how to use Curve Fiiting Express VI

[zubair]

hi,

Iam a new bee to LabView.

I want to use Curve Fiiting Express VI from (Anaylse>> Mathematics>>
Curve Fitting >> Curve Fitting). to get the best fit.

The actual formula is : I(L, λ) = Isp(λ)/G(λ) { exp [G(λ) · L ]
− 1}
where L is stripe Lenght

I is intensity

lamda is wave lenght

G is the net gain

given the values of L and I for respective lamda i have to find out the
net gain (g) by from the above formula.

the formula which i am using for non linear model is :
x/g*(exp(g*l)-1). is this formula right with respect to the above
formula. or shall i change it and i dont know how to use the Curve
Fitting Express VI, I will be very much thankful for any kind of
advice.

thank you.

[Evan]

Altenbach started to answer your question in what I am assuming is your other post.

http://forums.ni.com/ni/board/message?board.id=170&message.id=166529

If you can answer a few of his questions we should all be able to help you get to the solution you need.

Evan

[zubairkhan213]

Hi evan, thank you for your reply.

I(L, λ) = Isp(λ)/G(λ) { exp [G(λ) · L ] − 1} 

above formula is taken form Variable Stripe lenght method.

actually i have a very little idea about the formula.

my boss says Isp can be taken as a constant.

G is the net gain (Fitting parameter).

I is the intensity

L is the stripe lenght

i dont know what shall be the independent variable.

my task is related to spectrometer. using a labview vi, i get the data from the ccd camera and store it in a file.

please see the attachment to see how the data looks like

i have to find out netgain(G) using the above data. taking the G values i have to plot a graph. showing the net gain.

i am un aware weather this can be done using Curve fitting express.vi or shall i derive a formula for G.

if u can give any idea about this problem or show me some way, i will be very much thankfull to you.

bye.

[zubairkhan213]

Hi evan, thank you for your reply.

I(L, λ) = Isp(λ)/G(λ) { exp [G(λ) · L ] − 1} 

above formula is taken form Variable Stripe lenght method.

actually i have a very little idea about the formula.

my boss says Isp can be taken as a constant.

G is the net gain (Fitting parameter).

I is the intensity

L is the stripe lenght

i dont know what shall be the independent variable.

my task is related to spectrometer. using a labview vi, i get the data from the ccd camera and store it in a file.

please see the attachment to see how the data looks like

i have to find out netgain(G) using the above data. taking the G values i have to plot a graph. showing the net gain.

i am un aware weather this can be done using Curve fitting express.vi or shall i derive a formula for G.

if u can give any idea about this problem or show me some way, i will be very much thankfull to you.

bye.

[Andy F.]

I think I might see what you are trying to do.  You have I and L and you want to solve for G.  Is that correct?  You can certainly use the Curve Fitting Express VI.  If you place the VI down, you can select to have a non-linear fit.  This will put the fitted parameters into something that is very close to what you have for your equation (ie y=a*exp(b*x)+c).  Once you have the parameters all set up you can wire your data to the VI.  If I understand the formula correctly the b coefficient will be your Net Gain for that lambda (you can group your formula into the above equation where a=Isp/G, b=G, and c=-Isp/G).

Hope this helps!
Andy F.

[zubairkhan213]

hi andy,

thanks alot for the reply, it helped me. but still need to examine the wether the results are right. how ever thanks again for your valuable help.

bye.