Systems and methods for fault analysis

• US 10,037,026 B2
• Filed: 09/25/2014
• Issued: 07/31/2018
• Est. Priority Date: 09/25/2014
• Status: Active Grant

• Current Assignee: GE Infrastructure Technology, LLC (GE Vernova, Inc.)
• Original Assignee: General Electric Company
• Inventors: Shinkle, William Randolph, Singh, Hardev, Boudreaux, David Michael
• Primary Examiner(s): Breene, John
• Assistant Examiner(s): Islam, Mohammad
• Application Number:
  US14/497,156
• Publication Number:
  US 20160091397A1
• Time in Patent Office: 1,405 Days
• CPC Class Codes: G05B 23/0232 based on qualitative trend analysis, e.g. system evolution
• CPC Class Codes: G06F 17/18 for evaluating statistical data , e.g. average values, frequency distributions, probability functions, regression analysis forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04

Abstract

A system, includes machinery; and a protection monitoring system, comprising a processor configured to: analyze a trend of one or more data measurements of the machinery for one or more patterns indicative of a potential future fault within the machinery in the trend; and provide a fault prediction based upon the analysis of the trend.

Claims

1. A system, comprising:

sensors that capture one or more data measurements of machinery comprising a gas turbine having at least an inlet and a high pressure compressor, the one or more data measurements comprising;

an efficiency of the high pressure compressor, a temperature at the inlet, a pressure at the inlet, a discharge pressure of the high pressure compressor, a power output of the gas turbine, or any combination thereof; and

a protection monitoring system, comprising at least one processor configured to;

accumulate the one or more data measurements and determine a trend of the one or more measurements of the machinery;

analyze the trend of the one or more data measurements of the machinery to identify one or more patterns indicative of a potential future fault within the machinery in the trend; and

generate a graphical user interface that provides a graphical indication of a fault prediction, generate and present an alarm that provides an indication of the fault prediction, generate and present an alert that provides an indication of the fault prediction, control, via a controller, the machinery by performing a tapered deceleration of machinery operating speeds or shut down of the machinery in response to the fault prediction, prior to a fault in the machinery, or any combination thereof based upon the identified one or more patterns;

wherein;

the one or more data measurements comprise at least the temperature at the inlet and the pressure at the inlet, and the analysis comprises;

determining a quotient of a ratio of the temperature at the inlet to the pressure at the inlet, comparing the quotient with a predetermined threshold value, and when the quotient breaches the predetermined threshold value, predict a fault;

or

the one or more data measurements comprise at least the efficiency of the high pressure compressor and the analysis comprises;

identifying a spiked decrease in the efficiency of the high pressure compressor, and upon identifying the spiked decrease, predict a fault.

2. The system of claim 1, wherein the processor is configured to calculate a ratio of trends of at least two data measurements.

3. The system of claim 2, wherein a quotient of the ratio is used in the analysis.

4. The system of claim 1, comprising a controller configured to invoke one or more control actions comprising:

generation of the graphical user interface, generation and presentation of the alarm, generation and presentation of the alert, the control of the machinery, or any combination thereof based upon the fault prediction.

5. The system of claim 1, where in the processor is configured to determine a level of confidence of the fault prediction based upon a frequency of prior occurrences of the one or more patterns being attributed to a subsequent fault.

6. The system of claim 1, wherein the processor is configured to determine a level of urgency related to the fault prediction.

7. The system of claim 1, wherein the system comprises:

a plurality of independent industrial control systems; and

a cloud-computing environment configured to;

receive data measurements and fault data from at least one of the independent industrial control systems;

analyze the data measurements of the at least one independent industrial control systems to associate at least one data pattern of the data measurements with the future potential fault; and

provide the at least one data pattern to the other independent control systems, such that the other independent control systems are aware that the at least one data pattern is associated with the future potential fault.

