Control system using an adaptive neural network for target and path optimization for a multivariable, nonlinear process

• US 5,477,444 A
• Filed: 09/14/1992
• Issued: 12/19/1995
• Est. Priority Date: 09/14/1992
• Status: Expired due to Term

• Current Assignee: Aspentech Corporation, Texaco Incorporated (Chevron Corp.)
• Original Assignee: NeuralWare, Inc. (Aspentech Corporation), Texaco Incorporated (Chevron Corp.)
• Inventors: DuBose, Paul A., Graettinger, Timothy J., Federowicz, Alexander J., Bhat, Naveen V., Braden, William B., Heckendoorn, Kent E.
• Primary Examiner(s): Envall, Jr., Roy N.
• Assistant Examiner(s): BROWN, THOMAS
• Application Number:
  US07/944,645
• Time in Patent Office: 1,191 Days
• Field of Search: 364/148-152, 364/160-163, 395/21-24, 395/903-906
• US Class Current: 700/48
• CPC Class Codes: B01D 3/425 Head-, bottom- and feed stream
• CPC Class Codes: B01J 19/0033 Optimalisation processes, i.e. processes with adaptive control systems adaptive control systems per se G05B13/00
• CPC Class Codes: G05B 13/027 using neural networks only

Abstract

A control system having four major components: a target optimizer, a path optimizer, a neural network adaptation controller and a neural network. In the target optimizer, the controlled variables are optimized to provide the most economically desirable outputs, subject to operating constraints. Various manipulated variable and disturbance values are provided for modeling purposes. The neural network receives as inputs a plurality of settings for each manipulated and disturbance variable. For target optimization all the neural network input values are set equal to produce a steady state controlled variable value. The entire process is repeated with differing manipulated variable values until an optimal solution develops. The resulting target controlled and manipulated variable values are provided to the path optimizer to allow the manipulated variables to be adjusted to obtain the target output. Various manipulated variable values are developed to model moves from current to desired values. In this case trend indicating values of the manipulated and disturbance variables are provided to produce time varying values of the controlled variables. The process is repeated until an optimal path is obtained, at which time the manipulated variable values are applied to the actual process. On a periodic basis all of the disturbance, manipulated and controlled variables are sampled to find areas where the training of the neural network is sparse or where high dynamic conditions are indicated. These values are added to the set of values used to train the neural network.

Claims

1. A control system for a process having a plurality of disturbance variables as process input conditions, a plurality of manipulated variables as process control conditions and a plurality of controlled variables as process output conditions, the control system comprising:

means for periodically determining the values of the disturbance variables;

means for optimizing the values of the controlled variables based on the determined disturbance variable values and providing target manipulated variable values associated with said optimized controlled variable values;

means receiving said target manipulated variable values for optimizing the movement from the existing manipulated variable values to said target manipulated variable values and providing new manipulated variable values; and

a neural network coupled to said controlled variable optimizing means and said movement optimizing means and receiving manipulated variable values and disturbance variable values and producing the resulting controlled variable values representative of the process being controlled.

2. The control system of claim 1, wherein said neural network has a plurality of inputs of each manipulated variable, wherein when said plurality of inputs for each manipulated variable receive equal values, the resulting controlled variable values are steady state values and when said plurality of inputs for each manipulated variable receive different values, the resulting controlled variable values are future values.

3. The control system of claim 2, wherein said controlled variable optimizing means and said movement optimizing means include means for developing a plurality of manipulated variable values for provision to said neural network to produce a plurality of controlled variable values for the respective optimization, wherein said controlled variable optimizing means provides equal values to said plurality of inputs of each manipulated variable and wherein said movement optimizing means provides time dependent, varying values to said plurality of inputs of each manipulated variable.

4. The control system of claim 1, wherein said neural network has a plurality of inputs of each disturbance variable, wherein when said plurality of inputs for each disturbance variable receive equal values, the resulting controlled variable values are steady state values and when said plurality of inputs for each disturbance variable receive different values, the resulting controlled variable values are future values.

5. The control system of claim 4, wherein said neural network has a plurality of inputs of each manipulated variable, wherein when said plurality of inputs for each manipulated variable receive equal values, the resulting controlled variable values are steady state values and when said plurality of inputs for each manipulated variable receive different values, the resulting controlled variable values are future values.

6. The control system of claim 5, wherein said controlled variable optimizing means and said movement optimizing means include means for developing a plurality of manipulated variable and disturbance variable values for provision to said neural network to produce a plurality of controlled variable values for the respective optimization, wherein said controlled variable optimizing means provides equal values to said plurality of inputs of each manipulated and disturbance variable and wherein said movement optimizing means provides time dependent, varying values to said plurality of inputs of each manipulated and disturbance variable.

7. The control system of claim 1, wherein said controlled variable optimizing means includes means for setting limits of the controlled and manipulated variables to be utilized during optimizing and means for altering said controlled variable limits if an optimal solution cannot be developed using said limits.

