IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0051814
(2005-02-04)
|
등록번호 |
US-7400933
(2008-07-15)
|
발명자
/ 주소 |
- Rawlings,James B.
- Pannocchia,Gabriele
- Laachi,Nabil
|
출원인 / 주소 |
- Wisconsin Alumni Research Foundation
|
대리인 / 주소 |
|
인용정보 |
피인용 횟수 :
22 인용 특허 :
3 |
초록
▼
A method of predictive control for a single input, single output (SISO) system, including modeling the SISO system with model factors, detecting output from the SISO system, estimating a filtered disturbance from the output, determining a steady state target state from the filtered disturbance and a
A method of predictive control for a single input, single output (SISO) system, including modeling the SISO system with model factors, detecting output from the SISO system, estimating a filtered disturbance from the output, determining a steady state target state from the filtered disturbance and a steady state target output, populating a dynamic optimization solution table using the model factors and a main tuning parameter, and determining an optimum input from the dynamic optimization solution table. Determining an optimum input includes determining a time varying parameter, determining a potential optimum input from the time varying parameter, and checking whether the potential optimum input is the optimum input.
대표청구항
▼
We claim: 1. A method of predictive control for a single input, single output (SISO) system generating an optimum input for controlling the SISO system, comprising: modeling the SISO system with model factors; detecting output from the SISO system; estimating a filtered disturbance from the output;
We claim: 1. A method of predictive control for a single input, single output (SISO) system generating an optimum input for controlling the SISO system, comprising: modeling the SISO system with model factors; detecting output from the SISO system; estimating a filtered disturbance from the output; determining a steady state target state from the filtered disturbance and a steady state target output; populating a dynamic optimization solution table using the model factors and a main tuning parameter; determining the optimum input from the dynamic optimization solution table; and controlling the SISO system in response to the optimum input; wherein the estimating a filtered disturbance from the output comprises: determining filter gains from the model factors, an estimator tuning parameter, and an output noise covariance parameter; and estimating the filtered disturbance description="In-line Formulae" end="lead"{circumflex over (d)}k↑k={circumflex over (d)}k|k-1+Ld(yk-C {circumflex over (x)}k|k-1 (5).description="In-line Formulae" end="tail" 2. The method of claim 1 wherein the determining filter gains comprises determining the filter gains offline. 3. The method of claim 1 wherein the estimating the filtered disturbance and a filtered state comprises estimating the filtered disturbance online. 4. The method of claim 1 wherein the determining a steady state target state from the filtered disturbance and a steady state target output comprises: determining a constrained target matrix having elements M11, M12, M21, M22, from the model factors; determining constrained target gains G1, G2, and G3 from the model factors and a steady state factor; determining an unconstrained target input from the filtered disturbance and the steady state target output from description="In-line Formulae" end="lead"ū*=M21{circumflex over (d)}k|k+M22 y. (34a)description="In-line Formulae" end="tail" determining whether the unconstrained target input is on or within input constraints; when the unconstrained target input is on or within input constraints, setting a steady state target input equal to the unconstrained target input and determining the steady state target state description="In-line Formulae" end="lead" xk=M11{circumflex over (d)}k|k+M12 y (34c)description="In-line Formulae" end="tail" when the unconstrained target input is outside the input constraints, setting the steady state target input equal to a minimum input constraint when the steady state target input is less than the minimum input constraint, setting the steady state target input equal to a maximum input constraint when the steady state target input is greater than the maximum input constraint, and determining the steady state target state description="In-line Formulae" end="lead" xk=G1{circumflex over (d)}k|k+G2 y+G3ūk (34d).description="In-line Formulae" end="tail" 5. The method of claim 4 wherein the determining a constrained target matrix and determining constrained target gains comprises determining a constrained target matrix offline and determining constrained target gains offline. 6. The method of claim 4 wherein the determining an unconstrained target input comprises determining an unconstrained target input online. 7. The method of claim 1 wherein the populating a dynamic optimization solution table comprises populating a dynamic optimization solution table offline. 8. The method of claim 1 wherein the dynamic optimization solution table is ordered. 9. The method of claim 1 wherein the determining the optimum input from the dynamic optimization solution table comprises: determining a time varying parameter; determining a potential optimum input from the time varying parameter; and checking whether the potential optimum input is the optimum input. 10. The method of claim 9 further comprising determining a next potential optimum input from the time varying parameter when the potential optimum input is not the optimum input. 