- Choosing an initial guess, x=3, away from zero.
- Supplying an analytical gradient calculation using SpecifyObjectiveGradient=true
- Using log1p() to compute your objective with higher precision
- Scaling the objective by 1e6 to achieve more natural units
fsolve found wrong solution for easy equation
14 vues (au cours des 30 derniers jours)
Afficher commentaires plus anciens
Dear everyone, I have a stupid problem with fsolve:
my equation is : ![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/235021/image.png)
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/235021/image.png)
wehre a and b are constante. I would like to solve it for z=linspace (0,20,100) (for exemple)
if z=0, the solution is x=1.
i write :
close all
clc
clear all
c=3e8;
r=30e-6;
Aeff=pi.*r.*r;
lambda0=1030e-9;
om0=2.*pi.*c./lambda0;
C0=0.;
eta=4E-1
b2=-4e-28;
b2=b2;
b3=1.5e-40;
g=sign(b2);
T0=100./om0;
Tr=10./om0;
n2=1e-23;
gamma=2.*pi.*n2./(lambda0.*Aeff);
a=8.*b3.*Tr./(15.*T0.^4);
b= 2.*eta.*b3./(3.*gamma.*T0.^2);
sol= fsolve(@(x) x*x-1+a/b*log( (b*x*x -a) / (b-a) ),0)
return
%%% I also try with solve
syms u
eqn = u.*u-1+a./b.*log( (b.*u.*u -a) ./ (b-a) )==0;% + 2b.*z
solu = solve(eqn,u)
Solution for z=0 is not found...
Why?
Thanks
0 commentaires
Réponse acceptée
Matt J
le 23 Août 2019
Modifié(e) : Matt J
le 23 Août 2019
Choosing an initial guess of x=0 in fsolve is often a bad idea because the gradient of the objective will frequently be zero there (this is the case for your problem) or contain division-by-zero or log(0) operations.
In the implementation below, I've made a number of improvements including,
and get correspondingly better results.
function solveit
a = 0.049032234129438;
b = 6.209226846510589e-07;
opts=optimoptions('fsolve', 'SpecifyObjectiveGradient',true);
sol=fsolve(@fun,3,opts)
function [f,df]=fun(x)
f=x^2-1+a/b*log1p( b*(x^2 -1) / (b-a) );
f=f*1e6;
df=2*x*(1+a/(b*x^2-a));
df=df*1e6;
end
end
This produces,
sol =
1.000000000000032
Plus de réponses (1)
Matt J
le 22 Août 2019
Modifié(e) : Matt J
le 22 Août 2019
fzero works better
fun=@(x) x^2-1+a/b*log( (b*x^2 -a) / (b-a) );
[sol,fval]= fzero(fun,0)
sol =
-1.000000128628277
fval =
4.488965127631997e-12
Voir également
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!