System and method for dynamic modeling, parameter estimation and optimization for processes having operating targets

• US 6,816,822 B1
• Filed: 08/16/2000
• Issued: 11/09/2004
• Est. Priority Date: 08/16/2000
• Status: Active Grant

• Current Assignee: ABB Automation GmbH (ABB Limited)
• Original Assignee: ABB AG (ABB Limited), ABB Automation GmbH (ABB Limited)
• Inventors: Hess, Todd M., Richards, Martyn Glenn, Kroll, Andreas
• Primary Examiner(s): JONES, HUGH M
• Application Number:
  US09/639,543
• Time in Patent Office: 1,546 Days
• Field of Search: 703/2, 703/6, 700/32
• US Class Current: 703/2
• CPC Class Codes: G05B 13/042 in which a parameter or coefficient is automatically adjusted to optimise the performance
• CPC Class Codes: G05B 17/02 electric
• CPC Class Codes: G05B 23/0227 Qualitative history assessment, whereby the type of data acted upon, e.g. waveforms, images or patterns, is not relevant, e.g. rule based assessment; if-then decisions
• CPC Class Codes: G05B 23/0294 Optimizing process, e.g. process efficiency, product quality

Abstract

A technique to provide dynamic modeling, parameter estimation and optimization for processes having operating targets. The technique has an optimization cycle and a simulation cycle. The simulation cycle runs at a rate that is equal to or faster than the run rate of the optimization cycle. Both cycles obtain data from the field about the process. The optimization cycle provides a simulation execution of a dynamic process model and the results of that simulation are used to match the model response to the actual plant response. This matching determines if a dynamic parameter estimation and data reconciliation is needed for the model parameters. Optimized operating targets are determined using either the estimated and reconciled parameters and data or if the same is not needed the collected process data. Mathematical and heuristic assessment tools are used to determine if the optimized operating targets should or should not be invoked.

Claims

1. A method for dynamically optimizing a process comprising the steps of:

performing an optimization cycle at a first rate;

performing simultaneously with said optimization cycle a simulation cycle at a second rate which is equal to or faster than said first rate;

said simulation cycle comprising the steps of;

collecting data about said process;

using said collected process data to execute in a predefined time interval during said simulation cycle a dynamic simulation for a dynamic model of said process;

storing data for said dynamic simulation;

said optimization cycle comprising the steps of;

collecting data about said process;

determining if a dynamic parameter estimation and data reconciliation procedure is needed to minimize differences between said collected process data and said data stored for said dynamic simulation calculated during a simulation cycle corresponding to a period of time which is the same as the time period during which said process data was collected during said simulation cycle;

performing a dynamic parameter estimation and data reconciliation procedure to provide updated parameters and reconciled measurements for said data stored for said dynamic simulation when said procedure has been determined to be needed; and

calculating optimum operating targets for said dynamic process model either from said collected process data when said dynamic parameter estimation and data reconciliation procedure is not needed or from said updated parameters and reconciled measurements when said procedure has been determined to be needed.

2. The method of claim 1 wherein said optimization cycle further includes the steps of:

determining if said optimized operating targets should be invoked; and

determining if said optimization cycle should be exited cycle when said optimized operating targets should not be invoked.

3. The method of claim 2 wherein said optimization cycle further includes the step of waiting until the start of the next optimization cycle when said optimization cycle should not be exited.

4. The method of claim 1 wherein said optimization cycle further includes the steps of:

determining if said optimized operating targets should be invoked; and

updating a system for controlling said process when said optimized operating targets should be invoked.

5. The method of claim 4 further including the steps of:

determining after said system is updated if said optimization cycle should be exited; and

waiting until the start of the next optimization cycle when said optimization cycle should not be exited.

6. The method of claim 1 wherein said data about said process is collected in said optimization and simulation cycles from a system for controlling said process.

7. The method of claim 1 wherein said data about said process is collected in said optimization and simulation cycles from a real time database.

8. A method for dynamically optimizing a process comprising the steps of:

performing an optimization cycle at a first rate;

performing simultaneously with said optimization cycle a simulation cycle at a second rate which is equal to or faster than said first rate;

said optimization and simulation cycles both comprising the step of collecting data about said process;

said simulation cycle further comprising the steps of;

using said collected process data to execute in a predefined time interval during said simulation cycle a dynamic simulation for a dynamic model of said process; and

storing data for said dynamic simulation;

said optimization cycle further comprising the steps of;

performing a dynamic parameter estimation and data reconciliation procedure to provide updated parameters and reconciled measurements for said data stored for said dynamic simulation when said dynamic parameter estimation and data reconciliation procedure is needed to minimize differences between said collected process data and said data stored for said dynamic simulation calculated during a simulation cycle corresponding to a period of time which is the same as the time period during which said process data was collected during said simulation cycle;

calculating optimum operating targets for said dynamic process model from said updated parameters and reconciled measurements;

determining if said optimized operating targets should be invoked;

determining if said optimization cycle should be exited if said optimized operating targets should not be invoked;

waiting until the start of the next optimization cycle when said optimization cycle should not be exited; and

updating a system for controlling said process when said optimized operating targets should be invoked.

