Method and apparatus for minimizing error in dynamic and steady-state processes for prediction, control, and optimization

• US 20060259197A1
• Filed: 02/21/2006
• Published: 11/16/2006
• Est. Priority Date: 05/06/1996
• Status: Active Grant

• Current Assignee: Rockwell Automation Technologies Incorporated (Allen-Bradley Co., Inc.)
• Original Assignee: Rockwell Automation Technologies Incorporated (Allen-Bradley Co., Inc.)
• Inventors: Piche, Stephen, Martin, Gregory D., Boe, Eugene
• Granted Patent: US 8,311,673 B2
• Time in Patent Office: Days
• US Class Current: 700/246
• CPC Class Codes: G05B 13/027 using neural networks only
• CPC Class Codes: G05B 13/042 in which a parameter or coefficient is automatically adjusted to optimise the performance
• CPC Class Codes: G05B 13/048 using a predictor
• CPC Class Codes: G05B 17/02 electric
• CPC Class Codes: G09B 23/02 for mathematics for statics or dynamics G09B23/08

Abstract

A method for providing independent static and dynamic models in a prediction, control and optimization environment utilizes an independent static model (20) and an independent dynamic model (22). The static model (20) is a rigorous predictive model that is trained over a wide range of data, whereas the dynamic model (22) is trained over a narrow range of data. The gain K of the static model (20) is utilized to scale the gain k of the dynamic model (22). The forced dynamic portion of the model (22) referred to as the b_i variables are scaled by the ratio of the gains K and k. The b_i have a direct effect on the gain of a dynamic model (22). This is facilitated by a coefficient modification block (40). Thereafter, the difference between the new value input to the static model (20) and the prior steady-state value is utilized as an input to the dynamic model (22). The predicted dynamic output is then summed with the previous steady-state value to provide a predicted value Y. Additionally, the path that is traversed between steady-state value changes.

Claims

1. A dynamic controller for controlling operation of a plant or process by predicting a change in dynamic input values to the plant or process to effect a change in output of the plant or process from a current output value at a first time to a desired output value at a second time to achieve a specified objective of the dynamic controller, comprising:

a dynamic predictive model for receiving the current input value and the desired output value and predicting a plurality of input values at different times between the first time and the second time to define a dynamic operation path of the plant or process, between the current output value at the first time and the desired output value at the second time;

an error generator for comparing a predicted dynamic output value from the dynamic operation path to the desired output value and generating a primary error value as a difference therebetween for each of said times;

an error minimization device for determining a change in each input value to minimize the primary error value output by the error generator, wherein the error minimization device comprises a filter to determine operation of the error minimization device based upon accuracy of the dynamic model;

a summation device for summing the determined change in input value with an original input value, wherein the original input value comprises the input value before the determined change therein, for each time position to provide a future input value as a summed input value; and

a controller for controlling the operation of the error minimization device to operate under control of the dynamic controller to minimize the primary error value in accordance with the specified objective of the dynamic controller.

2. The dynamic controller of claim 1,

wherein the filter is operable to activate the error minimization device when the primary error value is below an error tolerance and to deactivate the error minimization device when the primary error value is above the error tolerance.

3. The dynamic controller of claim 2,

wherein the tolerance is based on a function of dynamic gain at the first time and dynamic gain at the second time.

4. The dynamic controller of claim 2,

wherein the tolerance is based on a ratio of the dynamic gain at the first time and the dynamic gain at the second time.

5. The dynamic controller of claim 2,

wherein the tolerance is based on a magnitude of the dynamic gain at the first time and/or the dynamic gain at the second time.

6. The dynamic controller of claim 2,

wherein the tolerance is based on a simple average of the dynamic gain at the first time and the dynamic gain at the second time.

7. The dynamic controller of claim 2,

wherein the tolerance is based on a weighted average of the dynamic gain at the first time and the dynamic gain at the second time.

8. The dynamic controller of claim 2,

wherein the tolerance is based on a linear interpolation between the dynamic gain at the first time and the dynamic gain at the second time, inclusively.

9. A computer-accessible memory medium that stores program instructions for controlling operation of a plant or process by predicting a change in dynamic input values to the plant or process to effect a change in output of the plant or process from a current output value at a first time to a desired output value at a second time to achieve a specified objective of the dynamic controller, wherein program instructions executable by a processor to implement:

a dynamic predictive model for receiving the current input value and the desired output value and predicting a plurality of input values at different times between the first time and the second time to define a dynamic operation path of the plant or process between the current output value at the first time and the desired output value at the second time;

an error generator for comparing a predicted dynamic output value from the dynamic operation path to the desired output value and generating a primary error value as a difference therebetween for each of said times;

an error minimization device for determining a change in each input value to minimize the primary error value output by the error generator, wherein the error minimization device comprises a filter to determine operation of the error minimization device based upon accuracy of the dynamic model;

a summation device for summing the determined change in input value with an original input value, wherein the original input value comprises the input value before the determined change therein, for each time position to provide a future input value as a summed input value; and

a controller for controlling the operation of the error minimization device to operate under control of the dynamic controller to minimize the primary error value in accordance with the specified objective of the dynamic controller.