11. The method of claim 9 wherein the determining a time varying parameter comprises: determining whether the steady state target output is reachable; when the steady state target output is reachable, determining an intermediate time varying parameter description="In-line Formulae" end="lead"{tilde over (c)}=Tw0+Tv-1 (14)description="In-line Formulae" end="tail" when the steady state target output is unreachable, determining the intermediate time varying parameter description="In-line Formulae" end="lead"{tilde over (c)}=Tw0+T+Tv-1, (23)description="In-line Formulae" end="tail" and; determining the time varying parameter description="In-line Formulae" end="lead"c={tilde over (c)}-Hūk (36c).description="In-line Formulae" end="tail" 12. The method of claim 9 wherein the dynamic optimization solution table comprises optimal input solution offset Bi and optimal input solution gain Ki, and the determining a potential optimum input comprises determining description="In-line Formulae" end="lead"ui=Kic+Bi. (52a).description="In-line Formulae" end="tail" 13. The method of claim 9 wherein the checking whether the potential optimum input is the optimum input comprises determining whether the potential optimum input meets a constraint condition. 14. The method of claim 9 wherein the checking whether the potential optimum input is the optimum input comprises determining whether any element of the active constraint Lagrange multiplier vector is zero or negative. 15. A system of predictive control for a single input, single output (SISO) system generating an optimum input for controlling the SISO system, comprising: means for modeling the SISO system with model factors; means for detecting output from the SISO system; means for estimating a filtered disturbance from the output; means for determining a steady state target state from the filtered disturbance and a steady state target output; means for populating a dynamic optimization solution table using the model factors and a main tuning parameter; means for determining the optimum input from the dynamic optimization solution table; and means for controlling the SISO system in response to the optimum input; wherein the means for estimating a filtered disturbance from the output comprises: means for determining filter gains from the model factors, an estimator tuning parameter, and an output noise covariance parameter; and means for estimating the filtered disturbance description="In-line Formulae" end="lead"{circumflex over (d)}k|k={circumflex over (d)}k|k-1+Ld(yk-C {circumflex over (x)}k|k-1) (5).description="In-line Formulae" end="tail" 16. The system of claim 15 wherein the means for determining filter gains comprises means for determining the filter gains offline. 17. The system of claim 15 wherein the means for estimating the filtered disturbance and a filtered state comprises means for estimating the filtered disturbance online. 18. The system of claim 15 wherein the means for determining a steady state target state from the filtered disturbance and a steady state target output comprises: means for determining a constrained target matrix having elements M11, M12, M21, M22, from the model factors; means for determining constrained target gains G1, G2, and G3 from the model factors and a steady state factor; means for determining an unconstrained target input from the filtered disturbance and the steady state target output from description="In-line Formulae" end="lead"ū*=M21{circumflex over (d)}k|k+M22 y. (34a)description="In-line Formulae" end="tail" means for determining whether the unconstrained target input is on or within input constraints; means for setting a steady state target input equal to the unconstrained target input and means for determining the steady state target state, when the unconstrained target input is on or within input constraints, from description="In-line Formulae" end="lead" xk=M11{circumflex over (d)}k|k+M12 y (34c)description="In-line Formulae" end="tail" means for setting the steady state target input equal to a minimum input constraint when the steady state target input is less than the minimum input constraint, means for setting the steady state target input equal to a maximum input constraint when the steady state target input is greater than the maximum input constraint, and means for determining the steady state target state, when the unconstrained target input is outside the input constraints, from description="In-line Formulae" end="lead" xk=G1{circumflex over (d)}k|k+G2 y+G3ūk (34d).description="In-line Formulae" end="tail" 19. The system of claim 18 wherein: the means for determining a constrained target matrix comprises means for determining a constrained target matrix offline; and the means for determining constrained target gains comprises means for determining constrained target gains offline. 20. The system of claim 18 wherein the means for determining an unconstrained target input comprises means for determining an unconstrained target input online. 21. The system of claim 15 wherein the means for populating a dynamic optimization solution table comprises means for populating a dynamic optimization solution table offline. 22. The system of claim 15 wherein the means for populating a dynamic optimization solution table further comprises means for ordering the dynamic optimization solution table. 