Collaborative sensing and prediction of source rock properties

Patent metadata

Field	Value
Publication number	US-11352879-B2
Application number	US-201815921003-A
Country	US
Kind code	B2
Filing date	Mar 14, 2018
Priority date	Mar 14, 2017
Publication date	Jun 7, 2022
Grant date	Jun 7, 2022

Abstract

Systems, apparatuses, and computer-implemented methods are provided for the sensing and prediction of properties of source rock. Disclosed here is a method of predicting the maturity of a source rock that includes obtaining a plurality of data of a sample source rock from a plurality of data acquisition devices placed in vicinity of the sample source rock and analyzing the received data using a predictive correlation to determine maturity of the sample source rock. The predictive correlation is generated by applying a machine learning model to correlate the plurality of data acquired from a plurality of representative source rocks with a plurality of properties of the plurality of representative source rocks.

First claim

What is claimed is:

1. A computer implemented method of determining maturity of a sample source rock, the method comprising the steps of:
establishing, by a data analysis engine, communication links with a source rock database and a plurality of data acquisition devices placed in vicinity of a sample source rock, the source rock database containing a first plurality of data acquired from a plurality of representative source rocks and a plurality of properties of the plurality of representative source rocks;
acquiring, by the data analysis engine, a second plurality of data of a sample source rock from the plurality of data acquisition devices; and
analyzing, by the data analysis engine, the second plurality of data using a predictive correlation to determine maturity of the sample source rock, the analyzing comprising determining a spectroscopic wavenumber band for the first plurality of data and at least one of the plurality of properties to differentiate each type of the plurality of representative source rocks, the determining comprising computing a plurality of non-zero weights for a respective plurality of wavenumber points of the spectroscopic wavenumber band, the plurality of non-zero weights based on a sensitivity of the respective plurality of wavenumber points of the spectroscopic wavenumber band to the at least one of the plurality of properties, such that a higher weight of the plurality of non-zero weights corresponds to a greater sensitivity of a respective wavenumber point;
wherein the predictive correlation is generated, by the data analysis engine, by applying a machine learning model to correlate the first plurality of data acquired from the plurality of representative source rocks with the plurality of properties of the plurality of representative source rocks.

2. The computer implemented method of claim 1 , wherein the plurality of data acquisition devices includes a spectrometer comprising a light source, a pyroelectric detector, and a component to reflect light from the sample source rock and direct reflected light to the pyroelectric detector.

3. The computer implemented method of claim 2 , wherein the pyroelectric detector is integrated with a tunable filter.

4. The computer implemented method of claim 2 , wherein the component to reflect light from the sample source rock and direct reflected light to the pyroelectric detector is an attenuated total reflectance unit.

5. The computer implemented method of claim 1 , further comprising the step of
preparing the second plurality of data before the step of analyzing the second plurality of data by the data analysis engine by implementation of one or more of outlier detection, baseline correction, peak enhancement, and normalization.

6. The computer implemented method of claim 1 , further comprising the steps of
storing, by the data analysis engine, the first plurality of data of a sample source rock and the determined maturity of the sample source rock in a source rock database.

7. The computer implemented method of claim 1 , wherein the plurality of properties of the plurality of representative source rocks includes kerogen typing and elemental compositions.

8. The computer implemented method of claim 1 , wherein the first plurality of data includes two or more of location data, spectral measurements, and optical measurements acquired from the plurality of representative source rocks.

9. The computer implemented method of claim 8 , wherein the spectral measurements include one or more of measurements obtained from Fourier Transform Infrared spectroscopy, Electron Spin Resonance spectroscopy, terahertz spectroscopy, and Ultraviolet spectroscopy.

10. The computer implemented method of claim 8 , wherein the first plurality of data further includes pyrolysis data.

11. The computer implemented method of claim 10 , wherein the pyrolysis data is obtained by Rock-Eval® pyrolysis analysis of the plurality of representative source rocks.

12. The computer implemented method of claim 1 , wherein the second plurality of data includes two or more of location data, spectral measurements, and optical measurements acquired from the sample source rock.

13. The computer implemented method of claim 12 , wherein the spectral measurements include one or more of measurements obtained from Fourier Transform Infrared spectroscopy, Electron Spin Resonance spectroscopy, terahertz spectroscopy, and Ultraviolet spectroscopy.

14. The computer implemented method of claim 12 , wherein the optical measurements include one or more of measurements obtained by fluorescence microscopy and confocal laser scanning microscopy.

15. The computer implemented method of claim 1 , wherein the machine learning model is based on one or more of support vector machine, Random Forest®, logistic regression, and Adaptive Boosting algorithms.

16. The computer implemented method of claim 1 , further comprising the step of:
selecting a spectroscopic wavenumber band for operation of the plurality of data acquisition devices in vicinity of the sample source rock.

17. The computer implemented method of claim 16 , wherein the spectroscopic wavenumber band for the sample source rock is selected in response to receiving, by the data analysis engine, one or more selections of desired maturity and desired organofacies profile of the sample source rock from a user interface.

18. A system to determine maturity of a sample source rock, the system comprising:
a plurality of data acquisition devices placed in vicinity of a sample source rock and communicatively coupled to a computing device;
the computing device coupled to a source rock database via a communication network and configured to:
obtain a first plurality of data of a sample source rock from the plurality of data acquisition devices; and
analyze the first plurality of data using a predictive correlation to determine maturity of the sample source rock,
wherein the predictive correlation is generated by applying a machine learning model to correlate a second plurality of data acquired from a plurality of representative source rocks with a plurality of properties of the plurality of representative source rocks; and

the source rock database containing the second plurality of data associated with the plurality of representative source rocks, the plurality of properties of the plurality of representative source rocks, and the predictive correlation, wherein the operation of analyze the first plurality of data comprises determining a spectroscopic wavenumber band for the second plurality of data and at least one of the plurality of properties that differentiates each type of the plurality of representative source rocks the determining comprising computing a plurality of non-zero weights for a respective plurality of wavenumber points of the spectroscopic wavenumber band, the plurality of non-zero weights based on a sensitivity of the respective plurality of wavenumber points of the spectroscopic wavenumber band to the at least one of the plurality of properties, such that a higher weight of the plurality of non-zero weights corresponds to a greater sensitivity of a respective wavenumber point.

19. The system of claim 18 , wherein the plurality of data acquisition devices is positioned to acquire data of the sample source rock using the spectroscopic wavenumber band.

20. The system of claim 18 , further comprising a sample source rock retrieving apparatus to obtain a portion of the sample source rock.

21. The system of claim 18 , where

Assignees

• Saudi Arabian Oil Co

Inventors

• Li Weichang
• Csutak Sebastian
• Jacobi David
• Mcalpin Tiffany Dawn
• Deffenbaugh Max
• Eichmann Shannon Lee

Classifications

• G06N20/20 Primary

  Ensemble learning · CPC title

• G01N33/24

  Earth materials (G01N33/42 takes precedence) · CPC title

• G16C20/30

  Prediction of properties of chemical compounds, compositions or mixtures · CPC title

• G01N21/3563

  for analysing solids; Preparation of samples therefor · CPC title

• G01N2021/3595

  using FTIR · CPC title