Automated inspection-plan based detection

Patent metadata

Field	Value
Publication number	US-11625821-B2
Application number	US-202017070447-A
Country	US
Kind code	B2
Filing date	Oct 14, 2020
Priority date	Oct 14, 2020
Publication date	Apr 11, 2023
Grant date	Apr 11, 2023

Abstract

An inspection-plan based inspection method includes receiving data characterizing an inspection plan associated with inspection of one or more nodes in an inspection site by an inspection device. A first step of the inspection plan includes a first set of operating parameters of the inspection device associated with the inspection of a first node of the one or more nodes and a first set of constraints associated with one or more inspection criteria at the first node by the inspection device. The method also includes generating a first control signal configured to instruct the inspection device to inspect the first node of the one or more nodes. The first control signal is based on one or more of the first set of operating parameters and a user input. The method further includes receiving data characterizing the inspection measurement of the first node by the inspection device.

First claim

What is claimed is:

1. A method comprising:
receiving data characterizing an inspection plan associated with an inspection of one or more nodes in an inspection site by an inspection device, wherein the inspection site comprises an industrial device and a first step of the inspection plan comprises a first set of operating parameters of the inspection device associated with the inspection of a first node of the one or more nodes and a first set of constraints associated with one or more inspection criteria at the first node by the inspection device;
generating a first control signal configured to instruct the inspection device to inspect the first node of the one or more nodes, wherein the first control signal is based on one or more of the first set of operating parameters defining a position of the inspection device relative to the first node of the industrial device during inspection of the first node;
receiving data, collected during the inspection of the first node, the data characterizing an inspection measurement of the first node by the inspection device; and
determining, by a first analytical model, one or more characteristics of the data characterizing the inspection measurement of the first node, wherein the first analytical model comprises an image recognition algorithm that extracts the one or more characteristics from images included in the data;
comparing the one or more characteristics extracted from the images included in the data with existing training images to determine a correlation between the one or more characteristics of the data with the first set of constraints indicative of a defect of the first node of the industrial device visible from the position of the inspection device relative to the first node; and
generating a next operation of the inspection device based on the correlation between the one or more characteristics of the data with the first set of constraints to analyze the defect of the first node of the industrial device.

2. The method of claim 1 , wherein determining one or more characteristics comprises identifying one or more of a site feature in the first node, a surrounding region associated with the first node, a machine part of the inspection site where the first node is located, and image acquisition parameters associated with the inspection measurement of the first node.

3. The method of claim 2 , further comprising:
determining a defect type associated with the site feature;
comparing the defect type with a constraint defect type in the first set of constraints to generate a comparison result; and
determining the next operation of the inspection device based on the comparison result.

4. The method of claim 3 , further comprising:
generating a notification comprising the next operation, wherein the notification comprises one or more of an indication that the inspection of the first node is successfully completed or repeating inspection of the first node, and an inspection summary, the inspection summary comprising one or more of the one or more characteristics, the first set of operating parameters and the first set of constraints.

5. The method of claim 4 , further comprising:
determining, an updated set of operating parameters for the inspection device associated with the inspection of the first node, the determining based on comparing the defect type and the constraint defect type.

6. The method of claim 5 , further comprising generating a second control signal based on the updated set of operating parameters, wherein the next operation of the inspection device is based on the second control signal.

7. The method of claim 4 , further comprising:
presenting the notification to a user;
receiving an input from the user in response to the notification; and
generating a second control signal based on the input, wherein the next operation of the inspection device is based on the second control signal.

8. The method of claim 7 , wherein the next operation of the inspection device comprises execution of a second step of the inspection plan that comprises a second set of operating parameters of the inspection device associated with the inspection of a second node of the one or more nodes and a second set of constraints associated with an inspection measurement at the second node by the inspection device.

9. The method of claim 8 , further comprising:
receiving data characterizing the inspection measurement of the second node by the inspection device; and
determining, by one of the first analytical model and a second analytical model, one or more characteristics of the data characterizing the inspection measurement of the second node, wherein the next operation of the inspection device is based on correlation between the determined one or more characteristics of the data and the second set of constraints.

10. The method of claim 9 , wherein the first analytic model and the second analytic model are included in the inspection plan.

11. The method of claim 8 , wherein the first analytic model is trained during a training performed prior to the inspection, the training comprising:
receiving training data associated with the first analytical model, the training data comprising historical inputs and outputs of the first analytical model; and
training, by a training model, the first analytical model based on one or more of the training data, data characterizing the inspection measurement and the inspection plan.

12. The method of claim 2 , further comprising generating a data file comprising one or more of information associated with the site feature in the first node, the first analytical model, data characterizing the inspection of the first node and the first set of operating parameters.

13. The method of claim 1 , further comprising:
presenting the first step of the inspection plan to a user;
receiving user input from a user, wherein the first control signal is generated based on the user input.

14. The method of claim 1 , wherein the first control signal is generated based on the first set of operating parameters in the first step of the inspection plan.

15. The method of claim 1 , wherein the determined one or more characteristics comprises a surrounding region associated with the first node.

16. The method of claim 1 , wherein the first analytical model can be based on the first set of constraints.

17. A system comprising:
at least one data processor;
a memory coupled to the at least one data processor, the memory storing instructions to cause the at least one data processor to perform operations comprising:
receiving data characterizing an inspection plan associated with an inspection of one or more nodes in an inspection site by an inspection device, wherein the inspection site comprises an industrial device and a first step of the inspection plan comprises a first set of operating parameters of the inspection device associated with the inspection of a first node of the one or more nodes and a first set of constraints associated with one or more inspection criteria at the first node by the inspection device;
generating a first control signal configured to instruct the inspection device to inspect the first node of the one or more nodes, wherein the first control signal is based on one or more of the first set of operating parameters defining a position of the inspection device relative to the first node of the industrial device during inspection of the first node;
receiving data, collected during the inspection of the first node, the data characterizing an inspection measurement of the first node by the inspe

Assignees

• Baker Hughes Oilfield Operations Llc

Inventors

• Coombs Kevin Andrew

Classifications

• G06N20/00

  Machine learning · CPC title

• G06T2207/10048

  Infrared image · CPC title

• G06Q10/06316

  Sequencing of tasks or work · CPC title

• G06T2207/20081

  Training; Learning · CPC title

• G06T7/0004 Primary

  Industrial image inspection · CPC title