# US 2019/0095765 A1 · Machine learning system for in-situ recognition of common locations in a rotatable body with repeating segments

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

- Number: US 2019/0095765 A1
- Document ID: US20190095765A1
- Status: Published application
- Kind: A1
- Application number: 15/714,208
- Filed: 2017-09-25
- Published: 2019-03-28
- Earliest claimed: 2017-09-25
- Assignees: General Electric Company
- Applicants: General Electric Company
- Inventors: Ser Nam Lim; David Scott Diwinsky
- Web page: https://patentgrep.com/patent/US20190095765A1

## CPC

- G01N 21/954: Inspecting the inner surface of hollow bodies, e.g. bores
- F01D 5/005: {Repairing methods or devices}
- F01D 21/003: {Arrangements for testing or measuring (for measuring vibrations G01H)}
- G06F 18/217: {Validation; Performance evaluation; Active pattern learning techniques}
- G06T 7/0006: {using a design-rule based approach}
- G06V 10/75: Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- F05D 2220/32: in gas turbines
- F05D 2230/80: Repairing, retrofitting or upgrading methods
- F05D 2260/80: Diagnostics
- G06T 7/0004: {Industrial image inspection}
- G06T 7/20: Analysis of motion (motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51)
- G06T 7/70: Determining position or orientation of objects or cameras (camera calibration G06T7/80)
- G06T 2207/30164: Workpiece; Machine component
- G06V 2201/06: Recognition of objects for industrial automation
- G01N: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES (measuring or testing apparatus or processes other than immunoassay, involving enzymes or microorganisms C12M, C12Q)
- F01D: NON-POSITIVE DISPLACEMENT MACHINES OR ENGINES, e.g. STEAM TURBINES (machines or engines for liquids F03; non-positive displacement pumps F04D)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- F05D: INDEXING SCHEME FOR ASPECTS RELATING TO NON-POSITIVE-DISPLACEMENT MACHINES OR ENGINES, GAS-TURBINES OR JET-PROPULSION PLANTS

## Abstract

A system includes one or more processors configured to automatically identify different distressed portions in repeating segments of a rotating body. At least one of a size and/or a shape of one or more of the distressed portions changes with respect to time. The one or more processors also are configured to determine a pattern of the different distressed portions in the repeating segments of the rotating body during rotation of the rotating body based on identifying the different distressed portions. The one or more processors also are configured to subsequently automatically identify locations of individual segments of the repeating segments in the rotating body using the pattern of the distressed portions that is determined.

## Claims

### Claim 1

A method comprising:
automatically identifying, using a machine learning system, different distressed portions in repeating segments of a rotating body, wherein at least one of a size or a shape of one or more of the distressed portions changes with respect to time;
determining a pattern of the different distressed portions in the repeating segments of the rotating body during rotation of the rotating body based on identifying the different distressed portions; and
subsequently automatically identifying, using the same machine learning system or another machine learning system, locations of individual segments of the repeating segments in the rotating body using the pattern of the distressed portions that is determined.

### Claim 2 (depends on 1)

The method of claim 1, further comprising tracking changes in the distressed portions of the repeating segments of the rotating body by consistently tracking the locations of the individual segments of the repeating segments in the rotating body during subsequent examinations of the distressed portions of the repeating segments using the pattern that is determined.

### Claim 3 (depends on 1)

The method of claim 1, wherein identifying the locations of the individual segments of the repeating segments occurs after the size, the shape, or both the size and the shape of the one or more distressed portions in the repeating segments of the rotating body changes.

### Claim 4 (depends on 1)

The method of claim 1, further comprising using the pattern to individually identify, label, or both identify and label each of the repeating segments of the rotating body.

### Claim 5 (depends on 1)

The method of claim 1, further comprising tracking changes in the size, the shape, or both the size and the shape of the distressed portions of the repeating segments of the rotating body using the pattern that is determined.

