Back to results
US 7,095,012 B2Granted
Methods and apparatus for determining chemical composition of reservoir fluids
AssigneeSchlumberger Technology Corporation
InventorsGo Fujisawa, Oliver C. Mullins, Toru Terabayashi, Fredrick A. Jenet, Maria A. van Agthoven, Philip A. Rabbito
Life of this patent
Estimating the term…
Abstract
Methods of analyzing formation fluids in an oilfield environment are near-infrared absorption spectroscopy. Indications of near-infrared absorptions are analyzed to determine the concentration of compounds in a formation fluid sample.
Claims26
A method of providing a chemical compositional analysis while sampling a formation fluid in an oilfield environment comprising:
- extracting a formation fluid sample;
- transmitting near-infrared light through the formation fluid sample;
- detecting indications of near-infrared absorptions from the formation fluid sample;
- analyzing the indications of near-infrared light absorptions using a principal component regression model, wherein the principal component regression model is established before sampling the formation fluid in the oilfield environment and establishing the principal component regression model comprises:
- constructing a calibration data matrix from a plurality of near-infrared absorption spectra of calibration samples;
- determining the principal components of the calibration data matrix;
- decomposing the calibration data matrix into a matrix of the principal components and a matrix of scores for the plurality of calibration spectra;
- and
- determining a plurality of regression vectors, each of which relates concentration of a constituent compound in the calibration samples to the matrix of scores;
- and
- determining concentrations of a plurality of compounds in the formation fluid sample, the plurality of compounds including:
- methane and carbon dioxide.
9 more elements
A method of providing a chemical compositional analysis while sampling a formation fluid in an oilfield environment comprising:
- extracting the formation fluid sample;
- transmitting near-infrared light through the formation fluid sample;
- detecting indications of near-infrared absorptions from the formation fluid sample at a plurality of wavelength channels;
- analyzing the indications of near-infrared light absorptions using a principal component regression model, wherein the principal component regression model is established before sampling the formation fluid in the oilfield environment and establishing the principal component regression model comprises:
- constructing a calibration data matrix from a plurality of near-infrared absorption spectra of calibration samples:
- determining the principal components of the calibration data matrix:
- decomposing the calibration data matrix into a matrix of the principal components and a matrix of scores for the plurality of calibration spectra;
- and
- determining a plurality of regression vectors, each of which relates concentration of a constituent compound in the calibration samples to the matrix of scores;
- and
- determining concentrations of a plurality of compounds in the formation fluid sample.
8 more elements
An optical fluid analysis module adapted to be housed within a borehole tool comprising:
- means for transmitting near-infrared light through a formation fluid sample;
- means for detecting indications of near-infrared absorptions from the sample;
- and
- means for analyzing the indications of near-infrared absorptions to determine concentrations of a plurality of compounds in the sample including methane, carbon dioxide, and a higher hydrocarbon, wherein the means for analyzing the indications of near-infrared absorptions includes a processor and memory means coupled with the processor, the processor being programmed with instructions which, when executed by the processor, cause the processor to apply a principal components regression model to the indications of near-infrared absorptions, wherein the principal component regression model is established before sampling the formation fluid in the oilfield environment and establishing the principal component regression model comprises:
- constructing a calibration data matrix from a plurality of near-infrared absorption spectra of calibration samples;
- determining the principal components of the calibration data matrix;
- decomposing the calibration data matrix into a matrix of the principal components and a matrix of scores for the plurality of calibration spectra;
- and
- determining a plurality of regression vectors, each of which relates concentration of a constituent compound in the calibration samples to the matrix of scores.
6 more elements
A borehole tool comprising:
- means for extracting a formation fluid sample from a subsurface region into the tool;
- an optical analyzer housed within the tool, the optical analyzer including:
- means for transmitting near-infrared light through the sample,
- means for detecting indications of near-infrared absorptions from the sample, and
- means for analyzing the indications of near-infrared absorptions to determine concentrations of a plurality of compounds in the sample, including methane, carbon dioxide, and at least one higher hydrocarbon, wherein the means for analyzing the indications of near-infrared absorptions includes a processor and memory means coupled with the processor, the processor being programmed with instructions which, when executed by the processor, cause the processor to apply a principal components regression model to the indications of near-infrared absorptions, wherein the principal component regression model is established before sampling the formation fluid in the oilfield environment and establishing the principal component regression model comprises:
- constructing a calibration data matrix from a plurality of near-infrared absorption spectra of calibration samples;
- determining the principal components of the calibration data matrix:
- decomposing the calibration data matrix into a matrix of the principal components and a matrix of scores for the plurality of calibration spectra;
- and
- determining a plurality of regression vectors, each of which relates concentration of a constituent compound in the calibration samples to the matrix of scores;
- and
- means for diverting the formation fluid sample into the optical analyzer.
9 more elements
Dependencies are inferred from the claim text.