I want to find the best coefficients for curve fitting for my experiment data

Greetings

I have experiment data as you can see below, I want to find the curve with the formula of y=K0+K1exp(-t/Tau) which is fit to my experiments data.

I have 'y' and 't' from experiment and I am looking for the best values for K0, K1 and Tau.

the bellow is the graph of my experiment data

Answer 1

[Image Analyst]

You can use fitnlm if you have the stats toolbox. See attached demo and make changes as needed, like reading in your actual data, and renaming variables if you want.

[Misi]

Many thanks it worked very well

Answer 2

[Torsten]

Try "lsqcurvefit".

[Misi]

Thank you

I used the lsqcurve fit but it couldnot fit on my curve and the final asnwer was zero for all coeffiecient.

This what I recieved in command window

Initial point is a local minimum.

Optimization completed because the size of the gradient at the initial point

is less than the default value of the optimality tolerance.

<stopping criteria details>

a =

0

b =

0

c =

0

Answer 3

[Torsten]

Your code ?

Essential in your case is a good starting guess for the parameters.

Try the following code and take the coefficients as initial values for the optimization.

fun = @(K0,K1,tau,t)  K0 + K1*exp(-t/tau);
t = 0:0.1:10;
K0 = 1e10;
K1 = 6.2e10;
tau = 4.5;
y = fun(K0,K1,tau,t);
plot(t,y)

[Misi]

Many thanks for your response