Electricity: measuring and testing – Particle precession resonance – Using well logging device
Reexamination Certificate
2000-10-17
2002-07-16
Lefkowitz, Edward (Department: 2862)
Electricity: measuring and testing
Particle precession resonance
Using well logging device
C324S300000
Reexamination Certificate
active
06420869
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This present invention relates to predicting nuclear magnetic resonance (NMR) values for a downhole formation fluid sample, and more particularly, relates to predicting NMR decay times, T
1
and T
2
for a hydrocarbon sample by employing measured near infrared spectra and predetermined correlations between near infrared spectra and NMR values.
2. Summary of Related Art
The measurement of near infrared (NIR) spectra is well known in the art. See for example, U.S. Pat. No. 5,939,717, by Mullins, entitled “Method and Apparatus for Determining Gas-Oil Ratio in a Geological Formation Through the Use of Spectroscopy”, incorporated herein by reference. The Mullins patent discloses a method and apparatus for illuminating formation samples with light and determining a gas-oil ratio from the NIR spectra. Earlier, a similar disclosure for using NIR spectra to estimate gas-oil ratio, API gravity, asphaltenes, and other properties was given in application GB 2,217,838A by DiFoggio. Conventional NMR measurement techniques are well known in the art, as taught in U.S. Pat. No. 5,055,787, Kleinberg et al., entitled Borehole Measurements Of NMR Characteristics Of Earth Formation, herein incorporated by reference. Thus, there are numerous tools available in the industry for measurement of properties of oil formations adjacent a wellbore.
As an alternative to specific measurement tools, mathematical techniques can been utilized to predict oil formation properties based on non-specific measurement tools to determine the desired property. Regression analysis is a popular mathematical technique that is frequently useful for correlation between measured values. Regression analysis is useful to build models explaining a dependent variable or variables in terms of a set of independent variables. The mathematical model or correlation is useful for extrapolating into new population sets other than those observed. Extrapolation or prediction can be performed by choosing values of the independent variables for new cases and thereby predicting the most likely value of the dependent variable. For example, one might wish to predict age at death (the dependent variable) on the basis of life habits, genetic characteristics, and physiology (the independent variables).
The basic bivariate linear regression equation is
Y=a+b
(
X−X
′),
Where Y is the dependent variable, a is a constant, b is the regression coefficient of X on Y, the Xs are independent observations, and X′ their mean. For example, Y might be height, b the coefficient with which height changes with age, and X the age. The equation is easily extended to multiple independent variables, in which case b would be a row vector, and X and X′ column vectors. Nonlinear regression is also possible. Many forms exist, including polynomial, quadratic, and higher order. An example of a polynomial regression equation of the order 3 (cubic) would be
Y=a+b
1
(
X−X
′)+
b
2
(
X
2
−X′
2
)+
b
2
(
X
3
−X′
3
).
Regression analysis is thus used to build models and, as an extension, to allow extrapolation from the model to predict future values of the dependent variable. While the calculations are basically simple, the presence of numerous variables and observations, particlularly if the independent variables are interrelated, can cause the total amount of calculations to be very large. Various methods of regression analysis are well known in the art and are widely available commercially as computer programs for model building and prediction.
There are, however, no known equations for predicting NMR decay times T
1
and T
2
from NIR spectra. Thus there is a need for a method and apparatus for predicting NMR decay times T
1
and T
2
from NIR spectra.
SUMMARY OF THE INVENTION
The present invention provides a method and apparatus for developing a correlation between NIR spectra and the Nuclear Magnetic Resonance (NMR) decay times, T
1
and T
2
for a library of known crude oil samples, and using the correlation and NIR spectral measurements to predict the NMR decay times, T
1
and T
2
for unknown crude samples extracted from a formation adjacent a well bore. One of the objects of the invention is to measure NIR spectra to predict the NMR decay times, T
1
and T
2
of crude oils during or immediately after being sampled down hole by a down hole formation tester such as the Baker Atlas Reservation Characterization Instrument (RCI). It is also an object of the present invention to use these predictions of NMR T
1
and T
2
decay times to improve log interpretation of NMR logs. One of the advantages of the present invention that it enables the use of existing down hole NIR spectrometers in down hole tools such as the RCI tool string, without requiring the additional expense and time to develop and deploy a separate and independent NMR sensor for the RCI tool or some other down hole tool. Thus, the present invention reduces the length and weight of the down hole tool string and the expense and necessity of an additional independent NMR tool.
REFERENCES:
patent: 5055787 (1991-10-01), Kleinberg et al.
patent: 5668374 (1997-09-01), DiFoggio et al.
patent: 5939717 (1999-08-01), Mullins
patent: 2217838 (1989-11-01), None
Article titled: “Guidelines for Applying Chemometrics to Spectra: Feasibility and Error Propagation”; Author: Rocco DiFoggio; Publication: Applied Spectroscopy; vol. 54, No. 3, Mar. 2000, pp.94A-113A.
Baker Hughes Incorporated
Madan Mossman & Sriram P.C.
Shrivastav Brij B.
LandOfFree
Method and apparatus for estimating NMR properties by near... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for estimating NMR properties by near..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for estimating NMR properties by near... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2868912