System and method for detecting an abnormal situation associated with a reactor

• US 7,181,654 B2
• Filed: 09/17/2004
• Issued: 02/20/2007
• Est. Priority Date: 09/17/2004
• Status: Active Grant

• Current Assignee: Fisher-Rosemount Systems Incorporated (Emerson Electric Company)
• Original Assignee: Fisher-Rosemount Systems Incorporated (Emerson Electric Company)
• Inventors: Eryurek, Evren, Ford, Ferrill E. Jr.
• Primary Examiner(s): BARNES-BULLOCK, CRYSTAL JOY
• Application Number:
  US10/944,610
• Publication Number:
  US 20060064182A1
• Time in Patent Office: 886 Days
• Field of Search: 702 33- 36, 702/44, 702/47, 702/50, 702/82, 702/138, 702179-181, 702/182, 702/185, 700/28, 700 32- 34, 700 49- 51, 700/79, 700/80, 700/103, 700/104, 700/108, 700/301, 706 45- 48, 706 52- 55, 706 59- 61, 706/906, 706/907, 706/914, 706/915, 714/1, 714/25, 714/26, 714/37, 714/47, 714/48, 714/57, 340/500, 340/501, 340/506, 340/511, 340/517, 340/521, 340/523, 366/342, 366/343, 366/348
• US Class Current: 714/47.2
• CPC Class Codes: G05B 23/024 Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks
• CPC Class Codes: G05B 23/0281 Quantitative, e.g. mathematical distance; Clustering; Neural networks; Statistical analysis

Abstract

A system for detecting abnormal situations associated with a reactor in a process plant receives statistical data associated with pressure within a reactor vessel. For example, a pressure sensor device disposed at least partially within the reactor vessel may generate the statistical data based on a pressure signal. The statistical data is analyzed to detect whether one or more abnormal situations associated with an agitator of the reactor exist. For example, the statistical data may be analyzed to detect whether the agitator is no longer turning, is turning at a rate that is different than expected, is out of balance, is broken, etc. If an abnormal situation is detected, an indicator of the abnormal situation may be generated.

Claims

1. A method for detecting an abnormal situation associated with a reactor in a process plant, the method comprising:

calculating first statistical data from a pressure signal indicative of pressure within a vessel of the reactor, wherein calculating the first statistical data comprises calculating at least one of a mean of the pressure signal, a standard deviation of the pressure signal, a variance of the pressure signal, a root-mean square of the pressure signal, a rate of change of the pressure signal, a range of the pressure signal, a filtered pressure signal, and a correlation associated with the pressure signal;

analyzing the first statistical data to detect whether one or more abnormal situations associated with an agitator of the reactor exist; and

generating an indicator of an abnormal situation if one or more of the one or more abnormal situations are detected.

2. A method as defined in claim 1, wherein the pressure signal is generated by a pressure sensor disposed at least partially within the vessel of the reactor.

3. A method as defined in claim 1, wherein calculating the first statistical data further comprises calculating at least one of an indication that a spike in the pressure signal occurred, an indication that the pressure signal includes a bias, an indication that a standard deviation of the pressure signal is greater than a first threshold, an indication that the standard deviation of the pressure signal is less than a second threshold, an indication that the pressure signal includes cyclic oscillations, and an indication that the pressure signal is erratic.

4. A method as defined in claim 1, wherein calculating the first statistical data comprises calculating at least one of an indication that a spike in the pressure signal occurred, an indication that the pressure signal includes a bias, an indication that a standard deviation of the pressure signal is greater than a first threshold, an indication that the standard deviation of the pressure signal is less than a second threshold, an indication that the pressure signal includes cyclic oscillations, and an indication that the pressure signal is erratic.

5. A method as defined in claim 1, wherein the first statistical data is based on a pressure signal generated by a pressure sensor disposed at least partially within a vessel of the reactor;

wherein analyzing the first statistical data comprises detecting spikes in the pressure signal.

6. A method as defined in claim 1, wherein analyzing the first statistical data comprises determining if a minimum number of spikes in the pressure signal occur during a time period.

7. A method as defined in claim 6, wherein determining if the minimum number of spikes in the pressure signal occur comprises determining if at least one spike in the pressure signal occurs during a particular time period.

8. A method as defined in claim 6, wherein determining if the minimum number of spikes in the pressure signal occur comprises determining if at least N spikes in the pressure signal occur during the particular time period, wherein N is greater than one.

9. A method as defined in claim 1, wherein analyzing the first statistical data comprises:

determining time periods between spikes in the pressure signal; and

comparing the determined time periods with expected time periods.

10. A method as defined in claim 1, wherein analyzing the first statistical data comprises:

determining a rate at which spikes occur in the pressure signal; and

comparing the determined rate with expected an expected rate.

11. A method as defined in claim 1, wherein analyzing the first statistical data comprises determining amplitudes of spikes in the pressure signal.

12. A method as defined in claim 11, wherein analyzing the first statistical data further comprises determining if each amplitude is within a range.

13. A method as defined in claim 11, wherein analyzing the first statistical data further comprises determining if differences between amplitudes are within a range.

14. A method as defined in claim 1, wherein the pressure signal is generated by a pressure sensor disposed at least partially within the vessel of the reactor, the method further comprising analyzing the pressure signal to detect whether the one or more abnormal situations exist.

