Error using + Matrix dimensions must agree.

how can i solve this error?

Below is my script:

function [y, cons] = TP_CONSTR_objfun(x)
    y = [0,0];
    cons = [0,0];
    theta1=[0:0.1:pi];theta2=[0:0.1:pi];
    for theta1=1:length(x(1))
        for theta2=1:length(x(2))
            y(1) = (-sqrt(x(1).^2+x(2).^2+x(3).^2)).*2.*(x(3)+x(1).*cos(x(4)));
            %%%%%
            e=(x(1).*(sin(theta1)-sin(theta2))./(2.*x(3)+x(1).*cos(theta2)-x(1).*cos(theta1)));
            f=x(1).*x(3).*(cos(theta2)+cos(theta1))./(2.*x(3)+x(1).*cos(theta2)-x(1).*cos(theta1));
            px=e.*y+f;
            d=1+e.^2;
            g=2.*(e.*f-e.*x(1).*cos(theta1)+e.*x(3)-x(1).*sin(theta1));
            h=f.^2-2.*f.*(x(1).*cos(theta1)-x(3))-2.*x(1).*x(3).*cos(theta1)+x(3).^2+x(1).^2-x(2).^2;
            py=-g+sqrt(g.^2-4.*d.*h)./2.*d;
            %%%%%%%%%%
            y(2)=(((py - x(1).*sin(theta2)).^2.*(sin(theta1).*(x(3) + px) - py.*cos(theta1)).^2)./(2.*(x(1).*py.*cos(theta1) - 2.*x(3).*py - x(1).*py.*cos(theta2) + x(1).*x(3).*sin(theta1) + x(1).*x(3).*sin(theta2) - x(1).*px.*sin(theta1) + x(1).*px.*sin(theta2) - x(1).^2.*cos(theta1).*sin(theta2) + x(1).^2.*cos(theta2).*sin(theta1)).^2) + ((py - x(1).*sin(theta1)).^2.*(py.*cos(theta2) + sin(theta2).*(x(3) - px)).^2)./(2.*(x(1).*py.*cos(theta1) - 2.*x(3).*py - x(1).*py.*cos(theta2) + x(1).*x(3).*sin(theta1) + x(1).*x(3).*sin(theta2) - x(1).*px.*sin(theta1) + x(1).*px.*sin(theta2) - x(1).^2.*cos(theta1).*sin(theta2) + x(1).^2.*cos(theta2).*sin(theta1)).^2) + ((sin(theta1).*(x(3) + px) - py.*cos(theta1)).^2.*(x(3) - px + x(1).*cos(theta2)).^2)./(2.*(x(1).*py.*cos(theta1) - 2.*x(3).*py - x(1).*py.*cos(theta2) + x(1).*x(3).*sin(theta1) + x(1).*x(3).*sin(theta2) - x(1).*px.*sin(theta1) + x(1).*px.*sin(theta2) - x(1).^2.*cos(theta1).*sin(theta2) + x(1).^2.*cos(theta2).*sin(theta1)).^2) + ((py.*cos(theta2) + sin(theta2).*(x(3) - px)).^2.*(x(3) + px - x(1).*cos(theta1)).^2)./(2.*(x(1).*py.*cos(theta1) - 2.*x(3).*py - x(1).*py.*cos(theta2) + x(1).*x(3).*sin(theta1) + x(1).*x(3).*sin(theta2) - x(1).*px.*sin(theta1) + x(1).*x.*sin(theta2) - x(1).^2.*cos(theta1).*sin(theta2) + x(1).^2.*cos(theta2).*sin(theta1)).^2)).^(1./2).*((py - x(1).*sin(theta1)).^2./(2.*(x(3).*sin(theta1) - py.*cos(theta1) + px.*sin(theta1)).^2) + (py - x(1).*sin(theta2)).^2./(2.*(py.*cos(theta2) + x(3).*sin(theta2) - px.*sin(theta2)).^2) + (x(3) + px - x(1).*cos(theta1)).^2./(2.*(x(3).*sin(theta1) - py.*cos(theta1) + px.*sin(theta1)).^2) + (x(3) - px + x(1).*cos(theta2)).^2./(2.*(py.*cos(theta2) + x(3).*sin(theta2) - px.*sin(theta2)).^2)).^(1./2);
        end
    end
    % constraint function
    c = (-sqrt(x(1).^2+x(2).^2+x(3).^2))-0.5;
    if(c<0)
        cons(1) = abs(c);
    end
    c = (2.*(x(3)+x(1).*cos(x(4))))-1.5;
    if(c<0)
        cons(2) = abs(c);
    end

Thank you

[Bob Thompson]

Can you post the specific error message, please. It helps to know which line is causing the problem.

Also, what is the structure of x? I assume that x is an array of doubles, as it looks like you're calling specific elements for your equations, but if that is the case then your for loops will both only loop once, because the length of an element is only 1.

[Walter Roberson]

It is the y(2) assignment.

The line is too long to understand easily. You should break it down into a series of sub-sections.

[KSSV]

Look at the expression for y(2)..it is huge........split it into small chynks and add them..you have error some where there in the expression.