23. The system of claim 15 wherein the means for determining the optimum input from the dynamic optimization solution table comprises: means for determining a time varying parameter; means for determining a potential optimum input from the time varying parameter; and means for checking whether the potential optimum input is the optimum input. 24. The system of claim 23 further comprising means for determining a next potential optimum input from the time varying parameter when the potential optimum input is not the optimum input. 25. The system of claim 23 wherein the means for determining a time varying parameter comprises: means for determining whether the steady state target output is reachable; means for determining an intermediate time varying parameter, when the steady state target output is reachable, from description="In-line Formulae" end="lead"{tilde over (c)}=Tw0+Tv-1 (14)description="In-line Formulae" end="tail" means for determining the intermediate time varying parameter, when the steady state target output is unreachable, from description="In-line Formulae" end="lead"{tilde over (c)}=Tw0+T+Tv-1, (23)description="In-line Formulae" end="tail" and; means for determining the time varying parameter description="In-line Formulae" end="lead"c={tilde over (c)}-Hūk (36c).description="In-line Formulae" end="tail" 26. The system of claim 23 wherein the dynamic optimization solution table comprises optimal input solution offset Bi and optimal input solution gain Ki, and the means for determining a potential optimum input comprises means for determining description="In-line Formulae" end="lead"ui=Kic+Bi. (52a).description="In-line Formulae" end="tail" 27. The system of claim 23 wherein the means for checking whether the potential optimum input is the optimum input comprises means for determining whether the potential optimum input meets a constraint condition. 28. The system of claim 23 wherein the means for checking whether the potential optimum input is the optimum input comprises means for determining whether any element of the active constraint Lagrange multiplier vector is zero or negative. 29. A computer readable medium storing a computer program for predictive control of a single input, single output (SISO) system generating an optimum input for controlling the SISO system, comprising: computer readable code for modeling the SISO system with model factors; computer readable code for detecting output from the SISO system; computer readable code for estimating a filtered disturbance from the output; computer readable code for determining a steady state target state from the filtered disturbance and a steady state target output; computer readable code for populating a dynamic optimization solution table using the model factors and a main tuning parameter; computer readable code for determining the optimum input from the dynamic optimization solution table; and computer readable code for controlling the SISO system in response to the optimum input; wherein the computer readable code for estimating a filtered disturbance from the output comprises: computer readable code for determining filter gains from the model factors, an estimator tuning parameter, and an output noise covariance parameter; and computer readable code for estimating the filtered disturbance description="In-line Formulae" end="lead"{circumflex over (d)}k|k={circumflex over (d)}k|k-1+Ld(yk-C {circumflex over (x)}k|k-1) (5).description="In-line Formulae" end="tail" 30. The computer readable medium of claim 29 wherein the computer readable code for determining filter gains comprises computer readable code for determining the filter gains offline. 31. The computer readable medium of claim 29 wherein the computer readable code for estimating the filtered disturbance and a filtered state comprises computer readable code for estimating the filtered disturbance online. 32. The computer readable medium of claim 29 wherein the computer readable code for determining a steady state target state from the filtered disturbance and a steady state target output comprises: computer readable code for determining a constrained target matrix having elements M11, M12, M21, M22, from the model factors; computer readable code for determining constrained target gains G1, G2, and G3 from the model factors and a steady state factor; computer readable code for determining an unconstrained target input from the filtered disturbance and the steady state target output from description="In-line Formulae" end="lead"ū*=M21{circumflex over (d)}k|k+M22 y. (34a)description="In-line Formulae" end="tail" computer readable code for determining whether the unconstrained target input is on or within input constraints; computer readable code for setting a steady state target input equal to the unconstrained target input and computer readable code for determining the steady state target state, when the unconstrained target input is on or within input constraints, from description="In-line Formulae" end="lead" xk=M11{circumflex over (d)}k|k+M12 y (34c)description="In-line Formulae" end="tail" computer readable code for setting the steady state target input equal to a minimum input constraint when the steady state target input is less than the minimum input constraint, computer readable code for setting the steady state target input equal to a maximum input constraint when the steady state target input is greater than the maximum input constraint, and computer readable code for determining the steady state target state, when the unconstrained target input is outside the input constraints, from description="In-line Formulae" end="lead" xk=G1{circumflex over (d)}k|k+G2 y+G3ūk (34d).description="In-line Formulae" end="tail" 33. The computer readable medium of claim 32 wherein: the computer readable code for determining a constrained target matrix comprises computer readable code for determining a constrained target matrix offline; and the computer readable code for determining constrained target gains comprises computer readable code for determining constrained target gains offline. 