Method for determining and tuning process characteristic parameters using a simulation system

• US 10,222,769 B2
• Filed: 10/12/2012
• Issued: 03/05/2019
• Est. Priority Date: 10/12/2012
• Status: Active Grant

• Current Assignee: Emerson Process Management Power & Water Solutions Incorporated (Emerson Electric Company)
• Original Assignee: Emerson Process Management Power & Water Solutions Incorporated (Emerson Electric Company)
• Inventors: Cheng, Xu
• Primary Examiner(s): Louis, Andre Pierre
• Application Number:
  US13/650,296
• Publication Number:
  US 20140107993A1
• Time in Patent Office: 2,335 Days
• Field of Search: 703 2, 703 13, 700 19, 700 29, 700 51
• CPC Class Codes: G05B 17/02 electric
• CPC Class Codes: Y02P 90/02 Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Abstract

A process characteristic parameter determination system uses a process model and a tuning module to accurately determine a value for a process characteristic parameter within a plant without measuring the process characteristic parameter directly, and may operate on-line or while the process is running to automatically determine a correct value of the process characteristic parameter at any time during on-going operation of the process. The process characteristic parameter value, which may be a heat transfer coefficient value for a heat exchanger, can then be used to enable the determination of a more accurate simulation result and/or to make other on-line process decisions, such as process control decisions, process operational mode decisions, process maintenance decisions such as implementing a soot blowing operation, etc.

Claims

1. A system for operating a process plant by determining a process characteristic parameter of the process plant, the system comprising:

one or more processors;

one or more program memories communicatively connected to the one or more processors and storing executable instructions that, when executed by the one or more processors, cause the one or more processors to implement a plurality of units, comprising;

a process control unit configured to generate a process control signal to control at least a portion of a controlled process within the process plant;

a process model unit configured to receive a set of process inputs related to the portion of the controlled process and to generate a predicted value of a process variable related to the portion of the controlled process using a process model that models the portion of the controlled process based on (i) the set of process inputs and (ii) a value of the process characteristic parameter, which process characteristic parameter is a parameter within the process model that corresponds to a process characteristic of the portion of the controlled process that is not directly measured, wherein the process characteristic is a property of the portion of the controlled process or a property of process equipment of the portion of the controlled process;

a difference unit configured to generate a difference signal indicating a difference between a first input indicating a process variable value for the process variable and a second input indicating the predicted value of the process variable, wherein the first input receives the process variable value for the process variable of the controlled process as measured or determined (i) by a component within the controlled process or (ii) by the process control unit, and wherein the second input receives the predicted value of the process variable as generated by the process model unit; and

a tuner unit configured to receive the difference signal from the difference unit, to determine a change to the value of the process characteristic parameter within the process model by evaluating the received difference signal using a tuner logic, and to automatically adjust the value of the process characteristic parameter within the process model based on the determined change,

wherein the adjusted value of the process characteristic parameter is used to determine and implement a control action to control the process plant.

2. The system of claim 1, wherein the tuner logic includes a predetermined set of rules or routine to drive the difference signal towards zero.

3. The system of claim 1, wherein the process model is a first principle-based model of the controlled process.

4. The system of claim 1, wherein process model is an empirical model of the controlled process.

5. The system of claim 1, wherein the tuner unit includes a feedback control logic to determine a manner in which to change the value of the process characteristic parameter within the process model.

6. The system of claim 1, wherein the process model unit includes a simulated process controller having control logic to produce a simulated control signal value as the predicted value of the process variable, wherein the process model produces a predicted value of a further process variable and wherein the simulated process controller has a first input to receive a set point and a second input to receive the predicted value of the further process variable from the process model and wherein the simulated process controller uses the control logic to develop the simulated control signal value as the predicted value of the process variable based on the set point and the predicted value of the further process variable.

7. The system of claim 6, wherein the further process variable is a controlled process variable within the controlled process and wherein the simulated process controller receives the predicted value of the further process variable as a controlled variable feedback signal, and receives the set point as a set point for the further process variable from the controlled process.

8. The system of claim 6, wherein the difference unit receives the simulated process control signal as the predicted value of the process variable at the second input and receives the process control signal from the process control unit as the process variable value at the first input.

