| Author |
 Topic  |
|
|
zhotihu
China
Posts |
 Posted - 09/25/2006 : 2:49:14 PM
|
Origin Version (Select Help-->About Origin): origin 7.0
Operating System: Windows xp
Dear all, I have a big problem
I want to obtain the best-fit parameters by fitting the plot y-x using the Frumkin-Davies equations (Eqs. [1] and [2]), with h, p, a and k as parameters.
y=-p*[ln(1-y1)-h*(y1*y1/2)]+(1-cosha) [1]
y1=[(x/k)*exp(-h*y1)exp(-a)]/[1+(x/k)*exp(-h*y1)exp(-a)] [2]
Fig y-x shows the experiment data. Independent variable x and Dependent variable y are not in the same function, I cannot take one function and determine all parameters.
Would you please give me the suggestions about how to determine the parameters h, p, a and k by means of fitting the Fig y-x according to Frumkin-Davies (Eqs. [1] and [2]).
I cannot updown the plot of y-x
Thank you all in advance!
Edited by - zhotihu on 09/25/2006 3:12:15 PM
Edited by - zhotihu on 09/25/2006 3:52:20 PM |
|
|
zachary_origin
China
Posts |
Posted - 09/26/2006 : 01:19:00 AM
|
It is diffcult to define the equations in Origin because Equ.2 is a implicit equation and Origin does not support the numeric solution of the equation. If you can get the experiment data for y1, you can set y1 as independent variable and do the fitting. However, it seems that y1 can not be gotten from experiment. Then the only resort is to write codes in Origin C to solve the equation with the mostly used algorithm, e.g. Newton method. This will be a little complicated.
Zachary
OriginLab Technical Services |
 |
|
| |
Topic |
|
|
|
how to fit nonlinear curve?