I developed a thermal model, its open loop system has complex RHP zeros. Is it possible?

If my model is correct, what could cause the system to have complex RHP zeros?

How the complex RHP zeros affect the system performance limits? Quantitively?

Thank you very much for your help.

 

Are you sure you really have RHP zeroes?

I would expect a process lag in most thermal processes, and this will show up as a time delay. You can model, as an approximation, a "lag" with a matching pole and zero as follows:

(-0.5Ls+1)/(0.5Ls+1)* K/(Ts+1)

Where L is the lag (this is called a first order pade approximation): this is not a very accurate model and I would advise against using it: you might try a higher order pade, but it can be quite cumbersome.

The RHP, unlike a simple delay however, will cause a change in direction so that initial process movement would be opposite its final direction. This "reverse initial action" limits the gain you can use because if it is too high, the controller will push the process initially in the wrong direction and the loop will tend toward instability.

EACH RHP zero will cause a single change in direction and they will normally come in odd pairs (1,3,5....7 etc)

 

Thanks for your reply.

I think I have a NMP zero in the system. My system is a heater to heat the air flowing through the heater. The transfer function of the outlet air temperature vs the inlet air flow rate has a NMP zero in it. Do you think this system truly has a NMP zero?

Thanks.