8. A tangible, non-transitory, machine-readable medium, comprising instructions to:

obtain data measurement trends related to one or more characteristics of a piece of machinery, wherein the machinery comprises a gas turbine having at least an inlet and a high pressure compressor, and wherein the one or more characteristics comprise;

an efficiency of the high pressure compressor, a temperature at the inlet, a pressure at the inlet, a discharge pressure of the high pressure compressor, a power output of the gas turbine, or any combination thereof and the one or more characteristics are captured via one or more data sensors associated with the piece of machinery and are accumulated to determine the data measurement trends;

analyze the data measurement trends to identify one or more patterns indicative of a potential future fault within the machinery; and

generate a graphical user interface that provides a graphical indication of a fault prediction, generate and present an alarm that provides an indication of the fault prediction, generate and present an alert that provides an indication of the fault prediction, control, via a controller, the piece of machinery by performing a tapered deceleration of machinery operating speeds or shut down of the machinery in response to the fault prediction, prior to a fault in the piece of machinery, or any combination thereof based upon the identified one or more patterns;

wherein;

the one or more characteristics comprise at least the temperature at the inlet and the pressure at the inlet, and the analysis comprises;

determining a quotient of a ratio of the temperature at the inlet to the pressure at the inlet, comparing the quotient with a predetermined threshold value, and when the quotient breaches the predetermined threshold value, predict a fault;

or

the one or more characteristics comprise at least the efficiency of the high pressure compressor and the analysis comprises;

identifying a spiked decrease in the efficiency of the high pressure compressor, and upon identifying the spiked decrease, predict a fault.

9. The tangible, non-transitory, machine-readable medium of claim 8, comprising instructions to:

receive historic fault data; and

identify the one or more patterns indicative of a potential future fault based at least in part on the historic fault data.

10. The tangible, non-transitory, machine-readable medium of claim 8, wherein the characteristics of the piece of machinery comprise:

the efficiency of the high pressure compressor, a temperate at a low-pressure inlet, a pressure at a burner inlet, or power output of the gas turbine.

11. The tangible, non-transitory, machine-readable medium of claim 8, wherein the instructions to analyze the data measurement trends to identify one or more patterns comprise instructions to:

identify data spikes in the data measurement trends over a particular threshold of change.

12. The tangible, non-transitory, machine-readable medium of claim 8, wherein the particular threshold of change is larger than 0.5%.

13. A method, comprising:

predicting, via a computer processor, a potential machinery fault by;

obtaining data measurement trends related to one or more characteristics of the machinery, wherein the machinery comprises a gas turbine having at least an inlet and a high pressure compressor, and wherein the one or more characteristics comprise;

an efficiency of the high pressure compressor, a temperature at the inlet, a pressure at the inlet, a discharge pressure of the high pressure compressor, a power output of the gas turbine, or any combination thereof and the one or more characteristics are captured via one or more data sensors associated with the piece of machinery and are accumulated to determine the data measurement trends;

analyzing the data measurement trends to identify one or more patterns indicative of a potential future fault within the machinery; and

generating a graphical user interface that provides a graphical indication of providing a fault prediction, generating and presenting an alarm that provides an indication of the fault prediction, generating and presenting an alert that provides an indication of the fault prediction, controlling, via a controller, the machinery by performing a tapered deceleration of machinery operating speeds or shut down of the machinery in response to the fault prediction, prior to a fault in the machinery, or any combination thereof based upon the identified one or more patterns;

wherein;

the one or more characteristics comprise at least the temperature at the inlet and the pressure at the inlet, and the analysis comprises;

determining a quotient of a ratio of the temperature at the inlet to the pressure at the inlet, comparing the quotient with a predetermined threshold value, and when the quotient breaches the predetermined threshold value, predict a fault;

or

the one or more characteristics comprise at least the efficiency of the high pressure compressor and the analysis comprises;

identifying a spiked decrease in the efficiency of the high pressure compressor, and upon identifying the spiked decrease, predict a fault.

14. The method of claim 13, comprising:

determining, via the computer processor, a confidence level of the fault prediction; and

selecting a particular control action from a set of control actions comprising;

generation of the graphical user interface, generation and presentation of the alarm, generation and presentation of the alert, the control of the machinery, or any combination thereof based upon the confidence level of the fault prediction.

15. The method of claim 13, comprising:

determining, via the computer processor, an urgency level of the fault prediction; and

selecting a particular control action from a set of control actions comprising;

generation of the graphical user interface, generation and presentation of the alarm, generation and presentation of the alert, the control of the machinery, or any combination thereof based upon the urgency level of the fault prediction.

16. The method of claim 13, wherein the one or more characteristics comprise:

the efficiency of the high pressure compressor, a temperate at a low-pressure inlet, a pressure at a burner inlet, a power output of the gas turbine, or any combination thereof.

17. The method of claim 13, comprising:

calculating a ratio of two or more of the one or more characteristics and using the ratio to identify the one or more patterns indicative of the potential future fault.