System and method for predicting future disturbances in model predictive control applications

• US 9,760,067 B2
• Filed: 09/10/2009
• Issued: 09/12/2017
• Est. Priority Date: 09/10/2009
• Status: Active Grant

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Havlena, Vladimir
• Primary Examiner(s): Lin, Jason
• Application Number:
  US12/556,811
• Publication Number:
  US 20110060424A1
• Time in Patent Office: 2,924 Days
• Field of Search: 700 29- 31, 703 7, 703 18, 703 20- 21
• CPC Class Codes: G05B 13/048 using a predictor

Abstract

A system and method for predicting future disturbance in MPC applications by segregating a transient part and a steady state value associated with the disturbance. A dynamic state space model that includes a variable disturbance prediction module can be utilized for analyzing a dynamic behavior of a physical process associated with a process model. The process model represents a dynamic behavior of the physical process being controlled and the dynamic state space model represents current deviations from the process model and future deviations over a predetermined prediction horizon. A predicted trajectory can be calculated as a response to the initial conditions estimated by a Kalman Filter for the process model extended by a disturbance model. The output of the dynamic state space model utilized for the disturbance prediction can be further provided as an estimated input to a MPC.

Claims

1. A method for predicting disturbances of a physical process, said method comprising:

analyzing the dynamic behavior of a physical process of a mineral refinery associated with a process model via a dynamic state space model thereby manipulating functions representing future disturbances;

predicting said future disturbances utilizing a disturbance prediction module associated with said dynamic state space model;

segregating a transient part and segregating a steady state value of said-disturbances associated with said process model, indicative of said dynamic behavior of said physical process associated with said mineral refinery being controlled thereto;

calculating a predicted trajectory as a response to an initial condition estimated by a Kalman Filter based on outputs of said process model and said disturbance model;

estimating a future deviation using said disturbance prediction module;

generating an estimated disturbance prediction output of said disturbance model based on the predicted trajectory;

inputting said estimated disturbance prediction output into a separate external model-based predictive controller wherein said model based predictive is implemented as an online optimization and uses equivalent lookup tables computed according to a hybrid multi-parametric algorithm;

providing manipulated variables from said model-based predictive controller to a process dynamics module;

wherein said manipulated variables are determined based on said estimated disturbance prediction output;

wherein said output of said process dynamic module is determined based on an actual disturbance variable and said manipulated variable;

wherein said output of said process model is determined based on said estimated disturbance prediction output and said manipulated variables;

determining process predictions by said model-based predictive controller based on said disturbance prediction output and thereafter providing an optimized output as a controlled input to said process model; and

controlling said physical process based on said optimized output as a controlled input.

2. The method of claim 1 further comprising:

inputting disturbance variables from said disturbance model into said process model wherein said disturbance model comprises a dynamic response of observed disturbances transformed to a dynamic state space model.

3. The method of claim 2 further comprising:

estimating said future deviation as a function of a past deviation and a current deviation from said process model by said disturbance prediction module; and

inputting said disturbance variables into said model predictive controller.

4. The method of claim 3 wherein said disturbance model provides data indicative of said current deviation from said process model and said future deviation over a predetermined prediction horizon.

5. The method of claim 4 further comprising transmitting an input signal indicative of a measured process parameter to said model-based predictive controller.

6. The method of claim 5 wherein said disturbance prediction module performs an independent process control decision to determine manipulated parameter values in response to said input signal.

7. The method of claim 6 further comprising transmitting a command signal from said model-based predictive controller to said process model external to said model-based predictive controller which include said manipulated parameter value to govern said process model.

8. A system for predicting disturbances, said system comprising:

a processor;

a data bus coupled to said processor; and

a computer-usable medium embodying computer code, said computer-usable medium being coupled to said data bus, said computer program code comprising instructions executable by said processor and configured for;

analyzing the dynamic behavior of a physical process of a mineral refinery associated with a process model via a dynamic state space model thereby manipulating functions representing future disturbances;

predicting said future disturbances utilizing a disturbance prediction module associated with said dynamic state space model;

segregating a transient part and segregating a steady state value of said disturbances associated with said process model, indicative of said dynamic behavior of said physical process associated with said mineral refinery being controlled thereto;

calculating a predicted trajectory as a response to an initial condition estimated by a Kalman Filter based on outputs of said process model and said disturbance model;

estimating a future deviation using sad disturbance prediction module;

generating an estimated disturbance prediction output of said disturbance model based on the predicted trajectory;

