Space Vector Control of a Three Phase Boost Rectifier
Space vector control is a popular technique used in the control of motor drives or three-phase rectifiers since it offers reduced switching losses and better utilization of the dc bus compared to conventional PWM control. This example model demonstrates space vector control of a three-phase boost-type rectifier using PLECS. The schematic diagram of the boost rectifier is shown below.

The control goals for the three-phase boost rectifier are to draw sinusoidal current from the input supply, vn, and to regulate the output voltage, Vdc. Current control is achieved using an inner current control loop that measures the phase current, In, and controls the inductor-neutral voltage, vn1, to force the phase current to track its reference value. The current reference is provided by outer control loops that implement dc voltage and power factor control.
With space vector control, the inductor-neutral voltage is controlled as a vector quantity in the αβ or dq domains. In this example, control is performed in the dq domain. The advantage of dq control is that ac quantities become dc quantities in the dq domain. Thus no tracking error exists when using a PI controller to regulate the ac input current.
The inductor-neutral voltage vector calculated by the space vector controller is generated by time-averaging voltage vectors that the rectifier produces at its ac terminals. In the example, the voltage vectors are time-averaged using a symmetrical PWM switching pattern.
In this model, the power electronics are implemented using PLECS and the controllers using Matlab/Simulink. The outer voltage loop is implemented using Simulink and the dq current control and switch modulation functions are implemented using Matlab .m script files. The switch modulation function is time-synchronized with the beginning of the switching cycle to mirror what would occur in a practical DSP controller.
Requirements
PLECS 2.1 or newer
Matlab 7.0 (R14) or newer
Download
The package file plecs_svpwm.zip (388 KB) contains the example model and the documentation in PDF format.
|