34. The computer readable medium of claim 32 wherein the computer readable code for determining an unconstrained target input comprises computer readable code for determining an unconstrained target input online. 35. The computer readable medium of claim 29 wherein the computer readable code for populating a dynamic optimization solution table comprises computer readable code for populating a dynamic optimization solution table offline. 36. The computer readable medium of claim 29 wherein the computer readable code for populating a dynamic optimization solution table further comprises computer readable code for ordering the dynamic optimization solution table. 37. The computer readable medium of claim 29 wherein the computer readable code for determining the optimum input from the dynamic optimization solution table comprises: computer readable code for determining a time varying parameter; computer readable code for determining a potential optimum input from the time varying parameter; and computer readable code for checking whether the potential optimum input is the optimum input. 38. The computer readable medium of claim 37 further comprising computer readable code for determining a next potential optimum input from the time varying parameter when the potential optimum input is not the optimum input. 39. The computer readable medium of claim 37 wherein the computer readable code for determining a time varying parameter comprises: computer readable code for determining whether the steady state target output is reachable; computer readable code for determining an intermediate time varying parameter, when the steady state target output is reachable, from description="In-line Formulae" end="lead"{tilde over (c)}=Tw0+Tv-1 (14)description="In-line Formulae" end="tail" computer readable code for determining the intermediate time varying parameter, when the steady state target output is unreachable, from description="In-line Formulae" end="lead"{tilde over (c)}=Tw0+T+Tv-1, (23)description="In-line Formulae" end="tail" and; computer readable code for determining the time varying parameter description="In-line Formulae" end="lead"c={tilde over (c)}-Hūk (36c).description="In-line Formulae" end="tail" 40. The computer readable medium of claim 37 wherein the dynamic optimization solution table comprises optimal input solution offset Bi and optimal input solution gain Ki, and the computer readable code for determining a potential optimum input comprises computer readable code for determining description="In-line Formulae" end="lead"ui=Kic+Bi, (52a).description="In-line Formulae" end="tail" 41. The computer readable medium of claim 37 wherein the computer readable code for checking whether the potential optimum input is the optimum input comprises computer readable code for determining whether the potential optimum input meets a constraint condition. 42. The computer readable medium of claim 37 wherein the computer readable code for checking whether the potential optimum input is the optimum input comprises computer readable code for determining whether any element of the active constraint Lagrange multiplier vector is zero or negative. 43. A method of predictive control for a single input, single output (SISO) system generating an optimum input for controlling the SISO system, comprising: modeling the SISO system with model factors; detecting output from the SISO system; estimating a filtered disturbance from the output; determining a steady state target state from the filtered disturbance and a steady state target output; populating a dynamic optimization solution table using the model factors and a main tuning parameter; determining the optimum input from the dynamic optimization solution table; and controlling the SISO system in response to the optimum input; wherein the determining a steady state target state from the filtered disturbance and a steady state target output comprises: determining a constrained target matrix having elements M11, M12, M21, M22, from the model factors; determining constrained target gains G1, G2, and G3 from the model factors and a steady state factor; determining an unconstrained target input from the filtered disturbance and the steady state target output from description="In-line Formulae" end="lead"ū*=M21{circumflex over (d)}k|k+M22 y. (34a)description="In-line Formulae" end="tail" determining whether the unconstrained target input is on or within input constraints; when the unconstrained target input is on or within input constraints, setting a steady state target input equal to the unconstrained target input and determining the steady state target state description="In-line Formulae" end="lead" xk=M11{circumflex over (d)}k|k+M12 y (34c)description="In-line Formulae" end="tail" when the unconstrained target input is outside the input constraints, setting the steady state target input equal to a minimum input constraint when the steady state target input is less than the minimum input constraint, setting the steady state target input equal to a maximum input constraint when the steady state target input is greater than the maximum input constraint, and determining the steady state target state description="In-line Formulae" end="lead" xk=G1{circumflex over (d)}k|k+G2 y+G3ūk (34d).description="In-line Formulae" end="tail" 44. The method of claim 43 wherein the determining a constrained target matrix and determining constrained target gains comprises determining a constrained target matrix offline and determining constrained target gains offline. 