Apparatus and method for screening data for kernel regression model building

Patent metadata

Field	Value
Publication number	US-10733533-B2
Application number	US-201715451601-A
Country	US
Kind code	B2
Filing date	Mar 7, 2017
Priority date	Mar 7, 2017
Publication date	Aug 4, 2020
Grant date	Aug 4, 2020

Abstract

Raw data is received from an industrial machine. The industrial machine includes one or more sensors that obtain the data, and the sensors transmit the raw data to a central processing center. The raw data is received at the central processing center and an unsupervised kernel-based algorithm is recursively applied to the raw data. The application of the unsupervised kernel-based algorithm is effective to learn characteristics of the raw data and to determine from the raw data a class of acceptable data. The class of acceptable data is data having a degree of confidence above a predetermined level that the data was obtained during a healthy operation of the machine. The acceptable data is successively determined and refined upon each application of the unsupervised kernel-based algorithm. The unsupervised kernel-based algorithm is executed until a condition is met.

First claim

What is claimed is:

1. A method, comprising:
receiving raw data from an industrial machine, the industrial machine including one or more sensors that obtain the data, the sensors transmitting the raw data to a central processing center;
receiving the raw data at the central processing center and recursively applying an unsupervised kernel-based algorithm to the raw data, the application of the unsupervised kernel-based algorithm being effective to learn characteristics of the raw data and to determine from the raw data a class of acceptable data, the class of acceptable data being data having a degree of confidence above a predetermined level that the data was obtained during a healthy operation of the machine, the acceptable data being successively determined and refined upon each application of the unsupervised kernel-based algorithm, the unsupervised kernel-based algorithm being executed until a condition is met, wherein the condition relates to a predetermined number of data points.

2. The method of claim 1 , wherein the condition relates to a number of iterations, and the number of iterations is adjustable between a first number and a second number.

3. The method of claim 1 , wherein the algorithm is a one-class SVM algorithm.

4. The method of claim 1 , wherein the condition is an integer number of application times.

5. The method of claim 1 , further comprising preprocessing the raw data before applying the unsupervised kernel-based algorithm.

6. The method of claim 1 , further comprising accepting user limits concerning the raw data.

7. A method, comprising:
receiving raw data from an industrial machine, the industrial machine including one or more sensors that obtain the data, the sensors transmitting the raw data to a central processing center;
receiving the raw data at the central processing center and recursively applying an unsupervised kernel-based algorithm to the raw data, the application of the unsupervised kernel-based algorithm being effective to learn characteristics of the raw data and to determine from the raw data a class of acceptable data, the class of acceptable data being data having a degree of confidence above a predetermined level that the data was obtained during a healthy operation of the machine, the acceptable data being successively determined and refined upon each application of the unsupervised kernel-based algorithm, the unsupervised kernel-based algorithm being executed until a condition is met, and further comprising receiving user information concerning data viability that identifies at least some acceptable data.

8. The method of claim 7 , further comprising preprocessing the raw data before applying the unsupervised kernel-based algorithm.

9. The method of claim 7 , further comprising accepting user limits concerning the raw data.

10. The method of claim 7 , wherein the algorithm is a one-class SVM algorithm.

11. The method of claim 7 , wherein the condition is an integer number of application times.

12. The method of claim 7 , wherein the condition relates to a number of iterations, and the number of iterations is adjustable between a first number and a second number.

13. An apparatus disposed at a central processing center, the apparatus comprising:
a receiver circuit that is configured to receive raw data from sensors at an industrial machine, the industrial machine including one or more sensors that obtain the data;
a data storage device coupled to the receiver circuit, the data storage device configured to store the raw data;
a control circuit coupled to the data storage device and the receiver circuit, the control circuit configured to receive the raw data and to recursively apply an unsupervised kernel-based algorithm to the raw data, the application of the unsupervised kernel-based algorithm being effective to learn characteristics of the raw data and to determine from the raw data a class of acceptable data, the class of acceptable data being data having a degree of confidence above a predetermined level that the data was obtained during a healthy operation of the machine, the acceptable data being successively determined and refined upon each application of the unsupervised kernel-based algorithm, the unsupervised kernel-based algorithm being executed until a condition is met, wherein the condition relates to a predetermined number of data points.

14. The apparatus of claim 13 , wherein the condition relates to a number of iterations, and the number of iterations is adjustable between a first number and a second number.

15. The apparatus of claim 13 , wherein the algorithm is a one-class SVM algorithm.

16. The apparatus of claim 13 , wherein the condition is an integer number of application times.

17. The apparatus of claim 13 , wherein the control circuit is further configured to preprocess the raw data before applying the unsupervised kernel-based algorithm.

18. The apparatus of claim 13 , wherein the receiver circuit is further configured to accept user limits concerning the raw data.

19. An apparatus disposed at a central processing center, the apparatus comprising:
a receiver circuit that is configured to receive raw data from sensors at an industrial machine, the industrial machine including one or more sensors that obtain the data;
a data storage device coupled to the receiver circuit, the data storage device configured to store the raw data;
a control circuit coupled to the data storage device and the receiver circuit, the control circuit configured to receive the raw data and to recursively apply an unsupervised kernel-based algorithm to the raw data, the application of the unsupervised kernel-based algorithm being effective to learn characteristics of the raw data and to determine from the raw data a class of acceptable data, the class of acceptable data being data having a degree of confidence above a predetermined level that the data was obtained during a healthy operation of the machine, the acceptable data being successively determined and refined upon each application of the unsupervised kernel-based algorithm, the unsupervised kernel-based algorithm being executed until a condition is met, wherein the receiver circuit is further configured to receive user information concerning data viability that identifies at least some acceptable data.

20. The apparatus of claim 19 , wherein the control circuit is further configured to preprocess the raw data before applying the unsupervised kernel-based algorithm.

21. The apparatus of claim 19 , wherein the receiver circuit is further configured to accept user limits concerning the raw data.

22. The apparatus of claim 19 , wherein the algorithm is a one-class SVM algorithm.

23. The apparatus of claim 19 , wherein the condition is an integer number of application times.

24. The apparatus of claim 19 , wherein the condition relates to a number of iterations, and the number of iterations is adjustable between a first number and a second number.

Assignees

• Gen Electric

Inventors

• Abado Shaddy
• Yang Jianbo
• Hu Xiahui
• Saxena Abhinav
• Patel Charmin

Classifications

• G06N20/10 Primary

  using kernel methods, e.g. support vector machines [SVM] · CPC title

• G06N20/00 Primary

  Machine learning · CPC title

• G05B13/048

  using a predictor · CPC title