# US 2022/0035068 A1 · Systems and methods for identifying subsurface features as a function of position in a subsurface volume of interest

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2022/0035068 A1
- Document ID: US20220035068A1
- Status: Published application
- Kind: A1
- Application number: 16/945,486
- Filed: 2020-07-31
- Published: 2022-02-03
- Earliest claimed: 2020-07-31
- Assignees: Chevron U.S.A. Inc.
- Applicants: Chevron U.S.A. Inc.
- Inventors: Shuxing Cheng; Zhao Zhang; Kellen Leigh Gunderson; Reynaldo Cardona
- Web page: https://patentgrep.com/patent/US20220035068A1

## CPC

- G01V 20/00: Geomodelling in general
- G01V 1/301: {for determining seismic cross-sections or geostructures}
- G06F 30/27: using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
- G06N 3/045: Combinations of networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- G06N 20/00: Machine learning
- G01V 2210/64: Geostructures, e.g. in 3D data cubes
- G06F 2113/08: Fluids
- G01V: GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS (means for indicating the location of accidentally buried, e.g. snow-buried, persons A63B29/02)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Systems, devices, and methods are disclosed for identifying subsurface features as a function of position in a subsurface volume of interest. A computer-implemented method may include obtaining training subsurface data and corresponding training subsurface feature data; obtaining an initial subsurface feature model including tiers of elements; generating a conditioned subsurface feature model by training the initial subsurface feature model using the training subsurface data and the corresponding training subsurface feature data; and storing the conditioned subsurface feature model in the non-transient electronic storage.

## Claims

### Claim 1

A computer-implemented method for training a subsurface model to identify subsurface features as a function of position in a subsurface volume of interest, the method being implemented in a computer system that comprises a physical computer processor and non-transient electronic storage, the method comprising:
obtaining, from the non-transient electronic storage, training subsurface data comprising a subsurface energy value and corresponding training subsurface feature data comprising a training structural identification value defined by structural identification parameters and a corresponding training categorization value defined by categorization parameters and the training structural identification value;
obtaining, from the non-transient electronic storage, an initial subsurface feature model comprising tiers of elements, wherein the tiers comprise a first tier comprising a structural identification element corresponding to the training structural identification value and a second tier, dependent on the structural identification value, comprising a categorization element corresponding to the training categorization value;
generating, with the physical computer processor, a conditioned subsurface feature model by training the initial subsurface feature model using the training subsurface data and the corresponding training subsurface feature data; and
storing the conditioned subsurface feature model in the non-transient electronic storage.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, further comprising:
obtaining, from the non-transient electronic storage, target subsurface data comprising a subsurface energy value as a function of position in the subsurface volume of interest; and
generating, with the physical computer processor, target subsurface feature data by applying the first conditioned subsurface feature model to the target subsurface data, wherein the target subsurface feature data comprises a target structural identification value defined by the structural identification parameters as a function of position in the subsurface volume of interest and a corresponding target categorization value defined by the categorization parameters and the target structural identification value as a function of position in the subsurface volume of interest.

### Claim 3 (depends on 2)

The computer-implemented method of claim 2, wherein the computer system comprises a display, and wherein the computer-implemented method further comprises:
generating, with the physical computer processor, a representation of geological structures as a function of position in the subsurface volume of interest using visual effects to depict at least some of the target subsurface feature data; and
displaying the representation via the display.

### Claim 4 (depends on 2)

The computer-implemented method of claim 2, wherein the structural identification values indicate whether or not there is a subsurface feature.

### Claim 5 (depends on 2)

The computer-implemented method of claim 2, wherein the categorization values indicate one of what category the subsurface feature is categorized as and what sub-category the subsurface feature is categorized as.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, wherein the structural identification parameters comprise one of a geometric shape of subsurface horizons and a truncation of the subsurface horizons against other geologic elements.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, wherein the categorization parameters comprise one of a histogram of oriented gradients (HOG), a scale-invariant feature transform (SIFT), an analog geologic feature.

### Claim 8 (depends on 1)

The computer-implemented method of claim 1, wherein different ones of the structural identification parameters and the categorization parameters are weighted differently based on the conditioned subsurface feature model.

### Claim 9 (depends on 1)

The computer-implemented method of claim 1, wherein a structure of the tiers corresponds to a hierarchy of geological taxonomy.

### Claim 10 (depends on 1)

The computer-implemented method of claim 1, wherein the categorization element comprises one of an extensional element, a faulted element, a contractional fold element, a salt element, a fault bend fold element, a tip-line fold element, a detachment fold element, and a gas chimney element.

### Claim 11 (depends on 1)

The computer-implemented method of claim 1, wherein the subsurface feature models comprise one of a mask R-CNN network, a feature pyramids network, an object detection network, a neural network, a random forest model, a support vector machine, a regression, and a Bayesian network.

