.FN versus spacing that are simulated for
each formation by MCNP. When the resulting values of L.sub.FN are
inserted into equation (5), it is apparent that these do not
represent a consistent set of values meeting the constraint that
resulting L.sub.IG values must exhibit the required inverse
proportionality to bulk density .rho..sub.b. Further, the values of
L.sub.FN do not exhibit a smooth relation to fast neutron response
C.sub.FN computed for r.sub.NE=50 cm. In addition, resulting
L.sub.IG values are smaller than expected and not unique for a
common bulk density .rho..sub.b.
[0060] As a first step toward improving the results obtained from
equation (5) for the range of formations under investigation, the
L.sub.FN values computed from equation (1) for liquid and gas
saturated formations are adjusted. This process of adjusting
L.sub.FN is best described as a process of linearizing the relation
between .PHI..sub.FN and L.sub.FN in such a way that a smooth curve
can be fitted to the result. The values of L.sub.FN computed from
.PHI..sub.FN and the fit coefficients, when used with the inelastic
gamma ratio in equation (5), produce good solutions which have no
ambiguity and which show inverse relation to bulk density
.rho..sub.b. The resulting linearized value, defined as
L.sub.FN,Li, is expressed mathematically as
L.sub.FN,Li=L.sub.FN+D.sub.i (6)
[0061] where D.sub.i is an adjustable parameter and i denotes
liquid (i=L) or gas (i=G) formation saturation. Differences between
L.sub.FN values for 100 percent liquid saturated (S.sub.W=1.00)
formations are small. As an example, the MCNP derived values of
L.sub.FN are 10.96, 10.95, and 10.9 cm for the three saturated
formations of porosity 18, 30, and 50 p.u., respectively. These
values of L.sub.FN were initially decreased by D.sub.L=-1.0 cm,
which is a substantial percentage adjustment. Smaller adjustments
in the value of DL were then made to give the values of
L.sub.FN,Li=9.9 cm, 9.7 cm and 9.5 cm, respectively, which show the
desired linear trend with .PHI..sub.FN. Then, for each formation
with a given bulk density and liquid saturated pore space (i=L),
the companion gas saturated porosity case is evaluated for
L.sub.FN,Li(i=G) using equation (5), where D.sub.G is iterated
until L.sub.IG,G, the inelastic gamma ray diffusion length for gas
saturated formations, agrees with L.sub.IG,L, the inelastic gamma
ray diffusion length for liquid saturated formation value at that
density.
[0062] L.sub.FN,Li values obtained by the above procedure, plotted
as a function of C.sub.FN at r.sub.NE=50 cm, are shown in FIG. 3.
Data points 62 for gas (density=0.0 gm/cm.sup.3) saturated
formations are denoted by circles, and data points 64 for liquid
(salt water of density 1.07 gm/cm.sup.3) saturated formations are
denoted by triangles. Numbers shown in parenthesis denote formation
porosity in p.u. The curve 60, fitted to all data points, exhibits
the desired linear trend and is expressed by the relationship
L.sub.NF,Li=0.033C.sub.FN+1.676 (7)
[0063] with a fit correlation of R.sup.2=0.998. Equation (7)
represents a linearized relationship that is independent of fluid
saturation, fluid type and pore space volume of the formation. When
used in equation (5), this set of L.sub.FN,Li values produces a set
of L.sub.IG,I values by the iterative solution.
[0064] FIG. 4 is a plot of 1/L.sub.IG,I versus bulk density
.rho..sub.b. Separate symbols are used for the salt-water saturated
porosity and the gas saturated porosity cases to show the nature of
the agreement and resulting minimization of the L.sub.RHO ambiguity
shown in FIG. 2. Specifically, the diamond points 76 indicate salt
water saturated formations, and the triangle points 74 indicate gas
saturated formations. A curve 72 is shown fitted through all data
points 74 and 76, although there is a very small separation in the
curves at low values of 1/L.sub.IG,I as indicated by the data
points 74 and 76. The curve 72 is expressed mathematically as a
first order polynomial
.rho..sub.b=72.99(1/L.sub.IG,I)-3.6225 (8)
[0065] with a correlation coefficient of R.sup.2=0.996. In view of
the goodness of fit, it is apparent that, in practice,
double-valued .rho..sub.b determinations have been eliminated by
the data processing method disclosed above. The iterative solution
method, when applied separately to gas and liquid saturated
formations with the same density, produces bulk density values
.rho..sub.b that agree to within 1 percent for .rho..sub.b ranging
from 1.86 g/cm.sup.3 to 2.65 g/cm.sup.3. The non-zero intercept
obtained from the fit represented by equation (8) and shown in FIG.