### Claim 6 (depends on 1)

The method of claim 1, further comprising automatically implementing a repair action to change a state of the rotating body from a damaged state to a repaired state by repairing at least one of the distressed portions.

### Claim 7 (depends on 1)

The method of claim 1, wherein the rotating body includes a turbine machine and the repeating segments include turbine blades.

### Claim 8 (depends on 1)

The method of claim 1, wherein the distressed portions of the repeating segments of the rotating body include one or more cracks, spalls, or pits in turbine blades of a turbine machine.

### Claim 9 (depends on 1)

The method of claim 1, wherein the pattern that is determined includes an order in which the different distressed portions of the repeating segments of the rotating body are encountered by the machine learning system during rotation of the rotating body.

### Claim 10 (depends on 1)

The method of claim 1, wherein the pattern that is determined includes a spatial separation gap, a temporal separation gap, or a combination of the spatial separation gap and the temporal separation gap between the different distressed portions of the repeating segments in the rotating body during rotation of the rotating body.

### Claim 11 (depends on 1)

The method of claim 1, wherein automatically identifying the distressed portions of the repeating segments in the rotating body and subsequently automatically identifying the locations of the segments in the rotating body include optically detecting the distressed portions using the machine learning system or the other machine learning system.

### Claim 12 (depends on 1)

The method of claim 1, wherein automatically identifying the distressed portions of the repeating segments in the rotating body and subsequently automatically identifying the locations of the segments of the rotating body occur during rotation of the rotating body in a common direction.

### Claim 12

A system comprising:
one or more processors configured to automatically identify different distressed portions in repeating segments of a rotating body, wherein at least one of a size or a shape of one or more of the distressed portions changes with respect to time, the one or more processors also configured to determine a pattern of the different distressed portions in the repeating segments of the rotating body during rotation of the rotating body based on identifying the different distressed portions, wherein the one or more processors also are configured to subsequently automatically identify locations of individual segments of the repeating segments in the rotating body using the pattern of the distressed portions that is determined.

### Claim 13 (depends on 12)

The system of claim 12, wherein the one or more processors also are configured to track changes in the distressed portions of the repeating segments of the rotating body by consistently tracking the locations of the individual segments of the repeating segments in the rotating body during subsequent examinations of the distressed portions of the repeating segments using the pattern that is determined.

### Claim 14 (depends on 12)

The system of claim 12, wherein the one or more processors are configured to identify the locations of the individual segments of the repeating segments occurs after the size, the shape, or both the size and the shape of the one or more distressed portions in the repeating segments of the rotating body changes.

### Claim 15 (depends on 12)

The system of claim 12, wherein the one or more processors are configured to use the pattern to individually identify, label, or both identify and label each of the repeating segments of the rotating body.

### Claim 16 (depends on 12)

The system of claim 12, wherein the one or more processors also are configured to track changes in the size, the shape, or both the size and the shape of the distressed portions of the repeating segments of the rotating body using the pattern that is determined.

### Claim 17

A method comprising:
automatically identifying, using a machine learning system, plural different damaged blades of a turbine engine, wherein at least one of a size or a shape of damage to one or more of the damaged blades changes with respect to time;
determining a sequential pattern of the different damaged blades of the turbine engine during rotation of the turbine engine based on identifying the different damaged blades; and
subsequently automatically identifying, using the same machine learning system or another machine learning system, the damaged blades of the turbine engine after the size, the shape, or both the size and the shape of the damage to the one or more damaged blades changes.

### Claim 18 (depends on 17)

The method of claim 17, further comprising automatically implementing a repair action to change a state of the turbine engine from a damaged state to a repaired state by repairing at least one of the damaged blades.

### Claim 19 (depends on 17)

The method of claim 17, wherein the damage to the damaged blades of the turbine engine include one or more cracks, spalls, or pits in the turbine blades.

### Claim 20 (depends on 17)

The method of claim 17, wherein the sequential pattern that is determined includes an order in which the different damaged blades of the turbine engine are encountered by the machine learning system during rotation of the blades.