[Dikra dikra]

x(1) x(2) and x(3) are the design variables of an optimization problem

y(1) and y(2) are two objectif functions

t1 and t2 are two angles varies between 0 and pi

my script can be written as:

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function [y, cons] = TP_CONSTR_objfun(x)
    y = [0,0];
    cons = [0,0];
    t1=[0:0.1:pi];t2=[0:0.1:pi];
    for t1=1:length(y(1))
        for t2=1:length(y(2))
            %%%%%%%%%
            e=(x(1).*(sin(t1)-sin(t2))./(2.*x(3)+x(1).*cos(t2)-x(1).*cos(t1)));
            f=(x(1).*x(3).*(cos(t2)+cos(t1))./(2.*x(3)+x(1).*cos(t2)-x(1).*cos(t1)));
            d=(1+e.^2);
            g=(2.*(e.*f-e.*x(1).*cos(t1)+e.*x(3)-x(1).*sin(t1)));
            h=(f.^2-2.*f.*(x(1).*cos(t1)-x(3))-2.*x(1).*x(3).*cos(t1)+x(3).^2+x(1).^2-x(2).^2);
            yp=(-g+sqrt(g.^2-4.*d.*h)./2.*d);
            xp=(e.*y+f);
            %objectif function
            y(1) = (sqrt(x(1).^2+x(2).^2+x(3).^2)).*2.*(x(3)-x(1).*sin(t1));
            y(2) =(((x(1).^2.*(yp - x(1).*sin(t2)).^2.*(sin(t1).*(x(3) + xp) - yp.*cos(t1)).^2)./(2.*(x(1).*yp.*cos(t1) - 2.*x(3).*yp - x(1).*yp.*cos(t2) + x(1).*x(3).*sin(t1) + x(1).*x(3).*sin(t2) - x(1).*xp.*sin(t1) + x(1).*xp.*sin(t2) - x(1).^2.*cos(t1).*sin(t2) + x(1).^2.*cos(t2).*sin(t1)).^2) + (x(1).^2.*(yp.*cos(t2) + sin(t2).*(x(3) - xp)).^2.*(x(3) + xp - x(1).*cos(t1)).^2)./(2.*(x(1).*yp.*cos(t1) - 2.*x(3).*yp - x(1).*yp.*cos(t2) + x(1).*x(3).*sin(t1) + x(1).*x(3).*sin(t2) - x(1).*xp.*sin(t1) + x(1).*xp.*sin(t2) - x(1).^2.*cos(t1).*sin(t2) + x(1).^2.*cos(t2).*sin(t1)).^2)).^(1./2).*((x(1).*yp.*cos(t1) - 2.*x(3).*yp - x(1).*yp.*cos(t2) + x(1).*x(3).*sin(t1) + x(1).*x(3).*sin(t2) - x(1).*xp.*sin(t1) + x(1).*xp.*sin(t2) - x(1).^2.*cos(t1).*sin(t2) + x(1).^2.*cos(t2).*sin(t1)).^2./(2.*x(1).^2.*(yp.*cos(t2) + x(3).*sin(t2) - xp.*sin(t2)).^2.*(x(3) + xp - x(1).*cos(t1)).^2) + (x(1).*yp.*cos(t1) - 2.*x(3).*yp - x(1).*yp.*cos(t2) + x(1).*x(3).*sin(t1) + x(1).*x(3).*sin(t2) - x(1).*xp.*sin(t1) + x(1).*xp.*sin(t2) - x(1).^2.*cos(t1).*sin(t2) + x(1).^2.*cos(t2).*sin(t1)).^2./(2.*x(1).^2.*(yp - x(1).*sin(t2)).^2.*(x(3).*sin(t1) - yp.*cos(t1) + xp.*sin(t1)).^2)).^(1./2));
        end
    end
    %Constraints
    c = (sqrt(x(1).^2+x(2).^2+x(3).^2))-100;
    if(c<0)
        cons(1) = abs(c);
    end
    c = 2.*(x(3)-x(1).*sin(t1))-200;
    if(c<0)
        cons(2) = abs(c);
    end
    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

The error message is:

Evaluating the objective function... Generation: 1 / 100 , Individual: 1 / 100

In an assignment A(I) = B, the number of elements in B and I must be the same.

Error in TP_CONSTR_objfun (line 24)

[madhan ravi]

why don't you just use size() ?? to verify the sizes??

Answer 1

[per isakson]

Observation:

    for t1=1:length(y(1))
        for t2=1:length(y(2))

Why loops? The length of one element is always ONE.

I made a simple test (R2018b) and got an error in line 17 (not 24), which is the very wide expression.

>> TP_CONSTR_objfun([1,2,3])
Unable to perform assignment because the left and right sides have a different number of elements.
Error in TP_CONSTR_objfun (line 17)
            y(2) =(((x(1).^2.*(yp - x(1).*sin(t2)).^2.*(sin(t1).*(x(3) + xp) - <snip>

This expression returns a vector, which cannot be assigned to one element of y, thus the error.

Set a break-point and start debugging

>> TP_CONSTR_objfun([1,2,3])
14              xp=(e.*y+f);
K>> yp
yp =
   1.6829 + 2.2361i
...
...            
K>> (((x(1).^2.*(yp - x(1).*sin(t2)).^2.*(sin(t1).*(x(3) + xp) - yp.*cos(t1)).^2) <snip>
K>> zzz=ans
zzz =
   1.0891 - 0.1597i   1.0891 - 0.1597i

See: Debug a MATLAB Program and Examine Values While Debugging

[Dikra dikra]

Thank you

I think the error is in the insertion of t1 and t2

even when i remove the loops i get this error:

Error using *

Inner matrix dimensions must agree.

Error in TP_CONSTR_objfun (line 24)

xp=(e*y+f);

Error in TP_CONSTR_objfun (line 24)

xp=(e*y+f);

Error in evaluate>evalIndividual (line 67)

[y, cons] = objfun( indi.var, varargin{:} );

Error in evaluate (line 45)

[pop(i), allTime(i)] = evalIndividual(pop(i), opt.objfun, varargin{:});

Error in nsga2 (line 81)

[pop, state] = evaluate(opt, pop, state, varargin{:});

Error in TP_CONSTR (line 25)

result = nsga2(options); % begin the optimization!