9. A method for dynamically optimizing a process comprising the steps of:

performing an optimization cycle at a first rate;

performing simultaneously with said optimization cycle a simulation cycle at a second rate which is equal to or faster than said first rate;

said optimization and simulation cycles both comprising the step of collecting data about said process;

said simulation cycle further comprising the steps of;

using said collected process data to execute in a predefined time interval during said simulation cycle a dynamic simulation for a dynamic model of said process; and

storing data for said dynamic simulation;

said optimization cycle further comprising the steps of;

performing a dynamic parameter estimation and data reconciliation procedure to provide updated parameters and reconciled measurements for said data stored for said dynamic simulation when said dynamic parameter estimation and data reconciliation procedure is needed to minimize differences between said collected process data and said data stored for said dynamic simulation calculated during a simulation cycle corresponding to a period of time which is the same as the time period during which said process data was collected during said simulation cycle; and

calculating optimum operating targets for said dynamic process model from said updated parameters and reconciled measurements.

10. The method of claim 9 wherein said optimization cycle further includes the steps of:

determining if said optimized operating targets should be invoked; and

updating a system for controlling said process when said optimized operating targets should be invoked.

11. The method of claim 9 wherein said optimization cycle further includes the steps of:

determining if said optimized operating targets should be invoked; and

determining if said optimization cycle should be exited cycle when said optimized operating targets should not be invoked.

12. The method of claim 11 wherein said optimization cycle further includes the step of waiting until the start of the next optimization cycle when said optimization cycle should not be exited.

13. The method of claim 9 wherein said data about said process is collected in said optimization and simulation cycles from a system for controlling said process.

14. The method of claim 9 wherein said data about said process is collected in said optimization and simulation cycles from a real time database.

15. A method for dynamically optimizing a process comprising the steps of:

performing an optimization cycle at a first rate;

performing simultaneously with said optimization cycle a simulation cycle at a second rate which is equal to or faster than said first rate;

said optimization and simulation cycles both comprising the step of collecting data about said process;

said simulation cycle further comprising the steps of;

using said collected process data to execute in a predefined time interval during said simulation cycle a dynamic simulation for a dynamic model of said process; and

storing data for said dynamic simulation;

said optimization cycle further comprising the steps of;

determining if a dynamic parameter estimation and data reconciliation procedure is needed to minimize differences between said collected process data and said data stored for said dynamic simulation calculated during a simulation cycle corresponding to a period of time which is the same as the time period during which said process data was collected during said simulation cycle;

calculating optimum operating targets for said dynamic process model from said collected process data when said dynamic parameter estimation and data reconciliation procedure is not needed;

determining if said optimized operating targets should be invoked;

determining if said optimization cycle should be exited if said optimized operating targets should not be invoked;

waiting until the start of the next optimization cycle when said optimization cycle should not be exited; and

updating a system for controlling said process when said optimized operating targets should be invoked.

16. A method for dynamically optimizing a process comprising the steps of:

performing an optimization cycle at a first rate;

performing simultaneously with said optimization cycle a simulation cycle at a second rate which is equal to or faster than said first rate;

said optimization and simulation cycles both comprising the step of collecting data about said process;

said simulation cycle further comprising the steps of;

using said collected process data to execute in a predefined time interval during said simulation cycle a dynamic simulation for a dynamic model of said process; and

storing data for said dynamic simulation;

said optimization cycle further comprising the steps of;

determining if a dynamic parameter estimation and data reconciliation procedure is needed to minimize differences between said collected process data and said data stored for said dynamic simulation calculated during a simulation cycle corresponding to a period of time which is the same as the time period during which said process data was collected during said simulation cycle; and

calculating optimum operating targets for said dynamic process model from said collected process data when said dynamic parameter estimation and data reconciliation procedure is not needed.

17. The method of claim 16 wherein said optimization cycle further includes the steps of:

determining if said optimized operating targets should be invoked; and

updating a system for controlling said process when said optimized operating targets should be invoked.

18. The method of claim 16 wherein said optimization cycle further includes the steps of:

determining if said optimized operating targets should be invoked; and

determining if said optimization cycle should be exited cycle when said optimized operating targets should not be invoked.

19. The method of claim 18 wherein said optimization cycle further includes the step of waiting until the start of the next optimization cycle when said optimization cycle should not be exited.

20. The method of claim 16 wherein said data about said process is collected in said optimization and simulation cycles from a system for controlling said process.

21. The method of claim 16 wherein said data about said process is collected in said optimization and simulation cycles from a real time database.