Granted Patent B2

US 12,392,678 · App. 17/971,189 · Granted Aug 19, 2025

Inventor: Quan Shen (Houston, TX)

Assignee: Rebellion Photonics, Inc.

G01M3/04G06N5/022

Abstract

Claims (48)

1. An apparatus comprising at least one processor and at least one non-transitory memory comprising program code, wherein the at least one non-transitory memory and the program code are configured to, with the at least one processor, cause the apparatus to at least:

obtain a plurality of gas column density maps associated with one or more gas leak tests, each gas column density map having an associated distance from a gas leak imaging device to a corresponding one of the one or more gas leak tests and an associated leak rate of a corresponding one of the one or more gas leak tests;

determine if the plurality of gas column density maps meet a predefined quality threshold;

if the plurality of gas column density maps does not meet the predefined quality threshold, filter the plurality of gas column density maps to remove any gas column density maps that do not meet one or more predefined quality metrics and generate a leak rate predictive model based at least in part on providing some or all of any remaining gas column density maps that do meet the one or more predefined quality metrics, associated distances from the gas leak imaging device to the corresponding one of the one or more gas leak tests, and associated leak rates of the corresponding one of the one or more gas leak tests to an artificial intelligence algorithm; and

if the plurality of gas column density maps does meet the predefined quality threshold, generate a leak rate predictive model based at least in part on providing some or all of the gas column density maps, associated distances from the gas leak imaging device to the corresponding one of the one or more gas leak tests, and associated leak rates of the corresponding one of the one or more gas leak tests to the artificial intelligence algorithm.

2. The apparatus of claim 1 , wherein each gas column density map comprises one or more discrete plumes;

wherein at least one of the predefined quality metrics comprises a number of discrete plumes in a gas column density map being less than a predefined plume threshold; and

wherein the at least one non-transitory memory and the program code are configured to, with the at least one processor, cause the apparatus to:

determine a number of discrete plumes for each gas column density map; and

remove each gas column density map in which the number of discrete plumes is greater than the predefined plume threshold.

3. The apparatus of claim 1 , wherein, if the plurality of gas column density maps meets the predefined quality threshold, the at least one non-transitory memory and the program code are configured to, with the at least one processor, cause the apparatus to:

obtain a plurality of gas column density maps from a non-test gas leak, each gas column density map having an associated distance from a gas leak imaging device to the non-test gas leak; and

provide the plurality of gas column density maps from the non-test gas leak and the corresponding distances from the gas leak imaging device to the non-test gas leak to the leak rate predictive model to generate a predicted leak rate for each of the plurality of gas column density maps from the non-test gas leak.

4. The apparatus of claim 3 , wherein the at least one non-transitory memory and the program code are configured to, with the at least one processor, cause the apparatus to apply a smoothing algorithm to some or all of the generated predicted leak rates for the plurality of gas column density maps from the non-test gas leak.

5. The apparatus of claim 1 , wherein the plurality of gas column density maps from one or more gas leak tests each have an associated leak location expressed as one or more pixels of the corresponding gas column density map; and

wherein the at least one non-transitory memory and the program code are configured to, with the at least one processor, cause the apparatus to generate a leak location predictive model based at least in part on providing any remaining gas column density maps and associated leak locations to an artificial intelligence algorithm.

6. A computer-implemented method comprising:

obtaining a plurality of gas column density maps associated with one or more gas leak tests, each gas column density map having an associated distance from a gas leak imaging device to a corresponding one of the one or more gas leak tests and an associated leak rate of a corresponding one of the one or more gas leak tests;

determining if the plurality of gas column density maps meet a predefined quality threshold;

if the plurality of gas column density maps does not meet the predefined quality threshold, filtering the plurality of gas column density maps to remove any gas column density maps that do not meet one or more predefined quality metrics and generating a leak rate predictive model based at least in part on providing some or all of any remaining gas column density maps that do meet the one or more predefined quality metrics, associated distances from the gas leak imaging device to the corresponding one of the one or more gas leak tests, and associated leak rates of the corresponding one of the one or more gas leak tests to an artificial intelligence algorithm; and

if the plurality of gas column density maps does meet the predefined quality threshold, generating a leak rate predictive model based at least in part on providing some or all of the gas column density maps, associated distances from the gas leak imaging device to the corresponding one of the one or more gas leak tests, and associated leak rates of the corresponding one of the one or more gas leak tests to the artificial intelligence algorithm.