9. The system of claim 1, wherein difference unit is a summer that subtracts a first one of the first and second inputs from a second one of the first and second inputs to create the difference signal.

10. The system of claim 1, wherein the process variable is an uncontrolled variable within the controlled process, and wherein the process model unit produces the predicted value of the process variable as a predicted value of the uncontrolled variable.

11. The system of claim 1, wherein the process variable is process control signal within the controlled process, and wherein the process model unit produces the predicted value of the process variable as a predicted value of the process control signal.

12. A method of operating a process plant by simulating a process including a process characteristic parameter of the process plant, comprising:

generating, by one or more processors, a process control signal to control at least a portion of a controlled process within the process plant;

simulating, by the one or more processors, the process to produce a predicted value of a process variable of the controlled process of the process plant using a process model unit to receive a set of process inputs related to the portion of the controlled process and to generate the predicted value of the process variable using a process model that models the portion of the controlled process based on (i) the set of process inputs and (ii) a value of the process characteristic parameter, which process characteristic parameter is a parameter within the process model that corresponds to a process characteristic of the portion of the controlled process that is not directly measured, wherein the process characteristic is a property of the portion of the controlled process or a property of process equipment of the portion of the controlled process;

obtaining, by the one or more processors, an actual value of the process variable from the controlled process as (i) a measurement or determination by a component within the controlled process or (ii) the generated process control signal;

determining, by the one or more processors, a difference signal indicating a difference between the actual value of the process variable and the predicted value of the process variable produced by the process model unit;

determining, by the one or more processors, a change to the value of the process characteristic parameter within the process model by evaluating the difference signal using a tuner logic;

automatically adjusting, by the one or more processors, the value of the process characteristic parameter of the process model based on the determined change to drive the difference signal towards zero;

determining, by the one or more processors, a control action to control the process plant; and

implementing, by the one or more processors, the determined control action.

13. The method of operating a process plant according to claim 12, wherein simulating the controlled process to produce the predicted value of the process variable includes using the process model to produce a predicted value of an uncontrolled process variable within the controlled process as the predicted value of the process variable and wherein obtaining an actual value of the process variable from the controlled process includes obtaining a measured value of the uncontrolled process variable in the controlled process.

14. The method of operating a process plant according to claim 12, wherein simulating the controlled process to produce the predicted value of the process variable includes using the process model to produce a predicted value of a controlled process variable within the controlled process, and simulating the control system of the controlled process using the predicted value of the controlled process variable to produce a simulated control signal value as the predicted value of the process variable and wherein obtaining an actual value of the process variable from the controlled process includes obtaining a value of the process control signal developed to control the controlled process variable from a process controller within the controlled process.

15. The method of operating a process plant according to claim 14, further including obtaining, by the one or more processors, a value of a set point used by the process controller within the controlled process and wherein simulating the control system of the controlled process includes using the set point value and the predicted value of the controlled process variable to produce the simulated control signal value as the predicted value of the process variable.

16. The method of operating a process plant according to claim 15, further including using a same control logic in the simulated process controller that is equivalent to control logic used in the process controller within the controlled process.

17. The method of operating a process plant according to claim 12, wherein adjusting the value of the process characteristic parameter of the process model includes using a feedback control routing to determine a manner in which to adjust the value of the process characteristic parameter of the process model based on the difference signal.

18. The method of operating a process plant according to claim 12, wherein simulating the controlled process to produce the predicted value of the process variable using the process model includes using a first principle-based model of the controlled process within the process model.

19. The method of operating a process plant according to claim 12, wherein simulating the controlled process to produce the predicted value of the process variable using the process model includes using an empirical-based model of the controlled process within the process model.

20. The method of operating a process plant according to claim 12, wherein simulating the controlled process includes producing a predicted value of an uncontrolled process variable as the predicted value of the process variable.

21. The method of operating a process plant according to claim 12, wherein simulating the controlled process includes producing a simulated value of the process control signal as the predicted value of the process variable.

22. The method of operating a process plant according to claim 12, wherein the process characteristic parameter is a heat transfer coefficient of a heat exchanger.

23. The method of operating a process plant according to claim 12, further including performing the steps of simulating the controlled process to produce the predicted value of the process variable, obtaining the actual value of the process variable from the controlled process, determining the difference signal indicating the difference between the actual value of the process variable and the predicted value of the process variable and adjusting the value of the process characteristic parameter of the process model in real-time while the controlled process is operating.