45. The method of claim 43 wherein the determining an unconstrained target input comprises determining an unconstrained target input online. 46. A method of predictive control for a single input, single output (SISO) system generating an optimum input for controlling the SISO system, comprising: modeling the SISO system with model factors; detecting output from the SISO system; estimating a filtered disturbance from the output; determining a steady state target state from the filtered disturbance and a steady state target output; populating a dynamic optimization solution table using the model factors and a main tuning parameter; determining the optimum input from the dynamic optimization solution table; and controlling the SISO system in response to the optimum input; wherein the determining the optimum input from the dynamic optimization solution table comprises: determining a time varying parameter; determining a potential optimum input from the time varying parameter; and checking whether the potential optimum input is the optimum input; and wherein the determining a time varying parameter comprises: determining whether the steady state target output is reachable; when the steady state target output is reachable, determining an intermediate time varying parameter description="In-line Formulae" end="lead"{tilde over (c)}=Tw0+Tv-1 (14);description="In-line Formulae" end="tail" when the steady state target output is unreachable, determining the intermediate time varying parameter description="In-line Formulae" end="lead"{tilde over (c)}=Two+T+Tv-1, (23)description="In-line Formulae" end="tail" and; determining the time varying parameter description="In-line Formulae" end="lead"c= c-Hūk (36c).description="In-line Formulae" end="tail" 47. A method of predictive control for a single input, single output (SISO) system generating an optimum input for controlling the SISO system, comprising: modeling the SISO system with model factors; detecting output from the SISO system; estimating a filtered disturbance from the output; determining a steady state target state from the filtered disturbance and a steady state target output; populating a dynamic optimization solution table using the model factors and a main tuning parameter; determining the optimum input from the dynamic optimization solution table; and controlling the SISO system in response to the optimum input; wherein the determining the optimum input from the dynamic optimization solution table comprises: determining a time varying parameter; determining a potential optimum input from the time varying parameter; and checking whether the potential optimum input is the optimum input; and wherein the dynamic optimization solution table comprises optimal input solution offset Bi and optimal input solution gain Ki, and the determining a potential optimum input comprises determining description="In-line Formulae" end="lead"ui=Kic+Bi. (52a).description="In-line Formulae" end="tail" 48. A system of predictive control for a single input, single output (SISO) system generating an optimum input for controlling the SISO system, comprising: means for modeling the SISO system with model factors; means for detecting output from the SISO system; means for estimating a filtered disturbance from the output; means for determining a steady state target state from the filtered disturbance and a steady state target output; means for populating a dynamic optimization solution table using the model factors and a main tuning parameter; means for determining the optimum input from the dynamic optimization solution table; and means for controlling the SISO system in response to the optimum input; wherein the means for determining a steady state target state from the filtered disturbance and a steady state target output comprises: means for determining a constrained target matrix having elements M11, M12, M21, M22, from the model factors; means for determining constrained target gains G1, G2, and G3 from the model factors and a steady state factor; means for determining an unconstrained target input from the filtered disturbance and the steady state target output from description="In-line Formulae" end="lead"ū*=M21{circumflex over (d)}k|k+M22 y. (34a)description="In-line Formulae" end="tail" means for determining whether the unconstrained target input is on or within input constraints; means for setting a steady state target input equal to the unconstrained target input and means for determining the steady state target state, when the unconstrained target input is on or within input constraints, from description="In-line Formulae" end="lead" xk=M11{circumflex over (d)}k|k+M12 y (34c)description="In-line Formulae" end="tail" means for setting the steady state target input equal to a minimum input constraint when the steady state target input is less than the minimum input constraint, means for setting the steady state target input equal to a maximum input constraint when the steady state target input is greater than the maximum input constraint, and means for determining the steady state target state, when the unconstrained target input is outside the input constraints, from description="In-line Formulae" end="lead" xk=G1{circumflex over (d)}k|k+G2 y+G3ūk (34d).description="In-line Formulae" end="tail" 49. The system of claim 48 wherein: the means for determining a constrained target matrix comprises means for determining a constrained target matrix offline; and the means for determining constrained target gains comprises means for determining constrained target gains offline. 