7. The computer-implemented method of claim 6 , wherein each gas column density map comprises one or more discrete plumes:

wherein at least one of the predefined quality metrics comprises a number of discrete plumes in a gas column density map being less than a predefined plume threshold; and

wherein the computer-implemented method further comprises:

determining a number of discrete plumes for each gas column density map; and

removing each gas column density map in which the number of discrete plumes is greater than the predefined plume threshold.

8. The computer-implemented method of claim 6 , wherein, if the plurality of gas column density maps meets the predefined quality threshold, the computer-implemented method further comprises:

obtaining a plurality of gas column density maps from a non-test gas leak, each gas column density map having an associated distance from a gas leak imaging device to the non-test gas leak; and

providing the plurality of gas column density maps from the non-test gas leak and the corresponding distances from the gas leak imaging device to the non-test gas leak to the leak rate predictive model to generate a predicted leak rate for each of the plurality of gas column density maps from the non-test gas leak.

9. The computer-implemented method of claim 8 , wherein the computer-implemented method further comprises applying a smoothing algorithm to some or all of the generated predicted leak rates for the plurality of gas column density maps from the non-test gas leak.

10. The computer-implemented method of claim 6 , wherein the plurality of gas column density maps from one or more gas leak tests each have an associated leak location expressed as one or more pixels of the corresponding gas column density map; and

wherein the computer-implemented method further comprises generating a leak location predictive model based at least in part on providing any remaining gas column density maps and associated leak locations to an artificial intelligence algorithm.

11. A computer program product comprising at least one non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions comprising an executable portion configured to:

obtain a plurality of gas column density maps associated with one or more gas leak tests, each gas column density map having an associated distance from a gas leak imaging device to a corresponding one of the one or more gas leak tests and an associated leak rate of a corresponding one of the one or more gas leak tests;

determine if the plurality of gas column density maps meet a predefined quality threshold;

if the plurality of gas column density maps does not meet the predefined quality threshold, filter the plurality of gas column density maps to remove any gas column density maps that do not meet one or more predefined quality metrics and generate a leak rate predictive model based at least in part on providing some or all of any remaining gas column density maps that do meet the one or more predefined quality metrics, associated distances from the gas leak imaging device to the corresponding one of the one or more gas leak tests, and associated leak rates of the corresponding one of the one or more gas leak tests to an artificial intelligence algorithm; and

if the plurality of gas column density maps does meet the predefined quality threshold, generate a leak rate predictive model based at least in part on providing some or all of the gas column density maps, associated distances from the gas leak imaging device to the corresponding one of the one or more gas leak tests, and associated leak rates of the corresponding one of the one or more gas leak tests to the artificial intelligence algorithm.

12. The computer program product of claim 11 , wherein each gas column density map comprises one or more discrete plumes;

wherein at least one of the predefined quality metrics comprises a number of discrete plumes in a gas column density map being less than a predefined plume threshold; and

wherein the computer-readable program code portions comprise the executable portion configured to:

determine a number of discrete plumes for each gas column density map; and

remove each gas column density map in which the number of discrete plumes is greater than the predefined plume threshold.

13. The computer program product of claim 11 , wherein, if the plurality of gas column density maps meets the predefined quality threshold, the computer-readable program code portions comprise the executable portion configured to:

obtain a plurality of gas column density maps from a non-test gas leak, each gas column density map having an associated distance from a gas leak imaging device to the non-test gas leak; and

provide the plurality of gas column density maps from the non-test gas leak and the corresponding distances from the gas leak imaging device to the non-test gas leak to the leak rate predictive model to generate a predicted leak rate for each of the plurality of gas column density maps from the non-test gas leak.

14. The computer program product of claim 13 , wherein the computer-readable program code portions comprise the executable portion configured to apply a smoothing algorithm to some or all of the generated predicted leak rates for the plurality of gas column density maps from the non-test gas leak.

15. The computer program product of claim 11 , wherein the plurality of gas column density maps from one or more gas leak tests each have an associated leak location expressed as one or more pixels of the corresponding gas column density map; and

wherein the computer-readable program code portions comprise the executable portion configured to generate a leak location predictive model based at least in part on providing any remaining gas column density maps and associated leak locations to an artificial intelligence algorithm.

Assignments (1)

Continuity (2)

References Cited (10)