15. A method as defined in claim 1, wherein the pressure signal is generated by a pressure sensor disposed at least partially within a vessel of the reactor;

the method further comprising analyzing second statistical data based on one or more process signals separate from the pressure signal to detect whether the one or more abnormal situations exist.

16. A method as defined in claim 1, wherein the pressure signal is generated by a first pressure sensor disposed at least partially within a vessel of the reactor;

the method further comprising analyzing at least second statistical data to detect whether the one or more abnormal situations exist, wherein the at least second statistical data is based on at least a second pressure signal generated by a at least a second pressure sensor disposed at least partially within the vessel of the reactor.

17. A system for detecting an abnormal situation associated with a reactor in a process plant, the system comprising:

a statistical parameter generator to generate one or more first statistical parameters based on a first pressure signal generated by a first pressure sensor disposed at least partially within a vessel of the reactor, wherein the one or more first statistical parameters comprise at least one of a mean value of the first pressure signal, a standard deviation of the first pressure signal, a variance of the first pressure signal, root-mean square value of the first pressure signal, a rate of change of the first pressure signal, a range of the first pressure signal, a filtered first pressure signal, and a correlation associated with the first pressure signal; and

an abnormal situation detector to detect at least one abnormal situation associated with an agitator of the reactor based on the one or more first statistical parameters, and to generate one or more indicators of one or more abnormal situations that are detected.

18. A system as defined in claim 17, wherein the abnormal situation detector is configured to determine if a minimum number of spikes occur in the first pressure signal during a time period.

19. A system as defined in claim 18, wherein the minimum number of spikes is one.

20. A system as defined in claim 18, wherein the minimum number of spikes is an integer greater than one.

21. A system as defined in claim 17, wherein the abnormal situation detector is configured to determine time periods between spikes in the first pressure signal and to compare the determined time periods to expected time periods.

22. A system as defined in claim 17, wherein the abnormal situation detector is configured to determine a rate at which spikes in the first pressure signal occur and to compare the determined rate to an expected rate.

23. A system as defined in claim 17, wherein the abnormal situation detector is configured to determine amplitudes of spikes in the first pressure signal.

24. A system as defined in claim 23, wherein the abnormal situation detector is configured to determine if each amplitude is within a range.

25. A system as defined in claim 23, wherein the abnormal situation detector is configured to determine if differences between the amplitudes are within a range.

26. A system as defined in claim 17, wherein the abnormal situation detector is configured to detect the one or more abnormal situations further based on the first pressure signal.

27. A system as defined in claim 17, wherein the abnormal situation detector is configured to detect the one or more abnormal situations further based on one or more second statistical parameters based on at least a second pressure signal generated by at least a second pressure sensor disposed at least partially within the vessel of the reactor.

28. A system as defined in claim 17, further comprising an event detector to generate one or more indicators of one or more statistical events based on the one or more first statistical parameters;

wherein the abnormal situation detector is configured to detect the one or more abnormal situations based on the one or more indicators of the one or more statistical events.

29. A system as defined in claim 28, wherein the one or more indicators of the one or more statistical events comprises at least one of an indicator that a spike in the first pressure signal occurred, an indicator that the first pressure signal includes a bias, an indicator that a standard deviation of the first pressure signal is greater than a first threshold, an indicator that the standard deviation of the first pressure signal is less than a second threshold, an indicator that the first pressure signal includes cyclic oscillations, and an indicator that the first pressure signal is erratic.

30. A tangible medium storing machine readable instructions, the machine readable instructions capable of causing one or more machines to:

receive statistical data associated with pressure within a vessel of the reactor, the statistical data calculated from a pressure signal generated by a pressure sensor;

analyze the statistical data to detect whether one or more abnormal situations associated with an agitator of the reactor exist; and

generate an indicator of an abnormal situation if one or more of the one or more abnormal situations are detected.

31. A tangible medium storing machine readable instructions as defined in claim 30, wherein the machine readable instructions are further capable of causing one or more machines to calculate the statistical data based on a pressure signal generated by a pressure sensor disposed at least partially within a vessel of the agitator.

32. A tangible medium storing machine readable instructions as defined in claim 31, wherein the machine readable instructions are further capable of causing one or more machines to calculate at least one statistical parameter based on the pressure signal, wherein the at least one statistical parameter comprises at least one of a mean of the pressure signal, a standard deviation of the pressure signal, a variance of the pressure signal, root-mean square of the pressure signal, a rate of change of the pressure signal, and a range of the pressure signal.

33. A tangible medium storing machine readable instructions as defined in claim 32, wherein the machine readable instructions are further capable of causing one or more machines to calculate at least one indication of a statistical event associated with the pressure signal based on the at least one statistical parameter, wherein the at least one indication of a statistical event comprises at least one of an indication that a spike in the pressure signal occurred, an indication that the pressure signal includes a bias, an indication that a standard deviation of the pressure signal is greater than a first threshold, an indication that the standard deviation of the pressure signal is less than a second threshold, an indication that the pressure signal includes cyclic oscillations, and an indication that the pressure signal is erratic.

34. A tangible medium storing machine readable instructions as defined in claim 31, wherein the machine readable instructions are further capable of causing one or more machines to calculate at least one of an indication that a spike in the pressure signal occurred, an indication that the pressure signal includes a bias, an indication that a standard deviation of the pressure signal is greater than a first threshold, an indication that the standard deviation of the pressure signal is less than a second threshold, an indication that the pressure signal includes cyclic oscillations, and an indication that the pressure signal is erratic.