The Plant of a Feedback System

[Wrydog]

Hello,

I'm having trouble understanding how to derive the plant for feedback control. For a dc-dc full bridge converter is it just the Vout equation in terms of Vin? Does it include the output filter?

Any help would be appreciated.

Thanks.

[moffy]

I am not sure what you mean by "the plant of a feedback system", but if you mean modelling the control system, then yes you have to include the output filter. They establish poles and zeros which you need to know about(laplace domain).

[Wrydog]

Yes, I mean in terms of modeling the control system. How is the plant derived?

I'm probably overthinking it but none of the control systems textbooks I've looked at say much about it.

[IanB]

By "plant" do you mean the process that is being controlled? That is to say, the part of the feedback loop that is not the control system?

If so, you have to come up with a suitable model of the process that adequately represents its behavior. One way to do this is to use suitable model equations for each element of the process, making simplifications as appropriate, and then combine them appropriately.

Any good textbook on process control should devote some space to process modeling. You can't compute the control system response if you don't know how the process behaves.

[T3sl4co1l]

The plant is whatever that block corresponds to, in the loop.

If the controller is sensing the output voltage, and the output voltage follows the filter, then yes.

If the controller is sensing the voltage into the filter, then no.

The controller outputs a control voltage, which is not the supply/input voltage.  But the control input is still an input, so your meaning is ambiguous.

This is an excellent justification for current mode control: only one pole in the control loop, so the controller can be compensated optimally.  The resulting transconductance amplifier (control voltage in, constant current out) is then enclosed in another loop, regulating output voltage.

Tim