Adaptive shed strategy for constraint enforcement in a multivariate control system

• US 6,349,236 B1
• Filed: 10/06/1998
• Issued: 02/19/2002
• Est. Priority Date: 10/06/1998
• Status: Expired due to Term

• Current Assignee: Schneider Electric Systems USA, Inc. (Schneider Electric SE)
• Original Assignee: The Foxboro Company (Schneider Electric SE)
• Inventors: Di Vittorio, Nicholas P., Hansen, Peter D.
• Primary Examiner(s): Grant, William
• Assistant Examiner(s): FRANK, ELLIOT L
• Application Number:
  US09/166,880
• Time in Patent Office: 1,232 Days
• Field of Search: 700/28, 700/32, 700/36, 700/47, 700/67, 700/295, 700/296, 705/8, 709/102-108
• US Class Current: 700/67
• CPC Class Codes: G05B 11/32 with inputs from more than one sensing element; with outputs to more than one correcting element
• CPC Class Codes: G05B 13/021 in which a variable is automatically adjusted to optimise the performance

Abstract

A method for changing the difference between the number of active controlled variables and the number of available manipulated variables in a multivariate control system. Controlled variables can be adaptively classified into one of an ordered plurality of classes, with each class having a rank that defines its position within the order. If the number of active valid controlled variables exceeds the number of available manipulated variables, the active controlled variables from the lowest ranking class having active controlled variables are then deactivated according to a predefined priority order by setting their associated setpoints equal to their measured values. Otherwise, inactive controlled variables from the highest ranking class having inactive controlled variables are reactivated according to the same predefined priority order by setting their associated setpoints to the desired setpoint values.

Claims

1. In a multivariate control system having

an ordered plurality of controlled variables, each of said controlled variables having

a measured value

a predefined value defining its position within the order, and

an associated setpoint having an original setpoint value, each of said controlled variables being selected from

active controlled variables and

inactive controlled variables, and

a plurality of manipulated variables, each of said manipulated variables being selected from available manipulated variables and unavailable manipulated variables,

a method for changing the difference between the number of active controlled variables and the number of available manipulated variables, said method comprising:

classifying each of said controlled variables as belonging to one of an ordered plurality of classes, each of said classes having a rank defining its position within the order, and,

changing the number of said active controlled variables, said changing further including at least one of,

relinquishing control over a lowest priority controlled variable, said lowest priority controlled variable being that active controlled variable which is in the class having the lowest rank and which has the lowest priority of all controlled variables in that class, and

asserting control over a highest priority controlled variable, said highest priority controlled variable being that inactive controlled variable which is in the class having the highest rank and which has the highest priority of all controlled variables in that class.

2. The method of claim 1 wherein said step of relinquishing control comprises the step of setting the setpoint of said lowest priority controlled variable equal to the measured value of said lowest priority controlled variable.

3. The method of claim 1 wherein the step of asserting control over a highest priority controlled variable comprises the step of setting the setpoint associated with said highest priority controlled variable equal to said original setpoint value.

4. The method of claim 1 wherein said step of relinquishing control over a lowest priority controlled variable comprises the step of setting the difference between the value of said lowest priority controlled variable and the value of the setpoint associated with said lowest priority controlled variable equal to zero.

5. The method of claim 1 wherein said classifying step comprises the step of determining whether the measured value of said active controlled variable is outside a selected range.

6. The method of claim 1 further comprising the step of determining a desired number of active controlled variables.

7. The method of claim 6 wherein the determining step comprises the steps of

determining whether the measured value of a selected controlled variable is valid or invalid,

setting the value of the selected controlled variable to be equal to its original setpoint value if the measured value is invalid and the controlled variable is an active controlled variable, and

setting the value of the selected controlled variable to the measured value is valid and the controlled variable is an inactive controlled variable.

8. The method of claim 7 wherein the step of setting the value of the selected controlled variable equal to the measured value comprises the steps of

defining an unshedding interval having

a first endpoint at which the value of the selected controlled variable is equal to its original setpoint value, and

a second endpoint at which the value of the selected controlled variable is equal to the measured value,

continuously varying the value of the selected controlled variable measurement variable during the unshedding interval.

