Method and system for analyzing gas leak based on machine learning

Patent metadata

Field	Value
Publication number	US-10031040-B1
Application number	US-201715472018-A
Country	US
Kind code	B1
Filing date	Mar 28, 2017
Priority date	Mar 28, 2017
Publication date	Jul 24, 2018
Grant date	Jul 24, 2018

Abstract

Embodiments of the present invention provide a system for estimating a location of a gas leak, based on machine learning from forward gas concentration data provided by an analog or scale model including a gas source. The system improves significantly over previous systems by providing high quality, physically accurate forward modeling data inexpensively. During operation, the system configures an aerosol source at a first location to emit a gaseous aerosol. The system then configures a laser source to illuminate the aerosol with a laser sheet. The system may then obtain an image of a reflection of the laser sheet from the aerosol. The system may then analyze the image to quantify a three-dimensional concentration distribution of the aerosol. The system may then estimate, based on solving an inverse problem and an observed second gas concentration, a second location of a second gas source.

First claim

What is claimed is:

1. A computer-implemented method for measuring a gas concentration distribution, comprising:
configuring, by a computer system, an aerosol source at a first location to emit a gaseous aerosol;
configuring, by the computer system, a laser source to illuminate the gaseous aerosol with a laser sheet;
obtaining, from a camera, an image of a reflection of the laser sheet from a plurality of particles of the gaseous aerosol;
analyzing the image to quantify a three-dimensional concentration distribution of the gaseous aerosol associated with the first location; and
solving, directly based on the three-dimensional concentration distribution and the first location, an inverse problem; and
estimating, based on the solved inverse problem and an observed second gas concentration, a second location of a second gas source.

2. The method of claim 1 , wherein the first location of the aerosol source is within a wind tunnel or a turbulent wind tunnel.

3. The method of claim 2 , further comprising selecting a relative wind direction by configuring a turn table to rotate the aerosol source within the wind tunnel or turbulent wind tunnel.

4. The method of claim 1 , wherein configuring the aerosol source at the first location further comprises selecting the aerosol source from a set of multiple sources located at predetermined locations.

5. The method of claim 1 , further comprising:
configuring the laser source to illuminate the gaseous aerosol with a plurality of laser sheets at different heights and/or orientations;
obtaining, from the camera, a plurality of images of reflections of the plurality of laser sheets from the gaseous aerosol; and
analyzing the plurality of images to quantify the three-dimensional concentration distribution of the gaseous aerosol.

6. The method of claim 1 , wherein analyzing the image further comprises one or more of:
correcting for a placement of the camera by a perspective transformation;
calibrating a spatial scale of the image using a set of fiduciary points at known fiduciary locations;
removing artifacts from the image by differencing an artifact image, wherein the artifact image is taken with the aerosol source configured not to emit the gaseous aerosol;
suppressing an aerosol signature originating out of a plane of the laser sheet by correlating a set of red, green, and blue (RGB) values for a respective pixel within the image; and
isolating a plume of the gaseous aerosol within the image by analyzing connectedness of shapes in the image.

7. The method of claim 1 :
wherein the first location of the aerosol source is in a scale model of a well-pad; and
wherein the second gas source is a natural gas leak.

8. A non-transitory computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method for measuring a gas concentration distribution, the method comprising:
configuring an aerosol source at a first location to emit a gaseous aerosol;
configuring a laser source to illuminate the gaseous aerosol with a laser sheet;
obtaining, from a camera, an image of a reflection of the laser sheet from a plurality of particles of the gaseous aerosol;
analyzing the image to quantify a three-dimensional concentration distribution of the gaseous aerosol associated with the first location; and
solving, directly based on the three-dimensional concentration distribution and the first location, an inverse problem; and
estimating, based on the solved inverse problem and an observed second gas concentration, a second location of a second gas source.

9. The non-transitory computer-readable storage medium of claim 8 , wherein the first location of the aerosol source is within a wind tunnel or a turbulent wind tunnel.

10. The non-transitory computer-readable storage medium of claim 9 , wherein the method further comprises selecting a relative wind direction by configuring a turn table to rotate the aerosol source within the wind tunnel or turbulent wind tunnel.

11. The non-transitory computer-readable storage medium of claim 10 , wherein configuring the aerosol source at the first location further comprises selecting the aerosol source from a set of multiple sources located at predetermined locations.

12. The non-transitory computer-readable storage medium of claim 8 , wherein analyzing the image further comprises one or more of:
correcting for a placement of the camera by a perspective transformation;
calibrating a spatial scale of the image using a set of fiduciary points at known fiduciary locations;
removing artifacts from the image by differencing an artifact image, wherein the artifact image is taken with the aerosol source configured not to emit the gaseous aerosol;
suppressing an aerosol signature originating out of a plane of the laser sheet by correlating a set of red, green, and blue (RGB) values for a respective pixel within the image; and
isolating a plume of the gaseous aerosol within the image by analyzing connectedness of shapes in the image.

13. The non-transitory computer-readable storage medium of claim 8 :
wherein the first location of the aerosol source is in a scale model of a well-pad; and
wherein the second gas source is a natural gas leak.

14. A computing system for measuring a gas concentration distribution, the system comprising:
a set of processors;
an aerosol source at a first location;
a laser source;
a camera; and
a non-transitory computer-readable medium coupled to the set of processors storing instructions thereon that, when executed by the processors, cause the processors to perform a method for measuring a gas concentration distribution, the method comprising:
configuring the aerosol source to emit a gaseous aerosol;
configuring the laser source to illuminate the gaseous aerosol with a laser sheet;
obtaining, from the camera, an image of a reflection of the laser sheet from a plurality of particles of the gaseous aerosol;
analyzing the image to quantify a three-dimensional concentration distribution of the gaseous aerosol associated with the first location; and
solving, directly based on the three-dimensional concentration distribution and the first location, an inverse problem; and
estimating, based on the solved inverse problem and an observed second gas concentration, a second location of a second gas source.

15. The computing system of claim 14 , wherein the first location of the aerosol source is within a wind tunnel or a turbulent wind tunnel.

16. The computing system of claim 14 , wherein the method further comprises selecting a relative wind direction by configuring a turn table to rotate the aerosol source within the wind tunnel or turbulent wind tunnel.

17. The computing system of claim 14 :
further comprising a set of multiple aerosol sources located at predetermined locations; and
wherein configuring the aerosol source at the first location further comprises selecting the aerosol source from the set of multiple sources.

18. The computing system of claim 14 , wherein the method further comprises:
configuring the laser source to illuminate the gaseous aerosol with a plurality of laser sheets at different heights and/or orientations;
obtaining, from the camera, a plurality of images of reflections of the plurality of laser sheets from the gaseous aerosol; and
analyzing the plurality of images to quantify the three-dimensional concentration distribution of the gaseous aerosol.

19. The computing system of claim 14 , wherein analyzing the image further compri

Assignees

• Palo Alto Res Ct Inc

Inventors

• Smith Clinton J
• Saha Bhaskar
• Beck Victor A
• Schwartz David E

Classifications

• G01M3/007 Primary

  Leak detector calibration, standard leaks (G01M3/207 takes precedence) · CPC title

• G06T2207/20081

  Training; Learning · CPC title

• G06T7/0004

  Industrial image inspection · CPC title

• G06T2207/10024

  Color image · CPC title

• G01M3/38

  by using light (G01M3/02 takes precedence) · CPC title