why integral is not done

function output=Hplus(Q,W,a,mutilde)

output=integral(@(u) sqrt(1-u.^2)./(exp(0.5.*abs(Q.*u+W)-a.*mutilde)+1),-1,1,'ArrayValued',true,'RelTol',1e-4);

end

syms Q W a mutlide

output=Hplus(Q,W,a,mutilde)

%% I am trying to run this program but error is come as

%%Error using integralCalc/iterateArrayValued (line 159)

%%Inputs must be floats, namely single or double.

%%Error in integralCalc/vadapt (line 130)

%% [q,errbnd] = iterateArrayValued(u,tinterval,pathlen);

%%Error in integralCalc (line 75)

%% [q,errbnd] = vadapt(@AtoBInvTransform,interval);

%%Error in integral (line 88)

%%Q = integralCalc(fun,a,b,opstruct);

%%Error in Hplus (line 2)

%%output=integral(@(u) sqrt(1-u.^2)./(exp(0.5.*abs(Q.*u+W)-a.*mutilde)+1),-1,1,'ArrayValued',true,'RelTol',1e-4);

%%Error in Untitled2 (line 2)

%%output=Hplus(Q,W,a,mutilde)

[madhan ravi]

because you didn’t define the inputs while calling ( has to be numbers not symbolic variables) , integral() cannot deal with symbolic variables

[SHARAD KUMAR UPADHYAY]

If i don't want to give number to variables then how can i run this

[Torsten]

You could try to use MATLAB's "int" (symbolic integration), but I doubt that your function has an analytical antiderivative.

[Bjorn Gustavsson]

If you dont manage with the exact symbolic integration you could try to make a Taylor-expansion of the root - then I got some explicit values for the integral: int(p(u)/(exp(u)+1),-1,1). You'd still might need to do some variable transformations to clean up your integrand a bit. Integrals like these is what we used handbooks for when I studied solid state physics (too long a time ago...)

HTH