System and method for detecting an abnormal situation associated with a process gain of a control loop

• US 20060020423A1
• Filed: 06/09/2005
• Published: 01/26/2006
• Est. Priority Date: 06/12/2004
• Status: Active Grant

• Current Assignee: Fisher-Rosemount Systems Incorporated (Emerson Electric Company)
• Original Assignee: Fisher-Rosemount Systems Incorporated (Emerson Electric Company)
• Inventors: Sharpe, Joseph H. Jr.
• Granted Patent: US 7,660,701 B2
• Time in Patent Office: Days
• US Class Current: 702/183
• CPC Class Codes: C10G 11/187 Controlling or regulating
• CPC Class Codes: G05B 13/024 in which a parameter or coefficient is automatically adjusted to optimise the performance
• CPC Class Codes: G05B 23/0254 based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks

Abstract

In a method for monitoring a control loop in a process plant, process gain data associated with a control loop may be collected. The collected process gain data may be used to determine an expected process gain behavior. For example, expected values of a process variable for given values of a load variable may be determined. As another example, expected changes in a process variable for given changes in a load variable may be determined. Then, during operation of the control loop, the process gain may be monitored. If the monitored process gain substantially deviates from the expected behavior, this may indicate an abnormal situation associated with the control loop.

Claims

1. A method for monitoring operation of a control loop in a process plant, comprising:

collecting process gain data associated with a first operating region of a control loop in a process plant, the control loop associated with a unit operation in the process plant;

determining an expected process gain behavior in the first operating region based on the collected process gain data;

monitoring the process gain during operation of the control loop in the first operating region;

determining when the monitored process gain substantially deviates from the expected process gain behavior in the first operating region; and

determining an abnormal situation associated with at least one of the control loop or the unit operation based at least on a substantial deviation from the expected process gain behavior in the first operating region.

2. A method according to claim 1, wherein collecting process gain data associated with the first operating region of the control loop comprises at least one of:

collecting data regarding a process variable versus a load variable;

collecting data regarding the load variable versus the process variable;

collecting data regarding a process gain versus the load variable;

collecting data regarding the load variable versus a process gain;

or

collecting data regarding the process variable versus one or more other process variables.

3. A method according to claim 2, wherein the load variable comprises at least one of a control output or another process variable.

4. A method according to claim 1, wherein determining the expected process gain behavior in the first operating region comprises at least one of:

determining expected values of a variable;

or

determining expected rates of change of the variable.

5. A method according to claim 1, wherein determining the abnormal situation comprises determining an abnormal situation associated with the unit operation based at least on a substantial deviation from the expected process gain behavior in the first operating region of the control loop and a substantial deviation from an expected process gain behavior of a different control loop associated with the unit operation.

6. A method according to claim 1, further comprising:

collecting process gain data associated with at least a second operating region of the control loop in the process plant;

determining an expected process gain behavior in the at least the second operating region based on the collected process gain data;

monitoring the process gain during operation of the control loop in the at least the second operating region;

determining when the monitored process gain substantially deviates from the expected process gain behavior in the at least the second operating region; and

wherein determining the abnormal situation comprises determining the abnormal situation based at least on a substantial deviation from the expected process gain behavior in the first operating region or a substantial deviation from the expected process gain behavior in the second operating region.

7. A method according to claim 1, further comprising:

determining when the control loop is operating in a second operating region for which process gain data has not yet been collected; and

collecting process gain data associated with the second operating region of the control loop in the process plant after determining that the control loop is operating the second operating region of the control loop;

monitoring the process gain during operation of the control loop in the at the second operating region; and

determining when the monitored process gain substantially deviates from the expected process gain behavior in the second operating region.

8. A method according to claim 7, further comprising;

prompting an operator whether to collect process gain data associated with the second operating region of the control loop in the process plant after determining that the control loop is operating the second operating region of the control loop;

wherein collecting process gain data associated with the second operating region of the control loop comprises process gain data associated with the second operating region if the operator indicates that process gain data associated with the second operating region of the control loop should be collected.

9. A method according to claim 7, wherein a unit of the process plant comprises the control loop;

wherein determining when the control loop is operating in the second operating region comprises determining when the unit of the process plant is operating in an operating region for which process gain data associated with the unit has not yet been collected.

10. A method according to claim 1, wherein collecting process gain data associated with the first operating region of the control loop comprises collecting process gain data regarding a plurality of process gains associated with the control loop, the plurality of process gains including at least a first process gain and a second process gain;

wherein determining the expected process gain behavior in the first operating region comprises determining an expected behavior of the first process gain with respect to at least the second process gain;

wherein monitoring the process gain during operation of the control loop in the first operating region comprises monitoring the first process gain and monitoring the second process gain;

wherein determining when the monitored process gain substantially deviates from the expected process gain behavior in the first operating region comprises determining when the monitored first process gain substantially deviates from the expected behavior of the first process gain with respect to at least the second process gain.