10. A method for controlling operation of a plant or process by predicting a change in dynamic input values to the plant or process to effect a change in output of the plant or process from a current output value at a first time to a desired output value at a second time to achieve a specified objective of the dynamic controller, the method comprising:

predicting a plurality of input values at different times between the first time and the second time based on the current input value and the desired output value to define a dynamic operation path of the plant or process between the current output value at the first time and the desired output value at the second time;

comparing a predicted dynamic output value from the dynamic operation path to the desired output value and generating a primary error value as a difference therebetween for each of said times;

determining a change in each input value to minimize the primary error value;

summing the determined change in input value with an original input value, wherein the original input value comprises the input value before the determined change therein, for each time position to provide a future input value as a summed input value; and

minimizing the primary error value in accordance with the specified objectives of the dynamic controller and based on accuracy of the dynamic model.

11. A dynamic controller for controlling operation of a plant or process by predicting a change in dynamic input values to the plant or process to effect a change in output of the plant or process from a current output value at a first time to a desired output value at a second time to achieve a specified objective of the dynamic controller, comprising:

a dynamic predictive model for receiving the current input value and the desired output value and predicting a plurality of input values at different times between the first time and the second time to define a dynamic operation path of the plant or process between the current output value at the first time and the desired output value at the second time;

an error generator for comparing a predicted dynamic output value from the dynamic operation path to the desired output value and generating a primary error value as a difference therebetween for each of said times;

an error minimization device for determining a change in each input value to minimize the primary error value output by the error generator, wherein the error minimization device is operable to minimize an objective function of the dynamic predictive model with respect to a constraint, and wherein the error minimization device comprises an error constraint to determine operation of the error minimization device with respect to the constraint of the objective function;

a summation device for summing the determined change in input value with an original input value, wherein the original input value comprises the input value before the determined change therein, for each time position to provide a future input value as a summed input value; and

a controller for controlling the operation of the error minimization device to operate under control of the dynamic controller to minimize the primary error value in accordance with the specified objective of the dynamic controller.

12. The dynamic controller of claim 11,

wherein the error minimization device is operable to minimize the objective function with respect to the constraint when the primary error value is above the error constraint, and wherein the minimization device is operable to minimize the objective function with respect to the constraint when the primary error value is below the error constraint.

13. The dynamic controller of claim 11,

wherein the error constraint is based on a function of dynamic gain at the first time and dynamic gain at the second time.

14. The dynamic controller of claim 11,

wherein the error constraint is based on a ratio of the dynamic gain at the first time and the dynamic gain at the second time.

15. The dynamic controller of claim 11,

wherein the error constraint is based on a magnitude of the dynamic gain at the first time and the dynamic gain at the second time.

16. The dynamic controller of claim 11,

wherein the error constraint is based on a simple average of the dynamic gain at the first time and the dynamic gain at the second time.

17. The dynamic controller of claim 11,

wherein the error constraint is based on a weighted average of the dynamic gain at the first time and the dynamic gain at the second time.

18. The dynamic controller of claim 11,

wherein the error constraint is based on a linear interpolation between the dynamic gain at the first time and the dynamic gain at the second time, inclusively.

19. A computer-accessible memory medium comprising program instructions for controlling operation of a plant or process by predicting a change in dynamic input values to the plant or process to effect a change in output of the plant or process from a current output value at a first time to a desired output value at a second time to achieve a specified objective of the dynamic controller, wherein the program instructions are executable by a processor to implement:

a dynamic predictive model for receiving the current input value and the desired output value and predicting a plurality of input values at different times between the first time and the second time to define a dynamic operation path of the plant or process between the current output value at the first time and the desired output value at the second time;

an error generator for comparing a predicted dynamic output value from the dynamic operation path to the desired output value and generating a primary error value as a difference therebetween for each of said times;

an error minimization device for determining a change in each input value to minimize the primary error value output by the error generator, wherein the error minimization device is operable to minimize an objective function of the dynamic predictive model with respect to a constraint, and wherein the error minimization device comprises an error constraint to determine operation of the error minimization device with respect to the constraint of the objective function;

a summation device for summing the determined change in input value with an original input value, wherein the original input value comprises the input value before the determined change therein, for each time position to provide a future input value as a summed input value; and

a controller for controlling the operation of the error minimization device to operate under control of the dynamic controller to minimize the primary error value in accordance with the specified objective of the dynamic controller.

20. A method for controlling operation of a plant or process by predicting a change in dynamic input values to the plant or process to effect a change in output of the plant or process from a current output value at a first time to a desired output value at a second time to achieve a specified objective of the dynamic controller, the method comprising:

predicting a plurality of input values at different times between the first time and the second time based on the current input value and the desired output value to define a dynamic operation path of the plant or process between the current output value at the first time and the desired output value at the second time;

comparing a predicted dynamic output value from the dynamic operation path to the desired output value and generating a primary error value as a difference therebetween for each of said times;

determining a change in each input value to minimize the primary error value;

determining whether the change in each input value is outside bounds of an error constraint to determine error minimization with respect to a constraint of the specified objective of the dynamic predictive model;

summing the determined change in input value with an original input value, wherein the original input value comprises the input value before the determined change therein, for each time position to provide a future input value as a summed input value; and

minimizing the primary error value in accordance with the specified objectives of the dynamic controller.