Problems about the nonlinear least squares

[chilly_charly]

Refresh508,
you will find that everything works much better when the coefficients and X array are fed at the right position in the formula vi : the inputs have been inverted, with regard to their use in the XuNon-linear Lev-Mar fit vi.

Hope this equilibrate my previous post, Christian ?

CC

[altenbach]

Good catch, CC! The world is back in balance. 🙂

[refresh508]

I can't get what you said. The problem is still not solved. I run the formula.vi alone ,it seems that there is no wrong .But when we run the BYSJ.vi,the formula.vi seems wrong
for example , I set the #sample =300,the the x,dx should be 300 points ,the cofficent is 6 points ,the I is 300 points ,but the result is not like I expect
what is the reason ? the subvi(formula.vi) is something wrong ? but i dont know where !
thanks

[altenbach]

Yes, there is something wrong!

Your formula VI has the inputs reversed! Inside the Xufit VIs, the upper left terminal is for the coefficents and the lower left terminal is for the x-arrays. In your formula VI the upper left terminal is for the x-arrays and the lower terminal is for the coefficients.

You need to switch these two connectors in your formula VI and swap the wire leading to it in your main VI. Now it works just fine!

See attached (LabVIEW 7.0).

[refresh508]

Altenchment:
Thank you very much for your helping ,this problem is solved. I just have been being learning Labview for about two months ,so I will meet many problems . If the problem I cant work it out ,I will ask you ,CC.

[refresh508]

Thanks for your helping last time.This time I met a problem again,as you see when I set the parameters Amplitude=10000 or greater ,then the output will not turn out .
Another problem is the derivative of the output on the waveform wobble after I click run for a while