Simulating on-grid three-phase inverter with SVPWM, but the alpha beta components diverge so much

I am simulating for on grid three-phase inverter using PI and PR(seperately control) using SVPWM, this figure aboce is the SVPWM reference signal, which are completely wrong. Somehow the alpha beta component diverge so much from what is intended. what could be the main problem of this? The carrier frequency is 10kHz, and i cant figure out the amplitude because the values kept on diverging and it doesnt show any steadiness

I also tried to switch my own SVPWM design with matlab's pwm generator and it worked perfectly, I know using the pwm generator is the easier choice, but i want to implement my own SVPWM approach.

This is the DQ component from the PI controller, just before entering into the PWM block, likewise the alpha beta component is also affected by the divergence.

Below is the intended shape of the reference signal for SVPWM, i obtained this from using pure sine waves as the input for the SVPWM, however for an actually inverter, i must use the vref from the PI controller as the input of SVPWM

[Shivam Gothi]

Hello @bieron surya,

I have few doubts. Could you please clearify them?

• As you said "I also tried to switch my own SVPWM design with matlab's pwm generator and it worked perfectly, I know using the pwm generator is the easier choice, but i want to implement my own SVPWM approach."

Which blocks in the above diagram are you replacing with "PWM generator" of MATLAB?

• If possible, can you please share your simulink model? It will be helpfull to debug the approach you are using to implement SVPWM and accordingly suggest a possible workaround to rectify it.

Thanks !

Answer 1

[Sabin]

There is a good chance that the angle is wrong because you are doing atan2 of input(2) and input(2). However, input seems to be the d and q voltages if I understand correctly. If that’s the case, then the angle is not computed correctly. To debug this you can directly use the measured electrical angle.

To learn more about Space Vector Modulation check this page:

https://mathworks.com/discovery/space-vector-modulation.html