how can I get rid of these errors?

hi I have a huge function to integrate

>> syms x y
>>  f=(228155022448185.*(cos((2.*pi).*y)./exp(131738205584307./(35184372088832.*x)) - 1)*(cos((8.*pi).*y)/exp(131738205584307./(8796093022208.*x)) - 1)*(cos((8.*pi).*y)/exp(131738205584307./(8796093022208.*x)) + cos((18.*pi).*y)/exp(1185643850258763./(35184372088832.*x)) - 2))/((18014398509481984.*(x.^2)).*exp(x.*((1981232555272083.*(y.^2))/2251799813685248 - y./16 + 1./16)))

I need integration of it( x:[0,inf) and y:[0,1] )

but for this function I receive error for quad2d and dblquad

quad2d(quadfun,0,100,0,1)
??? Error using ==> quad2d>tensor at 350
Integrand output size does not match the input size.

Error in ==> quad2d at 164
[Qsub,esub] = tensor(thetaL,thetaR,phiB,phiT);

and

   dblquad(quadfun,0,100,0,1)
  ??? Error using ==> dblquad>innerintegral at 74
  Inputs must be floats, namely single or double.

Error in ==> quad at 76
y = f(x, varargin{:});

Error in ==> dblquad at 53
Q = quadf(@innerintegral, ymin, ymax, tol, trace, intfcn, ...

could you explain why these errors appear? and how can I fix it?

Answer 1

[Roger Stafford]

The 'quad2d' and 'dblquad' functions are intended for use with numerical data, namely 'double' or 'single' variables as the error message has stated. For integration with symbolic variables you should be using the 'int' function - in this case you need an 'int' of an 'int' to accomplish double integration. Either that or switch to 'double' type for x and y. (In the latter case you may encounter numerical difficulties with the x variable ranging from zero to infinity within the exponential function.)