24. The method of operating a process plant according to claim 12, further including performing the steps of simulating the operation of the controlled process to produce the predicted value of the process variable, obtaining the actual value of the process variable from the controlled process, determining the difference signal indicating the difference between the actual value of the process variable and the predicted value of the process variable and adjusting the value of the process characteristic parameter of the process model based on process data stored in a database.

25. A non-transitory computer readable medium storing instructions for simulating a controlled process of a process plant that, when executed by at least one processor of a computer system, cause the computer system to execute a plurality of simulation modules, each of the simulation modules being communicatively connected to one or more other simulation modules of the plurality of simulation modules, wherein each of the simulation modules includes a process model that performs modeling of a different portion of the controlled process, wherein one of the simulation modules includes:

a process model unit configured to receive a set of process inputs related to the respective portion of the controlled process and to generate a predicted value of a process variable related to the portion of the controlled process using a process model based on (i) the set of process inputs and (ii) a value of a process characteristic parameter, which process characteristic parameter is a parameter within the process model that corresponds to a process characteristic of the portion of the controlled process that is not directly measured, wherein the process characteristic is a property of the portion of the controlled process or a property of process equipment of the portion of the controlled process;

a difference unit configured to generate a difference signal indicating a difference between a first input indicating a process variable value for the process variable and a second input indicating the predicted value of the process variable, wherein the first input receives the process variable value for the process variable of the controlled process as measured or determined within the controlled process and wherein the second input receives the predicted value of the process variable as generated by the process model unit; and

a tuner unit configured to receive the difference signal from the difference unit, to determine a change to the value of the process characteristic parameter within the process model by evaluating the received difference signal using a tuner logic, and to automatically adjust the value of the process characteristic parameter within the process model based on the determined change;

wherein the one of the simulation modules operates to communicate an output of the process model unit as an input to another one of the simulation modules; and

wherein the value of the process characteristic parameter is used to determine a control action implemented to control the process plant.

26. The non-transitory computer readable medium of claim 25, wherein the process model is a first principle-based model of the particular portion of the controlled process.

27. The non-transitory computer readable medium of claim 25, wherein the process model unit includes a simulated process controller having control logic to produce a simulated control signal value as the predicted value of the process variable, wherein the process model is configured to produce a predicted value of a further process variable and wherein the simulated process controller has a first input to receive a set point and a second input to receive the predicted value of the further process variable from the process model.

28. The non-transitory computer readable medium of claim 27, wherein the further process variable is a controlled process variable within the controlled process and wherein the simulated process controller is configured to receive the predicted value of the further process variable as a controlled variable feedback signal, and receives the set point as a set point for the further process variable from the controlled process.

29. The non-transitory computer readable medium of claim 28, wherein the difference unit is configured to receive the simulated process control signal value as the predicted value of the process variable at the second input and is configured to receive a control signal value from a process controller within the controlled process as the process variable value at the first input.

30. The non-transitory computer readable medium of claim 25, wherein the process variable is an uncontrolled variable within the controlled process, and wherein the process model unit is configured to produce the predicted value of the process variable as a predicted value of the uncontrolled variable.

31. The non-transitory computer readable medium of claim 25, wherein tuner unit includes a feedback control logic to determine a manner in which to change the value of the process characteristic parameter within the process model.

32. The non-transitory computer readable medium of claim 25, wherein the process variable is a process control signal within the controlled process, and wherein the process model unit is configured to produce the predicted value of the process variable as a predicted value of the process control signal.

33. The non-transitory computer readable medium of claim 25, wherein the plurality of simulation modules are stored in and executed in separate computer devices connected by one or more communication channels.

34. The non-transitory computer readable medium of claim 25, wherein the plurality of simulation modules operate on-line during operation of the process.

35. The non-transitory computer readable medium of claim 25, wherein the one of the simulation modules is stored in a device associated with controlling equipment associated with the particular portion of the controlled process.

36. The non-transitory computer readable medium of claim 25, further including a data historian that stores process data generated in the controlled process, wherein at least one of the plurality of simulation modules are communicatively coupled to the data historian to receive the process data for the process variable to operate off-line with respect to operation of the controlled process.