01-24-2022 11:34 AM
Hello,
I have a question. Is there any possible way to automatically determine the properties of the step response? I mean properties such as Time constant, delay and gain. Is there any block that can do that?
01-24-2022 12:00 PM - edited 01-24-2022 12:01 PM
Can you show us some typical data?
If you have a theory for the response you can do a nonlinear fit to the model. Do you?
(Maybe the exponential fit will even work. Make sure you fit for an offset if there is one)
01-24-2022 12:20 PM
Im doing a project in which i have to control a temperature of metal object. Below is a example of step response.
Red graph is a simulation of the step response, and the yellow one is the actual response of the temperature measured by Pt100. I can do the step response simulation by entering the parameters of time constant, gain and delay only with the trial and error method. I want to do this simulation automatically by reading the three parameters of the step response, but how?...
01-24-2022 12:23 PM
Also, as you said, Im doing the step simulation by entering the exponential model on the graph
01-24-2022 02:54 PM
Why would you need a formula node for this??
As I said you can do an exponential fit, just make sure you are selecting the correct subsection of the data.
(Also note that we strongly prefer VIs over pictures. Feel free to attach your VI instead. This way it is also more likely that somebody will actually help)
01-26-2022 10:26 AM
Of course you can easily write a model that fits the entire trace, where the switch time is just another parameter. Can you assume that the time constant for on and off are the same?