Coordination in multilayer process control and optimization schemes

• US 20040093124A1
• Filed: 11/12/2002
• Published: 05/13/2004
• Est. Priority Date: 11/12/2002
• Status: Active Grant

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Havlena, Vladimir
• Granted Patent: US 6,832,134 B2
• Time in Patent Office: Days
• US Class Current: 700/291
• CPC Class Codes: F22B 35/008 Control systems for two or more steam generators F22D5/36 takes precedence
• CPC Class Codes: G05B 13/026 using a predictor

Abstract

Loads 1, 2, . . . , N are controlled by predicting a total energy requirement for the loads 1, 2, . . . , N at prediction points k=0, 1, 2, . . . , K, by allocating the total energy requirement to the loads 1, 2, . . . , N at prediction points k=0, 1, 2, . . . , K, by determining a dynamic energy demand requirement for each of the loads 1, 2, . . . , N at prediction points k=0, 1, 2, . . . , K based on the allocated energy requirements, and by controlling the loads 1, 2, . . . , N based on the dynamic energy demand requirements.

Claims

1. A process control system for controlling a load comprising:

a predictive controller that predicts an energy requirement for the load at prediction points k=0, 1, 2, . . . , K based on a steady state target energy requirement for the load; and

,

a real time optimizer that determines an optimized dynamic energy demand requirement for the load at the prediction points k=0, 1, 2, . . . , K based on the predicted energy requirement and that controls the load based on the dynamic energy demand requirement.

2. The process control system of claim 1 wherein the real time optimizer is arranged to constrain the dynamic energy demand requirement to a limit and controls the load based on the constrained dynamic energy demand requirement.

3. The process control system of claim 2 wherein the limit comprises a maximum limit.

4. The process control system of claim 2 wherein the limit comprises a minimum limit.

5. The process control system of claim 2 wherein the limit comprises a minimum limit and a maximum limit.

6. The process control system of claim 2 wherein the real time optimizer determines Δ

F^dyn according to the following equation;

Δ

F^dyn=DF^dyn−

F^act

wherein D is a (K+1) by (K+1) matrix given by the following equation;

7. The process control system of claim 6 wherein the limit for Δ

F^dyn comprises a maximum limit.

8. The process control system of claim 6 wherein the limit for Δ

F^dyn comprises a minimum limit.

9. The process control system of claim 6 wherein the limit for Δ

F^dyn comprises a minimum limit and a maximum limit.

10. The process control system of claim 1 wherein the predicted energy requirement is designated F^unconstr, wherein the dynamic energy demand requirement is designated F^dyn, and wherein F^dyn is determined by minimizing a quadratic function based on a difference between F^unconstr and F^dyn.

11. The process control system of claim 10 wherein F^unconstr and F^dyn each have dimensions K+1.

12. The process control system of claim 10 wherein the real time controller is arranged to constrain F^dyn to a limit and to control the load based on the constrained F^dyn.

13. The process control system of claim 12 wherein the limit comprises a maximum limit.

14. The process control system of claim 12 wherein the limit comprises a minimum limit.

15. The process control system of claim 12 wherein the limit comprises a minimum limit and a maximum limit.

16. A process control method for controlling loads 1, 2, . . . , N comprising:

predicting, in a predictive controller, an energy requirement for each of the loads 1, 2, . . . , N at prediction points k=0, 1, 2, . . . , K based on a steady state target allocation for each of the loads 1, 2, . . . , N;

determining, in a real time optimizer, a dynamic energy demand requirement for each of the loads 1, 2, . . . , N at prediction points k=0, 1, 2, . . . , K based on the predicted energy requirements; and

,

controlling the loads 1, 2, . . . , N based on the dynamic energy demand requirements.

17. The process control method of claim 16 wherein the controlling of the loads based on the dynamic energy demand requirements comprises:

constraining each of the dynamic energy demand requirements to a corresponding limit; and

,

controlling the loads 1, 2, . . . , N based on the corresponding constrained dynamic energy demand requirements.

18. The process control method of claim 17 wherein the corresponding limits comprise corresponding maximum limits.

19. The process control method of claim 17 wherein the corresponding limits comprise corresponding minimum limits.

20. The process control method of claim 17 wherein the corresponding limits comprise corresponding minimum limits and corresponding maximum limits.

