Methods for controlling electrical inverters and electrical inverters and systems using the same

1. A method of controlling an electrical inverter including an inverter power train for outputting power to an electrical grid comprising:

generating a reference current waveform in response to a request for power, the reference current waveform shaped by a shape of a grid voltage waveform on the electrical grid;

comparing an inverter output current waveform with the reference current waveform to generate an error signal;

generating a corrective control action signal from the error signal;

generating a nominal duty cycle operating point signal with a feed forward function; and

adding the nominal duty cycle operating point signal and the corrective control action signal to generate a duty cycle control signal for controlling the duty cycle of the inverter power train and thereby provide an output current waveform with a shape substantially matching the shape of the grid voltage waveform.

2. The method of claim 1 , wherein the inverter power train outputs multiple-phase power and the duty cycle control signal controls the generation of an output current waveform of one selected phase such that a shape of the output current waveform of the selected phase substantially matches a shape of the voltage waveform of the selected phase on the electrical grid.

3. The method of claim 1 , wherein generating the reference current waveform comprises:

receiving a set point value proportional to the power requested;

multiplying the set point value and the grid voltage waveform to generate a waveform substantially matching the grid voltage waveform of the electrical grid;

calculating a square of a root-mean-square voltage of the grid voltage waveform; and

generating the reference current waveform shaped by the grid voltage waveform by dividing the waveform by the square of the root-mean-square voltage of the grid voltage waveform.

4. The method of claim 3 , further comprising:

shifting a phase of the grid voltage waveform prior to multiplication with the set point value to vary a phase difference between grid voltage waveform and the output current waveform and thereby vary a corresponding amount of reactive power output from the electrical inverter to the electrical aid.

5. The method of claim 1 , wherein comparing an inverter output current waveform with the reference current waveform to generate an error signal comprises:

scaling and filtering the inverter output current waveform to generate an image of the inverter output current waveform; and

taking a difference between the image of inverter output current waveform and the reference waveform to generate the error signal.

6. The method of claim 1 , wherein comparing an inverter output current, waveform with the reference current waveform to generate an error signal comprises:

scaling and low pass filtering the inverter output current waveform to generate an image of the inverter output current waveform;

scaling and high pass filtering an inverter input current waveform to generate an image of the inverter input current waveform;

adding the images of the input and output current waveforms to generate a sum;

taking a difference between the sum and the reference waveform to generate the error signal.

7. The method of claim 6 , further comprising:

generating a signal predicting a crossing point of the resulting product with the reference waveform; and

taking a difference between signal predicting the crossing point and the reference waveform to generate the error signal.

8. The method of claim 1 , wherein the power train generates a half cycle of a phase of an output voltage waveform with an instantaneous phase voltage V PhaseX by switching between a voltage V Plus and a voltage V Middle with a duty-cycle controlled by the duty cycle control signal, and wherein the feed forward function provides the nominal duty cycle operating point signal as:

Nominal Duty Cycle=( V PhaseX −V middle )/( V Plus −V Middle )

where the second voltage V Middle is determined by a minimum-maximum shaping of the output voltage waveform as:

V Middle =(Maximum( V PhaseX )−Minimum( V PhaseX ))/2.

9. The method of claim 1 , wherein the power train generates a half-cycle of a phase of an output voltage waveform with an instantaneous phase voltage V PhaseX by switching between a voltage V Plus and a voltage V Middle with a duty-cycle controlled by the duty cycle control signal, and wherein the feed-forward function provides the nominal duty cycle operating point signal as:

Nominal Duty Cycle=( V PhaseX −V middle )/( V Plus −V Middle )

where the second voltage V Middle is determined by third harmonic shaping as:

V

2

=

∏

X

⁢

V

Phase

⁢

⁢

X

[

Maximum

⁢

⁢

(

V

Phase

⁢

⁢

X

)

)

]

2

wherein the capital Greek letter pi (Π) is representative of a product.

10. A system for generating a duty cycle control signal for controlling an inverter power train providing an output current waveform to a grid operating with a grid voltage waveform comprising:

a control loop operable to

generate a reference current waveform shaped to the grid voltage waveform;

generate an error signal representing a difference between the reference current waveform and the output current waveform;

generate an corrective control action signal from the error signal;

generate a feed-forward signal representing a nominal duty cycle using a teed-forward function; and

generate the duty cycle control signal by combining the corrective control action signal and the feed-forward signal.

11. The system of claim 10 , wherein the control loop comprises a selected one of a plurality of identical control loops, wherein, each control loop controls the output current waveform of a corresponding phase of multi-phase power being provided by the power train to the grid.

12. The system of claim 11 , wherein the control loop further comprises set point circuitry for generating the reference current waveform and operable to:

generate a signal representing a square of a root-mean-square value of the grid voltage waveform;

scaling the grid voltage with a value representing power to be provided to the grid; and

generate the reference current waveform by dividing the scaled grid voltage waveform by the square of the root-mean-square value of the grid voltage waveform.

13. The system of claim 12 , wherein the set point circuitry is further operable to shift the phase of the grid voltage waveform prior to scaling with a value representing power to be provided to the grid.

14. The system of claim 10 , wherein at least a portion of the control loop is implemented with a digital signal processor.

15. The system of claim 10 , wherein at least a portion of the control loop is implemented in an analog integrated circuit.

16. The system of claim 10 , further comprising a current sensor for sensing the output current waveform and filtering and scaling circuitry for generating a signal representing the output current waveform for comparison with the reference current waveform to generate the error signal.

17. The system of claim 16 , further comprising another current sensor for sensing a power train input current waveform and filtering and scaling circuitry for generating a signal representing the power train input current waveform for combination with the signal representing the output current waveform to generate a signal for comparison with the reference current waveform to generate the error signal.

18. The stem of claim 10 , wherein:

the system further comprises:

a current sensor for sensing the output current waveform; and

another current sensor for sensing a power train input current waveform; and

the control loop is further operable to:

filter and scale the output of the current sensor to generate a first signal representing the output current waveform;

filter and scale and output of the another current sensor to generate a second signal representing the power train input current waveform;

combine the first and second signals;

implement an anticipation function to predict the crossing point of the combination of the first and second signals and the reference current waveform; and

in response to a result of the anticipation function, compare the combination of the first and second signals with the reference current waveform to generate the error signal.

19. The system of claim 18 , wherein the anticipation function is implemented with a one-pole, one-zero function.

20. The system of claim 10 , wherein generating a corrective control action comprises implementing a selected one of a proportional integral derivative function and a two-poles, two-zeros function.