Method of adapting and applying control parameters in non-linear process controllers

• US 5,748,467 A
• Filed: 02/21/1995
• Issued: 05/05/1998
• Est. Priority Date: 02/21/1995
• Status: Expired due to Term

• Current Assignee: Fisher-Rosemount Systems Incorporated (Emerson Electric Company)
• Original Assignee: Fisher-Rosemount Systems Incorporated (Emerson Electric Company)
• Inventors: Wojsznis, Wilhelm K., Ott, Michael G., Qin, S. Joe
• Primary Examiner(s): Elmore, Reba I.
• Assistant Examiner(s): Garland, Steven R.
• Application Number:
  US08/391,717
• Time in Patent Office: 1,169 Days
• Field of Search: 364/148, 364/150, 364/152-155, 364/157-163, 364/165, 364/166, 364/177, 364/180, 364/181, 395/61, 395/900, 395/903-907, 318/561, 318/609, 318/610
• US Class Current: 700/50
• CPC Class Codes: G05B 13/0275 using fuzzy logic only

Abstract

A non-linear process controller drives a process variable to be substantially equivalent to a set point based on the expected load disturbances within the process and on a measurement of the actual magnitude of set point changes. The controller uses control parameters developed in an optimal manner to control the process in response to expected load disturbances when no set point change has occurred. Whenever a set point change is detected, the controller derives a set of set point change control parameters and uses these set point change control parameters in responding to the set point change. Each of the set point change control parameters may be developed as a function of the magnitude of the actual change in the set point and according to a set of process characteristics which may include the time delay and the dominant time constant of the process.

Claims

1. A non-linear controller for controlling a process in response to load disturbances and a change in a set point, comprising:

means for developing a first set of control parameters for use in response to a load disturbance;

means for determining a change in the set point;

means responsive to the determining means for developing a second set of control parameters for use in response to the set point change, and

means coupled to the determining means and to the developing means for controlling the process in accordance with a non-linear control technique using the first set of control parameters in response to a load disturbance and for controlling the process in accordance with a non-linear control technique using the second set of control parameters in response to the change in the set point.

2. The controller of claim 1, wherein the determining means includes means for measuring the magnitude of the set point change and wherein the developing means includes further means for developing one of the second set of control parameters in accordance with the detected magnitude of the change in the set point.

3. The controller of claim 2, wherein the developing means further includes means for deriving the one of the second set of control parameters as a function of a process characteristic.

4. The controller of claim 3, wherein the process characteristic comprises a time delay of the process and a dominant time constant of the process.

5. The controller of claim 3, wherein the process characteristic comprises characteristics necessary to determine an integral time constant and a proportional gain of the controller.

6. The controller of claim 3, further including a tuner which measures the process characteristic.

7. The controller of claim 3, further including means for allowing the process characteristic to be input by an operator.

8. The controller of claim 1, wherein the controlling means comprises a fuzzy logic controller.

9. The controller of claim 8, wherein the one of the second set of control parameters comprises a change in an error scaling factor (S.sub.Δ

e) and wherein the developing means develops the change in the error scaling factor according to the equation;

space="preserve" listing-type="equation">S.sub.Δ

e =β

Δ

.sub.SP

where;

Δ

_SP =the magnitude of the change in the set point; and

β

=a multiplication factor.

10. The controller of claim 9, wherein the multiplication factor β

is a function of a time delay of the process and the dominant time constant of a process.

11. The controller of claim 9, wherein the multiplication factor β

is a constant.

12. The controller of claim 9, wherein the controlling means includes a fuzzification unit which uses two membership functions to translate a continuous input variable into a fuzzy logic variable.

13. A non-linear controller adapted to control a process in response to a change in a set point, comprising:

means for measuring the magnitude of the change in the set point;

means for determining a control parameter from the measured magnitude of the change in the set point and in accordance with a process characteristic associated with the process; and

means for developing a process control signal from the control parameter using a fuzz logic control technique, wherein the process control signal is adapted for use in controlling the process in response to the change in the set point.

14. The non-linear controller of claim 13, wherein the process characteristic includes a process time delay and a dominant time constant of the process and the determining means determines the control parameter as a function of the process time delay and the dominant time constant of the process.

15. The non-linear controller of claim 13, wherein the determining means determines the control parameter in proportion to the measured magnitude of the change in the set point.

16. The non-linear controller of claim 13, wherein the determining means determines an integral time constant of the controller and a proportional gain of the controller in accordance with the process characteristic and further determines the control parameter as a function of the integral time constant and the proportional gain of the controller.

