Thread Subject:

curve fitting of a double Wiebe Function

Hi everyone,
I'm working on the determination of the net heat released due to combustion in a Diesel Combustion engine. In order to do that, i'm using a double Wiebe Function that illustrates this heat release.

( double Wiebe Function):
F= [betta*ap*(mp+1)*(((tetta-td)./dOp).^(mp)).*exp(-ap*(((tetta- td)./dOp).^(mp+1)))] + (1-betta)*ad*(md+1)*(((tetta-td)./dOd).^(md)).*[exp(-ad*(((tetta-td)./dOd).^(md+1)))];

my objective is to fit this curve to the one extracted from experiments.
It contains 8 parameters as you can see:

-betta: fraction of fuel burned in the pre-mixed phase
-ap,ad: Wiebe parameters ( 'p': pre-mixed combustion , 'd': diffused combustion)
-mp,md: Wiebe shape factors (also 'p' for pre-mixed and 'd' for diffusion)
-td: cranck angle at which the combustion is ignited
-dOp: Combustion duration (pre-mixed)
-dOd: Combustion duration (diffusion)

tetta is the cranckangle and it varies between (td+0.1)?:0.1?:150?.

The problem is that i'm a not an expert of Matlab, and i don't know which function i need to use and how.

Any advice or help concerning this fitting problem i would be more than grateful.
best Regards.

> Any advice or help concerning this fitting problem i would be more than grateful.
> best Regards.

i've tried Ezyfit, but it seems that the equation is too much complex fot this function.
And the Weird thing is that when using Ezyfit, Each time i change the initial value of the parameters... it changes the output values. Which must be wrong.

There must be someone who have worked with the Wiebe Function ....
Help plz... this equation is making me go crazy :(((((((

"Charbel" wrote in message <gsprtp$82i$1@fred.mathworks.com>...
> > Any advice or help concerning this fitting problem i would be more than grateful.
> > best Regards.
>
> i've tried Ezyfit, but it seems that the equation is too much complex fot this function.
> And the Weird thing is that when using Ezyfit, Each time i change the initial value of the parameters... it changes the output values. Which must be wrong.
>
> There must be someone who have worked with the Wiebe Function ....
> Help plz... this equation is making me go crazy :(((((((


Did you guys ever find a solution. I am having the same problem but my supeviroe suggest that i use the experimetal results to find match. the idea is to solve for m sing simutenouse equations and goal seek in excel for a and the duration of the combustion. Did you ever find a diffrent way to solve this