50. The system of claim 48 wherein the means for determining an unconstrained target input comprises means for determining an unconstrained target input online. 51. A system of predictive control for a single input, single output (SISO) system generating an optimum input for controlling the SISO system, comprising: means for modeling the SISO system with model factors; means for detecting output from the SISO system; means for estimating a filtered disturbance from the output; means for determining a steady state target state from the filtered disturbance and a steady state target output; means for populating a dynamic optimization solution table using the model factors and a main tuning parameter; means for determining the optimum input from the dynamic optimization solution table; and means for controlling the SISO system in response to the optimum input; wherein the means for determining the optimum input from the dynamic optimization solution table comprises: means for determining a time varying parameter; means for determining a potential optimum input from the time varying parameter; and means for checking whether the potential optimum input is the optimum input; and wherein the means for determining a time varying parameter comprises: means for determining whether the steady state target output is reachable; means for determining an intermediate time varying parameter, when the steady state target output is reachable, from description="In-line Formulae" end="lead"{tilde over (c)}=Tw0+Tv-1 (14);description="In-line Formulae" end="tail" means for determining the intermediate time varying parameter, when the steady state target output is unreachable, from description="In-line Formulae" end="lead"{tilde over (c)}=Tw0+T+Tv-1, (23)description="In-line Formulae" end="tail" and; means for determining the time varying parameter description="In-line Formulae" end="lead"c={tilde over (c)}-Hūk (36c).description="In-line Formulae" end="tail" 52. A system of predictive control for a single input, single output (SISO) system generating an optimum input for controlling the SISO system, comprising: means for modeling the SISO system with model factors; means for detecting output from the SISO system; means for estimating a filtered disturbance from the output; means for determining a steady state target state from the filtered disturbance and a steady state target output; means for populating a dynamic optimization solution table using the model factors and a main tuning parameter; means for determining the optimum input from the dynamic optimization solution table; and means for controlling the SISO system in response to the optimum input; wherein the means for determining the optimum input from the dynamic optimization solution table comprises: means for determining a time varying parameter; means for determining a potential optimum input from the time varying parameter; and means for checking whether the potential optimum input is the optimum input; and wherein the dynamic optimization solution table comprises optimal input solution offset Bi and optimal input solution gain Ki, and the means for determining a potential optimum input comprises means for determining description="In-line Formulae" end="lead"ui=Kic+Bi. (52a).description="In-line Formulae" end="tail" 53. A computer readable medium storing a computer program for predictive control of a single input, single output (SISO) system generating an optimum input for controlling the SISO system, comprising: computer readable code for modeling the SISO system with model factors; computer readable code for detecting output from the SISO system; computer readable code for estimating a filtered disturbance from the output; computer readable code for determining a steady state target state from the filtered disturbance and a steady state target output; computer readable code for populating a dynamic optimization solution table using the model factors and a main tuning parameter; computer readable code for determining the optimum input from the dynamic optimization solution table; and computer readable code for controlling the SISO system in response to the optimum input; wherein the computer readable code for determining a steady state target state from the filtered disturbance and a steady state target output comprises: computer readable code for determining a constrained target matrix having elements M11, M12, M21, M22, from the model factors; computer readable code for determining constrained target gains G1, G2, and G3 from the model factors and a steady state factor; computer readable code for determining an unconstrained target input from the filtered disturbance and the steady state target output from description="In-line Formulae" end="lead"ū*=M21{circumflex over (d)}k|k+M22 y. (34a)description="In-line