8. The control system of claim 7, wherein said means for altering said limits removes all limits.

9. The control system of claim 7, wherein said means for altering said limits alters said limits in order of least economic effect.

10. The control system of claim 7, wherein the optimization of the values of the controlled variables utilizes a transfer function between measured values and values to be optimized and wherein said means for altering said limits alters said transfer function to allow for a continuous function outside of said limits.

11. The control system of claim 10, wherein said continuous function rapidly ramps to a zero optimizing value outside of said limits.

12. The control system of claim 1, wherein said movement optimizing means operates on a regular, periodic basis, and wherein said controlled variable optimizing means includes means for setting disturbance variable or controlled variable change limits and operates when the change in a disturbance or controlled variable exceeds its respective limit.

13. The control system of claim 1, further comprising:

means for monitoring the disturbance variable values, the manipulated variable values and the controlled variable values and storing said monitored variable values for use in training said neural network.

14. The control system of claim 13, wherein said monitoring and storing means determines the respective disturbance and controlled variable values and stores said monitored variable values if the number of stored monitored variable values within a predetermined region of said disturbance and controlled variable values is below a predetermined number.

15. The control system of claim 13, wherein said monitoring and storing means determines the dynamics of said disturbance and controlled variable values and stores said monitored variable values if said dynamics exceed predetermined limits.

16. A method of controlling a process having a plurality of disturbance variables as process input conditions, a plurality of manipulated variables as process control conditions and a plurality of controlled variables as process output conditions, the method comprising the steps of:

periodically determining the values of the disturbance variables;

optimizing the values of the controlled variables based on the determined disturbance variable values and providing target manipulated variable values associated with said optimized controlled variable values; and

optimizing the movement from the existing manipulated variable values to said target manipulated variable values and providing new manipulated variable values; and

said steps of controlled variable optimizing and movement optimizing utilizing a neural network receiving manipulated variable values and disturbance variable values and producing the resulting controlled variable values representative of the process being controlled.

17. The method of claim 16, wherein said neural network has a plurality of inputs of each manipulated variable, wherein when said plurality of inputs for each manipulated variable receive equal values, the resulting controlled variable values are steady state values and when said plurality of inputs for each manipulated variable receive different values, the resulting controlled variable values are future values.

18. The method of claim 17, wherein said steps of controlled variable optimizing and movement optimizing each include developing a plurality of manipulated variable values for provision to said neural network to produce a plurality of controlled variable values for the respective optimization, wherein said step of controlled variable optimizing provides equal values to said plurality of inputs of each manipulated variable and wherein said step of movement optimizing provides time dependent, varying values to said plurality of inputs of each manipulated variable.

19. The method of claim 16, wherein said neural network has a plurality of inputs of each disturbance variable, wherein when said plurality of inputs for each disturbance variable receive equal values, the resulting controlled variable values are steady state values and when said plurality of inputs for each disturbance variable receive different values, the resulting controlled variable values are future values.

20. The method of claim 19, wherein said neural network has a plurality of inputs of each manipulated variable, wherein when said plurality of inputs for each manipulated variable receive equal values, the resulting controlled variable values are steady state values and when said plurality of inputs for each manipulated variable receive different values, the resulting controlled variable values are future values.

21. The method of claim 20, wherein said steps of controlled variable optimizing and movement optimizing include developing a plurality of manipulated variable and disturbance variable values for provision to said neural network to produce a plurality of controlled variable values for the respective optimization, wherein said step of controlled variable optimizing provides equal values to said plurality of inputs of each manipulated and disturbance variable and wherein said step of movement optimizing provides time dependent, varying values to said plurality of inputs of each manipulated and disturbance variable.

22. The method of claim 16, wherein said step of controlled variable optimizing includes setting limits of the controlled and manipulated variables to be utilized during optimizing and altering said controlled variable limits if an optimal solution cannot be developed using said limits.

23. The method of claim 22, wherein said altering of said limits removes all limits.

24. The method of claim 22, wherein said altering said limits alters said limits in order of least economic effect.

25. The method of claim 22, wherein the optimization of the values of the controlled variables utilizes a transfer function between measured values and values to be optimized and wherein said step of altering said limits alters said transfer function to allow for a continuous function outside of said limits.

26. The method of claim 16, wherein said continuous function rapidly ramps to a zero optimizing value outside of said limits.

27. The method of claim 16, wherein said step of movement optimizing operates on a regular, periodic basis, and wherein said step of controlled variable optimizing includes setting disturbance or controlled variable change limits and operating when the change in a disturbance or controlled variable exceeds its respective limit.

28. The method of claim 16, further comprising the step of:

monitoring the disturbance variable values, the manipulated variable values and the controlled variable values and storing said monitored variable values for use in training said neural network.

