SYSTEM AND METHOD FOR ADVANCED CONDITION MONITORING OF AN ASSET SYSTEM

• US 20090295561A1
• Filed: 05/29/2008
• Published: 12/03/2009
• Est. Priority Date: 05/29/2008
• Status: Active Grant

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Mitchell, Robert James JR., Bonissone, Piero Patrone, Hershey, John Erik, Subbu, Rajesh Venkat, Hu, Xiao, Taware, Avinash Vinayak
• Granted Patent: US 8,144,005 B2
• Time in Patent Office: Days
• US Class Current: 340/501
• CPC Class Codes: G05B 19/41875 characterised by quality surveillance of production
• CPC Class Codes: G05B 2219/31455 Monitor process status
• CPC Class Codes: G05B 2219/32201 Build statistical model of past normal proces, compare with actual process
• CPC Class Codes: G05B 2219/33027 Artificial neural network controller
• CPC Class Codes: Y02P 90/02 Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Abstract

A method for advanced condition monitoring of an asset system includes sensing actual values of an operating condition for an operating regime of the asset system using at least one sensor; estimating sensed values of the operating condition by using an auto-associative neural network; determining a residual vector between the estimated sensed values and the actual values; and performing a fault diagnostic on the residual vector. In another method, an operating space of the asset system is segmented into operating regimes; the auto-associative neural network determines estimates of actual measured values; a residual vector is determined from the auto-associative neural network; a fault diagnostic is performed on the residual vector; and a change of the operation of the asset system is determined by analysis of the residual vector. An alert is provided if necessary. A smart sensor system includes an on-board processing unit for performing the method of the invention.

Claims

1-19. -19. (canceled)

20. A method for advanced condition monitoring of an asset system, the method comprising the steps of:

monitoring at least one variable of an asset system using the at least one sensor of a smart sensor system;

determining whether the asset system has departed from normal operation;

identifying the at least one variable of the asset system indicating the departure from normal operation; and

providing an alert if the asset system departs from normal operation.

21. The method of claim 20, wherein a Hotelling'"'"'s T² statistic technique is used to determine whether the asset system has departed from normal operation.

22. The method of claim 21, further comprising the step of applying a post-processing technique to the Hotelling'"'"'s T² statistic technique.

23. The method of claim 22, wherein the post-processing technique is selected from the group consisting of a curve fitting technique, a wavelet technique, a norm minimization technique, a model fitting technique, a median filtering technique, a calculation of excess kurtosis testing technique, a higher order moment statistical technique, an entropy measurement, a minority-decision measurement, and a combination thereof.

24. The method of claim 20, wherein the identifying step comprises the steps of:

expressing a Hotelling'"'"'s T² statistic technique in terms of its principal components;

normalizing the scores of the principal components;

calculating the contributions of variables appearing in the normalized scores;

normalizing the contributions of the variables;

ordering a ranking of the normalized contributions of the variables from largest magnitude to smallest magnitude; and

interpreting the ranking order as an order of the significance of the variables.

25. The method of claim 20, wherein the alerting step comprises distributing a message via an information transport medium selected from the group consisting of electrical cabling, power line conduction, an intranet, the Internet, and wireless transmission.

30. The method of claim 20, wherein the at least one sensor comprises a smart sensor.

31. The method of claim 30, wherein the smart sensor is connected to a smart sensor system.

32. The method of claim 31, wherein the smart sensor system comprises an on-board processing unit for advanced condition monitoring of the asset system.

26. A method for advanced condition monitoring of an asset system, the method comprising the steps of:

monitoring a variable of the system asset using a smart sensor;

analyzing the time sequential values of the monitored variable by using the group consisting of a Rank Permutation Transformation test, a Hotelling'"'"'s T² statistic test, and a Likelihood Ratio Test;

determining a change of an operating condition of the asset system using the analyzed values; and

alerting of the change of the operating condition of the asset system.

27. The method of claim 26, further comprising the step of applying a post-processing technique to the Hotelling'"'"'s T² statistic test.

28. The method of claim 27, wherein the post-processing technique is selected from the group consisting of a curve fitting technique, a wavelet technique, a norm minimization technique, a model fitting technique, a median filtering technique, a calculation of excess kurtosis testing technique, a higher order moment statistical technique, an entropy measurement, a minority-decision measurement, and a combination thereof.

29. The method of claim 26, wherein the alerting step comprises distributing a message via an information transport medium selected from the group consisting of an electrical cable, power line conduction, an intranet, an Internet, and a wireless transmission.