21. The process control method of claim 17 further comprising;

determining Δ

F^dyn according to the following equation;

Δ

F_i^dyn=DF_i^dyn−

F_i^act

wherein D is a (K+1) by (K+1) matrix given by the following equation;

$D=\left(\begin{array}{cccc}1& \text{\hspace{1em}<br><br>}& \text{\hspace{1em}<br><br>}& \text{\hspace{1em}<br><br>}\\ -1& 1& \text{\hspace{1em}<br><br>}& \text{\hspace{1em}<br><br>}\\ \text{\hspace{1em}<br><br>}& \mathrm{\ddots <br><br>}& \mathrm{\ddots <br><br>}& \text{\hspace{1em}<br><br>}\\ \text{\hspace{1em}<br><br>}& \text{\hspace{1em}<br><br>}& -1& 1\end{array}\right)$

and wherein F_i^act is a (K+1) dimensional vector having an actual energy consumption of a corresponding one of the loads in the first component and zeros elsewhere as given by the following equation;

$F_i^{\mathrm{act}}=\left(\begin{array}{c}{F}_i^{\mathrm{act}}\\ 0\\ \mathrm{⋮<br><br>}\\ 0\end{array}\right)$

wherein i=1, 2, . . . , N designates the loads, wherein F_i^dyn designate the dynamic energy demand requirements of the loads 1, 2, . . . , N, and wherein Δ

F_i^dyn are constrained to corresponding limits.

22. The process control method of claim 21 wherein the corresponding limits for Δ

F_i^dyn comprise corresponding maximum limits.

23. The process control method of claim 21 wherein the corresponding limits for AF^dyn comprise corresponding minimum limits.

24. The process control method of claim 21 wherein the corresponding limits for Δ

F^dyn comprise corresponding minimum limits and corresponding maximum limits.

25. The process control method of claim 16 wherein the predicted energy requirement for each of the loads 1, 2, . . . , N is designated F^unconstr, wherein the dynamic energy demand requirement for each of the loads 1, 2, . . . , N is designated F^dyn, wherein F^dyn is determined by minimizing the following quadratic function:

f(F,z)=∥

F^dyn−

F^unconstr∥

²Q(1)+∥

z∥

²Q(2)

with respect to variables F^dyn and z, wherein F^dyn and F^unconstr are vectors each having a dimension 2N(K+1), wherein Q(1) and Q(2) are each square N(K+1) by N(K+1) norm matrices given by the following equations;

26. The process control method of claim 25 wherein F^unconstr and F^dyn each have dimensions K+1.

27. The process control method of claim 25 wherein the controlling of the loads 1, 2, . . . , N based on the dynamic energy demand requirements comprises:

constraining F_i^dyn to corresponding limits; and

,

controlling the load based on the constrained F_I^dyn.

28. The process control method of claim 27 wherein the corresponding limits comprise corresponding maximum limits.

29. The process control method of claim 27 wherein the corresponding limits comprise corresponding minimum limits.

30. The process control method of claim 27 wherein the corresponding limits comprise corresponding minimum limits and maximum limits.

31. A process control method for controlling loads 1, 2, . . . , N comprising:

predicting, by way of a predictive controller, a total energy requirement for the loads 1, 2, . . . , N at prediction points k=0, 1, 2, . . . , K;

allocating, by way of a real time optimizer, the total energy requirement to the loads 1, 2, . . . , N at the prediction points k=0, 1, 2, . . . , K based on a steady state target for each of the loads 1, 2, . . . , N;

determining, by way of the real time optimizer, a dynamic energy demand requirement for each of the loads 1, 2, . . . , N at the prediction points k=0, 1, 2, ., K based on the allocated energy requirements; and

, controlling the loads 1, 2, . . . , N based on the dynamic energy demand requirements.

32. The process control method of claim 31 wherein the allocating of the total energy requirement to the loads 1, 2, . . . , N comprises:

determining a change Δ

F^tot(k) in the total energy requirement according to the following equation;

$\mathrm{\Delta <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>F^{\mathrm{tot}}<br><br>\left(k\right)=F^{\mathrm{tot}}<br><br>\left(k\right)-\underset{i=1}{\overset{N}{\sum <br><br>}}<br><br>F_i^{\mathrm{targ}}<br><br>\left(k\right)$

for each of the loads i=1, 2, . . . , N and for each of the prediction points k=0, 1, . . . , K; and

, allocating the total energy requirement to the loads 1, 2, . . . , N at prediction points k=0, 1, 2, . . . , K according to the following equation;

F_i^allocated(k)=F_i^{t arg}(k)+w_i^dyn(k)Δ

F^tot(k)

wherein F_i^arg(k) are the steady state targets for the loads 1, 2, . . . , N, and wherein w_i^dyn(k) are weights determining the allocation of the total energy requirement.

