Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
04/30/03 22:43
Read: times


 
#44573 - RE: Auto-tuning pid
Responding to: ???'s previous message
Quite HOW it all works I don't know, but when the system is stimulated by a PRBS of signals that keep the output UNsaturated, a record is made of the system output. A cross-correlation algorithm is then used to extract the impulse response for the system from the data. With the impulse response, it is possible to apply a control strategy.

We have played with a Ziegler- Nichols method with sampled data systems and it didn't really work well for us - and as Alfredo points out, you can't force a PID solution onto a system with significant third or fourth order poles, or other non-linear systems.

Since Jez seems to have far more maths than the rest of us, I hope he can inject some comments in this thread.
Steve

List of 9 messages in thread
TopicAuthorDate
Auto-tuning pid            01/01/70 00:00      
   RE: Auto-tuning pid            01/01/70 00:00      
      RE: Auto-tuning pid            01/01/70 00:00      
         RE: Auto-tuning pid Bahri            01/01/70 00:00      
            RE: Auto-tuning pid Bahri            01/01/70 00:00      
               RE: Auto-tuning pid            01/01/70 00:00      
   RE: Auto-tuning pid            01/01/70 00:00      
      RE: Auto-tuning pid            01/01/70 00:00      
   RE: Auto-tuning pid            01/01/70 00:00      

Back to Subject List