Predictive control system and method

• US 7,243,006 B2
• Filed: 02/24/2003
• Issued: 07/10/2007
• Est. Priority Date: 02/28/2002
• Status: Expired due to Fees

• Current Assignee: Zetacon Corporation
• Original Assignee: Zetacon Corporation
• Inventors: Richards, Michael J.
• Primary Examiner(s): Picard; Leo
• Assistant Examiner(s): Garland; Steven R.
• Application Number:
  US10/370,476
• Publication Number:
  US 20030163296A1
• Time in Patent Office: 1,597 Days
• Field of Search: 700/286, 703/14, 703/18, 318/801, 318/254
• US Class Current: 700/286
• CPC Class Codes: G06F 30/367 Design verification, e.g. using simulation, simulation program with integrated circuit emphasis [SPICE], direct methods or relaxation methods
• CPC Class Codes: H02M 7/53873 with digital control
• CPC Class Codes: H02P 21/09 Field phase angle calculation based on rotor voltage equation by adding slip frequency and speed proportional frequency
• CPC Class Codes: H02P 21/16 Estimation of constants, e.g. the rotor time constant

Abstract

A control system and method selects a switch configuration for a power circuit having N binary switches, based in part on a finite state machine. The control system includes an embedded simulator, and present and next state contemplators. The various switch states of the power circuit are modeled by the finite state machine such that at any time, the power circuit switches are in a Present State and there are a plurality of Next States which are one or more switch transitions away from the Present State. The embedded simulator estimates the operating conditions of the load based on measured operational characteristics and the Present State. The present state contemplator determines, based on the operating conditions, whether a switch state transition should be contemplated. If so, the next state contemplator determines the optimal next state based on performance criteria and sends a state switch command to the power circuit.

Claims

1. A control system for controlling the operation of a power circuit, said power circuit having N binary switches for one of providing power to and receiving power from a load circuit, said N binary switches having various configurations being represented by a finite state machine having up to 2^N switch states, a present configuration of said N binary switches being associated with a present switch state and having a plurality of next switch states according to state transition constraints of the finite state machine, said control system comprising:

a) an acquisition module for acquiring a set of operational signals related to the load circuit associated with the present switch state;

b) an embedded simulator coupled to the acquisition module for estimating a set of estimated operational characteristics of the load circuit based on the set of operational signals and the present switch state;

c) a next state contemplator coupled to said embedded simulator for estimating a set of estimated operational characteristics of the load circuit for each of the plurality of next switch states, for comparing and selecting the next switch state having estimated operational characteristics that best meet a first set of criteria, and for actuating the power circuit to change the switch configuration of the N binary switches therein to the switch configuration associated with said selected next switch state of the plurality of next switch states.

2. The system of claim 1, further comprising a present state contemplator coupled to the embedded simulator for activating the next state contemplator if the set of estimated operational signals have exceeded a second set of criteria.

3. The system of claim 2, wherein said load circuit is a motor.

4. The system of claim 3, wherein the motor is one of an induction motor, a synchronous motor, a permanent magnet motor, and a synchronous reluctance motor.

5. The system of claim 3, wherein said power circuit is a three phase inverter having three binary switches.

6. The system of claim 5, wherein the embedded simulator includes a main flux predictor/corrector which determines a main flux linkage using a modified Gear Method as applied to time shifted values of estimated air gap voltages.

7. The system of claim 6, wherein the modified gear method comprises determining the main flux linkage for a next time step of duration h using one of a one-step calculation and a two-step calculation, wherein, when the next switch state is equal to the present switch state, the one-step calculation is used to calculate the main flux linkage, and when the next switch state is not equal to the present switch state, the two-step calculation is used to calculate an intermediate main flux linkage based on an interlock voltage for a first time step of duration h_int associated with an interlock delay time, and the intermediate main flux linkage is then used to calculate the main flux linkage based on a voltage associated with the next switch state for a second time step of duration h-h_int.

8. The system of claim 6, wherein the embedded simulator includes a loss compensator coupled to the main flux predictor/corrector, wherein the loss compensator produces a decay in stator flux by using a compensating parameter for representing a combined effect of all unmodeled losses of the motor.

9. The system of claim 8, wherein the loss compensator produces a decay for a quadrature component of the stator flux by summing first and second values and subtracting a third value, the first value being a quadrature component of stator current multiplied by stator resistance, the second value being a derivative of the quadrature component of stator current multiplied by stator leakage inductance and the third value being the compensating parameter multiplied by the quadrature component of the stator flux.

10. The system of claim 8, wherein the system further comprises a parameter module coupled to the embedded simulator for providing parameter and structural details of the load circuit.

