ControlLawMacro.h Assistance

Tim Love

Champs,

A customer of mine has an inquiry regarding the ControlLawMacro.h that I could use some assistance with:

I'm using the ControlLawMacro.h 2p2z controller to regulate voltage. How do I convert between the P, I, and D based gain values to the B2, B1, B0, A2, and A1 coefficients? There is documentation in the "Digital Telecom Rectifier: GUI Control Program" pdf. It states that the b2, b1, and b0 and calculated a certain way based on PID gains, but mentions nothing about the a2 or a1. Do those equations only apply to that particular setup? Any help in converting regular PID gains into these coefficients would help.

Thanks in advance for any assistance provided.

Regards,

Tim Love

over 17 years ago

0 David M. Alter over 17 years ago

TI__Guru 61235 points

Tim,

PID is a subset of a general 2P2Z controller where a1=-1, and a2=0, C(z) = (b0*z^2 + b1*z + b2)/(z^2 + a1*z + a2). If you have a 2P2Z controller with other a1 and a2 values, it doesn't theoretically convert to PID. With a1=-1 and a2=0, the poles of C(z) are p1=1, p2=0. To have PID-like equivalence for your 2P2Z you need to have the poles close to these (i.e., one very fast pole near z=0, and the other pole near the unit circle). Then, people just convert the b's to get the PID-like equivalence.

Regards,

David

0 Manish Bhardwaj over 15 years ago in reply to David M. Alter

TI__Mastermind 43730 points

In addition to David's explaination, the enclosed pdf goes over the maths of how this is done. you may find this useful.

3036.DiscretePIDcontroller.pdf

C2000™︎ microcontrollers

C2000 microcontrollers forum

ControlLawMacro.h Assistance