Re: Question: PID Parameter Adjustment for no overshoot

Duane Mattern (d.mattern@ieee.org)
Mon, 12 Jul 1999 01:08:09 +0200 (MET DST)

On 11 Jul 1999 10:39:01 GMT, you wrote:
> I must also mention that I use the derivative part in such a way
>that the input to the derivative is the Actual value only,
>not the controller error.

You need to change that to include the derivative of the measurement
since that affects the closed loop properties of the system and determines
the system response. The derivative of the input can give you lead information
about command changes, but does not affect the closed loop response of the
system.

>Another problem is that the system, which is actually a temperature
>control system with large dead time, should not require a set value
>change to be re-tuned. A disturbance on the system should be enough to
>initiate a tuning step.

The problem with this is you won't know the magnitude or waveform
of the disturbance input. If you command a step change, then you know
what the value is and can thus use that information in your tuning algorithm.

>If anyone knows some link on the web or has done such a work himself, I
>would thank you very much for your help.

Astrom and Wittenmark have done a lot of work on self tuning controllers
back in the 1980's. They maintain webpages at their universities. I don't
know what those are off hand, but that would be the place to start.....see
what has already been done. I did some work on Model Reference
adaptive control back in the late 1980's, but I've not followed
much of the recent work because I believe the problem has
mostly been solved, from a practical point of view.

Duane Mattern (d.mattern@ieee.org)
Columbus, OH USA

############################################################################
This message was posted through the fuzzy mailing list.
(1) To subscribe to this mailing list, send a message body of
"SUB FUZZY-MAIL myFirstName mySurname" to listproc@dbai.tuwien.ac.at
(2) To unsubscribe from this mailing list, send a message body of
"UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL yoursubscription@email.address.com"
to listproc@dbai.tuwien.ac.at
(3) To reach the human who maintains the list, send mail to
fuzzy-owner@dbai.tuwien.ac.at
(4) WWW access and other information on Fuzzy Sets and Logic see
http://www.dbai.tuwien.ac.at/ftp/mlowner/fuzzy-mail.info
(5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html