# US 2026/0188532 A1 · Robust automatic tracking of individual triso-fueled pebbles through a novel application of x-ray imaging and machine learning

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

- Number: US 2026/0188532 A1
- Document ID: US20260188532A1
- Status: Published application
- Kind: A1
- Application number: 19/545,572
- Filed: 2026-02-20
- Published: 2026-07-02
- Earliest claimed: 2020-03-10
- Assignees: University of Florida Research Foundation, Inc.
- Applicants: University of Florida Research Foundation, Inc.
- Inventors: Kyle Hartig; Hongcheng Liu; Emily Kwapis
- Web page: https://patentgrep.com/patent/US20260188532A1

## CPC

- G21C 17/06: Devices or arrangements for monitoring or testing fuel or fuel elements outside the reactor core, e.g. for burn-up, for contamination (G21C17/08, G21C17/10 take precedence; detecting leaking fuel elements during reactor operation G21C17/04)
- G06F 16/55: Clustering; Classification
- G06V 10/10: Image acquisition (document image scanning and transmission H04N1/00; control of digital cameras H04N23/60)
- G06V 10/82: using neural networks
- G06V 20/693: {Acquisition}
- G21C 17/063: {Burn-up control (G21C17/066 takes precedence)}
- G21C 17/066: {Control of spherical elements}
- Y02E 30/30: Nuclear fission reactors
- G21C: NUCLEAR REACTORS (fusion reactors, hybrid fission-fusion reactors G21B; nuclear explosives G21J)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- Y02E: REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION

## Abstract

The present disclosure presents systems and methods of tagging TRISO-fueled pebbles. One such method comprises acquiring an ionizing radiation image of a TRISO-fueled pebble; analyzing, using a machine learning algorithm, the acquired image of the TRISO-fueled pebble to identify a unique pattern of particle distributions that is visible in the acquired image of the TRISO-fueled pebble; and matching the unique pattern of particle distributions with a unique pattern of particle distributions previously stored in an image database for the TRISO-fueled pebble. Other systems and methods are also presented.

## Claims

### Claim 1

A method for tagging a plurality of TRISO-fueled pebbles flowing throughout a pebble bed reactor core, the method comprising:
before the plurality of TRISO-fueled pebbles exit the pebble bed reactor core, for each of the plurality of TRISO-fueled pebbles:
acquiring a first ionizing radiation image of the TRISO-fueled pebble;
analyzing, using a machine learning algorithm, the acquired first ionizing radiation image of the TRISO-fueled pebble to identify a first unique pattern of TRISO-fuel particle distributions within the TRISO-fueled pebble that is visible in the acquired first ionizing radiation image of the TRISO-fueled pebble; and
storing the first ionizing radiation image of the TRISO-fueled pebble in an image database;
acquiring a second ionizing radiation image of one of the plurality of TRISO-fueled pebbles exiting the pebble bed reactor core;
analyzing the acquired second ionizing radiation image of the exiting TRISO-fueled pebble to identify, using the machine learning algorithm, a second unique pattern of TRISO-fuel particle distributions within the exiting TRISO-fueled pebble that is visible in the acquired second ionizing radiation image of the exiting TRISO-fueled pebble;
matching the second unique pattern of TRISO-fuel particle distributions with the first unique pattern of TRISO-fuel particle distributions visible in the first ionizing radiation image stored in the image database;
obtaining a current measurement of a burnup value for the exiting TRISO-fueled pebble;
obtaining a stored burnup value associated with the exiting TRISO-fueled pebble; and
validating that the second unique pattern of TRISO-fuel particle distributions was successfully matched with the first unique pattern of TRISO-fuel particle distributions for the exiting TRISO-fueled pebble by comparing the stored burnup value with the current measurement of the burnup value.

### Claim 2 (depends on 1)

The method of claim 1, wherein the acquired first ionizing radiation image of each of the plurality of TRISO-fueled pebbles comprises an X-ray image.

### Claim 3 (depends on 1)

The method of claim 1, wherein the acquired first ionizing radiation image of each of the plurality of TRISO-fueled pebbles comprises a neutron tomography image.

### Claim 4 (depends on 1)

The method of claim 1, further comprising:
after storing the first ionizing radiation image for each of the plurality of TRISO-fueled pebbles in the image database, introducing the plurality of TRISO-fueled pebbles into the pebble bed reactor core.

### Claim 5 (depends on 1)

The method of claim 1, wherein comparing the stored burnup value with the current measurement of the burnup value comprises determining whether the current measurement of the burnup value is within a set range of the stored burnup value.

### Claim 6 (depends on 1)

The method of claim 1, further comprising:
obtaining a measurement of a burnup value for each of the plurality of TRISO-fueled pebbles before the plurality of TRISO-fueled pebbles exit the pebble bed reactor core; and
storing the measurement of the burnup value for each of the plurality of TRISO-fueled pebbles.

### Claim 7 (depends on 1)

The method of claim 1, further comprising deriving a fingerprint identifier for the TRISO-fueled pebble from the first unique pattern of TRISO-fuel particle distributions within the TRISO-fueled pebble and storing the fingerprint identifier in the image database.

### Claim 8

A system for tagging a plurality of TRISO-fueled pebbles flowing throughout a pebble bed reactor core, the system comprising:
an imaging system that is configured to capture ionizing radiation images of the plurality of TRISO-fueled pebbles; and
a computing device having a memory and a processor,
wherein, before the plurality of TRISO-fueled pebbles exit the pebble bed reactor core, for each of the plurality of TRISO-fueled pebbles, the processor is configured to:
analyze, using a machine learning algorithm, a first captured image of the TRISO-fueled pebble to identify a first unique pattern of TRISO-fuel particle distributions within the TRISO-fueled pebble that is visible in the first captured image of the TRISO-fueled pebble, wherein the first captured image is a first ionizing radiation image obtained from the imaging system; and
store the first captured image of the TRISO-fueled pebble in an image database;
wherein, after the plurality of TRISO-fueled pebbles exit the pebble bed reactor core, for each of the plurality of TRISO-fueled pebbles, the processor is configured to:
acquire a second image of one of the plurality of TRISO-fueled pebbles exiting the pebble bed reactor core;
analyze the acquired second image of the exiting TRISO-fueled pebble to identify, using the machine learning algorithm, a second unique pattern of TRISO-fuel particle distributions within the exiting TRISO-fueled pebble that is visible in the acquired second image of the exiting TRISO-fueled pebble, wherein the acquired second image is a second ionizing radiation image obtained from the imaging system;
match the second unique pattern of TRISO-fuel particle distributions with the first unique pattern of TRISO-fuel particle distributions visible in the first ionizing radiation image stored in the image database;
obtain a current measurement of a burnup value for the exiting TRISO-fueled pebble;
obtain a stored burnup value associated with the exiting TRISO-fueled pebble; and
validate that the second unique pattern of TRISO-fuel particle distributions was successfully matched with the first unique pattern of TRISO-fuel particle distributions for the exiting TRISO-fueled pebble by comparing the stored burnup value with the current measurement of the burnup value.

### Claim 9 (depends on 8)

The system of claim 8, wherein the first captured image of each of the plurality of TRISO-fueled pebbles comprises an X-ray image.

### Claim 10 (depends on 8)

The system of claim 8, wherein the first captured image of each of the plurality of TRISO-fueled pebbles comprises a neutron tomography image.

### Claim 11 (depends on 8)

The system of claim 8, wherein comparing the stored burnup value with the current measurement of the burnup value comprises determining whether the current measurement of the burnup value is within a set range of the stored burnup value.

### Claim 12 (depends on 8)

The system of claim 8, wherein the computing device is configured to train a deep neural network to analyze the first captured image of each of the plurality of TRISO-fueled pebbles using the machine learning algorithm, wherein the deep neural network is executed by the processor of the computing device.