11. The system of claim 8, wherein the embedded simulator includes a motor characteristic calculator coupled to the main flux predictor/corrector for calculating a plurality of motor characteristics, said motor characteristic calculator being coupled to the next state contemplator and the present state contemplator for providing at least a portion of said plurality of motor characteristics thereto.

12. The system of claim 3, wherein said power circuit is one of a full-bridge inverter and a half-phase inverter.

13. The system of claim 2, wherein said second set of criteria utilized by the present state contemplator includes at least one of a torque tolerance and a stator flux tolerance.

14. The system of claim 13, wherein said second set of criteria further comprises a time duration for remaining in a given switch state.

15. The system of claim 2, wherein the load circuit comprises a first stator winding and a second stator winding, the power circuit being connected to the first stator winding, wherein the embedded simulator, the next state contemplator and the present state contemplator define a master controller for controlling the power circuit, and wherein the control system further comprises a second power circuit being connected to the second stator winding and a slave controller connected to and controlling the second power circuit, the slave controller comprising an additional embedded simulator, an additional next state contemplator and an additional present state contemplator.

16. The system of claim 15, wherein the master and slave controllers are adapted for providing similar currents to each stator winding.

17. The system of claim 2, wherein the power circuit is coupled to a second load via a link supply, and the load circuit is a rotating system, the second load being controlled by the control system for receiving or providing energy to the rotating system and the rotating system being adapted to receive or provide energy to the second load.

18. The system of claim 17, wherein the control system controls the rotating system for providing a first reactive power to the second load for canceling a second reactive power that is present in the second load.

19. The system of claim 17, wherein the link supply is one of a capacitor, a source and a combination of a capacitor and a source.

20. The system of claim 2, wherein the load circuit is a three-phase grid source connected to the power circuit via a power grid, and the power circuit includes a DC link, the control system being adapted to adjust power delivered by the three-phase grid.

21. The system of claim 20, wherein the load circuit is connected to the power circuit via line reactors.

22. The system of claim 1, wherein said first set of criteria consists of an ideal stator flux and a related tolerance.

23. The system of claim 22, wherein the set of estimated operational characteristics includes an estimated stator flux and the selected next switch state is the next switch state having a magnitude squared error between the estimated stator flux and the ideal stator flux that is a minimum.

24. The system of claim 22, wherein the next state contemplator comprises an ideal stator flux vector calculator that calculates quadrature components of an ideal stator flux vector (λ

_ds* and λ

_qs*) according to λ

_ds*=| λ

_s|*∘

cos(A) and λ

_qs*=| λ

_s|*∘

sin(A) where | λ

_s|* is a stator flux magnitude reference,

$A={\sin }^{-1}<br><br>\left[\left(\frac{3·<br><br>\mathrm{Lm}·<br><br>\mathrm{\sigma <br><br>}}{P<br><br>\left(1-\mathrm{\sigma <br><br>}\right)}\right)<br><br>\left(\frac{T_e^{*}}{{<br><br>{\stackrel{\_}{\mathrm{\lambda <br><br>}}}_s<br><br>}^{*}·<br><br><br><br>{\stackrel{\_}{\mathrm{\lambda <br><br>}}}_r<br><br>}\right)\right]+{\tan }^{-1}<br><br>\left(\frac{{\mathrm{\lambda <br><br>}}_{\mathrm{qr}}}{{\mathrm{\lambda <br><br>}}_{\mathrm{dr}}}\right),$

Lm is magnetization inductance of the load circuit,

$\mathrm{\sigma <br><br>}=1-\frac{{\mathrm{Lm}}^2}{L_S<br><br>L_R},$

L_S is stator inductance, L_R is rotor inductance, P is a number of poles of the motor, T_e* is torque reference, | λ

_r| is a magnitude of rotor flux, and λ

_qr and λ

_dr are quadrature components of rotor flux.

