Someone please help to rectify this error: Error using barrier Objective function is undefined at initial point. Fmincon cannot continue. Optimization problem

3 visualizaciones (últimos 30 días)
I want to fit the mathematical model containing double numerical integration with the data. Please help me!
This is the driver code:
clc;
clear all;
u = load("thoracicdata11.mat");
xdata11 = u.Figure9S11(1:225,1);
ydata11 = u.Figure9S11(1:225,2);
x0=[5 2 1 0.01 0.1];%
lb=[0.001 0.001 0.001 1 0.0];
ub=[100.0 200.0 100.0 100 pi/2];
options = optimset('Display', 'iter', 'MaxIter', 100, 'TolFun', 1e-6);
obj = obj_GOHGR_planar(x0,xdata11,ydata11)
Warning: Non-finite result. The integration was unsuccessful. Singularity likely.
Warning: Non-finite result. The integration was unsuccessful. Singularity likely.
Index in position 2 exceeds array bounds. Index must not exceed 1.

Error in solution>obj_GOHGR_planar (line 19)
obj = sum((ydata11(:,2)-BiaxialResponse).^2);
%[x, fval]=fmincon(@(x)obj_GOHGR_planar(x,xdata11,ydata11),x0,[],[],[],[],lb,ub,[],options);
This is the objective function:
function obj=obj_GOHGR_planar(x,xdata11,ydata11)
%xdata11 = Figure9S11(1:225,1);
%ydata11 = Figure9S11(1:225,2);
BiaxialResponse=zeros(size(xdata11(:,1)));
for i=1:size(xdata11(:,1))
BiaxialResponse(i)= BiaxialNewInvariantFunc11(x,xdata11(i,1));
end
obj = sum((ydata11(:,2)-BiaxialResponse).^2);
end
Calling function:
function K1 = BiaxialNewInvariantFunc11(par,lambda1)
integrand=@(par,lambda1,theta,phi) (3/4).*exp(1).^((1/64).*lambda1.^(-12).*(1+(-2).*lambda1.^4+ ...
lambda1.^6+(-1).*((-1)+lambda1.^6).*cos(2.*theta)+ ...
lambda1.^4.*(lambda1.^(-8).*(2.*((-1)+lambda1.^2).^2.*(1+2.* ...
lambda1.^2+lambda1.^4+lambda1.^8)+(-2).*((-1)+2.*lambda1.^4+ ...
(-2).*lambda1.^10+lambda1.^12).*cos(2.*theta))).^(1/2)).^3.* ...
par(4)+2.*cos(phi+(-1).*par(5)).^2.*par(2).*sin( ...
theta).^2).*lambda1.^(-8).*(2.*pi).^(-1/2).*(1+(-2).* ...
lambda1.^4+lambda1.^6+(-1).*((-1)+lambda1.^6).*cos(2.*theta) ...
+lambda1.^4.*(lambda1.^(-8).*(2.*((-1)+lambda1.^2).^2.*(1+ ...
2.*lambda1.^2+lambda1.^4+lambda1.^8)+(-2).*((-1)+2.* ...
lambda1.^4+(-2).*lambda1.^10+lambda1.^12).*cos(2.*theta))) ...
.^(1/2)).^2.*erfi(2.^(1/2).*par(2).^(1/2)).^(-1).*par( ...
1).*par(2).^(1/2).*par(3).*sin(theta).*((1/2).* ...
lambda1.^2.*cos(phi).^2.*sin(theta).^2.*(1+((-1)+lambda1.^2) ...
.*(lambda1.^(-8).*((-1)+lambda1.^2).^2.*((1+lambda1.^2).^2.* ...
cos(theta).^2+lambda1.^8.*sin(theta).^2)).^(-1/2))+(-1/2).* ...
lambda1.^(-4).*cos(theta).^2.*(1+lambda1.^(-4).*(1+(-1).* ...
lambda1.^4).*(lambda1.^(-8).*((-1)+lambda1.^2).^2.*((1+ ...
lambda1.^2).^2.*cos(theta).^2+lambda1.^8.*sin(theta).^2)).^( ...
-1/2)));
K1 = 0.3E0.*((-1).*lambda1.^(-4)+lambda1.^2)+ integral2(@(phi,theta) integrand(par,lambda1,theta,phi),0,2*pi,0,pi);
end
Thanks in advance! :)
  1 comentario
Torsten
Torsten el 22 de Feb. de 2023
Editada: Torsten el 22 de Feb. de 2023
Besides the unsuccessful integration one syntax error: ydata11(:,2) does not exist because ydata11 has only one column, not two.

Iniciar sesión para comentar.

Respuestas (1)

Steven Lord
Steven Lord el 22 de Feb. de 2023
We can't run your code as we don't have all of the functions it uses, but what do you receive as an answer when you call obj_GOHGR_planar using the x0 you pass into fmincon as the input? I've commented out the call below because I want to run later code in this answer, but I wanted to show what I want you to run.
%{
y = obj_GOHGR_planar(x0)
%}
But I see another problem with your code that may be related. Does your initial point satisfy your lower and upper bounds?
x0=[5 2 1 0.01 0.1];%
lb=[0.001 0.001 0.001 1 0.0];
ub=[100.0 200.0 100.0 100 pi/2];
isx0InBounds = (lb <= x0) & (x0 <= ub)
isx0InBounds = 1×5 logical array
1 1 1 0 1
Are you sure the fourth element of x0 should be 0.01?

Categorías

Más información sobre Get Started with Curve Fitting Toolbox en Help Center y File Exchange.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by