matlab nonlinear equation solver

i have a set of3 nonlinear equations and i need to solve them by using fsolve in matlab

      function F = root2d(y)
     syms b1 b2 b3 w21 w31 theta1 theta2 theta3 a1 a2 a3;
     F(1) = (1+exp(-b1*(w21*y(2)+w31*y(3)-theta1)))^(-1) - a1*y(1);
     F(2) = (1/(1+exp(-b2*(y(1)-theta2))))-a2*y(2);
     F(3)=   (1/(1+exp(-b3*(y(1)-theta3))))-a3*y(3);

this is my function matlab file. I call this matlab function while using the matlab file

          fun = @root2d;
          x0 = [0,0,0];
          x = fsolve(fun,x0)

But it is giving me the error

Error using fsolve (line 258)
FSOLVE requires all values returned by user functions to be of data type double.

    Error in Untitled5 (line 3)
    x = fsolve(fun,x0)

can somebody help ?

Answer 1

[Matt J]

FSOLVE does not handle symbolic variables, so get rid of the line

   syms b1 b2 b3 w21 w31 theta1 theta2 theta3 a1 a2 a3

and substitute these quantities with actual numeric values.

[myetceteramail myetceteramail]

actually i want to get the answer in terms of variables only. how can it be done

[Matt J]

I doubt that's possible for these particular equations, but if it is possible it would be with the SOLVE command, but not with FSOLVE.

[myetceteramail myetceteramail]

with solve i get the error

Error in solve (line 227)
[eqns,vars,options] = getEqns(varargin{:});

[Matt J]

You have not shown your code nor the complete error message so, hard to advise.

[myetceteramail myetceteramail]

   Error using root2d (line 3)
   Not enough input arguments.

 Error in sym>funchandle2ref (line 1211)
     S = sym(x());

    Error in sym>tomupad (line 1114)
    x = funchandle2ref(x);

    Error in sym (line 151)
                S.s = tomupad(x);

    Error in solve>getEqns (line 410)
        a = formula(sym(a));

      Error in solve (line 227)
      [eqns,vars,options] = getEqns(varargin{:});

       Error in Untitled5 (line 3)
       x = solve(fun,x0);

[Matt J]

Still don't see the new code you're running.

[myetceteramail myetceteramail]

       function F = root2d(y)
       syms b1 b2 b3 w21 w31 theta1 theta2 theta3 a1 a2 a3;
       F(1) = (1+exp(-b1*(w21*y(2)+w31*y(3)-theta1)))^(-1) - a1*y(1);
       F(2) = (1/(1+exp(-b2*(y(1)-theta2))))-a2*y(2);
       F(3)=   (1/(1+exp(-b3*(y(1)-theta3))))-a3*y(3);

       fun = @root2d;
x0 = [0,0,0];
x = solve(fun,x0);

[Matt J]

To clarify what I said before, you must use the SOLVE command, but you should first read its documentation so you can use it correctly!