29. The method of claim 28, wherein said step of monitoring and storing variable values determines the respective disturbance and controlled variable values and stores said monitored variable values if the number of stored monitored variable values within a predetermined region of said disturbance and controlled variable values is below a predetermined number.

30. The method of claim 28, wherein said step of monitoring and storing variable values determines the dynamics of said disturbance and controlled variable values and stores said monitored variable values if said dynamics exceed predetermined limits.

31. A system for performing a distillation process in an oil refinery having a plurality of disturbance variables as process input conditions, a plurality of manipulated variables as process control conditions and a plurality of controlled variables as process output conditions, the system comprising:

a distillation column having a means for receiving an input feed stream and providing overhead and bottom output streams;

means for cooling said overhead output stream;

means for controllably returning a portion of said cooled overhead output stream to said distillation column, said rate of controllably returning being a manipulated variable;

means for controllably maintaining the pressure of materials in said distillation column, said level of controllably maintaining the pressure being a manipulated variable;

means for controllably heating a portion of material in said distillation column, said rate of controllably heating being a manipulated variable;

means for periodically determining the values of the disturbance variables;

means for optimizing the values of the controlled variables based on the determined disturbance variable values and providing target manipulated variable values associated with said optimized controlled variable values;

means receiving said target manipulated variable values for optimizing the movement from the existing manipulated variable values to said target manipulated variable values and providing new manipulated variable values; and

a neural network coupled to said controlled variable optimizing means and said movement optimizing means and receiving manipulated variable values and disturbance variable values and producing the resulting controlled variable values representative of the distillation process being controlled.

32. The system of claim 31, wherein said neural network has a plurality of inputs of each manipulated variable, wherein when said plurality of inputs for each manipulated variable receive equal values, the resulting controlled variable values are steady state values and when said plurality of inputs for each manipulated variable receive different values, the resulting controlled variable values are future values.

33. The system of claim 32, wherein said controlled variable optimizing means and said movement optimizing means include means for developing a plurality of manipulated variable values for provision to said neural network to produce a plurality of controlled variable values for the respective optimization, wherein said controlled variable optimizing means provides equal values to said plurality of inputs of each manipulated variable and wherein said movement optimizing means provides time dependent, varying values to said plurality of inputs of each manipulated variable.

34. The control system of claim 31, wherein said neural network has a plurality of inputs of each disturbance variable, wherein when said plurality of inputs for each disturbance variable receive equal values, the resulting controlled variable values are steady state values and when said plurality of inputs for each disturbance variable receive different values, the resulting controlled variable values are future values.

35. The control system of claim 34, wherein said neural network has a plurality of inputs of each manipulated variable, wherein when said plurality of inputs for each manipulated variable receive equal values, the resulting controlled variable values are steady state values and when said plurality of inputs for each manipulated variable receive different values, the resulting controlled variable values are future values.

36. The control system of claim 35, wherein said controlled variable optimizing means and said movement optimizing means include means for developing a plurality of manipulated variable and disturbance variable values for provision to said neural network to produce a plurality of controlled variable values for the respective optimization, wherein said controlled variable optimizing means provides equal values to said plurality of inputs of each manipulated and disturbance variable and wherein said movement optimizing means provides time dependent, varying values to said plurality of inputs of each manipulated and disturbance variable.

37. The system of claim 31, wherein said controlled variable optimizing means includes means for setting limits of the controlled and manipulated variables to be utilized during optimizing and means for altering said controlled variable limits if an optimal solution cannot be developed using said limits.

38. The system of claim 37, wherein said means for altering said limits removes all limits.

39. The system of claim 37, wherein said means for altering said limits are alters said limits in order of least economic effect.

40. The control system of claim 37, wherein the optimization of the values of the controlled variables utilizes a transfer function between measured values and values to be optimized and wherein said means for altering said limits alters said transfer function to allow for a continuous function outside of said limits.

41. The control system of claim 40, wherein said continuous function rapidly ramps to a zero optimizing value outside of said limits.

42. The system of claim 31, wherein said movement optimizing means operates on a regular, periodic basis, and wherein said controlled variable optimizing means includes means for setting disturbance variable change limits and operates when the change in a disturbance variable exceeds its respective limit.

43. The system of claim 31, further comprising:

means for monitoring the disturbance variable values, the manipulated variable values and the controlled variable values and storing said monitored variable values for use in training said neural network.

44. The system of claim 43, wherein said monitoring and storing means determines the respective disturbance and controlled variable values and stores said monitored variable values if the number of stored monitored variable values within a predetermined region of said disturbance and controlled variable values is below a predetermined number.

45. The system of claim 43, wherein said monitoring and storing means determines the dynamics of said disturbance and controlled variable values and stores said monitored variable values if said dynamics exceed predetermined limits.

46. The system of claim 31, wherein said means for controllably heating includes:

means for receiving materials from approximately the bottom of said distillation column;

means for heating said received materials; and

means for returning said heated received materials to said distillation column.