11. A method according to claim 1, wherein determining when the monitored process gain substantially deviates from the expected process gain behavior in the first operating region comprises at least one of:

determining when the monitored process gain is below an expected process gain for a specified period of time;

or

determining when the monitored process gain is above an expected process gain for the specified period of time.

12. A method according to claim 1, wherein determining the expected process gain behavior in the first operating region comprises determining a confidence interval for the first operating region;

wherein determining when the monitored process gain substantially deviates from the expected process gain behavior in the first operating region comprises determining at least when the monitored process gain is outside of the confidence interval in the first operating region.

13. A method according to claim 12, wherein determining when the monitored process gain substantially deviates from the expected process gain behavior in the first operating region comprises determining when the monitored process gain is outside of the confidence interval in the first operating region for a specified period of time.

14. A method according to claim 1, if an abnormal situation is determined, further comprising at least one of:

generating an alert;

adjusting a control parameter associated with the control loop;

initiating a diagnostic procedure;

or

shutting down equipment associated with the control loop.

15. A tangible medium having stored thereon machine executable instructions, the machine executable instructions capable of causing one or more machines to:

collect process gain data associated with a first operating region of a control loop in a process plant;

determine an expected process gain behavior in the first operating region based on the collected process gain data;

monitor the process gain during operation of the control loop in the first operating region;

determine when the monitored process gain substantially deviates from the expected process gain behavior in the first operating region; and

determine an abnormal situation associated with at least one of the control loop or the unit operation based at least on a substantial deviation from the expected process gain behavior in the first operating region.

16. A method for monitoring operation of a control loop in a process plant, comprising:

collecting process gain data associated with a first operating region of a control loop in a process plant;

determining an expected process gain behavior in the first operating region based on the collected process gain data;

providing data indicative of the expected process gain behavior in the first operating region to an expert engine;

providing process gain data associated with the control loop during operation of the control loop to the expert engine; and

utilizing the expert engine to detect an abnormal situation associated with the control loop based on the data indicative of the expected process gain behavior and the process gain data associated with the control loop during operation of the control loop.

17. A method according to claim 16, wherein utilizing the expert engine to detect the abnormal situation associated with the control loop comprises determining whether the process gain in the first operating region substantially deviates from the expected process gain behavior in the first operating region.

18. A method according to claim 16, further comprising:

providing process variable statistical data associated with the control loop during operation of the control loop to the expert engine; and

wherein utilizing the expert engine to detect the abnormal situation associated with the control loop comprises utilizing the expert engine to detect the abnormal situation associated with the control loop further based on the process variable statistical data associated with the control loop.

19. A method according to claim 18, wherein the process variable statistical data associated with the control loop comprises statistical data generated by field devices associated with the control loop.

20. A system for monitoring operation of a control loop in a process plant, the system comprising:

a process gain signature generator configured to generate a signature of expected process gain behavior associated with a control loop in a process plant;

a process gain evaluator configured to determine if an actual process gain substantially deviates from an expected process gain based on the signature of expected process gain behavior; and

an abnormal situation detector configured to detect an abnormal situation associated with a process unit associated with the control loop based at least in part on whether the actual process gain substantially deviates from the expected process gain.

21. A system according to claim 20, further comprising:

an interval generator configured to generate an interval associated with the signature of expected process gain behavior;

wherein the process gain evaluator is configured to determine if the actual process gain substantially deviates from the expected process gain based on the signature of expected process gain behavior and the interval associated with the signature of expected process gain behavior.

22. A system according to claim 20, further comprising an expert system, wherein the expert system comprises at least one of the process gain evaluator or the abnormal situation detector.

23. A system according to claim 22, wherein the expert system is configured to detect an abnormal situation associated with a process unit associated with the control loop based at least in part on whether the actual process gain substantially deviates from the expected process gain.

24. A system according to claim 20, further comprising a process gain data collector configured to collect data to be used by the process gain signature generator to generate the signature of expected process gain behavior associated with the control loop.

25. A method for facilitating monitoring operation of at least a portion of a process plant, comprising:

collecting process gain data indicative of process gains associated with respective unit operations in a process plant;

determining expected process gains associated with respective unit operations based on the collected process gain data; and

providing a common set of criteria for determining for each unit operation whether an abnormal situation associated with the unit operation exists based at least on whether the process gain associated with the unit operation substantially deviates from the expected process gain associated with the unit operation.

26. A method according to claim 25, wherein the common set of criteria comprises expert rules to be applied by an expert engine.

27. A method according to claim 25, further comprising:

permitting a user to modify the common set of criteria for a particular unit operation to generate a modified set of criteria;

utilizing the modified set of criteria to determine for the particular unit operation whether the abnormal situation associated with the particular unit operation exists; and

utilizing the common set of criteria to determine for each of at least one other unit operation whether the abnormal situation associated with the other unit operation exists.