9. The method of claim 8 wherein the step of continuously varying the value of the selected controlled variable measurement value during the unshedding interval comprises the step of linearly varying the value of the controlled variable between its measured value and its original setpoint value.

10. The method of claim 6 wherein the step of determining the desired number of active controlled variables includes the step of determining the number of active manipulated variables.

11. The method of claim 10 wherein the step of determining the number of active manipulated variables comprises the step of determining whether a selected manipulated variable is responsive to control.

12. The method of claim 1 wherein said relinquishing step further comprises the step of storing the original setpoint value.

13. A system for changing the difference between a number of active controlled variables and a number of available manipulated variables in a multivariate control system having

an ordered plurality of controlled variables, each of said controlled variables having

a measured value,

a predefined priority value defining its position within the order, and

an associated setpoint having an original setpoint value,

each of said controlled variables being selected from

active controlled variables, and

inactive controlled variables, and

a plurality of manipulated variables, each of said manipulated variables being selected from available manipulated variables and unavailable manipulated variables,

said system comprising

classifier means for classifying each of said controlled variables as belonging to one of an ordered plurality of classes, each of said classes having a rank,

control relinquishing means for relinquishing control over a lowest priority controlled variable, said lowest priority controlled variable being that active controlled variable which is in the class having the lowest rank and which has the lowest priority of all controlled variables in that class, and

control acquisition means for asserting control over a highest priority controlled variable, said highest priority controlled variable being that inactive controlled variable which is in the class having the highest rank and which has the highest priority of all controlled variables in that class.

14. The system of claim 13 wherein said control relinquishing means comprises shedding means for setting the setpoint of said lowest priority controlled variable equal to the measured value of said lowest priority controlled variable.

15. The system of claim 13 wherein said control acquisition means comprises means for setting the setpoint associated with said highest priority controlled variable equal to said original setpoint value.

16. The system of claim 13 wherein said control relinquishing means comprises means for setting the difference between the value of said lowest priority controlled variable and the value of the setpoint associated with said lowest priority controlled variable equal to zero.

17. The system of claim 13 wherein said classifier means comprises means for determining whether the measured value of the active controlled variable is outside a selected range.

18. The system of claim 13 further comprising means for determining a desired number of active controlled variables.

19. The system of claim 18 wherein said means for determining the desired number of active controlled variables comprises

validation means for validating the measured value of a selected controlled variable, thereby determining whether its measured value is valid or invalid,

controlled variable shedding means for setting the value of the selected controlled variable to be equal to its original setpoint value if the measured value is invalid and the controlled variable is an active controlled variable, and

controlled variable unshedding means for setting the value of the selected controlled variable to be equal to its measured value if the test measurement is valid and the controlled variable is an inactive controlled variable.

20. The system of claim 19 wherein said means for setting the value of the selected controlled variable equal to its measured value comprises

means for defining an unshedding interval having

a first endpoint at which the value of the selected controlled variable is equal to its original setpoint value, and

a second endpoint at which the value of the selected controlled variable is equal to its measured value, and

means for continuously varying the value of the selected controlled variable during said unshedding interval.

21. The system of claim 20 wherein the means for continuously varying the value of the selected controlled variable during shedding interval comprises means for linearly varying the measured value of the selected controlled variable between its measured value and its original setpoint value.

22. The system of claim 18 wherein the means for determining the desired number of active controlled variables includes means for determining the number of active manipulated variables.

23. The system of claim 22 wherein the means for determining the number of active manipulated variables comprises means for determining whether a selected manipulated variable is responsive to control.

