continuous curve fitting

[altenbach]

---

@brico33 wrote:

-I've made this study on qtiplot before and it gave the following parameters : -2,3,0,-2 . When I put those parameters in the pogram I got a nice erf function 

---

You might want to look at the order of parameters. If the third parameter is zero, all your x are the same and you get a horizontal line out of the model. As a first step you should use the model subVI using the parameter guesses and an x-ramp over the typical range to see if the output looks as expected. Did you do that?

[brico33]

Yes I've tested my subVI model and it works fine, I got a good erf function. I've also found other parameters more appropriate than the others one. But I've noticed something , when I start the program (with the run button sorry) all the first parameters are close to the best fit parameters except the parameter C which fluctuates a lot.

[altenbach]

---

@brico33 wrote:

Yes I've tested my subVI model and it works fine, I got a good erf function. I've also found other parameters more appropriate than the others one. But I've noticed something , when I start the program (with the run button sorry) all the first parameters are close to the best fit parameters except the parameter C which fluctuates a lot.

---

Is that still the same as you uploaded earlier? How much is "a lot"?

How does the data look like?

Can you attach a new VI that contains default data (e.g. as diagram constant) and your typical starting values set to default of the control.