adaptive contro theroyl

[cosmin]

In the VI provided, the stimulus signal (PID output) goes into the same model created initially, so there is no way that you will get any model changes by system identification (besides that from errors regarding the algorithm used for identification).

cosmin 

[Michelle_N]

There are a few things I notice-

1. You are doing some seemingly unneccessary conversions. You have a state-space, convert it to a transfer function and then pass that to a discrete transfer function. Is there a reason you are doing this over using the "Discrete  State-Space" function directly - such that you will not need to convert to a transfer function?

2. You didn't provide us any of your front panel control values. What are you using for A, B, C, D, sampling time, PID gains?

[aswwwa]

hi

is there a way to communicate directly with NI engineers

or any one with experiance in control ??

[Michelle_N]

One last thing that I missed is that you are not properly converting your model from ARX model - this could certainly cause an issue.

The correct way to do this is to replace the unbundle and the CD Set Data to Model and replace it with SI Convert to Models of CDT:

Please try this, even if this isn't the root cause of your issue, it is something worth changing as it could be causing other problems.

[aswwwa]

thanx alot for your help

[Michelle_N]

You're welcome.  Hope that helps.

[aswwwa]

IM JUST WINDERING IF THERE IS ANY WAY TO GET HELP LIKE AN ONLINE HELP FROM NI ENGINEERS , CONTROL AND SIMULATION

[jareds]

Hi aswwwa,

I'm not aware of a way of directly speaking with an NI Engineer online, but you can try to call into technical support.  Find more information at ni.com/support.