Simatic PID controller explanation

Hello,

there are many people here familiar with PID controllers. I hope they can explain one thing I read in Siemens software catalog. It was written something like: The Simatic controller always work according to position algorithm. The speed algorithm known from old Simatic S5 is no longer executed in S7.

Can anyone explain what this means? PID controller can be used also for temperature and flow not only for position and speed.
I notice that there are special temperature controllers in Simatic library. Does this mean that other PID controllers are only to be used with position or maybe this is something else...?


Thank you

No, you've got it all wrong. It is actually the way how control law (Controller output - CO) is calculated in discrete systems. There are basically two forms:
1. Position form
2. velocity (not speed) form

Please, read this and ask question if you still don't understand.

Best regards,
Pandiani

The terms "position PID" and "velocity PID" are bad terms to use.

I understand what it meant but they really have nothing to do with position and velocity. How is that for confusing?
The "position PID" is what you see all the time

Error=SP-PV
CO=Ki*Δt*integrate(Error)+Kp*Error+(Kd/Δt)*(Error-Error(n-1))

The "velocity PID" is how PIDs are implemented on many DSPs.
K0=Ki*Δt+Kp+(Kd/Δt)
K1=-(Kp+2*Kd/Δt)
K2=Kd/Δt
CO=CO(n-1)+K0*Error+K1*Error(n-1)+K2*Error(n-2)
It doesn't look much like a PID does it?
This form requires only 3 multiplies and adds so is very efficient. This form also doesn't suffer from integrator wind up problems and the best part is that one can change the gains on-the-fly and not get a discontinuous step in the control output. Because of this I like to call this the incremental form because you can see the new control output is just an incremental change from the previous control output. The "Position PID" or non incremental PID will have a step change in the control output if the gains are changed on-the-fly.

The disadvantage of the "Velocity PID" is that the integrator must be used and it must be tuned properly. If not the "Velocity PID" will not work well at all. Since there are too many customers, almost 100%, that are totally clueless about tuning the integrator properly the "Velocity PID" tends to be a tech support headache. Another disadvantage is that one can't see the output due to the integrator term, the proportional term and the derivative term. This is educational and using the integrator term is helpful setting up the feed forward gains.

Thank you gentlemen,now I understand what is this all about.