17. The non-linear controller of claim 13, further including a tuner which measures the process characteristic.

18. The non-linear controller of claim 13, further including means for allowing the process characteristic to be input by an operator.

19. The non-linear controller of claim 18, wherein the allowing means allows the operator to input the process characteristic in terms of an integral time constant and a proportional gain of the controller.

20. The non-linear controller of claim 13, wherein the developing means includes a fuzzification unit which uses two membership functions to translate a continuous variable into a fuzzy logic variable.

21. The non-linear controller of claim 13, wherein the developing means comprises a fuzzy logic controller, wherein the control parameter comprises a set of scaling factors including a change in an error scaling factor (S.sub.Δ

e) and wherein the determining means determines the change in the error scaling factor (S.sub.Δ

e) according to the equation

space="preserve" listing-type="equation">S.sub.Δ

e =β

Δ

.sub.SP

where;

Δ

_SP =the magnitude of the change in the set point; and

β

=a multiplication factor.

22. The non-linear controller of claim 21, wherein the set of scaling factors further includes a second scaling factor and a third scaling factor and wherein the determining means determines the second scaling factor as a function of the change in the error scaling factor (S.sub.Δ

e) and an integral time constant of the controller and determines the third scaling factor as function of the change in the error scaling factor (S.sub.Δ

e) and a proportional gain of the controller.

23. The non-linear controller of claim 21, wherein the multiplication factor β

is a constant.

24. The non-linear controller of claim 23, wherein the developing means includes a fuzzy logic proportional-integral-derivative (PID) controller and the multiplication factor β

is approximately 0.28.

25. The non-linear controller of claim 23, wherein the developing means is a fuzzy logic proportional-integral (PI) controller and the multiplication factor β

is approximately 0.4.

26. The non-linear controller of claim 21, wherein the multiplication factor β

is a function of the process characteristic.

27. The non-linear controller of claim 26, wherein the process characteristic includes a time delay of the process and a dominant time constant of the process and wherein the multiplication factor β

is a function of the ratio of the time delay of the process to the dominant time constant of the process.

28. The non-linear controller of claim 27, further including means for detecting the time delay and the dominant time constant of the process.

29. The non-linear controller of claim 13, wherein the developing means comprises a fuzzy logic controller and the control parameter comprises a set of scaling factors and wherein the determining means is responsive to a base set point change factor and includes means for deriving a set of nominal scaling factors in response to the process characteristic and the base set point change factor.

30. The non-linear controller of claim 29, wherein the determining means further includes means for altering one of the nominal scaling factors to develop one of the set of scaling factors.

31. The non-linear controller of claim 30, wherein the altering means multiplies the one of the nominal scaling factors by the ratio of the measured set point change to the base set point change value.

32. The non-linear controller of claim 30, wherein the altering means multiplies the one of the nominal scaling factors by a constant.

33. The non-linear controller of claim 30, wherein the developing means includes further means for developing the process control signal from the nominal set of scaling factors for use in controlling the process in the absence of the change in the set point.

34. The non-linear controller of claim 33, wherein the developing means uses the set of scaling factors to develop the process control signal from a time when a set point change occurs until a time when the process variable remains within a predetermined range of the set point for a predetermined amount of time.

35. A method of controlling a process in response to a change in a set point and in response to a process characteristic, comprising the steps of:

measuring the magnitude of the change in the set point;

developing a set of control parameters in response to the process characteristic and the measured magnitude of the change in the set point; and

controlling the process in accordance with a fuzzy logic control technique using the set of control parameters when the set point change has occurred.

36. The method of claim 35, further including the step of measuring the process characteristic.

38. The method of claim 36, wherein the control parameters include a change in an error scaling factor (S.sub.Δ

e), and wherein the step of developing includes the step of developing the change in the error scaling factor according to the equation;

space="preserve" listing-type="equation">S.sub.Δ

e =β

Δ

.sub.SP

where;

Δ

_SP =the magnitude of the change in the set point; and

β

=a multiplication factor.

39. The method of claim 38, wherein the multiplication factor β

is a constant.

40. The method of claim 38, wherein the multiplication factor β

is a function of a time delay of the process and a dominant time constant of the process.

41. The method of claim 36, further including the steps of determining a further set of control parameters and controlling the process in response to the further set of control parameters in response to a load disturbance in the process.

37. The method of 36, wherein the step of measuring a process characteristic includes the steps of determining a time delay and a dominant time constant of the process and wherein the step of developing includes the step of developing one of the set of control parameters as a function of the time delay and the dominant time constant of the process.