construct bode plot using control design toolkit

[leongchuan]

hello

how can i construct bode plot using the control design toolkit?
it seems that i need to get a model to do so
but i only have a set of gain-frequency values collected from frequency response experiment
is there anyway to convert the gain-frequency values into a model that i can use in control design toolkit?

[leongchuan]

does anybody know how to construct a model using a set of freq-response (gain vs frequency) data?

[Barp]

Hi!

I work on the Control Design Group in LabVIEW and, first of all, thank you for your interest in using the Control Design Toolkit.

If I understand the problem that you are trying to solve, you have a set of data (gain vs frequency) and you want to know the frequency response of you system. Well, the gain vs frequency is the frequency response of it. To obtain a graph of this, you can use a XY graph in LabVIEW and plot the frequency against the gain to obtain the bode plot. In general you need to change the X-Scale to Log and modify gain to be in dB (multiply the gain by 20*log(G)), but this should be enough to give you the bode response.

The CD Bode.vi, as most of the other functions available in Control Design Toolkit, needs to have a model in State-Space, transfer function or zero-poles-gain format. So, if you already have a model in Laplace or z-transform, you can easily use the Control Design Toolkit to obtain the bode plot. However, for what you are trying to do, it will not help you to plot the frequency response.

Now, if you are interested to know the transfer function coefficients of a specific system, the right toolkit is the System Identification Toolkit. It would allow you to obtain the coefficients from input/output signals of a system. To obtain better results for this toolkit, a square wave or noise signal should be applied to the system and measured at the same time. The function ARX, for example, gives you the coefficients based on this signals. I think this is the right toolkit to obtain the linear model of a system that you only can do experiment with it.

I hope this can help you to understand the different uses of our tools. Please let me know if you have any more question and I will be happy to help you!

[leongchuan]

Hi Barp
Thanks for your explanation.
Currently I am trying to make a comparision of the bode plots constructed using System Identification Toolkit and by taking the gain-frequency values. My second task is to determine the transfer function of the system by analysing the bode plot created from the set of gain-frequency values.

I have sucessfully implemented the 1st task by automating the process of collecting gain-frequency results for a number of frequency points and using the xy graph to obtain the bode plot.
My second bode plot created using the SI toolkit was not much of a problem also.

However for my second task I want to use the bode plot obtained from the gain-frequency set of values to determined the transfer function of the system automatically. Hence I thought of creating a system model from the gain-frequency results and after which obtain the transfer function using the SI toolkit.

But it seems I can't create a system model from the gain-frequency values and therefore I can't get my transfer function in this way. Is there any easier way to find the transfer function from the bode plot automatically without the user having to derive the transfer function from the bode plot?

[karthi_286]

Is it possible to obtain the bode magnitude and phase plot without the transfer function from the output FFTed frequency response curve?

Thanks,

Karthick

[karthi_286]

Hello,

In this case,having the output frequency response curve,from this, can you please help how to obtain the bode phase plot curve (phase vs frequency)?

Thanks & Regards,

Karthick