25. The system of claim 22, wherein the next state contemplator comprises an ideal stator flux vector calculator that calculates quadrature components of an ideal stator flux vector (λ

_ds* and λ

_qs*) according to λ

_ds*=| λ

_s|*∘

cos(A) and λ

_qs*=| λ

_s|*∘

sin(A) where

${<br><br>{\stackrel{\_}{\mathrm{\lambda <br><br>}}}_s<br><br>}^{*}=<br><br>\frac{3·<br><br>\mathrm{Lm}·<br><br>\mathrm{\sigma <br><br>}·<br><br>T_e^{*}}{P<br><br>\left(1-\mathrm{\sigma <br><br>}\right)}<br><br>$

is a stator flux magnitude reference,

$A=\frac{\mathrm{\pi <br><br>}·<br><br>T_e^{*}}{2·<br><br><br><br>T_e^{*}<br><br>}+{\tan }^{-1}<br><br>\left(\frac{{\mathrm{\lambda <br><br>}}_{\mathrm{qr}}}{{\mathrm{\lambda <br><br>}}_{\mathrm{dr}}}\right),$

Lm is magnetization inductance of the load circuit,

$\mathrm{\sigma <br><br>}=1-\frac{{\mathrm{Lm}}^2}{L_S<br><br>L_R},$

L_S is stator inductance, L_R is rotor inductance, P is a number of poles of the motor, T_e* is torque reference, λ

_qr and λ

_dr are quadrature components of rotor flux.

26. The system of claim 1, wherein the control system is adapted to update the state transition constraints of the finite state machine with new state transition constraints during operation of the control system.

27. The system of claim 26, wherein the control system includes a set of new state transition constraints to compensate for failure of at least one of the N binary switches, wherein the set of new state transition constraints does not include transitions to a next switch state that involves switching on or off of a failed binary switch.

28. The system of claim 1, wherein the number of next switch states associated with a present switch state defined according to the state transition constraints of the finite state machine is less than the 2^N switch states for reducing computational complexity.

29. The system of claim 28, wherein one of the next switch states associated with a present switch state includes remaining in the present switch state.

30. The system of claim 1, wherein transition to the next switch state from the present switch state involves switching at most one of said N binary switches thereby reducing switching frequency.

31. The system of claim 1, wherein the system includes a sequential processing event manager for providing a system time step h for synchronizing the estimation of the set of estimated operational characteristics by the embedded simulator and a transition from the present switch state to one of the plurality of next switch states.

32. The system of claim 1, wherein each binary switch comprises an open complementary switch and a closed complementary switch connected at a common node in a totem pole configuration, the embedded simulator being adapted to perform interlock delay compensation wherein during transition from the present switch state to the next selected switch state if one of the binary switches comprises two complimentary switches being simultaneously open during an interlock period, the interlock delay compensation comprising determining an interlock voltage at the common node that depends on a direction of current at the common node.

33. The system of claim 32, wherein the closed complementary switch is at a top of the totem pole configuration and when the closed complementary switch opens a present switch state voltage is associated with the present switch state and a next switch state voltage is associated with the next switch state, and wherein the interlock voltage is equal to the present switch state voltage if the current flows into the common node and the interlock voltage is equal to the next switch state voltage if the current flows out of the common node.

34. The system of claim 32, wherein the closed complementary switch is at a bottom of the totem pole configuration and when the closed complementary switch opens a present switch state voltage is associated with the present switch state and a next switch state voltage is associated with the next switch state, and wherein the interlock voltage is equal to the next switch state voltage if the current flows into the common node and the interlock voltage is equal to the present switch state voltage if the current flows out of the common node.

35. The system of claim 25, wherein the embedded simulator comprises a rotating restart module for providing rotating restart functionality, wherein the rotating restart module provides initial motor characteristics including establishment of state variables and utilizing the interlock voltage for estimating operational characteristics of the motor.

36. The system of claim 35, wherein the rotating restart module further provides a torque command of zero and a stator flux command of zero when enabling the power circuit and thereafter providing a stator flux command that ramps from zero to a nominal stator flux reference value.

37. The system of claim 36, wherein the rotating restart module further provides a stator resistance value that is ramped down to a nominal stator resistance during rotating restart thereby improving motor performance during rotating restart.

38. The system of claim 35, wherein the load circuit is an induction motor and stator flux and rotor flux state variables are reset to zero.

39. The system of claim 35, wherein the load circuit is one of a synchronous reluctance motor and a permanent magnet motor and stator flux and rotor flux state variables are reset to an arbitrary initial vector having a constant magnitude.

40. The system of claim 1, wherein the next state contemplator is further adapted to iteratively estimate operational characteristics of the load circuit for each of the plurality of next switch states for selecting a subset of the next switch states which meet a third criteria, and for each next switch state in the subset, the next state contemplator estimates operational characteristics of the load circuit for a second set of next switch states thereby generating a plurality of second sets from which the next state contemplator compares and selects the next switch state having estimated operational characteristics that best meets the first set of criteria.