### Claim 12 (depends on 2)

The computer-implemented method of claim 2, further comprising generating, with the physical computer processor, uncertainty corresponding to the target subsurface feature data, wherein the conditioned subsurface feature model further comprises a subsurface feature uncertainty model, and wherein applying the uncertainty model to target subsurface data generates target subsurface feature data probabilities indicating a structural identification value probability and a corresponding target categorization value probability.

### Claim 13

A computer-implemented method for identifying subsurface features as a function of position in a subsurface volume of interest, the method being implemented in a computer system that comprises a physical computer processor and electronic storage, the method comprising:
obtaining, from the electronic storage, a conditioned subsurface feature model comprising tiers of elements, the conditioned subsurface feature model having been generated by applying the initial subsurface feature model to the training subsurface data and the training subsurface feature data, wherein the tiers comprise a first tier comprising structural identification elements corresponding to training structural identification values and a second tier, dependent on the structural identification values, comprising categorization elements corresponding to training categorization values, wherein the training subsurface data comprises subsurface energy values, wherein the training subsurface feature data comprises the training structural identification values defined by structural identification parameters and corresponding training categorization values defined by categorization parameters and one of the training structural identification values;
obtaining, from the non-transient electronic storage, target subsurface data comprising a subsurface energy value as a function of position in the subsurface volume of interest; and
generating, with the physical computer processor, target subsurface feature data by applying the first conditioned subsurface feature model to the target subsurface data, wherein the target subsurface feature data comprises a target structural identification value defined by the structural identification parameters as a function of position in the subsurface volume of interest and a corresponding target categorization value defined by the categorization parameters and the target structural identification value as a function of position in the subsurface volume of interest.

### Claim 14 (depends on 13)

The computer-implemented method of claim 13, wherein the computer system comprises a display, and wherein the computer-implemented method further comprises:
generating, with the physical computer processor, a representation of geological structures as a function of position in the subsurface volume of interest using visual effects to depict at least some of the target subsurface feature data; and
displaying the representation via the display.

### Claim 15 (depends on 13)

The computer-implemented method of claim 13, wherein the categorization element comprises one of an extensional element, a faulted element, a contractional fold element, a salt element, a fault bend fold element, a tip-line fold element, a detachment fold element, and a gas chimney element.

### Claim 16 (depends on 13)

The computer-implemented method of claim 13, wherein the subsurface feature models comprise one of a mask R-CNN network, a feature pyramids network, an object detection network, a neural network, a random forest model, a support vector machine, a regression, and a Bayesian network.

### Claim 17 (depends on 13)

The computer-implemented method of claim 13, further comprising generating, with the physical computer processor, uncertainty corresponding to the target subsurface feature data based on
H
⁡
(
P
)
=
-
∑
c
-
1
c
=
M
⁢
P
⁡
(
SubFeat
=
C
)
⁢
⁢
log
⁢
⁢
P
⁡
(
SubFeat
=
C
)
where c represents a given category of the subsurface feature, M represents different categories of subsurface features, and P(SubFeat=C) represents a probability of the recognized subsurface feature of a given category.

### Claim 18

A system for identifying subsurface features as a function of position in a subsurface volume of interest, the system comprising:
electronic storage; and
a physical computer processor configured by machine readable instructions to:
obtain, from the electronic storage, a conditioned subsurface feature model comprising tiers of elements, the conditioned subsurface feature model having been generated by applying the initial subsurface feature model to the training subsurface data and the training subsurface feature data, wherein the tiers comprise a first tier comprising structural identification elements corresponding to training structural identification values and a second tier, dependent on the structural identification values, comprising categorization elements corresponding to training categorization values, wherein the training subsurface data comprises subsurface energy values, wherein the training subsurface feature data comprises the training structural identification values defined by structural identification parameters and corresponding training categorization values defined by categorization parameters and one of the training structural identification values;
obtain, from the non-transient electronic storage, target subsurface data comprising a subsurface energy value as a function of position in the subsurface volume of interest; and
generate, with the physical computer processor, target subsurface feature data by applying the first conditioned subsurface feature model to the target subsurface data, wherein the target subsurface feature data comprises a target structural identification value defined by the structural identification parameters as a function of position in the subsurface volume of interest and a corresponding target categorization value defined by the categorization parameters and the target structural identification value as a function of position in the subsurface volume of interest, wherein the target structural identification value and the corresponding target categorization value are based on a subset of the target subsurface data.

### Claim 19 (depends on 18)

The system of claim 18, wherein the system further comprises a display, and wherein the physical computer processor is further configured by machine readable instructions to:
generate, with the physical computer processor, a representation of geological structures as a function of position in the subsurface volume of interest using visual effects to depict at least some of the target subsurface feature data; and
display the representation via the display.

### Claim 20 (depends on 18)

The system of claim 18, wherein the subsurface feature models comprise one of a mask R-CNN network, a feature pyramids network, an object detection network, a neural network, a random forest model, a support vector machine, a regression, and a Bayesian network.
