In-pipeline optical interference-based cognitive system for leak and defect detection

Patent metadata

Field	Value
Publication number	US-10876919-B2
Application number	US-201815966685-A
Country	US
Kind code	B2
Filing date	Apr 30, 2018
Priority date	Apr 30, 2018
Publication date	Dec 29, 2020
Grant date	Dec 29, 2020

Abstract

Methods, systems, computer program products, and devices for in-pipeline optical interference-based cognitive systems for leak and defect detection are provided herein. A computer-implemented method includes obtaining optical interference-related data pertaining to at least one interior portion of a pipeline; obtaining multiple items of sensor data pertaining to the at least one interior portion of the pipeline; generating one or more pipeline-irregularity predictions by applying an inference-based algorithm to the optical interference-related data, the multiple items of sensor data, and one or more additional items of data, wherein each of the one or more pipeline-irregularity predictions comprises an identified location within the interior surface of the pipeline of a predicted irregularity and an estimated size of the predicted irregularity; and outputting the one or more pipeline-irregularity predictions to one or more users.

First claim

What is claimed is:

1. A computer-implemented method, the method comprising:
obtaining optical interference-related data pertaining to at least one interior portion of a pipeline;
obtaining multiple items of sensor data pertaining to the at least one interior portion of the pipeline;
generating one or more pipeline-irregularity predictions by processing, using a neural network, (i) the optical interference-related data, and (ii) the multiple items of sensor data, wherein generating the one or more pipeline-irregularity predictions comprises (a) determining at least one surface discontinuity within the interior surface of the pipeline by analyzing, using the neural network, one or more changes in a fringing pattern associated with the optical interference-related data and the multiple items of sensor data, and (b) determining, using the at least one determined surface discontinuity and additional items of data, an identified location within the interior surface of the pipeline of a predicted irregularity and an estimated size of the predicted irregularity, wherein the additional items of data comprise crowd-sourced information comprising social media commentary and one or more public complaints pertaining to one or more negative issues associated with at least one particular region relevant to the pipeline; and
outputting the one or more pipeline-irregularity predictions to one or more users;
wherein the method is carried out by at least one computing device.

2. The computer-implemented method of claim 1 , wherein the multiple items of sensor data comprise flow-related data.

3. The computer-implemented method of claim 1 , wherein the multiple items of sensor data comprise pressure-related data.

4. The computer-implemented method of claim 1 , wherein the multiple items of sensor data comprise location-related data.

5. The computer-implemented method of claim 1 , wherein the multiple items of sensor data comprise acceleration-related data.

6. The computer-implemented method of claim 1 , wherein the additional items of data comprise one or more items of maintenance history data related to the pipeline.

7. The computer-implemented method of claim 1 , wherein each of the one or more pipeline-irregularity predictions comprises an estimated probability value attributed to the pipeline-irregularity prediction.

8. The computer-implemented method of claim 1 , wherein each of the one or more pipeline-irregularity predictions comprises an estimated severity value attributed to the pipeline-irregularity prediction.

9. The computer-implemented method of claim 8 , comprising:
determining a response time estimation in connection with one or more maintenance activities, wherein said determining the response time estimation is based on the estimated severity value attributed to the pipeline-irregularity prediction.

10. The computer-implemented method of claim 1 , comprising:
outputting the one or more pipeline-irregularity predictions to a feedback mechanism related to training the neural network.

11. A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computing device to cause the computing device to:
obtain optical interference-related data pertaining to at least one interior portion of a pipeline;
obtain multiple items of sensor data pertaining to the at least one interior portion of the pipeline;
generate one or more pipeline-irregularity predictions by processing, using a neural network, (i) the optical interference-related data, and (ii) the multiple items of sensor data, wherein generating the one or more pipeline-irregularity predictions comprises (a) determining at least one surface discontinuity within the interior surface of the pipeline by analyzing, using the neural network, one or more changes in a fringing pattern associated with the optical interference-related data and the multiple items of sensor data, and (b) determining, using the at least one determined surface discontinuity and additional items of data, an identified location within the interior surface of the pipeline of a predicted irregularity and an estimated size of the predicted irregularity, wherein the additional items of data comprise crowd-sourced information comprising social media commentary and one or more public complaints pertaining to one or more negative issues associated with at least one particular region relevant to the pipeline; and
output the one or more pipeline-irregularity predictions to one or more users.

12. A system comprising:
a memory; and
at least one processor operably coupled to the memory and configured for:
obtaining optical interference-related data pertaining to at least one interior portion of a pipeline;
obtaining multiple items of sensor data pertaining to the at least one interior portion of the pipeline;
generating one or more pipeline-irregularity predictions by processing, using a neural network, (i) the optical interference-related data, and (ii) the multiple items of sensor data, wherein generating the one or more pipeline-irregularity predictions comprises (a) determining at least one surface discontinuity within the interior surface of the pipeline by analyzing, using the neural network, one or more changes in a fringing pattern associated with the optical interference-related data and the multiple items of sensor data, and (b) determining, using the at least one determined surface discontinuity and additional items of data, an identified location within the interior surface of the pipeline of a predicted irregularity and an estimated size of the predicted irregularity, wherein the additional items of data comprise crowd-sourced information comprising social media commentary and one or more public complaints pertaining to one or more negative issues associated with at least one particular region relevant to the pipeline; and
outputting the one or more pipeline-irregularity predictions to one or more users.

Assignees

• IBM

Inventors

• Randhawa Sukanya
• Sathaye Ninad
• SRINIVAS Ashwin

Classifications

• G01M3/005 Primary

  using pigs or moles (G01M3/246, G01M3/2823 take precedence) · CPC title

• G01M3/2807

  for pipes (G01M3/2892, G01M3/30 take precedence) · CPC title

• G01M3/38 Primary

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

• E03B7/003

  Arrangement for testing of watertightness of water supply conduits · CPC title

• G01M3/08

  for pipes, cables or tubes; for pipe joints or seals; for valves; {for welds} · CPC title