Re: [foxboro] PIDTAU & PID Tuning.

Will FBTUNE generate reasonable tuning for aything other than a
first-order-plus-lag-and-delay process?  I was thinking about using a PIDA
w/PIDTAU on a level control we have here that has a huge amount of deadtime
and pretty high process gain (in case anyone hasn't guessed, we are trying
to work around some dubious process design :).

It seems physically possible to get reasonable if not outstanding control
on it, as my neural network controller (i.e., me and the output ramp keys)
can control it pretty well.

I'd also like to see a HH, white paper, etc.  on using the
PIDA/FBTUNE/FFTUNE blocks on some real-world applications.  I hear a lot of
"it works great....about half the time" from people who have tried them,
and wonder if a little expert guidance from Invensys -- on both successful
applications and potential gotchas -- might improve that perception.

Corey Clingo
BASF Corp.



|---------+---------------------------->
|         |           <MUGARTET@REPSOLY|
|         |           PF.COM>          |
|         |           Sent by:         |
|         |           foxboro-bounce@fr|
|         |           eelists.org      |
|         |                            |
|         |                            |
|         |           05/29/2003 12:20 |
|         |           PM               |
|         |           Please respond to|
|         |           foxboro          |
|         |                            |
|---------+---------------------------->
  
>------------------------------------------------------------------------------------------------------------------------------|
  |                                                                             
                                                 |
  |              To:  foxboro                                                   
                                                 |
  |              cc:                                                            
                                                 |
  |       Subject:  Re: [foxboro] PIDTAU & PID Tuning.                          
                                                 |
  
>------------------------------------------------------------------------------------------------------------------------------|




Rick,

In our case, with MODOPT 5 we had a SPLLAG of 1 to have a good initial =
response in the output to STP changes. To our surprise, the SPLLAG =
parameter disappeared when changed MODOPT to 8, and seemed to have no =
effect since the response was as if we had SPLLAG 0. As regards the lack =
of response to the sustained error, the INT value was 3 (not too large =
considering it is a temperature control of a fired heater). As regards =
the derivative, we used KD to damp the noise in the MEAS, so that =
shouldn't be bothering. We did not try the FBTUNE, but can do it to =
improve the tuning once we are sure it works fine.
The other concern is whether the tuning parameters should be changed =
given the fact that the algorithm for MODOPT 8 is non-interactive vs. =
product of factors for MODOPT 5. If that's the case, does anyone have =
the conversion table?
Regards,

Martin

-----Mensaje original-----
De: Rick Rys [mailto:rys@xxxxxxxxxxxxxx]
Enviado el: Jueves, 29 de Mayo de 2003 11:06
Para: foxboro@xxxxxxxxxxxxx
Asunto: Re: [foxboro] PIDTAU & PID Tuning.




 
 
_______________________________________________________________________
This mailing list is neither sponsored nor endorsed by Invensys Process
Systems (formerly The Foxboro Company). Use the info you obtain here at
your own risks. Read http://www.thecassandraproject.org/disclaimer.html
 
foxboro mailing list:             http://www.freelists.org/list/foxboro
to subscribe:         mailto:foxboro-request@xxxxxxxxxxxxx?subject=join
to unsubscribe:      mailto:foxboro-request@xxxxxxxxxxxxx?subject=leave

Other related posts: