The double integral of the charge density across the membrane gives the dipole potential; I'm not quite sure if that's what you want or not. It is strongly dependent on the force field.
By default CHARMM assumes a time base, so the stated definition of INTEgrate is wrt. to time starting at zero. For a z profile, you have to change the independent variable to match the z values via EDIT; for a range of -40 to 40 in 0.1 A intervals, something like
edit chg delta 0.1 skip 1 offset -40.
would be needed.
Perhaps you should reconsider writing your own program for this.