inputting said estimated disturbance prediction output into a separate external model predictive controller wherein said model based predictive controller implemented as an online optimization and using equivalent lookup tables computed according to a hybrid multi-parametric algorithm;

providing manipulated variables from said model-based predictive controller to a process dynamics module;

wherein said manipulated variables are determined based on said estimated disturbance prediction output;

wherein said output of said process dynamic module is determined based on an actual disturbance variable and said manipulated variable;

wherein said output of said process model is determined based on said estimated disturbance prediction output and said manipulated variables;

determining process predictions by said model-based predictive controller based on said disturbance prediction output and thereafter providing an optimized output as a controlled input to said process model, and

controlling said physical process based on said optimized output as a controlled input.

9. The system of claim 8 further comprising:

inputting disturbance variables from said disturbance model into said process model wherein said disturbance model comprises a dynamic response of observed disturbances transformed to a dynamic state space model.

10. The system of claim 9 wherein said instructions are further configured for:

estimating said future deviation as a function of a past deviation and a current deviation from said process model by said disturbance prediction module; and

inputting said disturbance variables into said model predictive controller.

11. The system of claim 10 wherein said instructions are further configured for generating from said disturbance model, data indicative of said current deviation from said process model and said future deviation over a predetermined prediction horizon.

12. The system of claim 11 wherein said instructions are further configured for transmitting an input signal indicative of a measured process parameter to said model predictive controller.

13. The system of claim 12 wherein said instructions are further configured for performing via said disturbance prediction module, an independent process control decision to determine manipulated parameter values in response to said input signal.

14. The system of claim 13 wherein said instructions are further configured for transmitting a command signal from said model predictive controller to said process model external to said model-based predictive controller which includes said manipulated parameter value to govern said process model.

15. A non-transitory computer-readable medium for authorizing access to a secure location, said computer-readable medium embodying a computer program code, said computer program code comprising computer executable instructions configured for;

analyzing the dynamic behavior of a physical process of a mineral refinery associated with a process model via a dynamic state space model thereby manipulating functions representing future disturbances;

predicting said future disturbances utilizing a disturbance prediction module associated with said dynamic state space model;

segregating a transient part and segregating a steady state value of said-disturbances associated with said process model, indicative of said dynamic behavior of said physical process associated with said mineral refinery being controlled thereto;

calculating a predicted trajectory as a response to an initial condition estimated by a Kalman Filter based on outputs of said process model and said disturbance model;

estimating a future deviation using said disturbance prediction module;

generating an estimated disturbance prediction output of said disturbance model based on the predicted trajectory;

inputting said estimated disturbance prediction output into a separate external model predictive controller wherein said model predictive controller is implemented as an online optimization and uses equivalent lookup tables computed according to a hybrid multi-parametric algorithm;

providing manipulated variables from said model-based predictive controller to a process dynamics module;

wherein said manipulated variables are determined based on said estimated disturbance prediction output;

wherein said output of said process dynamic module is determined based on an actual disturbance variable and said manipulated variable;

wherein said output of said process model is determined based on said estimated disturbance prediction output and said manipulated variables;

determining process predictions by said model-based predictive controller based on said disturbance prediction output and thereafter providing an optimized output as a controlled input to said process model; and

controlling said physical process based on said optimized output as a controlled input.

16. The non-transitory computer-readable medium of claim 15 further comprising:

inputting disturbance, variables from said disturbance model into said process mode wherein said disturbance model;

comprises a dynamic response of observed disturbances transformed to a dynamic state space model.

17. The computer-usable medium of claim 16 wherein said embodied computer program code further comprises computer executable instructions configured for estimating said future deviation as a function of a past deviation and a current deviation from said process model by said disturbance prediction module;

and

inputting said disturbance variables into said model predictive controller.

18. The non-transitory computer-readable medium of claim 17 wherein said embodied computer program code further comprises computer executable instructions configured for generating via said disturbance model, data indicative of said current deviation from said process model and said future deviation over a predetermined prediction horizon.

19. The non-transitory computer-readable medium of claim 18 wherein said embodied computer program code further comprises computer executable instructions configured for performing via said disturbance prediction module, an independent process control decision to determine manipulated parameter values in response to said input signal.

20. The non-transitory computer-readable medium of claim 19 wherein said embodied computer program code further comprises computer executable instructions configured for:

transmitting an input signal indicative of a measured process parameter to said model predictive controller; and

transmitting a command signal from said model predictive controller to said process model external to said model-based predictive controller which includes said manipulated parameter value to govern said process model.