Function or constraints have too few arguments on the right to satisfy the left

1 Ansicht (letzte 30 Tage)
I'm trying to optimize this equation:
f=@(x) 0.25*7850*pi^2*x(1).^2.*x(2).*x(3)
with this set of constraints:
g=@(x) [8*75./x(5)-7; 3.5-2*pi*sqrt(75^2/(4*75.*x(4)-x(5).^2)); 100*exp(-pi.*x(5)./sqrt(4*75.*x(4)-x(5).^2))-15; x(2)-0.75; 75*9.81.*(1.566.*x(1).^2-0.9295.*x(1).*x(2)-2.546.*x(2).^2)./(x(1).^3.*(x(1)-x(2)))-517.1*10^6; 78*10^9.*x(1).^4./(8.*x(2).^3.*x(3))-x(4)];
I'm using the fmincon function in the Optimization toolbox, but no matter what algorithm I select, I see the error "The right hand side of this assignment has too few values to satisfy the left hand side" with no further details. I have been trying to figure out the cause of this error, but I'm having difficulty figuring out what, specifically, the error is referring to.
  2 Kommentare
Richard Brown
Richard Brown am 26 Apr. 2012
could you provide the full call to fmincon, with any other constraints that you are passing in
Scott Bacvinskas
Scott Bacvinskas am 26 Apr. 2012
I was literally entering those two functions into the Optimization toolbox, but otherwise I use:
>> f=@(x) 0.25*7850*pi^2*x(1).^2.*x(2).*x(3);
>> g=@(x) [8*75./x(5)-7; 3.5-2*pi*sqrt(75^2/(4*75.*x(4)-x(5).^2)); 100*exp(-pi.*x(5)./sqrt(4*75.*x(4)-x(5).^2))-15; x(2)-0.75; 75*9.81.*(1.566.*x(1).^2-0.9295.*x(1).*x(2)-2.546.*x(2).^2)./(x(1).^3.*(x(1)-x(2)))-517.1*10^6; 78*10^9.*x(1).^4./(8.*x(2).^3.*x(3))-x(4)];;
>> x0=[1;1;1;1;1];
>> [X,FVAL,EXITFLAG,OUTPUT,LAMBDA] = fmincon(f,x0,[],[],[],[],[],[],g)
It highlights some kind of issue with my constraint function (g), but I cannot see what the issue is.

Melden Sie sich an, um zu kommentieren.

Akzeptierte Antwort

Richard Brown
Richard Brown am 26 Apr. 2012
Your nonlinear constraints need to be in a function of the form
function [c, ceq] = g(x)
c = [8*75./x(5) .......
ceq = [];
end
MATLAB was complaining because there was no equality constraint information being passed out (even there are no equality constraints).
  7 Kommentare
Teja Muppirala
Teja Muppirala am 26 Apr. 2012
There's a neat trick you can do with anonymous functions to make them return more than one value.
F = @(x) deal( x(1) + x(2), x(1) - x(2) )
[a,b] = F([5 2])
I sometimes use this when trying simple things out in scripts and I don't want to have to save a file.

Melden Sie sich an, um zu kommentieren.

Weitere Antworten (0)

Kategorien

Mehr zu Nonlinear Optimization finden Sie in Help Center und File Exchange

Community Treasure Hunt

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

Start Hunting!

Translated by