33. The process control method of claim 32 wherein w_i^dyn(k) are set by an operator.

34. The process control method of claim 32 wherein w_i^dyn(k) are set by an economic load allocation module.

35. The process control method of claim 32 wherein F_i^{t arg}(k) are set by an operator.

36. The process control method of claim 32 wherein F_i^{t arg}(k) are set by an economic load allocation module.

37. The process control method of claim 31 wherein the controlling of the loads 1, 2, . . . , N based on the dynamic energy demand requirements comprises:

constraining each of the dynamic energy demand requirements to a corresponding limit; and

,

controlling the loads 1, 2, . . . , N based on the corresponding constrained dynamic energy demand requirements.

38. The process control method of claim 37 wherein the corresponding limits comprise corresponding maximum limits.

39. The process control method of claim 37 wherein the corresponding limits comprise corresponding minimum limits.

40. The process control method of claim 37 wherein the corresponding limits comprise corresponding minimum limits and corresponding maximum limits.

41. The process control method of claim 31 wherein the total energy requirements allocated to the loads 1, 2, . . . , N at prediction points k=0, 1, 2, . . . , K are designated F^dyn, and wherein the process control method further comprises;

determining a change Δ

F^dyn in the total energy requirements allocated to the loads 1, 2, . . . , N according to the following equation;

Δ

F_i^dyn=DF_i^dyn−

F^act

wherein D is a (K+1) by (K+1) matrix given by the following equation;

$D=\left(\begin{array}{cccc}1& \text{\hspace{1em}<br><br>}& \text{\hspace{1em}<br><br>}& \text{\hspace{1em}<br><br>}\\ -1& 1& \text{\hspace{1em}<br><br>}& \text{\hspace{1em}<br><br>}\\ \text{\hspace{1em}<br><br>}& \mathrm{\ddots <br><br>}& \mathrm{\ddots <br><br>}& \text{\hspace{1em}<br><br>}\\ \text{\hspace{1em}<br><br>}& \text{\hspace{1em}<br><br>}& -1& 1\end{array}\right)$

and wherein F_i^act is a (K+1) dimensional vector having an actual energy consumption of the corresponding loads in the first component and zeros elsewhere as given by the following equation;

$F_i^{\mathrm{act}}=\left(\begin{array}{c}{F}_i^{\mathrm{act}}\\ 0\\ \mathrm{⋮<br><br>}\\ 0\end{array}\right)$

wherein i=1, 2, . . . , N designates the loads, and wherein Δ

F^dyn is constrained to a corresponding limit.

42. The process control method of claim 41 wherein the corresponding limits for Δ

F^dyn comprise corresponding maximum limits.

43. The process control method of claim 41 wherein the corresponding limits for Δ

F_i^dyn comprise corresponding minimum limits.

44. The process control method of claim 39 wherein the corresponding limits for Δ

F_i^dyn comprise corresponding minimum limits and corresponding maximum limits.

45. The process control method of claim 31 wherein the total energy requirement allocated to each of the loads 1, 2, . . . , N is designated F_i^unconstr, wherein the dynamic energy demand requirement for each of the loads 1, 2, . . . , N is designated F_i^dyn, and wherein F^dyn is determined by minimizing the following quadratic function:

f(F,z)=∥

F^dyn−

F^unconstr∥

²Q(1)+∥

z∥

²Q(2)

with respect to variables F^dyn and z, wherein F^dyn and F^unconstr are vectors each having a dimension 2N(K+1), wherein Q(1) and Q(2) are each square N(K+1) by N(K+1) norm matrices given by the following equations;

$Q^{\left(j\right)}=\left(\begin{array}{ccc}{q}_1^{\left(j\right)}& \text{\hspace{1em}<br><br>}& \text{\hspace{1em}<br><br>}\\ \text{\hspace{1em}<br><br>}& \mathrm{⋰<br><br>}& \text{\hspace{1em}<br><br>}\\ \text{\hspace{1em}<br><br>}& \text{\hspace{1em}<br><br>}& q_N^{\left(j\right)}\end{array}\right)$q_i(j)=w_i^(j)I

wherein j 1, 2, wherein I is a (K+1) by (K+1) unity matrix, wherein w_i^(j) and w_i⁽²⁾ are penalty weights for i=1, . . . , N, and wherein z is a penalty vector having a 2N(K+1) dimension.

46. The process control method of claim 45 wherein F^unconstr and F^dyn each have dimensions K+1.

47. The process control method of claim 45 wherein the controlling of the loads 1, 2, . . . , N based on the dynamic energy demand requirements comprises:

constraining F_i^dyn to corresponding limits; and

,

controlling the load based on the constrained F_i^dyn.

48. The process control method of claim 47 wherein the corresponding limits comprise corresponding maximum limits.

49. The process control method of claim 47 wherein the corresponding limits comprise corresponding minimum limits.

50. The process control method of claim 47 wherein the corresponding limits comprise corresponding minimum limits and maximum limits.