# US 2022/0035334 A1 · Technologies for producing training data for identifying degradation of physical components

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

- Number: US 2022/0035334 A1
- Document ID: US20220035334A1
- Status: Published application
- Kind: A1
- Application number: 16/941,594
- Filed: 2020-07-29
- Published: 2022-02-03
- Earliest claimed: 2020-07-29
- Assignees: ABB Schweiz AG
- Applicants: ABB Schweiz AG
- Inventors: Nolan Nicholas; Remus Boca; Mirrasoul Mousavi; Mithun Acharya
- Web page: https://patentgrep.com/patent/US20220035334A1

## CPC

- G01M 99/002: {Thermal testing (flaw detection G01N25/72)}
- G05B 19/4065: Monitoring tool breakage, life or condition
- G01M 11/081: {by using a contact-less detection method, i.e. with a camera}
- G01N 17/002: {Test chambers}
- G06F 30/27: using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
- G06N 3/0475: Generative networks
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- G06N 3/094: Adversarial learning
- G01N 2021/8883: {involving the calculation of gauges, generating models}
- G06F 2119/04: Ageing analysis or optimisation against ageing
- G01M: TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G)
- 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)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Technologies for producing training data for identifying degradation of physical components include a system. The system includes circuitry configured to apply an accelerated degradation process to a physical component of an industrial plant. Additionally, the circuitry of the system is configured to obtain measurement data indicative of visual characteristics of the physical component at each of multiple phases of degradation, wherein the measurement data is usable to train a neural network to identify a phase of degradation of another physical component.

## Claims

### Claim 1

A system comprising:
circuitry configured to:
apply an accelerated degradation process to a physical component of an industrial plant; and
obtain measurement data indicative of visual characteristics of the physical component at each of multiple phases of degradation, wherein the measurement data is usable to train a neural network to identify a phase of degradation of another physical component.

### Claim 2 (depends on 1)

The system of claim 1, wherein to apply the accelerated degradation process comprises to apply the accelerated degradation process to the physical component in a degradation chamber configured to produce a target environment within the degradation chamber.

### Claim 3 (depends on 1)

The system of claim 1, wherein the circuitry is further configured to:
determine a degradation model from the measurement data; and
produce simulated measurement data using the degradation model, wherein the simulated measurement data is indicative of characteristics of the physical component at multiple phases of degradation and is usable as training data for the neural network.

### Claim 4 (depends on 1)

The system of claim 1, wherein the circuitry is further configured to obtain measurement data that is additionally indicative of a performance characteristic of the physical component at each of the multiple phases of degradation.

### Claim 5 (depends on 1)

The system of claim 1, wherein to obtain the measurement data comprises to obtain the measurement data with a robot having a sensor configured to produce the measurement data.

### Claim 6 (depends on 1)

The system of claim 1, wherein to apply the accelerated degradation process comprises to subject the physical component to vibration, gas, vapor, abrasive conditions, impacts, thermal cycling, thermal shock, liquid spray, liquid soak, humidity, light, radiation, mold, fungus, or an electric arc in a degradation chamber.

### Claim 7 (depends on 1)

The system of claim 1, wherein to obtain measurement data indicative of visual characteristics comprises to obtain measurement data indicative of rust, corrosion, discoloration, decomposition, wear, weathering, leaching, crazing, pitting, or cracking.

### Claim 8 (depends on 1)

The system of claim 1, wherein the circuitry is further configured to obtain measurement data that is additionally indicative of a performance characteristic of the physical component at each of the multiple phases of degradation by performing at least one of strength testing, cycle fatigue resistance testing, vibration resistance testing, modulus testing, and softening point testing.

### Claim 9 (depends on 1)

The system of claim 1, wherein to obtain measurement data comprises to perform destructive measurements on multiple samples of the physical component.

### Claim 10 (depends on 1)

The system of claim 1, wherein to apply an accelerated degradation process to a physical component of an industrial plant comprises to apply an accelerated degradation process to a representative subsection of the physical component.

### Claim 11 (depends on 1)

The system of claim 1, wherein the circuitry is further to:
determine a degradation model from the measurement data, including performing feature extraction to identify characteristics of corresponding phases of degradation and utilizing a symbolic regression engine to identify correlations in feature development as a function of time; and
produce simulated measurement data using the degradation model, wherein the simulated measurement data is indicative of characteristics of the physical component at multiple phases of degradation and is usable as training data for the neural network.

### Claim 12 (depends on 11)

The system of claim 11, wherein the circuitry is further to incorporate, with the regression engine, a known equation that describes a degradation process of the physical component.

### Claim 13 (depends on 11)

The system of claim 11, wherein the circuitry is further to determine the degradation model for one or more local geometries of the physical component.

### Claim 14 (depends on 13)

The system of claim 13, wherein to determine the degradation model for one or more local geometries comprises to determine the degradation model for raised features or inset features.

### Claim 15 (depends on 1)

The system of claim 1, wherein the circuitry is further to produce the simulated measurement data with a neural network that has been trained with the measurement data.

### Claim 16 (depends on 1)

The system of claim 1, wherein to produce the simulated measurement data with a neural network comprises to produce the simulated measurement data with a generative adversarial network.

### Claim 17

A method comprising:
applying, by a system for producing training data, an accelerated degradation process to a physical component of an industrial plant; and
obtaining, by the system, measurement data indicative of visual characteristics of the physical component at each of multiple phases of degradation, wherein the measurement data is usable to train a neural network to identify a phase of degradation of another physical component.

### Claim 18 (depends on 17)

The method of claim 17, wherein applying the accelerated degradation process comprises applying the accelerated degradation process to the physical component in a degradation chamber configured to produce a target environment within the degradation chamber.

### Claim 19 (depends on 17)

The method of claim 17, further comprising:
determining, by the system, a degradation model from the measurement data; and
producing, by the system, simulated measurement data using the degradation model, wherein the simulated measurement data is indicative of characteristics of the physical component at multiple phases of degradation and is usable as training data for the neural network.

### Claim 20

One or more machine-readable storage media comprising a plurality of instructions stored thereon that, in response to being executed, cause a system to:
apply an accelerated degradation process to a physical component of an industrial plant; and
obtain measurement data indicative of visual characteristics of the physical component at each of multiple phases of degradation, wherein the measurement data is usable to train a neural network to identify a phase of degradation of another physical component.