24. The system of claim 13 further comprising means for storing the original setpoint value.

25. A method for changing the number of active controlled variables in a multivariate control system, the multivariate control system including active and inactive controlled variables and available and unavailable manipulated variables, the method comprising,

providing a predefined value for at least two active and inactive controlled variables,

ordering the at least two active and inactive controlled variables based on the predefined values,

providing at least two classes,

ranking the at least two classes based on priority of control of the at least two classes,

classifying the at least two active and inactive controlled variables into one of the at least two classes while preserving, within the at least two classes, the order based on the predefined values, and,

changing the number of active controlled variables by performing at least one of,

relinquishing control over an active controlled variable having a lowest priority in one of the at least two classes having a lowest priority rank, and

asserting control over an inactive controlled variable having a highest priority in one of the at least two classes having a highest priority rank.

26. A method according to claim 25, wherein relinquishing control includes setting a setpoint of the lowest priority active controlled variable to the measured value of the lowest priority active controlled variable.

27. A method according to claim 25, wherein asserting control includes setting a setpoint of the highest priority inactive controlled variable to an original setpoint value.

28. A method according to claim 25, wherein relinquishing control includes causing a difference between the lowest priority active controlled variable and a setpoint associated with the lowest priority active controlled variable, to be zero.

29. A method according to claim 25, wherein classifying includes determining whether the measured value of the active and inactive controlled variable is outside a selected range.

30. A method according to claim 25, further including determining a desired number of active controlled variables.

31. A method according to claim 25, further including,

determining whether a measured value of a selected active or inactive controlled variable is valid or invalid, and,

if the selected controlled variable is an active controlled variable, setting the value of the selected active controlled variable to an original setpoint value if the measured value is invalid, or

if the selected controlled variable is an inactive controlled variable, setting the value of the selected inactive controlled variable to the measured value if the measured value is valid.

32. A method according to claim 31, wherein setting the value of the selected inactive controlled variable to the measured value comprises the steps of

defining an unshedding interval having

a first endpoint at which the value of the selected controlled variable is equal to its original setpoint value, and

a second endpoint at which the value of the selected controlled variable is equal to its measured value,

varying the selected controlled variable measurement variable during the unshedding interval.

33. A method according to claim 32, wherein varying the selected controlled variable measurement value during the unshedding interval includes linearly varying the controlled variable between the second endpoint and the first endpoint.

34. A method according to claim 30, wherein determining a desired number of active controlled variables includes determining a number of active manipulated variables.

35. A method according to claim 30, wherein determining a desired number of active controlled variables includes determining whether a selected manipulated variable is responsive to control.

36. A method according to claim 25, wherein relinquishing further includes storing the original setpoint value.

37. A method according to claim 25, wherein asserting control further includes retrieving a previously stored original setpoint value.

38. A system for changing the number of active controlled variables in a multivariate control system, the multivariate control system including active and inactive controlled variables and available and unavailable manipulated variables, the system comprising,

a processor with instructions to,

provide a predefined value for at least two active and inactive controlled variables,

order the at least two active and inactive controlled variables based on the predefined values,

provide at least two classes,

rank the at least two classes based on priority of control of the at least two classes,

classify the at least two active and inactive controlled variables into one of the at least two classes while preserving, within the at least two classes, the order based on the predefined values,

relinquish control over an active controlled variable having a lowest priority in one of the at least two classes having a lowest priority rank, and

assert control over an inactive controlled variable having a highest priority in one of the at least two classes having a highest priority rank.

39. A computer product for changing the number of active controlled variables in a multivariate control system, the multivariate control system including active and inactive controlled variables and available and unavailable manipulated variables, the computer product including a processor and instructions for causing the processor to,

provide a predefined value for at least two active and inactive controlled variables,

order the at least two active and inactive controlled variables based on the predefined values,

provide at least two classes,

rank the at least two classes based on priority of control of the at least two classes,

classify the at least two active and inactive controlled variables into one of the at least two classes while preserving, within the at least two classes, the order based on the predefined values, and,

change the number of active controlled variables by performing at least one of,

relinquish control over an active controlled variable having a lowest priority in one of the at least two classes having a lowest priority rank, and

assert control over an inactive controlled variable having a highest priority in one of the at least two classes having a highest priority rank.