4 is due to the fact that as formation density gets very small,
approaching a void space, the borehole materials remain and thereby
produce an effective upper limit on L.sub.IG,I. This effect is
illustrated by setting bulk density .rho..sub.b=0.00 and solving
equation (8) for the intrinsic inelastic scatter gamma ray
diffusion length L.sub.IG,I. The solution yields L.sub.IG,I=20.15
cm, which is a value consistent with the relatively large inelastic
scatter gamma ray diffusion length expected for the predominant
borehole material, salt water.
[0066] The two-group diffusion analysis represented by equation (5)
still produces an uncertainty in the magnitude of the converged
L.sub.IG,I, value but it does produce the desired unambiguous or
"single-valued" inverse proportionality to formation bulk density.
The proportionality constant can be determined by calibration of
the response of the tool in formations with known lithology,
porosity, density, fluid saturation and borehole conditions. This
"physical model" calibration process can be supplemented with
computer simulations of the tool's response. In the following
discussions, it will be assumed that the tool has been calibrated
and that the inverse of the converged L.sub.IG,I has been
normalized so that it is proportional to formation bulk density
.rho..sub.b.
[0067] Once the bulk density .rho..sub.b is determined, formation
porosity and fluid saturation (expressed as liquid saturation
fraction S.sub.W) can be inferred from a plot of C.sub.FN versus
the reciprocal of atom density of the formations. FIG. 5
illustrates an example of such a plot, which was derived using
Monte Carlo simulation for a limestone formation with a bulk
density of 2.08 g/cm.sup.3. Data points 82 represent formations
saturated with a combination of salt water (liquid density=1.07
gm/cm.sup.3) and gas (fluid density=0.00 gm/cm.sup.3). Data points
84 represent formations saturated with a combination of low density
oil (liquid density=0.61 gm/cm.sup.3) and gas (fluid density=0.00
gm/cm.sup.3). Numbers within parenthesis by each data point
represent formation porosity and liquid saturation fraction,
respectively. S.sub.W and S.sub.O denote saturation fractions for
the salt water and oil liquids, respectively. Curve 80 is fitted
through all data points 82 and 84.
[0068] Still referring to FIG. 5, a series of similar curves would
be needed for the full bulk density range and lithology expected
from field data. As illustrated with the curve 80, note the near
linear dependence of the fast neutron response C.sub.FN on the
reciprocal of atom density. This property facilitates the
interpolation of values from measured fast neutron response as will
be illustrated in a subsequent section of this disclosure. Also
note that different types of fluids, and mixtures of fluids such as
oil and gas, also fall on the curve 80. This means that once bulk
density .rho..sub.b has been determined as discussed above from the
solution of equation (5), then a single curve 80 of the type shown
in FIG. 5 along with a measure of C.sub.FN is sufficient to
determine formation porosity and saturation for that bulk density,
regardless of fluid type.
[0069] FIG. 6 illustrates conceptually a graphical solution for
fluid saturation, expressed as S.sub.W, using a family of constant
bulk density curves of the type shown in FIG. 5. The curves shown
in FIG. 6 are for a specific formation lithology, such as sandstone
or limestone. Curves 90 represent constant densities increasing in
magnitude to the left as shown by the .rho..sub.b arrow. Constant
liquid fluid saturations (S.sub.W) are denoted by broken curves 93,
and increase to the left over a range of S.sub.W=0.00 to
S.sub.W=1.00. Assume that the measured fast neutron count rate is
C.sub.FN,m and the accompanying bulk density determination is
.rho..sub.b,m. The subscript "m" simply denotes specific "measured"
quantities used in this particular illustration. The previously
determined value of .rho..sub.b,m is entered at 98 and projected
along a constant bulk density as shown by the dotted curve 99. The
measured fast neutron count rate C.sub.FN,m is entered at 96 and
projected horizontally as illustrated by the dotted curve 97.
Curves 97 and 99 intersect at the point 100. A broken curve 94 is
then projected along a constant S.sub.W line thereby indicating
formation liquid fluid saturation S.sub.W of about 0.40. Once a
single-valued bulk density .rho..sub.b is unambiguously determined,
single-valued fluid saturation S.sub.W can also be unambiguously
determined using the analysis methods illustrated graphically in
FIG. 6. A single-valued formation porosity can then be determined
from .rho..sub.b and S.sub.W using the relationship
Porosity=(.rho..sub.ma-.rho..sub.b)/(.rho..sub.ma-(S.sub.W.rho..sub.liq))
(9)
[0070] where .rho..sub.ma=the density of the formation matrix and
.rho..sub.liq=the density of the liquid saturating the pore space
of the formation. It is assumed that .rho..sub.ma and .rho..sub.liq
are known from other sources.
