| Author |
 Topic  |
|
|
Juan_Español
4 Posts |
 Posted - 11/10/2015 : 11:58:59 AM
|
Hi,
I'm trying to adjust a characteristic Intensity-Voltage curve of a solar cell. I have the experimental data:
and I'm trying to adjust with the diode equation, which is I = IL - Io*( exp(q*(V+I*Rs)/(A*k*T) ) -1) - (V+I*Rs)/Rsh
where "I" is intensity, "V" is voltage, q,k,T,IL,Rsh,Io are constants, and I don't know A and Rs.So I define the function with the FFT:
and I fix the constants parametres, and also bound A and Rs, i.e. those I want Origin calculates for me:
but when I press the Fit button, it makes no iteration and it doesn´t converge:
and it returns me no values of A and Rs
Does somebody know which the problem is? Maybe the Function Form, or the upper and lower bound on the parametres?
Thank you! |
|
|
KenLefebvre
USA
66 Posts |
Posted - 11/10/2015 : 5:32:33 PM
|
Hi Juan--
Just a quick formality, could you tell us what version of Origin you're using, as well as what version of Windows?
With thanks,
Ken Lefebvre
Technical Service Representative
OriginLab |
 |
|
|
Juan_Español
4 Posts |
Posted - 11/14/2015 : 08:25:24 AM
|
| I think I'm using Origin 8 and Windows 8.1, because a friend has let me the PC and it's not mine. |
|
|
|
Juan_Español
4 Posts |
Posted - 11/17/2015 : 07:29:14 AM
|
I've seen on your website that only the Pro version allows to fit implicit functions, is that correct? I think because of the function is not explicit (i.e. the dependent variable "I" appears on both sides of the equation), the program could not adjust data to that function. Does not exist any option for fitting that kind of implicit function?
Thanks. |
|
|
|
KenLefebvre
USA
66 Posts |
Posted - 11/17/2015 : 3:44:36 PM
|
Regrettably this is such an old version of Origin, that the support I can offer you is limited. That aside, implicit functions are indeed only supported by 2016 Pro version, so I have little doubt this was the case in 2007.
Your hands might be tied but, even if you don't buy it, I'd suggest trying to build this same function in the evaluation version of OriginPro 2016. It's fully functional for 21 days so give it a try and if that still doesn't work perhaps one of my colleagues can help you.
I'm sorry that I can't offer you better advice than this. |
|
|
|
Juan_Español
4 Posts |
Posted - 12/02/2015 : 11:31:07 AM
|
Thank you for your help. Finally I have achieved that Origin performs some iterations, but I don't know why it only gets me one value of the undefined variable (that which I call "A"), but it doesn't seem to fit the Rs parameter.
Why could Origin ignores one of the adjustable parameters, despite of not fixing it?
I´m not fixing the A and Rs parameters but giving them an initial value (which I think is approximated to the final value)
but as you can see, the Rs parameter seems to be fixed for the program
so finally I "fit until converge" and it gives me the right value of A, but not for Rs.
Does someone know why the Rs parameter is not adjusted (I have proved giving so many different initial values of Rs, but the behaviour is always the same)?
Thanks.
|
|
|
| |
Topic |
|
|
|
Nonlinear Fitting Function doesn´t converge