### Claim 13 (depends on 8)

The system of claim 8, wherein the processor is further configured to:
obtain a measurement of a burnup value for each of the plurality of TRISO-fueled pebbles before the plurality of TRISO-fueled pebbles exit the pebble bed reactor core; and
store the measurement of the burnup value for each of the plurality of TRISO-fueled pebbles.

### Claim 14

A non-transitory, tangible computer-readable storage medium having instructions stored thereon that, in response to execution by a computer-based system, cause the computer-based system to perform operations for tagging a plurality of TRISO-fueled pebbles flowing throughout a pebble bed reactor core, the operations comprising:
before a plurality of TRISO-fueled pebbles exit a pebble bed reactor core, for each of the plurality of TRISO-fueled pebbles,
analyzing, using a machine learning algorithm, a first captured image of the TRISO-fueled pebble to identify a first unique pattern of TRISO-fuel particle distributions within the TRISO-fueled pebble that is visible in the first captured image of the TRISO-fueled pebble, wherein the first captured image is a first ionizing radiation image obtained from an imaging system; and
storing the first captured image of the TRISO-fueled pebble in an image database; and
after the plurality of TRISO-fueled pebbles exit the pebble bed reactor core, for each of the plurality of TRISO-fueled pebbles,
acquiring a second image of one of the plurality of TRISO-fueled pebbles exiting the pebble bed reactor core;
analyzing the acquired second image of the exiting TRISO-fueled pebble to identify, using the machine learning algorithm, a second unique pattern of TRISO-fuel particle distributions within the exiting TRISO-fueled pebble that is visible in the acquired second image of the exiting TRISO-fueled pebble, wherein the acquired second image is a second ionizing radiation image obtained from the imaging system;
matching the second unique pattern of TRISO-fuel particle distributions with the first unique pattern of TRISO-fuel particle distributions visible in the first ionizing radiation image stored in the image database;
obtaining a current measurement of a burnup value for the exiting TRISO-fueled pebble;
obtaining a stored burnup value associated with the exiting TRISO-fueled pebble; and
validating that the second unique pattern of TRISO-fuel particle distributions was successfully matched with the first unique pattern of TRISO-fuel particle distributions for the exiting TRISO-fueled pebble by comparing the stored burnup value with the current measurement of the burnup value.

### Claim 15 (depends on 14)

The non-transitory, tangible computer-readable storage medium of claim 14, wherein the first captured image of each of the plurality of TRISO-fueled pebbles comprises an X-ray image.

### Claim 16 (depends on 14)

The non-transitory, tangible computer-readable storage medium of claim 14, wherein the first captured image of each of the plurality of TRISO-fueled pebbles comprises a neutron tomography image.

### Claim 17 (depends on 14)

The non-transitory, tangible computer-readable storage medium of claim 14, wherein comparing the stored burnup value with the current measurement of the burnup value comprises determining whether the current measurement of the burnup value is within a set range of the stored burnup value.

### Claim 18 (depends on 14)

The non-transitory, tangible computer-readable storage medium of claim 14, wherein the operations further comprise training a deep neural network to analyze the first captured image of each of the plurality of TRISO-fueled pebbles using the machine learning algorithm.

### Claim 19 (depends on 14)

The non-transitory, tangible computer-readable storage medium of claim 14, wherein the operations further comprise:
obtaining a measurement of a burnup value for each of the plurality of TRISO-fueled pebbles before the plurality of TRISO-fueled pebbles exit the pebble bed reactor core; and
storing the measurement of the burnup value for each of the plurality of TRISO-fueled pebbles.

### Claim 20 (depends on 14)

The non-transitory, tangible computer-readable storage medium of claim 14, wherein the operations further comprise deriving a fingerprint identifier for the TRISO-fueled pebble from the first unique pattern of TRISO-fuel particle distributions within the TRISO-fueled pebble and storing the fingerprint identifier in the image database.