[0071] In the disclosure of analysis methods of the invention,
graphical solutions such as the ones illustrated in FIGS. 4 and 6
have been used to illustrate basic principles. It should be
understood that mathematical relationships have been developed for
the various steps of the analysis methodology, and that parameters
of interest are preferably computed using these mathematical
relationships along with appropriate measured quantities.
Computations are performed using a suitably programmed
processor.
[0072] The initial development of the tool response is summarized
in the form of a flow chart in FIG. 7. Values of
.PHI..sub.IG(r.sub.N), .PHI..sub.IG(r.sub.L), and
.PHI..sub.FN(r.sub.NE) are computed at steps 120, 122 and 124,
respectively, using simulation models. Values of L.sub.FN are
computed from numerical simulations. These values are incremented
by D.sub.i at step 126 according to equation (6) to yield
L.sub.FN,Li. These values of L.sub.FN,Li, along with
.PHI..sub.IG(r.sub.N) and .PHI..sub.IG(r.sub.L), are used in
equation (5) to compute L.sub.IG,L and L.sub.IG,G at step 128.
Resulting values are compared at step 130, and D.sub.i is
iteratively adjusted until L.sub.IG,L=L.sub.IG,G. The ratio
R=.PHI..sub.IG(r.sub.N)/.PHI..sub.IG(r.s- ub.L)=C.sub.G,N/C.sub.G,L
is computed at step 132. A converged "intrinsic" value of inelastic
gamma ray diffusion length L.sub.IG,I for the related .rho..sub.b
value is computed at step 134 using R and L.sub.FN,Li in equation
(5). The relationship between .rho..sub.b and 1/L.sub.IG,I is
determined at step 136. This step is shown graphically in FIG. 4
and expressed analytically in equation (8). This allows formation
bulk density .rho..sub.b to be determined from measured parameters.
An analytical expression relating formation fluid saturation to
.rho..sub.b and C.sub.FN (see FIG. 6) is developed at step 137. An
analytical expression relating formation porosity to .rho..sub.b
and S.sub.W (see equation (9)) is developed at step 138.
[0073] 7. Field Data Processing
[0074] In field applications of the system, measured quantities
are:
[0075] C.sub.FN=fast neutron count rate measured by a neutron
detector at an axial spacing from the neutron source of
r.sub.NE;
[0076] C.sub.IG,N=inelastic scatter gamma radiation count rate
measured by a gamma ray detector at an axial spacing from the
neutron source of r.sub.N; and
[0077] C.sub.IG,L=inelastic scatter gamma radiation count rate
measured by a gamma ray detector at an axial spacing from the
neutron source of r.sub.L.
[0078] The data processing methodology for field application of the
system is summarized in the flow chart of FIG. 8. The tool response
is calibrated under known conditions as discussed previously. As an
example, the measured fast neutron response C.sub.FN is converted
to a L.sub.FN,L value using an equation of the form
L.sub.FN,L=A+(BC.sub.FN) (10)
[0079] which is the same form shown in equation (7) and graphically
in FIG. 3. The subscript "i" has been omitted since a smooth
"linearized" relation representing all L.sub.FN,L points has been
established at this point, regardless of whether formation material
is gas or liquid saturated. A and B are calibration constants. The
calibration constants take into account detector efficiency, source
strength, tool geometry and the like. The count rates C.sub.FN,
C.sub.IG,N and C.sub.IG,L are measured at step 152. The parameter
L.sub.FN,L is computed at step 154. The ratio R of measured count
rates C.sub.IG,N/C.sub.IG,L is computed at step 156. The iterated,
intrinsic value of inelastic scatter gamma radiation diffusion
length, L.sub.IG,I, is computed at step 158. Formation bulk density
is then determined at step 160 using the method illustrated
graphically in FIG. 4. This value of formation bulk density and the
measured fast neutron count rate are combined at step 162 to obtain
formation porosity and fluid saturation. Step 162 is illustrated
graphically in FIG. 6 for determining formation fluid saturation
S.sub.W, and is illustrated analytically in equation (9) for
determining formation porosity. Finally, formation parameters of
interest are recorded as a function of well depth at which they are
determined at step 164, and the process is repeated at step 166 as
the logging tool is conveyed along the borehole.
[0080] While the foregoing disclosure is directed toward the
preferred embodiments of the invention, the scope of the invention
is defined by the claims, which follow.
* * * * *