41. The system of claim 40, wherein the third criteria comprises a hysteresis band centered at a stator flux value.

42. A method for controlling the operation of a power circuit, said power circuit having N binary switches for one of providing power to and receiving power from a load circuit, said N binary switches having various configurations being represented by a finite state machine having up to 2^N switch states, a present configuration of said N binary switches being associated with a present switch state and having a plurality of next switch states according to state transition constraints of the finite state machine, said method comprising:

a) acquiring a set of operational signals related to the load circuit associated with the resent switch state;

b) performing a simulation to estimate a set of estimated operational characteristics of the load circuit based on the set of operational signals and on the present switch state;

c) estimating a set of estimated operational characteristics of the load circuit for each of the plurality of next switch states;

d) comparing and selecting the next switch state having estimated operational characteristics that best meet a first set of criteria; and

,

e) changing the configuration of the N binary switches to the switch configuration associated with said selected next switch state of the plurality of next switch states.

43. The method of claim 42, wherein steps (d) and (e) are executed once the estimated operational characteristics have exceeded a second set of criteria.

44. The method of claim 43, wherein said load circuit is a motor and the power circuit is a three phase inverter having three binary switches and step (b) includes determining main flux linkage in the motor using a modified Gear Method as applied to the time shifted values of estimated air gap voltages.

45. The method of claim 44, wherein the modified gear method comprises determining the main flux linkage for a next time step of duration h using one of a one-step calculation and a two-step calculation, wherein, when the next switch state is equal to the present switch state, the one-step calculation is used to calculate the main flux linkage, and when the next switch state is not equal to the present switch state, the two-step calculation is used to calculate an intermediate main flux linkage based on an interlock voltage for a first time step of duration h_int associated with an interlock delay time, and the intermediate main flux linkage is then used to calculate the main flux linkage based on a voltage associated with the next switch state voltage a second time step of duration h-h_int.

46. The method of claim 44, wherein step (b) further includes calculating a decay in stator flux by using a compensating parameter for representing a combined effect of all unmodeled losses of the motor.

47. The method of claim 46, wherein calculating the decay for the stator flux for a quadrature component of the stator flux comprises summing a first and a second value and subtracting a third value, the first value being a quadrature component of stator current multiplied by stator resistance, the second value being a derivative of the quadrature component of stator current multiplied by stator leakage inductance and the third value being the compensating parameter multiplied by the quadrature component of the stator flux.

48. The method of claim 43, wherein step (d) includes determining whether the estimated operational characteristics have exceeded one of the group consisting of a torque tolerance, a stator flux tolerance, and a time duration for remaining in a given switch state.

49. The method of claim 43, wherein the power circuit is coupled to a second load via a link supply, and the load circuit is a rotating system, the method comprising controlling the operation of the rotating system via the power circuit for receiving or providing energy to the second load.

50. The system of claim 49, wherein the method comprises controlling the operation of the rotating system via the power circuit for providing a first reactive power to the second load for canceling a second reactive power that is present in the second load.

51. The method of claim 43, wherein the load circuit is a three-phase grid source, the power circuit being connected via a power grid to the three-phase grid source and the method comprises controlling the power circuit for adjusting power delivered by the three-phase grid.

52. The method of claim 51, wherein the method comprises controlling the power circuit for producing reactive currents in the power grid for achieving unity power factor in the power grid.

53. The method of claim 51, wherein the method comprises controlling the power circuit for producing negative currents in the power grid for removing harmonic currents in the power grid.

54. The method of claim 51, wherein the method comprises controlling the power circuit for providing power to the power grid for grid-power regeneration.

55. The method of claim 42, wherein step (d) includes comparing and selecting the next state switch state having an estimated stator flux value that is closest to an ideal stator flux vector.

56. The method of claim 55, wherein the method comprises calculating d-q axis quadrature components of the ideal stator flux vector (λ

_ds* and λ

_qs*) according to λ

_ds*=| λ

_s|*∘

cos(A) and λ

_qs=| λ

_s|*∘

sin(A) where | λ

_s|* is a stator flux magnitude reference,

$A={\sin }^{-1}<br><br>\left[\left(\frac{3·<br><br>\mathrm{Lm}·<br><br>\mathrm{\sigma <br><br>}}{P<br><br>\left(1-\mathrm{\sigma <br><br>}\right)}\right)<br><br>\left(\frac{T_e^{*}}{{<br><br>{\stackrel{\_}{\mathrm{\lambda <br><br>}}}_s<br><br>}^{*}·<br><br><br><br>{\stackrel{\_}{\mathrm{\lambda <br><br>}}}_r<br><br>}\right)\right]+{\tan }^{-1}<br><br>\left(\frac{{\mathrm{\lambda <br><br>}}_{\mathrm{qr}}}{{\mathrm{\lambda <br><br>}}_{\mathrm{dr}}}\right),$

Lm is magnetization inductance of the load circuit,

$\mathrm{\sigma <br><br>}=1-\frac{{\mathrm{Lm}}^2}{L_S<br><br>L_R},$

L_S is stator inductance, L_R is rotor inductance P is a number of poles of the motor, T_e* is torque reference, | λ

_r| is a magnitude of rotor flux and λ

_qr and λ

_dr are quadrature components of rotor flux.

