Error while declaring a fit-type

Can somebody tell me why using the following expression:

myfittype = fittype(...

'(integral(@(xi)( (1/pi*(0.1375*sqrt(2*pi)))*exp(-(xi-2.75).^2/(2.0*0.01890625))./((xval+0.09793600169767440*2*xi-1.34484937381768000).^2+d^2) ),0,1/0))',...

'dependent',{'y'},'independent',{'xval'},'coefficients',{'d'})

Gives me the following error

Error using fittype/evaluate (line 106)

Error in fittype expression ==> (integral(@(xi)(

(1./pi.*(0.1375.*sqrt(2.*pi))).*exp(-(xi-2.75).^2./(2.0.*0.01890625))./((xval+0.09793600169767440.*2.*xi-1.34484937381768000).^2+d.^2)

),0,1./0))

??? Output of the function must be the same size as the input. If FUN is an array-valued integrand, set

the 'ArrayValued' option to true.

when i turn ArrayValued option on it gives another error

myfittype = fittype(...

'(integral(@(xi)( (1/pi*(0.1375*sqrt(2*pi)))*exp(-(xi-2.75).^2/(2.0*0.01890625))./((xval+0.09793600169767440*2*xi-1.34484937381768000).^2+d^2) ),''ArrayValue'',true,0,1/0))',...

'dependent',{'y'},'independent',{'xval'},'coefficients',{'d'})

it gives another error

Error using fittype/evaluate (line 106)

Error in fittype expression ==> (integral(@(xi)(

(1./pi.*(0.1375.*sqrt(2.*pi))).*exp(-(xi-2.75).^2./(2.0.*0.01890625))./((xval+0.09793600169767440.*2.*xi-1.34484937381768000).^2+d.^2)

),'ArrayValue',true,0,1./0))

??? A and B must be floating-point scalars.

i can't understand what's wrong with the expression since using it as a stand alone function doesn't produce errors.