Formulae" end="tail" computer readable code for determining whether the unconstrained target input is on or within input constraints; computer readable code for setting a steady state target input equal to the unconstrained target input and computer readable code for determining the steady state target state, when the unconstrained target input is on or within input constraints, from description="In-line Formulae" end="lead" xk=M11{circumflex over (d)}k|k+M12 y (34c)description="In-line Formulae" end="tail" computer readable code for setting the steady state target input equal to a minimum input constraint when the steady state target input is less than the minimum input constraint, computer readable code for setting the steady state target input equal to a maximum input constraint when the steady state target input is greater than the maximum input constraint, and computer readable code for determining the steady state target state, when the unconstrained target input is outside the input constraints, from description="In-line Formulae" end="lead" xk=G1{circumflex over (d)}k|k+G2 y+G3ūk (34d).description="In-line Formulae" end="tail" 54. The computer readable medium of claim 53 wherein: the computer readable code for determining a constrained target matrix comprises computer readable code for determining a constrained target matrix offline; and the computer readable code for determining constrained target gains comprises computer readable code for determining constrained target gains offline. 55. The computer readable medium of claim 53 wherein the computer readable code for determining an unconstrained target input comprises computer readable code for determining an unconstrained target input online. 56. A computer readable medium storing a computer program for predictive control of a single input, single output (SISO) system generating an optimum input for controlling the SISO system, comprising: computer readable code for modeling the SISO system with model factors; computer readable code for detecting output from the SISO system; computer readable code for estimating a filtered disturbance from the output; computer readable code for determining a steady state target state from the filtered disturbance and a steady state target output; computer readable code for populating a dynamic optimization solution table using the model factors and a main tuning parameter; computer readable code for determining the optimum input from the dynamic optimization solution table; and computer readable code for controlling the SISO system in response to the optimum input; wherein the computer readable code for determining the optimum input from the dynamic optimization solution table comprises: computer readable code for determining a time varying parameter; computer readable code for determining a potential optimum input from the time varying parameter; and computer readable code for checking whether the potential optimum input is the optimum input; and wherein the computer readable code for determining a time varying parameter comprises: computer readable code for determining whether the steady state target output is reachable; computer readable code for determining an intermediate time varying parameter, when the steady state target output is reachable, from description="In-line Formulae" end="lead"{tilde over (c)}=Tw0+Tv-1 (14);description="In-line Formulae" end="tail" computer readable code for determining the intermediate time varying parameter, when the steady state target output is unreachable, from description="In-line Formulae" end="lead"{tilde over (c)}=Two+T+Tv-1, (23)description="In-line Formulae" end="tail" and; computer readable code for determining the time varying parameter description="In-line Formulae" end="lead"c={tilde over (c)}-Hūk (36c).description="In-line Formulae" end="tail" 57. A computer readable medium storing a computer program for predictive control of a single input, single output (SISO) system generating an optimum input for controlling the SISO system, comprising: computer readable code for modeling the SISO system with model factors; computer readable code for detecting output from the SISO system; computer readable code for estimating a filtered disturbance from the output; computer readable code for determining a steady state target state from the filtered disturbance and a steady state target output; computer readable code for populating a dynamic optimization solution table using the model factors and a main tuning parameter; computer readable code for determining the optimum input from the dynamic optimization solution table; and computer readable code for controlling the SISO system in response to the optimum input; wherein the computer readable code for determining the optimum input from the dynamic optimization solution table comprises: computer readable code for determining a time varying parameter; computer readable code for determining a potential optimum input from the time varying parameter; and computer readable code for checking whether the potential optimum input is the optimum input; and wherein the dynamic optimization solution table comprises optimal input solution offset Bi and optimal input solution gain Ki, and the computer readable code for determining a potential optimum input comprises computer readable code for determining description="In-line Formulae" end="lead"ui=Kic+Bi. (52a).description="In-line Formulae" end="tail"
※ AI-Helper는 부적절한 답변을 할 수 있습니다.