57. The method of claim 55, wherein the method comprises calculating quadrature components of the ideal stator flux vector (λ

_ds* and λ

_qs*) according to λ

_ds*=| λ

_s|*∘

cos(A) and λ

_qs*=| λ

_s|*∘

sin(A) where

${<br><br>{\stackrel{\_}{\mathrm{\lambda <br><br>}}}_s<br><br>}^{*}=<br><br>\frac{3·<br><br>\mathrm{Lm}·<br><br>\mathrm{\sigma <br><br>}·<br><br>T_e^{*}}{P<br><br>\left(1-\mathrm{\sigma <br><br>}\right)}<br><br>$

is a stator flux magnitude reference,

$A=\frac{\mathrm{\pi <br><br>}·<br><br>T_e^{*}}{2·<br><br><br><br>T_e^{*}<br><br>}+{\tan }^{-1}<br><br>\left(\frac{{\mathrm{\lambda <br><br>}}_{\mathrm{qr}}}{{\mathrm{\lambda <br><br>}}_{\mathrm{dr}}}\right),$

Lm is magnetization inductance of the load circuit,

$\mathrm{\sigma <br><br>}=1-\frac{{\mathrm{Lm}}^2}{L_S<br><br>L_R},$

L_S is stator inductance, L_R is rotor inductance, P is a number of poles of the motor, T_e* is torque reference, and λ

_qr and λ

_dr are quadrature components of rotor flux.

58. The method of claim 42, wherein the method comprises updating the state transition constraints of the finite state machine with new state transition constraints while controlling the power circuit.

59. The method of claim 58, wherein updating the state transition constraints of the finite state machine occurs when at least one of the N binary switches fails, wherein the new state transition constraints do not include transitions to a next switch state that involves turning on a failed binary switch.

60. The method of claim 42, wherein the method further includes synchronizing steps (b) and (e) with respect to one another for reducing numerical instability during estimation.

61. The method of claim 42, wherein each binary switch comprises an open complementary switch and a closed complementary switch connected at a common node in a totem pole configuration, and step (b) comprises performing interlock delay compensation wherein during transition from the present switch state to the next selected switch state one of the binary switches comprises two complimentary switches being simultaneously open during an interlock period, the interlock delay compensation comprising determining an interlock voltage at the common node that depends on a direction of current at the common node.

62. The method of claim 61, wherein the closed complementary switch is at a top of the totem pole configuration and when the closed complementary switch opens a present switch state voltage is associated with the present switch state and a next switch state voltage is associated with the next switch state, and wherein determining the interlock voltage involves determining the direction of the current flow wherein the interlock voltage is equal to the present switch state voltage if the current flows into the common node and the interlock voltage is equal to the next switch state voltage if the current flows out of the common node.

63. The method of claim 61, wherein the closed complementary switch is at a bottom of the totem pole configuration and when the closed complementary switch opens a present switch state voltage is associated with the present switch state and a next switch state voltage is associated with the next switch state, and wherein determining the interlock voltage involves determining the direction of the current flow wherein the interlock voltage is equal to the next switch state voltage if the current flows into the common node and the interlock voltage is equal to the present switch state voltage if the current flows out of the common node.

64. The method of claim 61, wherein the method further comprises performing a rotating restart by establishing initial motor characteristics and performing the interlock delay compensation during step (b).

65. The method of claim 61, wherein performing the rotating restart further comprises providing a torque command of zero and a stator flux command of zero when enabling the power circuit and thereafter providing a stator flux command that ramps from zero to a nominal stator flux reference value.

66. The method of claim 65, wherein performing the rotating restart further comprises providing a stator resistance value that is ramped down to a nominal stator resistance.

67. The method of claim 42, wherein step (c) of the method further comprises:

(f) selecting a several next switch states having estimated operational characteristics that best meet the first set of criteria to form a subset of next switch states;

(g) after step (f), estimating a set of estimated operational characteristics associated with a second set of next switch states for each next switch state in the subset thereby generating a plurality of second sets; and

,

(h) after step (g), comparing and selecting the next switch state from the subset that is associated with the next switch state from the plurality of second sets having estimated operational characteristics that best meets the first set of criteria.