Simulink Model for control of Full-Bridge DC-DC

[Wrydog]

Hello,

I've some questions regarding my Simulink model shown below. This is primarily for controlling the output voltage of a full-bridge DC-DC converter.

For the top model, I have a pulse generator (340V pulses) to represent the output of the two diodes on the secondary side of the transformer. For the plant, I have the transfer function of the inductor and the output capacitor which also has an 18 Ohm resistor b/c that's what was needed to get a good phase margin.

The bottom model is mostly the same except for a step input that goes to 340V.

I used Matlab's handy pidTuner to find values of kp and ki using the transfer function as the system. These values came to kp=1.022 and ki=4685.

With the lower model, the output looks good, i.e. goes to 340V. However, the top model levels off a 666V. I assume that this is due to the pulsed input and that this top model is not correct. Is that right? Since the output node has a capacitor that can't change quickly can I assume that the step input is ok to use?

What can I do to improve this? For instance, add a controller before the plant to vary the duty cycle based on output V? How would that work in the s-domain?

Thanks.

[Wrydog]

Thanks Blue,

Can you explain why it's 340+340? Doesn't subtracting the output from the reference voltage in this case work?