Information analysis system

Patent metadata

Field	Value
Publication number	US-9836449-B2
Application number	US-201514855517-A
Country	US
Kind code	B2
Filing date	Sep 16, 2015
Priority date	Sep 16, 2015
Publication date	Dec 5, 2017
Grant date	Dec 5, 2017

Abstract

The problem solved by this invention is to convert text information in a geology report to numerical values which reflects geological characteristics of a well's subsurface. Prior art referred above cannot be applicable to this problem. Since text information in the geology report is in the natural language form. This information is not widely used in this industry, due to the fact that the text information can be hardly extracted and summarized into numerical values and integrated into current physical geology models or statistical models. This invention makes the text information in geology report, which is often in a natural language form, easier to be integrated into current geology physical models or statistical models. Also, the numerical values extracted from the geology report can be integrated with other kinds of data, such as seismic data and well-logging data, to obtain more accurate and comprehensive analysis results.

First claim

The invention claimed is:

1. A system which processes a geology report about a subsurface of a well and visualizes information describing a plurality of rock samples collected along a drilled borehole of the well, comprising:
a processor;
a database that stores phrase database for rock properties including a phrase dictionary, a plurality of labeled phrases, and a wellbore database including wellbore information of a plurality of wellbores; and
a memory that stores instructions that cause the processor to execute:
a report receiver unit that receives a geology report from a user selection interface including text information of a plurality of phrases for rock properties describing the rock samples, which are collected at depth intervals of the drilled borehole;
a phrase extraction unit that, for each type of the rock properties, extracts a plurality of phrases about the rock properties and a corresponding plurality of depth intervals of the drilled borehole from the geology report based on the phrase dictionary in the phrase database;
a phrase classification unit that, for each type of the rock properties, classifies the extracted phrases into a plurality of probabilities of the labeled phrases over the plurality of depth intervals;
a numerical generation unit that, for each type of the rock properties, transforms the probabilities of the labeled phrases for the extracted phrases over the plurality of depth intervals into a numerical vector or a numerical scalar; and
an output unit that receives a selection of one of the rock properties, and displays a bird's eye viewpoint of the numerical vector or numerical scalar of the selected rock property of the drilled borehole in relation with different numerical vectors or numerical scalars of other individual wellbores stored in the wellbore database,

wherein the phrase classification unit is trained to classify the extracted phrases in to the plurality of probabilities of labeled phrases using the labeled phrases stored in the database, and
wherein the numerical generation unit uses a trained hidden Markov model to transform the probabilities of the labeled phrases into the numerical vector or numerical scalar.

2. The system according to claim 1 ,
wherein the text information of the geology report is in a natural language form.

3. The system according to claim 1 ,
wherein the geology report is in an image format, and the phrase extraction unit performs optical character recognition on the geology report to extract the depth intervals and the text information of the rock properties from the geology report.

4. The system according to claim 1 ,
wherein the numerical generation unit transforms the probabilities, which are frequencies, of the labeled phrases for the extracted phrases over the plurality of depth intervals into the numerical scalar or the numerical vector.

5. The system according to claim 1 ,
wherein the numerical generation unit calculates a plurality of transition frequencies among the labeled phrases and serializes the transition frequencies into the numerical vector or the numerical scalar.

6. The system according to claim 1 ,
wherein the numerical generation unit converts the probabilities of the labeled phrases describing the rock properties over the plurality of depth intervals into the numerical vector or the numerical scalar based on changes of the probabilities of the labeled phrases over the plurality of depth intervals.

7. The system according to claim 1 ,
wherein the numerical generation unit converts the probabilities of the labeled phrases over the plurality of depth intervals into the numerical vector or the numerical scalar based on physical information of the wellbores in the wellbore database.

8. The system according to claim 7 ,
wherein the physical information of the wellbores includes wellbore shape, and the transformation of the probabilities of the labeled phrases is performed over the depth intervals along a horizontal section of the drilled borehole.

9. The system according to claim 7 ,
wherein the physical information of the wellbores includes a plurality of locations of fracturing points, and the transformation of the probabilities of the labeled phrases is performed over the depth intervals near the fracturing points.

10. The system according to claim 7 ,
wherein the physical information of the wellbores includes a depth-interval measure scale, and the transformation of the probabilities of the labeled phrases is performed according to a ratio of a current depth-interval scale of the drilled borehole to an average depth-interval scale of the wellbores.

11. A method to process a geology report about a subsurface of a well and visualize information describing a plurality of rock samples collected along a drilled borehole of the well, the method comprising:
storing a phrase database for rock properties including a phrase dictionary, a plurality of labeled phrases, and a wellbore database including wellbore information of a plurality of wellbores;
receiving a selection of the geology report from a user interface including text information of a plurality of phrases for rock properties describing the rock samples, which are collected at depth intervals of the drilled borehole;
extracting, for each of the rock properties, a plurality of phrases about the rock properties and a corresponding plurality of depth intervals of the drilled borehole from the geology report based on the phrase dictionary in the phrase database;
classifying, for each of the rock properties, the extracted phrases into a plurality of probabilities of the labeled phrases over the plurality of depth intervals;
transforming, for each of the rock properties, the probabilities of the labeled phrases for the extracted phrases over the plurality of depth intervals into a numerical vector or a numerical scalar; and
receiving a selection of one of the rock properties, and displaying a bird's eye viewpoint of the numerical vector or numerical scalar of the selected rock property of the drilled borehole in relation with different numerical vectors or numerical scalars of other individual wellbores stored in the wellbore database,
wherein the classification of the extracted phrases in to the plurality of probabilities of labeled phrases uses a classifier that is trained with the labeled phrases stored in the database, and
wherein the numerical generation unit uses a trained hidden Markov model to transform the probabilities of the labeled phrases into the numerical vector or numerical scalar.

12. The method according to claim 11 ,
wherein the text information of the geology report is in a natural language form.

13. The method according to claim 11 ,
wherein the geology report is in an image format, and the extraction of the phrases includes performing optical character recognition on the geology report to extract the depth intervals and the text information of the rock properties from the geology report.

14. The method according to claim 11 ,
wherein the probabilities of the labeled phrases are frequencies which are transformed into the numerical scalar or the numerical vector.

15. The method according to claim 11 ,
wherein the transforming of the probabilities of the labeled phrases includes calculating a plurality of transition frequencies among the labeled phrases and serializing the transition frequencies into the numerical vector or the numerical scalar.

16. The method according to claim 11 ,
wherein the transforming of the probabilities of the labeled phrases includes converting the probabilities of the lab

Assignees

• Hitachi Ltd

Inventors

• Tong Bin
• Ozaki Hiroaki
• Iwayama Makoto
• Kobayashi Yoshiyuki
• Vennelakanti Ravigopal
• Sahu Anshuman

Classifications

• G06F40/279

  Recognition of textual entities · CPC title

• G06F17/18 Primary

  for evaluating statistical data {, e.g. average values, frequency distributions, probability functions, regression analysis (forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04)} · CPC title

• G06F17/2765 Primary

  Physics · mapped topic

• E21B47/00

  Survey of boreholes or wells (monitoring pressure or flow of drilling fluid E21B21/08) · CPC title